Granite Creek Watershed Assessme

Document Sample
Granite Creek Watershed Assessme Powered By Docstoc
					 GRANITE CREEK WATERSHED
       ASSESSMENT

          Prepared For



KALISPEL TRIBE AND PANHANDLE
      NATIONAL FOREST



          April 8, 2009



          Prepared By
            GRANITE CREEK WATERSHED ASSESSMENT




                                      Prepared for:

                     Kalispel Tribe and Panhandle National Forest




                                      Prepared By:

                           Watershed Professionals Network, LLC
                                      Boise, Idaho

                                   www.watershednet.com


WPN Analysis Team: Steve Bauer, Fish Biologist; Ed Salminen, Hydrologist; and Chris
                              Heider, Ecologist


Acknowledgements:

This watershed assessment was envisioned by fish biologists, Todd Anderson, Kalispel
Tribe of Indians, and Matt Fairchild, Panhandle National Forest, as a need to
understand limiting factors and develop a road map for restoration of Granite Creek.
We want to acknowledge Matt for his vision and effort in getting this project underway.
We also want to acknowledge the great effort undertaken by Todd Anderson and
Michele Wingert in collecting habitat and fish data, and their guidance during the
project.




Granite Creek Assessment                   WPN                                    Page ii
                              GRANITE CREEK
                          WATERSHED ASSESSMENT

                                                Table of Contents
1.0     EXECUTIVE SUMMARY .................................................................................................... 1
2.0     INTRODUCTION .................................................................................................................. 4
  2.1          GOALS AND OBJECTIVES .................................................................................................. 4
  2.2          ORGANIZATION OF THIS DOCUMENT ................................................................................ 4
3.0     WATERSHED OVERVIEW.................................................................................................. 6
  3.1          LOCATION & SUBWATERSHEDS ....................................................................................... 6
  3.2          WATER FEATURES ........................................................................................................... 7
  3.3          ROSGEN CHANNEL TYPES ................................................................................................ 9
  3.4          CLIMATE ........................................................................................................................ 11
  3.5          GEOLOGY & SOILS ......................................................................................................... 16
  3.6          HYDROLOGY .................................................................................................................. 19
  3.7          VEGETATION .................................................................................................................. 21
  3.8          LAND OWNERSHIP & USE .............................................................................................. 21
4.0     HYDROLOGY ..................................................................................................................... 24
  4.1          CRITICAL QUESTIONS .................................................................................................... 24
  4.2          METHODS ...................................................................................................................... 24
    4.2.1          SPATIAL INPUT DATA............................................................................................. 26
    4.2.2          MODEL CALIBRATION ............................................................................................ 42
  4.3          RESULTS ........................................................................................................................ 44
    4.3.1          ANNUAL PEAK FLOW MAGNITUDES ........................................................................ 44
    4.3.2          LOW FLOWS ........................................................................................................... 46
  4.4          INFORMATION GAPS AND MONITORING NEEDS ............................................................. 47
  4.5          RECOMMENDATIONS ...................................................................................................... 47
5.0     RIPARIAN CONDITIONS .................................................................................................. 48
  5.1          CRITICAL QUESTIONS .................................................................................................... 48
  5.2          METHODS ...................................................................................................................... 48
    5.2.1          RIPARIAN ZONE MAPPING...................................................................................... 48
    5.2.2          FIELD SAMPLING .................................................................................................... 48
    5.2.3          GROWTH AND MORTALITY MODELING .................................................................. 49
    5.2.4          LWD RECRUITMENT POTENTIAL ........................................................................... 50
    5.2.5          STREAM SHADE ..................................................................................................... 51
  5.3          RESULTS ........................................................................................................................ 51
    5.3.1          RIPARIAN ZONE DISTRIBUTION .............................................................................. 51
    5.3.2          MAJOR VEGETATION TYPES................................................................................... 52
    5.3.3          SUBWATERSHED SCALE GROWTH & MORTALITY PROJECTIONS ............................ 53


Granite Creek Assessment                                         WPN                                                                Page iii
    5.3.4         REACH SCALE GROWTH & MORTALITY PROJECTIONS ........................................... 56
    5.3.5         MODELING CONCLUSIONS ..................................................................................... 59
    5.3.6         STREAM SHADE RESULTS ...................................................................................... 60
  5.4          AREAS OF OPPORTUNITY: LWD ENHANCEMENT ........................................................... 65
    5.4.1         MAINSTEM GRANITE CREEK .................................................................................. 67
    5.4.2         NORTH FORK GRANITE CREEK .............................................................................. 70
    5.4.3         SOUTH FORK GRANITE CREEK ............................................................................... 71
    5.4.4         AREAS OF OPPORTUNITY: LWD ENHANCEMENT SUMMARY ................................. 72
    5.4.5         STREAM SHADE CONCLUSIONS AND RECOMMENDATIONS ..................................... 73
6.0     SEDIMENT SOURCES ....................................................................................................... 74
  6.1          CRITICAL QUESTIONS .................................................................................................... 74
  6.2          METHODS ...................................................................................................................... 74
    6.2.1          FIELD DATA COLLECTION ...................................................................................... 75
    6.2.2          WEPP ROAD MODELING ........................................................................................ 76
    6.2.3          CULVERT RISK ASSESSMENT ................................................................................. 77
  6.3          RESULTS ........................................................................................................................ 79
    6.3.1          WEPP ROAD MODELING ........................................................................................ 79
    6.3.2          CULVERT RISK ASSESSMENT ................................................................................. 80
  6.4          INFORMATION GAPS AND MONITORING NEEDS ............................................................. 82
  6.5          CONCLUSIONS AND RECOMMENDATIONS ...................................................................... 82
7.0     WATER QUALITY.............................................................................................................. 83
  7.1          CRITICAL QUESTIONS .................................................................................................... 83
  7.2          METHODS ...................................................................................................................... 83
  7.3          RESULTS ........................................................................................................................ 83
  7.4          INFORMATION GAPS AND MONITORING NEEDS ............................................................. 92
  7.5          CONCLUSIONS AND RECOMMENDATIONS ...................................................................... 92
8.0     FISHERIES ........................................................................................................................... 94
  8.1          INTRODUCTION .............................................................................................................. 94
  8.2          CRITICAL QUESTIONS .................................................................................................... 94
  8.3          FISH SPECIES AND LIFE HISTORIES ................................................................................ 95
  8.4          HISTORICAL FISHERIES INFORMATION ........................................................................... 96
  8.5          FISHERIES MANAGEMENT PROGRAMS ......................................................................... 104
    8.5.1           IDAHO DEPARTMENT OF FISH AND GAME ............................................................ 104
    8.5.2           US FISH AND WILDLIFE SERVICE ......................................................................... 105
  8.6          METHODS .................................................................................................................... 107
    8.6.1           DATA COLLECTION .............................................................................................. 107
    8.6.2           DATA EVALUATION ............................................................................................. 109
    8.6.3           REFERENCE BENCHMARKS FOR HABITAT VARIABLES ......................................... 112
  8.7          RESULTS ...................................................................................................................... 114
    8.7.1           AQUATIC HABITAT CONDITIONS .......................................................................... 114
    8.7.2           FISH DISTRIBUTION, ABUNDANCE, AND CONNECTIVITY ...................................... 125
    8.7.3           FISH DENSITY ...................................................................................................... 126
  8.8          SUMMARY .................................................................................................................... 133



Granite Creek Assessment                                         WPN                                                                 Page iv
9.0        REFERENCES ................................................................................................................... 135



                                                          List of Tables
Table 1. Subwatershed characteristics. .......................................................................................... 7
Table 2. National Wetland Inventory (NWI ) summary. ............................................................... 8
Table 3. Summary of channel types (miles of stream by type). .................................................... 9
Table 4. Stream gages in and around the Granite Creek watershed. ........................................... 19
Table 5. Descriptions of vegetation classification codes used in the DHSVM analysis. ............ 27
Table 6. Crosswalk between IPNF Forest type, LANFIRE Biophysical Setting, and Biophysical
    Environment classification.................................................................................................... 28
Table 7. Crosswalk between IPNF Size and Age Classes, and Size Description. ....................... 29
Table 8. Crosswalk between LANFIRE Existing Vegetation Height classes and Majore
    Structure type. ....................................................................................................................... 30
Table 9. Current Conditions of Biophysical Environments by Structural Stage. ........................ 32
Table 10. Reference Conditions of Biophysical Environments by Structural Stage (non-forest
    and unknown BPE‟s are excluded). ...................................................................................... 33
Table 11. Fractional coverage and canopy height assumptions for each DHSVM classification.
    ............................................................................................................................................... 34
Table 12. Annual mean and monthly LAI overstory and values for each DHSVM classification.
    ............................................................................................................................................... 36
Table 13. Summary of surveyed roads that are hydrologically connected. Percentages are
    percent of surveyed roads. .................................................................................................... 41
Table 14. Summary of DHSVM results. Values are median, minimum and maximum for
    thirteen annual peak flow events (WY 1994 – 2006) at the mouths of significant tributaries
    and the watershed outlet. ....................................................................................................... 45
Table 15. Modeled mean daily flow for the month of September. Values are modeled mean
    September flows for WY 1994 – 2006 at the mouths of significant tributaries and the
    watershed outlet. ................................................................................................................... 46
Table 16. Sampled stands within the watershed following a modified Common Stand Exam
    procedure. .............................................................................................................................. 49
Table 17. Tree diameter size class groupings. ............................................................................. 50
Table 18. Riparian forest and non-forest functional groups derived from the 3,783 stands within
    the Granite Creek watershed. ................................................................................................ 52
Table 19. The distribution of representative acres sampled or expanded by like functional groups
    for the watershed. .................................................................................................................. 52


Granite Creek Assessment                                               WPN                                                                   Page v
Table 20. A summary of plant associations and sampled acres ................................................... 53
Table 21. Major plant association groups identified as part of the field effort. Values are
    expressed as percentages of sampled acres within each subwatershed. ............................... 53
Table 22. Stream Shade Summary. Modeled average (minimum-maximum) effective shading
    (%) for principal stream systems in the watershed in 2008 and projected 50 and 100 years in
    the future. .............................................................................................................................. 62
Table 23. The summary of instream LWD in reaches that does not meet the 75th percentile target
    (of reference reaches), and how many pieces are required to meet those targets. ................ 67
Table 24. The summary of instream LWD in reaches that does not meet the 75th percentile target
    (of reference reaches), and how many pieces are required to meet those targets. ................ 70
Table 25. The summary of instream LWD in reaches that does not meet the 75th percentile target
    (of reference reaches), and how many pieces are required to meet those targets. ................ 71
Table 26. Road densities and surveyed road drainage connectivity. ........................................... 79
Table 27. WEPP Road modeling summary. ................................................................................ 80
Table 28. Environmental Risk Scores and overall rating. ........................................................... 80
Table 29. IDEQ Assessment Units in the Granite Creek watershed, and 2002 report status. ..... 84
Table 30. Water temperature standards applicable in the Granite Creek watershed. .................. 86
Table 31. Site location, responsible agency, and period of record for thermographs within the
    Granite Creek watershed. Thermograph data and data plots are available for all stations and
    years in Appendix E: Thermograph Data and Data Plots. ................................................... 91
Table 32. Regression statistics for the MWMT prediction equation. ........................................... 92
Table 33. Background information on fisheries in Granite Creek. ............................................ 100
Table 34. Spawning Gravel Condition Rating. .......................................................................... 109
Table 35. Habitat Variables summarized at the stream reach scale. .......................................... 111
Table 36. Habitat quality benchmarks from reference streams in the Upper Priest River
    watershed. ........................................................................................................................... 112
Table 37. North Fork Granite Subwatershed: Comparison of Channel and Pool Habitat Variables
    to Reference Conditions. ..................................................................................................... 115
Table 38. South Fork Granite Subwatershed: Comparison of Channel and Pool Habitat Variables
    to Reference Conditions. ..................................................................................................... 116
Table 39. Main Granite Subwatershed: Comparison of Channel and Pool Habitat Variables to
    Reference Conditions. ......................................................................................................... 117
Table 40. North Fork Subwatershed: Large Woody Debris variables. Stream reaches compared
    to the “Large” Stream LWD reference benchmark are indicated by bold text. .................. 119
Table 41. South Fork Subwatershed: Large Woody Debris variables. Stream reaches compared
    to the “Large” Stream LWD reference benchmark are indicated by bold text. .................. 120



Granite Creek Assessment                                             WPN                                                               Page vi
Table 42. Main Granite Subwatershed: Large Woody Debris variables. Stream reaches compared
    to the “Large” Stream LWD reference benchmark are indicated by bold text. .................. 121
Table 43. North Fork Subwatershed: Substrate and spawning habitat comparison to reference
    conditions. ........................................................................................................................... 122
Table 44. South Fork Subwatershed: Substrate and spawning habitat comparison to reference
    conditions. ........................................................................................................................... 123
Table 45. Main Granite Subwatershed: Substrate and spawning habitat comparison to reference
    conditions. ........................................................................................................................... 124
Table 46. Fish Density in Reference Reaches from Upper Priest River Watershed. (Cedar and
    Jackson Creek by electrofishing, Upper Priest Rive by snorkel counts.) ........................... 126
Table 47. Density benchmarks for all age classes of trout from sampled reference reaches. ... 127
Table 48. Fish Density in North Fork Granite Creek Subwatershed. ......................................... 128
Table 49. Fish Density in South Fork Granite Creek Subwatershed. ........................................ 129
Table 50. Fish Density in Main Stem Granite Creek Subwatershed ......................................... 130
Table 51. Average fish densities, fish/100m2, reported in 1983-1984 (Irving, 1987) compared to
    fish densities for the current study. ..................................................................................... 132



                                                       List of Figures
Figure 1. Location map. ................................................................................................................. 6
Figure 2. Cumulative percent watershed area by slope class......................................................... 7
Figure 3. Water features. ................................................................................................................ 8
Figure 4. Rosgen channel types. ................................................................................................... 10
Figure 5. Cumulative channel length by Rosgen type and contributing drainage area. .............. 11
Figure 6. Mean annual air temperature at the Priest River Experiment Station. ......................... 12
Figure 7. Mean annual precipitation isohyets (inches). ............................................................... 13
Figure 8. Mean monthly precipitation by subwatershed. ............................................................. 14
Figure 9. Annual precipitation at the Priest River Experiment Station. ....................................... 14
Figure 10. Cumulative standardized departure from normal for the Priest River Experiment
    Station annual precipitation record. ...................................................................................... 15
Figure 11. Snowpack (in inches of snow water equivalent) at the Bunchgrass Meadows
    SNOTEL station. ................................................................................................................... 16
Figure 12. General landtypes found in the Granite Creek watershed (Miller et al., 1999; IPNF,
    2006). .................................................................................................................................... 17
Figure 13. Soil characteristics (IPNF, 2006). .............................................................................. 18


Granite Creek Assessment                                             WPN                                                               Page vii
Figure 14. Relationship between mean daily flows at the Upper Priest River and Granite creek
    gages. .................................................................................................................................... 19
Figure 15. Mean daily flow and annual peak flow summary at the Priest River gage (period of
    record 5/7/1986 - 9/30/1998;10/1/1999- 9/30/2004). ........................................................... 20
Figure 16. Magnitude and recurrence interval for annual peak flow events at the Upper Priest
    River gage. ............................................................................................................................ 21
Figure 17. Vegetation size classes (IPNF, 2007). ........................................................................ 22
Figure 18. Vegetation size class distribution. .............................................................................. 22
Figure 19. Land ownership (IPNF, 2006). ................................................................................... 23
Figure 20. Management classes (IPNF, 2004). ............................................................................ 23
Figure 21. Model representation of watershed soil, vegetation and topography as discrete pixels
    (from Vanshaar and Lettenmaier, 2001). .............................................................................. 25
Figure 22. Stages of Structural Development. ............................................................................. 31
Figure 23. Current and potential vegetation conditions. .............................................................. 32
Figure 24. Soil depth and textural class. ...................................................................................... 37
Figure 25. Relationship between bankfull width (right) and depth (left) and contributing area for
    streams in the North Fork Coeur d‟Alene subbasin. ............................................................. 39
Figure 26. Surveyed roads having road ditches that are hydrologically connected to the stream
    system. .................................................................................................................................. 40
Figure 27. Observed and modeled streamflow at the Granite Creek gage. ................................. 43
Figure 28. Observed vs. modeled mean daily discharge at Granite Creek gage. ......................... 43
Figure 29. Trees per acre 20 inches DBH for a 100 year time period. ...................................... 54
Figure 30. Projected tree mortality per acre for trees 20 inches DBH. ..................................... 55
Figure 31. The ratio of projected dead to projected live trees 20 inches DBH. ........................ 55
Figure 32. Basal area (ft2 per acre) of trees 20 inches DBH...................................................... 56
Figure 33. The degree of departure from the subwatershed average, for standing TPA 20 inches
    DBH in 2008. ........................................................................................................................ 57
Figure 34. The average number of dead trees 20 inches in diameter likely to be produced in
    10-year increments per 100m of stream for year 10, 50 and 100 (1st, 5th and 10th time step),
    separated by reach. ................................................................................................................ 58
Figure 35. Current effective shade (top), and potential future shade conditions in 50 (middle) and
    100 (bottom) years along principal streams, Granite Creek watershed. ............................... 61
Figure 36. Longitudinal effective shade profiles for South Fork Granite Creek (top) and North
    Fork Granite Creek for current and modeled future conditions. ........................................... 63
Figure 37. Longitudinal effective shade profiles for Lower Mainstem Granite Creek (top) and
    Blacktail Creek for current and modeled future conditions. ................................................. 64


Granite Creek Assessment                                             WPN                                                               Page viii
Figure 38. Measured LWD sites. Light shading represents reaches that meet or exceed the 75th
    percentile as compared with reference reaches. Black lines indicate the number of pieces
    and/or dams required to meet the 75th percentile. Assuming active restoration, the colored
    lines indicate the remaining standing conifer trees 20 inches DBH, expressed as a
    percentage of the current and projected values per 100 m. Green is a >75% retention,
    orange is 50 – 75% and red is <50% retention or there are insufficient trees on site. .......... 66
Figure 39. The projected remaining standing live trees 20 inches DBH per 100 m if instream
    LWD restoration projects sourced wood from the immediate riparian zone (Table 23). ..... 68
Figure 40. The projected remaining standing live trees 20 inches DBH per 100 m if instream
    LWD restoration projects sourced wood from the immediate riparian zone (Table 23). ..... 69
Figure 41. The projected remaining standing live trees 20 inches DBH per 100 m if instream
    LWD restoration projects sourced wood from the immediate riparian zone (Table 24). ..... 71
Figure 42. The projected remaining standing live trees 20 inches DBH per 100 m if instream
    LWD restoration projects sourced wood from the immediate riparian zone (Table 25). ..... 72
Figure 43. System roads in the Granite Creek watershed by maintenance category, and roads and
    culverts surveyed in 2006-2007. ........................................................................................... 75
Figure 44. Miles of road by subwatershed and survey status. ..................................................... 76
Figure 45. Environmental Risk Ratings for 169 surveyed culverts. ............................................ 81
Figure 46. Summary of Environmental Risk ratings for surveyed culverts. ............................... 81
Figure 47. IDEQ assessment map units within the Granite Creek watershed. ............................ 85
Figure 48. Bull trout temperature standards and thermograph locations. ..................................... 87
Figure 49. Example thermograph for Granite Creek at the Road 302 bridge. ............................. 88
Figure 50. Percentage of days that the Maximum Weekly (7-day average) Maximum
    Temperature (MWMT) exceeded 10 degrees C. .................................................................. 89
Figure 51. Location of bull trout redds in surveyed sections of the North Fork Granite Creek,
    Priest Lake, 1983 (from Mauser and Ellis 1985). ............................................................... 103
Figure 52. Length-frequency distribution for bull trout at the weir in Granite Creek, mid-July to
    early September, 1983 (from Mauser and Ellis 1985). ....................................................... 104
Figure 53. Box plot of relative fish abundance from thirteen reference reaches in Upper Priest
    River watershed. (25th to 75th percentile indicated by the box, median by a horizontal line,
    and whiskers indicate 10th and 90th percentiles.) ................................................................ 127
Figure 54. Length frequency of cutthroat trout in Tillicum Creek, a tributary in Upper Granite
    Creek, to Cedar Creek, a tributary in Upper Priest River watershed. ................................. 128




Granite Creek Assessment                                     WPN                                                         Page ix
                                       Appendices

Appendix A:    DHSVM model configuration file
Appendix B:    Digital Appendices to the Riparian Assessment
Appendix C:    WEPP Inputs and Outputs
Appendix D:    Culvert Risk Assessment Data
Appendix E:    Thermograph Data and Data Plots
Appendix F:    Habitat Benchmarks derived from the Upper Priest River watershed
Appendix G:    Tables for Habitat Variables in Granite Creek
Appendix H:    Fish Habitat Distribution Maps




Granite Creek Assessment                     WPN                                  Page x
                             1.0     EXECUTIVE SUMMARY


The Kalispel Tribe Natural Resource Department (KNRD) and the Idaho Panhandle National
Forest (IPNF) are working jointly to restore native fish populations throughout the Priest Lake
basin. Granite Creek, a major tributary on the east side of the lake, which is primarily under
National Forest management, is a high priority for enhancement of native salmonid habitat. The
watershed assessment identified limiting factors for native fish, assessed hydrologic and riparian
function, and identified opportunities for restoration.

The reduction in fish populations in Granite Creek has been attributed to a legacy of land
management activities in combination with displacement and hybridization with nonnative
salmonids. The goal of this project was to identify habitat conditions at the stream reach scale,
evaluate watershed processes that are limiting native fish populations, and identify restoration
actions that will comprise a restoration strategy.

Granite Creek is located in Bonner County, in northwestern Idaho, and Pend Oreille County, in
northeast Washington. Elevations range from approximately 2,400 feet to 6,500 feet in the
headwaters of the North Fork. Despite being located over 300 miles from the Pacific Ocean, the
Granite Creek area is influenced by maritime air masses that are moved eastward by the
prevailing westerly winds. The area is relatively warmer and wetter in wintertime than mid-
continental areas of the same latitude and elevation. Mean annual precipitation ranges from 30
to 45 inches depending on elevation. Precipitation is lowest in August, and greatest in
November. Over 95% of the watershed area is forested, primarily with evergreen species. The
majority of the watershed (91%) is administered by the U.S. Forest Service with 90% of the
watershed area in some category of “Timber” management.

The watershed is approximately 99 square miles in size, and consists of three sixth field
subwatersheds. Stream channels within the subwatersheds – Main Granite, North Fork, and
South Fork – were classified by morphological channel characteristics using the Rosgen
classification scheme. Of the 132 stream miles classified the majority are distributed
approximately equally between B and C channels.

The watershed assessment evaluated current fish populations in relation to historic conditions
and reference areas located in the Upper Priest River watershed, and addressed the main drivers
of native fish habitat: 1) Habitat connectivity relating both to natural and man-made barriers, 2)
Hydrologic Regime as it is influenced by management activities, 3) Sediment Supply, 4)
Thermal Regime, and 5) Riparian Vegetation.

Westslope cutthroat trout (cutthroat) and brook trout comprise the current trout populations in
Granite Creek. In some stream reaches cutthroat trout predominate, in many others native fish
have been entirely replaced by brook trout. The current populations of cutthroat trout in Granite
Creek are likely composed predominately of resident fish, since the migratory component has
been nearly eliminated due to predation by lake trout. The relative abundance of cutthroat trout
in comparison to brook trout provides a mechanism for prioritizing stream reaches. Stream
reaches targeted for restoration, such as LWD enhancement, can be prioritized based on the


Granite Creek Assessment                       WPN                                           Page 1
relative abundance of cutthroat trout to brook trout, with cutthroat trout only waters receiving the
highest priority.

Bull trout are almost now entirely absent from Granite Creek, as evidenced by bull trout not
being encountered in recent electrofishing or snorkel counts completed by the Tribe and Forest
Service (2006 – 2008). A few bull trout redds were counted in the North Fork of Granite Creek
in 2008.

The distribution of cutthroat trout and brook trout in relationship to natural and man made
barriers was identified. This information will be used to develop recommendations for treatment
of fish passage barriers. The tradeoff between genetic isolation of the cutthroat trout population
versus the need to reconnect cutthroat populations needs to be balanced against the threat of
further brook trout invasion.

The Distributed Hydrology Soil Vegetation Model (DHSVM) was used to evaluate the extent to
which stream flows are affected by land management activities. The model uses spatial data on
topography, vegetation, soils, streams, and roads to evaluate effect on peak/base flows. Modeled
harvest effects on peak or base flows were minimal, given that the current vegetation is similar to
the historic range of variability. Hydrologically connected roads segments, particularly those in
the North Fork Granite and Athol Creek drainages, are estimated to have a moderate impact on
peak flows, and should be decoupled from the stream system to eliminate possible impacts.

Potential sediment sources in Granite Creek include mass wasting, upland erosion associated
with forest harvest, and road-related sediment inputs. Mass failure potential is generally very low
in the watershed, and harvest rates in the Granite Creek watershed have been low in recent years,
so these sources were not further quantified. The evaluation of sediment sources focused on
road-related sediment using two primary methods, the Water Erosion Prediction Project (WEPP)
model and a culvert risk assessment. The model results predicted that current road inputs are
estimated to be negligible. We assume that fines that are currently detected in aquatic habitats
are probably a legacy effect of past forest management activity.

The riparian assessment focused on evaluating the current condition and trajectory of riparian
vegetation over a 100-year scenario in providing riparian functions of large wood and shade. A
sample of forest stands within the riparian zones were randomly selected to represent the range
of cover types that would likely contribute LWD to the stream channel. The Forest Vegetation
Simulator (FVS) was used to model tree mortality, height, diameter and crown cover to predict
future LWD and shade. The conclusions from the modeling run relative to LWD enhancement
are: 1) There appears to a steady supply of large conifer trees in the system, 2) available large
dead trees steadily increases and levels to a steady range, and 3) ample opportunities exist to
source LWD for instream placement without high levels of modifications to forest structure
(trees per acre/ basal area) for forested sites.

Water temperature exceeds the 10 degrees Celsius EPA temperature criteria for bull trout in most
streams, including tributary and headwater locations, most of the time. Results of modeling of
effective shade using HeatSource showed that current shade levels are generally high and do not
appear to change dramatically over the next 100 years. Given these model results, it is unlikely


Granite Creek Assessment                       WPN                                            Page 2
that summertime water temperatures within the Granite Creek watershed are currently
significantly different than what they were historically. However, given the limited temperature
and flow data currently available, additional monitoring and analysis would be necessary prior to
delisting these streams as being water quality limited for temperature.

Aquatic habitat conditions were evaluated in relationship to reference streams from the Upper
Priest River watershed. Habitat limiting factors were identified by stream reach by ranking the
habitat parameter in comparison to reference condition on a scale from one to four. These
individual rankings provide a method to link restoration action needs, such as road treatments, to
limiting factors in the aquatic habitat at the stream reach scale. Habitat surveys generally
indicate that poor habitat in Granite Creek stream reaches are due to high levels of fine
sediments, poor pool habitats or lack of LWD. However, streambank stability and undercut
banks were generally similar to reference conditions, indicating that the physical integrity of the
stream channels is currently stable or in a recovery trend.




Granite Creek Assessment                       WPN                                            Page 3
                                    2.0     INTRODUCTION

2.1     GOALS AND OBJECTIVES
The decline in native salmonid populations in Granite Creek has been attributed to a legacy of
land management activities; and displacement and hybridization with nonnative salmonids. The
Kalispel Natural Resource Department (KNRD) and Idaho Panhandle National Forest (IPNF)
have assembled existing data on fisheries and watershed processes, and collected additional
habitat and riparian condition data from 2006 to 2008. KNRD and IPNF are working together to
restore native populations of fish in the Granite Creek watershed. This project will assist that
effort by assessing watershed conditions and making recommendations for habitat restoration.

The purpose of the project is to identify limiting factors and watershed processes that are limiting
native fish populations, identify restoration actions, and summarize these actions into a
defensible and cost-effective restoration strategy. Specific goals of the project are:

      1. Summarize the existing data so that it is meaningful at the stream reach, subwatershed
         and watershed scales,

      2. Evaluate the in-stream factors that are limiting fish populations,

      3. Identify causes and existing sources that impact aquatic habitats and connectivity,

      4. Identify tractable remedies to abate problem sources and enhance/restore fish habitats,
         fish passage, and address the impacts of nonnative fish, and

      5. Synthesize these elements into a Granite Creek Restoration Strategy.


2.2     ORGANIZATION OF THIS DOCUMENT
The document is organized to address the main drivers of salmonid habitat using data collected
by the Kalispel Tribe and the Forest Service. These primary drivers include: 1) Habitat
connectivity relating both to natural and man-made barriers, 2) Hydrologic Regime as it is
influenced by management activities, 3) Sediment Supply, 4) Thermal Regime, and 5) Riparian
Vegetation. In addition, current fish populations are compared to historic and reference
conditions to identify priorities for restoration actions. Sections of the report include:

Section 3.0, Watershed Overview. The watershed overview describes the biophysical
environment that provides the foundation for aquatic habitat.

Section 4.0, Hydrology. The hydrology assessment evaluates the extent to which stream flows
are affected by land management activities.




Granite Creek Assessment                         WPN                                           Page 4
Section 5.0, Riparian Conditions. The riparian assessment focuses on evaluating the current
condition and trajectory of adjacent forests to provide riparian ecosystem services of large wood
and shade.

Section 6.0, Sediment Sources. The sediment assessment evaluates sediment inputs from the
watershed with an emphasis on quantifying road-related sediment.

Section 7.0, Water Quality. Idaho DEQ has identified support status for beneficial uses of
aquatic life in Granite Creek. The assessment focuses on evaluation of water temperature, which
has been identified as the cause for not supporting the cold water aquatic life use in reaches of
Granite Creek.

Section 8.0, Fisheries. The fisheries section summarizes fish habitat and population historic
information, evaluates habitat limiting factors, and current fish population distribution and
abundance.

The Restoration Action Plan will be developed as a separate document which builds on the
watershed assessment




Granite Creek Assessment                      WPN                                           Page 5
                           3.0   WATERSHED OVERVIEW

3.1   LOCATION & SUBWATERSHEDS
The Granite Creek Watershed is located in Bonner County, in northwestern Idaho, and Pend
Oreille County, in northeast Washington (Figure 1). The watershed is approximately 99 square
miles in size, and consists of three sixth field subwatersheds. Elevations range from
approximately 2,400 feet at the watershed outlet at Priest Lake, to approximately 6,500 feet in
the headwaters of the North Fork (Table 1). Slopes are relatively gentler in the South Fork
drainage, where approximately 10% of the watershed area has slopes of 5% or less (Figure 2).




Figure 1. Location map.



Granite Creek Assessment                     WPN                                         Page 6
Table 1. Subwatershed characteristics.

                                                              Drainage area
                                                                     2
                                         Subwatershed             (mi )                     Mean elevation (min-max) in feet
North Fork Granite Creek                                          29.5                                4,477 ( 3,013-6,461)
South Fork Granite Creek                                          34.6                                4,063 (3,014-6,264)
                                        Main Granite Creek        35.1                                3,623 (2,442-5,693)



                                        100%

                                         90%

                                         80%
 Cumulative percent subwatershed area




                                         70%

                                         60%
                                                                                           Mainstem - Cumulative %
                                                                                           North Fork - Cumulative %
                                         50%                                               South Fork - Cumulative %

                                         40%

                                         30%

                                         20%

                                         10%

                                          0%
                                               0     10      20     30        40      50         60         70         80    90   100
                                                                                   Slope (%)

Figure 2. Cumulative percent watershed area by slope class1.

3.2                                       WATER FEATURES
The National Hydrography Dataset (NHD) identifies 177 miles of stream in the Granite Creek
watershed, 112 miles of which are identified as Perennial, and 65 miles identified as Intermittent
(Figure 3). Approximately 1% of the watershed area is comprised of forested / shrub dominated
wetland, and another 1% in emergent wetland types. Emergent wetland complexes are located
primarily in the lower reaches of the North and South Forks, and in some low-gradient headwater
locations (Figure 3).




1 Data source: USGS 1/3 arc-second digital elevation model (http://seamless.usgs.gov)


Granite Creek Assessment                                                           WPN                                            Page 7
Figure 3. Water features2.
Table 2. National Wetland Inventory (NWI ) summary.

                           Freshwater Freshwater
                            Emergent Forest/Shrub Freshwater
     Subwatershed            Wetland   Wetland       Pond                Lake         Riverine   Grand Total
Main Granite Creek             94            307            22             0             7          429
North Fork Granite
Creek                          114            62             6             -              -         182
South Fork Granite
Creek                          286           155            33             -              -         474
Grand Total                    494           524            60             0             7         1,085




2 Sources: National Hydrography Data Set (http://nhd.usgs.gov/); National Wetland Inventory
(http://www.fws.gov/wetlands/).


Granite Creek Assessment                             WPN                                               Page 8
3.3        ROSGEN CHANNEL TYPES
Classification of stream channels within a watershed is an important part of understanding the
inherent spatial variation in aquatic habitat conditions, and is important in prioritizing and
understanding the limitations to possible restoration activities. The underlying assumption in
any channel-typing scheme is that the morphological channel characteristics are the result of
geologic, climatic, and vegetative interactions. Furthermore, similar channel types can be
expected to respond in a similar manner to natural or human-caused changes within a watershed
in the supply of water, sediment, or wood inputs.

The classification scheme used in this analysis is commonly referred to as the Rosgen
methodology (Rosgen 1994 and 1996). The Rosgen methodology utilizes a hierarchical
approach to channel classification. The Rosgen classification methodology was applied to the
principal streams within the Granite Creek watershed using data collected by USFS and Tribal
survey crews. Rosgen channel types found within the watershed are summarized in Table 3, and
are shown in Figure 4.

Table 3. Summary of channel types (miles of stream by type).
                                                                                              Main
Channel                                                                      North   South   Granite Grand
 type                            General description                         Fork    Fork    Creek   Total
             Steep, entrenched, cascading, step/pool streams. High
      A      energy/debris transport associated with depositional soils.      0.4     4.0     1.4     5.8
             Very stable if bedrock or boulder dominated channel.
             Moderately entrenched, moderate gradient, riffle
      B      dominated channel, with infrequently spaced pools. Very         11.1     7.5     9.9    28.6
             stable plan and profile. Stable banks.
             Low gradient, meandering, point-bar, riffle/pool, alluvial
      C                                                                       4.6    10.5     15.7   30.8
             channels with broad, well defined floodplains
             Low gradient, meandering riffle/pool stream with low
      E      width/depth ratio and little deposition. Very efficient and      1.3     4.4     1.7     7.4
             stable. High meander width ratio.
             Entrenched meandering rime/pool channel on low
      F                                                                        -       -      0.5     0.5
             gradients with high width/depth ratio.
 Beaver Beaver dam complex (not a Rosgen classification).                      -       -      0.4     0.4
      na     Not assigned a channel type                                     18.7    24.4     16.2   59.3
                                                               Grand Total   36.1    50.8     45.9   132.8



The A stream types are steep (4%+ channel gradient) streams located primarily in headwater
areas. Transport processes dominate in these reaches, as they are often source areas for
downstream deposition. In Granite Creek A type channels occur primarily in headwater areas
having 10 square miles or less of contributing drainage area (Figure 4, Figure 5).




Granite Creek Assessment                               WPN                                           Page 9
Figure 4. Rosgen channel types.
Rosgen B type streams are typically found positioned downstream of A type channels, or as a
higher gradient, more confined reach between areas of lower confinement. Gradients are
typically in the 2-4% range. Although these streams are morphologically dominated by hillslope
(as opposed to floodplain) processes, they often contain some areas of floodplain development
and may be both transport and depositional reaches. Type B channels are found in all portions of
the Granite Creek watershed, from headwater streams to the mainstem of lower Granite Creek
(Figure 4, Figure 5).

Rosgen C type channels consist of relatively low-gradient streams with well-developed
floodplains and are typically highly responsive to sediment and wood inputs. Type C channels
make up the largest grouping of typed channels in the Granite Creek watershed (Table 3), and
are found typically in the mainstem and large tributaries (Figure 4, Figure 5).




Granite Creek Assessment                     WPN                                         Page 10
Figure 5. Cumulative channel length by Rosgen type and contributing drainage area.
Rosgen E type streams are very low gradient, meandering riffle/pool streams with low
width/depth ratio and little deposition. These streams are very efficient at transporting sediment
(despite their low gradient), and are stable in terms of bank erosion. E type streams are often
found flowing through meadow areas. Type E channels make up a small proportion of the total
miles of typed channels in the Granite Creek watershed (Table 3), and are found typically in the
mid- to upper-portions of tributaries (Figure 4, Figure 5).

Rosgen F type streams are similar to E type channels but they are much more entrenched, have
high width/depth ratios, and are laterally unstable with eroding banks. Type F channels are only
found in one location along the mainstem of Granite Creek (Figure 4, Figure 5). The Large
beaver dam complex along Fedar Creek was not classified as a Rosgen type (Figure 4).

3.4   CLIMATE
Despite being located over 300 miles from the Pacific Ocean, the Granite Creek area is
influenced by maritime air masses that are moved eastward by the prevailing westerly winds.
The area is relatively warmer and wetter in wintertime than mid-continental areas of the same
latitude and elevation.

Minimum air temperatures occur in the months of December and January, and maximum
temperatures occur in the months of July and August. Mean minimum January temperatures
range from 12 to 18 degrees Fahrenheit, and mean maximum July temperatures range from 68 to


Granite Creek Assessment                      WPN                                          Page 11
78 degrees Fahrenheit (PRISM, 2006). Temperatures within the watershed vary primarily with
elevation.

Long-term climatic records are available from the Priest River Experiment Station (NOAA ID
#107386, elevation 2,380 feet), located approximately 20 miles south of the Granite Creek
watershed outlet (Williams et al., 2008). Values for mean annual air temperature (by water year;
October 1 to September 30) are shown in Figure 6. Data indicate a 1.6 degrees Fahrenheit per
decade increase in temperature for the period of record, however, this apparent trend has not
been analyzed to determine its statistical significance.




Figure 6. Mean annual air temperature at the Priest River Experiment Station.
The Oregon Climate Service has published digital maps of mean annual and monthly
precipitation for the United States, based on available precipitation records for the period 1971-
2000 (PRISM, 2006). Mean annual precipitation within the watershed generally increases with
increasing elevation, ranging from approximately 29 inches in the middle portion of the Main
Granite Creek subwatershed to approximately 50 inches in the middle portion of the North Fork
subwatershed, and is 40 inches overall (Figure 7).




Granite Creek Assessment                      WPN                                          Page 12
Figure 7. Mean annual precipitation isohyets (inches).
Mean monthly precipitation was estimated for each subwatershed using data available from the
OCS (Figure 8). Mean monthly precipitation increases with elevation among the subwatersheds.
Mean monthly precipitation is lowest in the month of August for all subwatersheds and is
greatest in the month of November for subwatersheds.

Year-to-year variability in precipitation was assessed using the long-term record from the Priest
River Experiment Station (Williams et al., 2008). Values for annual precipitation (by water
year) are shown in Figure 9. These data indicate a slight (0.3 inch) per decade increase in
precipitation for the period of record.

Data from the Priest River Experiment Station was further analyzed to determine if any cyclical
patterns were apparent in the data set. These data were processed as follows:

    1. Mean and standard deviation were calculated for annual precipitation over the period of
       record.




Granite Creek Assessment                      WPN                                         Page 13
Figure 8. Mean monthly precipitation by subwatershed.




Figure 9. Annual precipitation at the Priest River Experiment Station.




Granite Creek Assessment                   WPN                           Page 14
    2. A standardized departure from normal was calculated for each year by subtracting mean
       annual precipitation from annual precipitation for a given year and dividing by the
       standard deviation.

    3. A cumulative standardized departure from normal was then calculated by adding the
       standardized departure from normal for a given year to the cumulative standardized
       departure from the previous year (the cumulative standardized departure from normal for
       the first year in a station record was set to zero).

This approach of using the cumulative standardized departure from normal better illustrates
patterns of increasing or decreasing precipitation over time by reducing year-to-year variations in
precipitation, thus compensating for the irregular nature of the data set. Values for the
cumulative standardized departure from normal increase during wet periods and decrease during
dry periods. Results for the Priest River Experiment Station are given in Figure 10. Regionally
there appears to have been a long warm/dry period from the beginning of the climate record to
approximately 1945. This was followed by a long cool/wet phase that lasted until the mid 1970s
or mid 1980s. The recent record has been much more erratic, with a relatively short warm/dry
phase from the mid 1980s to mid 1990s, followed by an equally short cool/wet phase that may
have ended around the year 2000.




Figure 10. Cumulative standardized departure from normal for the Priest River
Experiment Station annual precipitation record.



Granite Creek Assessment                       WPN                                          Page 15
Data on snowpack are available from the Bunchgrass Meadow SNOTEL station, located at an
elevation of 5,000 feet on the ridge separating the South Fork of Granite Creek and Harvey
Creek. The period of record for the station is from October 1981 to the present. A snowpack is
generally in place from Mid-October through the end of May in the higher elevation areas of the
watershed, reaching its maximum depth during the middle of April (Figure 11).




Figure 11. Snowpack (in inches of snow water equivalent) at the Bunchgrass Meadows
SNOTEL station3.

3.5   GEOLOGY & SOILS
The underlying bedrock of the Granite Creek watershed is approximately evenly divided
between older Precambrian Belt Supergroup series, and granitics from the cretaceous period
(Figure 12; Miller et al., 1999; IPNF, 2006). The belt series consists of metamorphosed
sedimentary rocks, and areas of extrusive lithology, consisting of Leola Volcanics of the
Windemere group. The belt series is highly faulted, and many stream features are located along
fault lines. Glaciation during the Quaternary period has resulted in significant areas of
unconsolidated surficial deposits overlying areas of bedrock.          Soils in the Granite Creek
watershed are of glacial till, glacial outwash, and residual origin. Soils are generally moderately
permeable to well drained.

Surface erosion hazards are generally low, and subsurface erosion hazards low to moderate,
across all landform types (Figure 13). Mass failure potential is generally low, with the exception


3 Data source: http://www.wcc.nrcs.usda.gov/snotel/snotel.pl?sitenum=376&state=wa


Granite Creek Assessment                           WPN                                      Page 16
of steep headwalls and incised stream canyons. Areas of high mass failure potential comprise
approximately 7% of the watershed area. Areas of high sediment delivery potential are located
generally along streams and valley bottoms, and make up about 18% of the watershed area. Soil
productivity is rated as moderate for 70% of the watershed. Natural sediment load ranges from
1-173 tons/mi2/year among individual landtypes. Natural sediment load ranges from 28
tons/mi2/year in the South Fork Granite Creek to 31 tons/mi2/year in the North Fork Granite
Creek and is 30 tons/mi2/year in Main Granite Creek and for the watershed as a whole.




Figure 12. General landtypes found in the Granite Creek watershed (Miller et al., 1999;
IPNF, 2006).




Granite Creek Assessment                    WPN                                       Page 17
Figure 13. Soil characteristics (IPNF, 2006).



Granite Creek Assessment                    WPN   Page 18
3.6   HYDROLOGY
Few data are available to characterize streamflow within the Granite Creek watershed. The
Idaho Department of Environmental Quality (IDEQ) maintained a continuous flow station at the
mouth of Granite Creek for only two water years (Table 4; G. Rothrock, pers comm., 6/19/2008).
The longest-term gage in the vicinity of the project area is the USFS gage on Upper Priest River.
The Upper Priest River gage is located approximately 5 miles north of the Granite Creek
watershed. The similarity in watershed characteristics (Table 4), and the good relationship
between mean daily flow values at each station (Figure 14) suggest that the Upper Priest River
gage may adequately represent flow characteristics in the Granite Creek watershed.

Table 4. Stream gages in and around the Granite Creek watershed.
                                Gage       Mean annual     Mean (min-max)
               Contributing   elevation    precipitation    elevation (ft)
                        2
   Gage         area (mi )      (feet)         (in)                            Period of record
  Granite                                                   4,031 (2,442-        10/1/1993 –
                     99        2,470            40
Creek gage                                                     6,461)             9/30/1995
                                                                                  5/7/1986 -
Upper Priest                                                4,769 (2,583-         9/30/1998;
                     71        2,520            53
River gage                                                     7,528)             10/1/1999-
                                                                                  9/30/2004




Figure 14. Relationship between mean daily flows at the Upper Priest River and Granite
creek gages.


Granite Creek Assessment                      WPN                                         Page 19
Given the location and climate, it is likely that both snowmelt and winter rain-on-snow (ROS)
are important peak flow-generating processes in the Granite Creek watershed. Examination of
the hydrograph for the Upper Priest River gage (Figure 15) shows a hydrograph typical for a
snowmelt-dominated system. All of the largest annual instantaneous peak flow events at the
gage occur during the spring snowmelt season, however, the spiky pattern of maximum mean
daily flows indicate that ROS events are prevalent.




Figure 15. Mean daily flow and annual peak flow summary at the Priest River gage
(period of record 5/7/1986 - 9/30/1998;10/1/1999- 9/30/2004).
Data on annual peak flows from the Upper Priest River gage (Figure 15) were used to illustrate
recent flood history. The magnitudes of the annual peak flow events are presented as a time
series in Figure 16. In addition, recurrence intervals were calculated for the period of record
using techniques described by the Interagency Advisory Committee on Water Data (1982), and
the recurrence intervals are also shown in Figure 16. The largest recorded peak flow event,
estimated to have a recurrence interval of approximately 25 years, occurred in water year 2002.




Granite Creek Assessment                     WPN                                        Page 20
Figure 16. Magnitude and recurrence interval for annual peak flow events at the Upper
Priest River gage.

3.7   VEGETATION
Information on vegetation types within the watershed is available from the Idaho Panhandle
National Forest (IPNF, 2007). Over 95% of the watershed area is forested; primarily with
evergreen species. The majority of forest lands are in the “sawtimber” size class (Figure 17,
Figure 18).

3.8   LAND OWNERSHIP & USE
The majority of the watershed (91%) is owned by the United States of America, administered by
the US Forest Service (Figure 19). An additional 7% is owned by private timber companies, and
the remaining 2% is in small private ownership. Over 90% of the watershed area is in some
category of “Timber” management, with approximately half of this area having a wildlife habitat
focus (Figure 20). Six percent of the watershed is in a “Minimum Management” class, and less
than 15% is in “Wilderness” or “Special Areas” designation.




Granite Creek Assessment                     WPN                                        Page 21
Figure 17. Vegetation size classes (IPNF, 2007).




Figure 18. Vegetation size class distribution.



Granite Creek Assessment                    WPN    Page 22
Figure 19. Land ownership (IPNF, 2006).




Figure 20. Management classes (IPNF, 2004).




Granite Creek Assessment                  WPN   Page 23
                                        4.0      HYDROLOGY

4.1       CRITICAL QUESTIONS
         To what extent have stream flows changed due to management-related activities (harvest
          and roads)?

4.2       METHODS
The Distributed Hydrology Soil Vegetation Model4 (DHSVM) is a distributed hydrologic model
originally developed to evaluate the effects of topography and vegetation on water movement
through a watershed (Wigmosta et al., 1994). Spatially distributed models such as DHSVM
provide a dynamic representation of the spatial distribution of soil moisture, snow cover,
evapotranspiration, and runoff production, at the scale of digital elevation model (DEM) pixel
(Figure 21). DHSVM has been used to assess changes in flood peaks due to enhanced rain-on-
snow and spring radiation melt response (e.g., Thyer et al., 2004), effects of forest roads and road
drainage (e.g., Lamarche and Lettenmaier, 2001), and the prediction of sediment erosion and
transport (Doten and Lettenmaier. 2004).




4An overview of the DHSVM model, source code, and details of the model application, can be found at
http://www.hydro.washington.edu/SurfaceWaterGroup/Models/DHSVM/index.shtml


Granite Creek Assessment                             WPN                                              Page 24
Figure 21. Model representation of watershed soil, vegetation and topography as discrete
pixels (from Vanshaar and Lettenmaier, 2001).



We used the DHSVM model to assess management-related impacts on stream flows from forest
harvest and roads in the Granite Creek watershed. The model was first constructed for the
current condition (i.e., current vegetation, and current road network). We then evaluated
management-related impacts on stream flows by selectively removing each management impact
(i.e., replacing areas currently occupied by roads and harvest units with the potential land cover
appropriate for the area), and rerunning the model. Results from these iterations allowed us to
compare peak flow magnitudes for selected storm events under three scenarios:

       Current conditions (i.e., existing vegetation and road conditions)

       Current vegetation conditions with road effects removed

       Potential vegetation conditions (no management) and no roads



Granite Creek Assessment                       WPN                                         Page 25
The DHSVM model requires several types of spatial and temporal data inputs. All spatial data
was processed using ArcView 3.3 GIS software, with the Spatial Analyst extension. The
DHSVM model itself was run on a Linux workstation running Debian Linux version 4.0.

4.2.1 Spatial Input Data

Spatial data inputs include a digital representation of topography, vegetation, soils, streams, and
roads. Topography, soils, and stream data were constant among model runs. Vegetation and
roads varied by management scenario.

The DHSVM model runtime is dependent on the length of the model run, as well as the
resolution of the raster data sets. Typical values for raster data range from 25- to 150-meter pixel
resolution. A pixel resolution of 30 meters was chosen for the Granite Creek watershed.

4.2.1.1           Topography

One-third arc second DEM data (~7 meter native resolution) was acquired for the project area5
and resampled to 30-meter resolution. The resulting DEM was then filled to eliminate “sinks”;
locations where the model would not “drain” in a downstream direction. A mask file was also
created from the resultant DEMs that is used by DHSVM to identify which pixels are inside and
outside of the watershed. These DEM files (and all subsequent raster files) were then exported
from the GIS in a binary format for input to the DHSVM model.

4.2.1.2           Vegetation

Vegetation input data were generated primarily from available Forest Service databases 6,
supplemented with data from the Landscape Fire and Resource Management Planning Tools
Project (LANDFIRE7) for areas where Forest Service data were unavailable. These data were
assigned structural and compositional classifications that follow (1) the major Biophysical
Environments (BPEs) of North Idaho ecosystems, and (2) the major stages of stand development
(structural stages). Combined, the classifications were made to place current conditions in
context with the Historic Range of Variability (HRV) for North Idaho Ecosystems (Smith and
Fisher, 1997). Four available metrics were used to create a suite of classification codes used for
the DHSVM analysis:

        Stand Type/ Major Species
        Size Class (immature, sawtimber, saplings, etc.; Figure 17)
        Canopy height
        Stand Age/ Origin


5 http://seamless.usgs.gov/
6 GIS data: http://www.fs.fed.us/ipnf/eco/yourforest/gis/index.html, stand attribute data from “Stands” and
“Components” tables: http://www.fs.fed.us/ipnf/eco/yourforest/gis/veg/vegetation_readme.html
7 http://www.landfire.gov/


Granite Creek Assessment                               WPN                                                    Page 26
The classification codes and their general descriptions are found in Table 5. Biophysical
Environment classification was based on the IPNF Forest Type code, supplemented with the
LANFIRE Biophysical Setting code for those stands where no Forest Type values were available
(Table 6). BPE classifications were made on the basis of major species and plant associations
identified in the available datasets, which did not take into account fire exclusion effects or
species shifts (e.g. ponderosa pine/ Douglas-fir forest types converting to grand-fir/ western
redcedar types). When classifying BPE‟s at landscape scales, it becomes problematic to
determine the likely compositional range of vegetation based upon the current conditions – field
crews typically identify species and plant associations on the basis of the regenerating layers and
fire exclusion effects often promote the establishment of fire intolerant species (i.e. late seral
species) in the understory strata. As such, the BPE classifications are probably biased toward the
cooler-moist ecotypes and warm-dry/ warm-moist types are probably underrepresented in the
classification scheme.

Table 5. Descriptions of vegetation classification codes used in the DHSVM analysis.

     Biophysical           Major Structure Type     North Idaho Structural   DHSVM Classification
                                                                  8
  Environment (BPE)                                         Stage                 Code
        Cool-Dry               Large Trees                   6-7                    CDLT
        Cool-Dry              Medium Trees                   4-5                    CDMT
        Cool-Dry              Sapling/ Pole                  1-3                    CDSP
       Cool-Moist              Large Trees                   6-7                    CMLT
       Cool-Moist             Medium Trees                   4-5                    CMMT
       Cool-Moist             Sapling/ Pole                  1-3                    CMSP
       Warm-Dry                Large Trees                   6-7                    WDLT
       Warm-Dry               Medium Trees                   4-5                    WDMT
       Warm-Dry               Sapling/ Pole                  1-3                    WDSP
      Warm-Moist               Large Trees                   6-7                    WMLT
      Warm-Moist              Medium Trees                   4-5                    WMMT
      Warm-Moist              Sapling/ Pole                  1-3                    WMSP
       Unknown                  Seedlings                    1-3                    SEED
       Unknown                 Non-Forest                     0                     NONF




8 See Figure 22


Granite Creek Assessment                          WPN                                       Page 27
Table 6. Crosswalk between IPNF Forest type, LANFIRE Biophysical Setting, and
Biophysical Environment classification.

  IPNF Forest
     Type                  BPE              LANFIRE Biophysical Setting                 BPE
                                  Rocky Mountain Subalpine Dry-Mesic Spruce-Fir
Larch                 Cool-Dry                                                        Cool-Dry
                                  Forest and Woodland
                                  Northern Rocky Mountain Mesic Montane Mixed
Lodgepole pine        Cool-Dry                                                       Cool-Moist
                                  Conifer Forest
Spruce,
                      Cool-Dry
Subalpine fir                     Rocky Mountain Subalpine Mesic-Wet Spruce-Fir
                                                                                     Cool-Moist
Western                           Forest and Woodland
                     Cool-Moist
Redcedar
Western Hemlock      Cool-Moist
                                  Northern Rocky Mountain Dry-Mesic Montane Mixed
Western White                                                                        Warm-Dry
                     Cool-Moist   Conifer Forest - Ponderosa Pine-Douglas-fir
Pine
Aspen                Warm-Dry     Northern Rocky Mountain Dry-Mesic Montane Mixed
                                                                                     Warm-Moist
Douglas Fir          Warm-Dry     Conifer Forest - Grand Fir
Ponderosa Pine       Warm-Dry     Rocky Mountain Montane Riparian Systems            Cool-Moist
                                  Northern Rocky Mountain Lower Montane-Foothill-
Grand Fir           Warm-Moist                                                       Non forest
                                  Valley Grassland
Non-forest           Non forest   Open Water                                         Non forest



Major structure type (Table 5) was estimated based on Age Class and Size Class values available
in the IPNF data set (Table 7), with missing values estimated from the LANFIRE Existing
Vegetation Height data set (Table 8). A major constraint involves forest structural attributes,
where vertical stand metrics were either not available or incomplete. The constraint with this
data gap involves the determination of single- or multi-strata forest types; this is especially
important for determining the current shift from reference conditions for the warm-dry forest
types following the Historic Range of Variability (i.e. “Stage 6” and “Stage 7” in Figure 22
cannot be differentiated). Hence, the most reliable source of stand structural information
involved only coarse codes (sawtimber, immature, pole, etc), and stand age was needed to
estimate and refine the structural stage condition. A common example in the original data
involved an “immature stand” classification that was 80-100 years old. The DHSVM
classifications presented here incorporated this stand age to classify the stand as having “large
trees” (e.g. “WMLT in Table 5). Despite the data gaps, the classification scheme provides a
reasonable estimate for a landscape-scale Current Condition analysis for use in the DHSVM
analysis. A summary of the Current Conditions by acres and land area is presented in Table 9
and Figure 23.




Granite Creek Assessment                       WPN                                        Page 28
Table 7. Crosswalk between IPNF Size and Age Classes, and Size Description.
                  Size Class                 Age class (years)    Major Structure
                                               Uneven-age            Seedlings
Nonstocked (NONS)
                                                 Unknown             Seedlings
                                                   0-20              Seedlings
Seedlings (SEED)                                  21-50              Seedlings
                                                 Unknown             Seedlings
                                                   0-20            Sapling/Pole
Overtopped with Brush (OSAP)
                                                    80+            Sapling/Pole
                                                   0-20            Sapling/Pole
                                                  21-50            Sapling/Pole
Saplings (SAPL)
                                                  51-80            Medium Trees
                                                 Unknown           Sapling/Pole
                                                  21-50            Sapling/Pole
                                                  51-80            Medium Trees
Immature Pole (IPOL)
                                                    80+             Large Trees
                                                 Unknown           Medium Trees
                                                  21-50            Sapling/Pole
                                                  51-80            Medium Trees
Immature (IMSA)                                     80+             Large Trees
                                               Uneven-age           Large Trees
                                                 Unknown           Medium Trees
                                                  21-50            Sapling/Pole
                                                  51-80            Sapling/Pole
Poletimber (POLE)
                                                    80+            Medium Trees
                                                 Unknown           Sapling/Pole
                                                  21-50            Medium Trees
                                                  51-80             Large Trees
Sawtimber (SAWT)
                                                    80+             Large Trees
                                                 Unknown            Large Trees
                                                    80+             Large Trees
Mature High Risk (MHRS)
                                                 Unknown            Large Trees
                                                    80+             Large Trees
Mature Low Risk (MLRS)
                                                 Unknown            Large Trees
Multisized - 2 age classes (MULS)                   80+             Large Trees
                                                  21-50            Medium Trees
                                                  51-80             Large Trees
Multisized - 3 or more age classes. (MULT)          80+             Large Trees
                                               Uneven-age           Large Trees
                                                 Unknown            Large Trees
Bare Soil (BARE)
Sod Occupying Site (SOD)
High Brush Occupying Site (HGHB)                     -              Non-Forest
Low Brush Occupying Site (LOWB)
Not Applicable (nonforest) (NA)




Granite Creek Assessment                      WPN                                   Page 29
Table 8. Crosswalk between LANFIRE Existing Vegetation Height classes and Majore
Structure type.
                   Class Names                       Major Structure
            Forest Height 0 to 5 meters                Sapling/Pole
           Forest Height 10 to 25 meters              Medium Trees
           Forest Height 25 to 50 meters               Large Trees
            Herb Height 0 to 0.5 meters                 Non-Forest
           Herb Height 0.5 to 1.0 meters                Non-Forest
                    Pasture/Hay                         Non-Forest




Granite Creek Assessment                   WPN                            Page 30
Figure 22. Stages of Structural Development.




Granite Creek Assessment                  WPN   Page 31
Table 9. Current Conditions of Biophysical Environments by Structural Stage.
                                                               9
                                             Structural Stage by BPE
                     Non-Forest       1-3               4-5             6-7
       BPE          acres    %    acres     %     acres       %     acres     %     Total acres
    Cool-Dry                      2,344     21%   3,908       35%   4,963     44%     11,214
    Cool-Moist                    5,232     13%   14,902      38%   19,493    49%     39,626
    Warm-Dry                      1,570     21%   2,118       29%   3,666     50%     7,355
   Warm-Moist                     521       25%       539     26%   1,018     49%     2,077
    Unknown         1,309   47%   1,473     53%                                       2,782




Figure 23. Current and potential vegetation conditions.


9 See Figure 22


Granite Creek Assessment                        WPN                                        Page 32
In the assessment of the historic conditions of the landscape, it is important to consider the native
disturbance patterns that helped to define those conditions, and to extrapolate a landscape-level
view of how those disturbances affected the distribution of vegetation structures. The intent of
the Historic Range of Variability (HRV) analysis is to describe the landscape-level patterns of
forest structure and composition, assuming the natural (or historical) patterns of disturbance. An
HRV analysis is not intended to be a stand-level analysis; rather it is used as a guide to evaluate
how a landscape is functioning with respect to forest structure and composition, given a suite of
disturbance patterns. For ease of interpretation, seven structural stages are identified that
describe the age structure of a given stand (Figure 22).

With the assumption that the BPE‟s defined in the Current Condition were accurately described,
it is possible to compare and contrast the distribution of forest structures at landscape scales.
Distribution percentiles of structural stages are presented in Table 10; HRV distribution
percentiles represent likely median values for the analysis area, though these percentages are
typically expressed as a range10.

Table 10. Reference Conditions of Biophysical Environments by Structural Stage (non-
forest and unknown BPE’s are excluded).
                                                                11
                                             Structural Stage        by BPE
                                 1-3                    4-5                       6-7
         BPE               acres       %         acres           %            acres     %     Total acres
      Cool-Dry             1,629       15%       3,413          30%           6,172     55%     11,214
     Cool-Moist            7,926       20%       15,849         40%       15,851        40%     39,626
      Warm-Dry             735         10%       1,471          20%           5,149     70%     7,355
     Warm-Moist            314         15%        520           25%           1,243     60%     2,077



Though the HRV analysis is a landscape-scale analysis, the DHSVM inputs require stand-
specific assignments of reference vegetation classifications. The primary assumption for this
analysis was that historic disturbance followed stochastic events on the landscape; as such, stand-
level Structural Stage classifications were altered from the Current Condition following a
random distribution. For example, the current condition of the warm-moist young forest (stage
1-3, Table 9) included 521 acres of land area (25% of the warm-moist BPE). On average, the
landscape-scale HRV would suggest only 314 acres, or 15% of the warm-moist BPE (Table 10).
Hence, a random selection of these acres were reassigned to other structural stages to alter the
distribution of the current condition to better match the distribution described in the HRV.
Following this method, the entire analysis area was assigned a DHSVM classification code
(Table 5, Figure 23) to represent a “snapshot” of a likely historic vegetative condition.




10 HRV target values are composites from the North Idaho USDA FS guidance and Smith & Fisher (1997).
11 See Figure 22.


Granite Creek Assessment                                WPN                                              Page 33
Each unique vegetation classification code (Table 5) was assigned a suite of parameters that the
DHSVM model utilizes to predict the influence of vegetation on stream flows. Among all
classifications, the primary differentiators included:

       Fractional Coverage (% cover in overstory trees)
       Canopy and understory heights
       Overstory monthly Leaf Area Index (LAI)

Other values were determined through published literature (Bowling and Lettenmaier, 1997;
Storck and Lettenmaier, 2000), with minor adjustments explained in this section. Appendix A:
DHSVM model configuration file contains all parameter values for all vegetation types.

Fractional Coverage is defined as the percent canopy cover of overstory trees. A review of
available data12 indicated that canopy cover values were available for only 3% of the stands in
Granite Creek. Consequently, canopy cover values for similar stands located in the North Fork
Coeur d‟Alene watershed were used in this analysis (Salminen and Heider, 2007; Table 11).
Similarly, tree height data was very limited in the “Components” dataset for Granite Creek;
consequently, values were used from the North Fork Coeur d‟Alene dataset (Table 11).

Table 11. Fractional coverage and canopy height assumptions for each DHSVM
classification.

                              Fractional Coverage
 DHSVM Classification              (% cover)                  Canopy Ht (m)   Canopy Ht (ft)
           CDLT                         0.9                         23.0          75.3
          CDMT                          0.8                         17.5          57.3
          CDSP                          0.6                         11.9          39.2
          CMLT                          0.9                         26.7          87.5
          CMMT                          0.8                         19.4          63.7
          CMSP                          0.66                        10.4          34.3
          WDLT                          0.9                         22.9          75.1
          WDMT                          0.75                        21.5          70.5
          WDSP                          0.5                          9.0          29.4
          WMLT                          0.9                         25.7          84.5
          WMMT                          0.8                         21.1          69.2
          WMSP                          0.66                         8.5          27.8
          SEED                          0.3                          2.6           8.5
          NONF                          0.05                        12.5          41.1




12 http://www.fs.fed.us/ipnf/eco/yourforest/gis/veg/vegetation_readme.html


Granite Creek Assessment                              WPN                                 Page 34
Leaf Area Index (LAI) is the ratio of leaf area (needle surface) per unit of ground area, and is
useful to describe the primary productivity of a system, or the photosynthetic activity throughout
the year. Because the DHSVM model utilizes time-series climate data, it is important to have an
understanding of plant activity throughout the year to calculate the amount of moisture lost from
evapotranspiration. Direct measures of LAI are complex and unavailable for this analysis;
values were obtained from previous DHSVM studies (Bowling and Lettenmaier, 1997) and from
the Land Data Assimilation System (LDAS)13 to estimate appropriate ranges of monthly LAI for
each of the vegetation types.

Canopy height and LAI measures reported elsewhere (Bowling and Lettenmaier, 1997) indicated
there was a significant relationship14, allowing us to use canopy height as a predictor for LAI.
Though not a complete sample, the relationship provides a basis to “fine tune” LAI estimates
with canopy height data specific to the IPNF.

Because LAI varies throughout the year, LDAS data were used to evaluate the seasonal flux for
forest and non-forest ecosystems. Three general forest ecosystem types were selected to
“bracket” the forested types represented in the analysis area (evergreen needle leaf, mixed cover,
and deciduous needle leaf). The average of the monthly values provided a profile by which a
monthly correction factor could be derived. This monthly correction factor was applied to the
modeled mean LAI generated from canopy height information from the North Fork Coeur
d‟Alene to provide an estimate of seasonal flux in LAI among the DHSVM vegetation
classifications. Understory LAI was applied in a similar manner, grouping all DHSVM classes
as either forested or non-forest/seedling components (closed shrub model from LDAS). The set
of monthly LAI parameters for overstory and understory components, including mean annual
LAI and the monthly correction factors are presented in Table 12.




13 http://ldas.gsfc.nasa.gov/LDAS8th/MAPPED.VEG/web.veg.monthly.table.html
14 Leaf Area Index = 0.4811*Height + 0.419; r2 = 0.89, P<0.01


Granite Creek Assessment                         WPN                                       Page 35
Table 12. Annual mean and monthly LAI overstory and values for each DHSVM classification.

                 Mean
  DHSVM          Annual
   Class          LAI      Jan     Feb     Mar      Apr     May          Jun    Jul    Aug       Sep     Oct     Nov     Dec
 Correction
  Factor                   84%     87%     94%      100%    109%     118%      117%    113%      108%    98%     88%     84%
                                                            Overstory
   CDLT           11.48    9.58    10.01   10.83    11.48   12.48    13.58     13.41   12.93     12.35   11.22   10.11   9.58
   CDMT           8.84     7.37    7.70    8.33     8.84    9.60     10.45     10.32   9.95      9.50    8.63    7.78    7.37
   CDSP           6.14     5.15    5.38    5.82     6.17    6.71        7.30   7.21    6.95      6.64    6.03    5.43    5.15
   CMLT           13.46    11.08   11.57   12.52    13.28   14.43    15.70     15.51   14.95     14.28   12.97   11.69   11.08
   CMMT           9.75     8.15    8.52    9.21     9.77    10.62    11.55     11.41   11.00     10.51   9.54    8.60    8.15
   CMSP           5.42     4.55    4.75    5.14     5.45    5.93        6.45   6.37    6.14      5.86    5.33    4.80    4.55
   NONF           6.43     0.87    0.91    0.99     1.05    1.14        1.24   1.22    1.18      1.13    1.02    0.92    0.87
   SEED           1.67     1.40    1.46    1.58     1.67    1.82        1.98   1.96    1.88      1.80    1.64    1.47    1.40
   WDLT           11.44    9.55    9.98    10.80    11.45   12.44    13.54     13.37   12.89     12.31   11.18   10.08   9.55
   WDMT           10.76    8.99    9.39    10.16    10.78   11.71    12.74     12.58   12.13     11.59   10.53   9.49    8.99
   WDSP           4.75     3.95    4.13    4.47     4.74    5.15        5.60   5.53    5.33      5.09    4.63    4.17    3.95
   WMLT           12.78    10.70   11.18   12.10    12.82   13.94    15.17     14.98   14.44     13.79   12.53   11.29   10.70
   WMMT           10.57    8.83    9.22    9.98     10.58   11.50    12.51     12.36   11.91     11.38   10.34   9.32    8.83
   WMSP           4.51     3.76    3.92    4.25     4.50    4.89        5.32   5.26    5.07      4.84    4.40    3.96    3.76
                                                            Understory
 Seed/ Non-       0.58     0.63    0.63    0.63     0.92    1.77        2.55   2.55    1.73      0.97    0.73    0.63    0.58
   forest
  Forested        0.36     0.46    0.60    0.60     0.60    0.60        1.13   1.08    0.64      0.36    0.36    0.36    0.36




Granite Creek Assessment                      WPN                                      Page 36
4.2.1.3           Soil

The DHSVM model requires spatially distributed information on both soil textural
characteristics and soil depth. Model parameters associated with textural class determine the rate
at which moisture moves through the soil profile under saturated and non-saturated conditions,
while soil depth controls the volume of soil moisture, as well as the interception of soil moisture
by stream and road cuts. Soil characteristics were held constant among all model runs.

The Idaho Panhandle National Forest (IPNF) “Landtypes” GIS coverage 15 (Figure 12) was used
to map soil depths and textural classes (Figure 24). Landtype descriptions16 that accompany the
GIS coverage provided details on textural type and typical profile depth. Typically several soil
types may be contained within a single Landtype. Conditions were averaged among soil types
and among soil layers within a single soil type. Multiple land types with similar textural types
were combined. Four distinct textural types were defined for the assessment area (Figure 24).




Figure 24. Soil depth and textural class.


15 http://www.fs.fed.us/ipnf/eco/yourforest/gis/index.html#soils
16 http://www.fs.fed.us/ipnf/eco/yourforest/gis/soils/landtype_descriptions.zip


Granite Creek Assessment                                WPN                                 Page 37
Sixteen soil parameters need to be defined for each of the seven soil textural types:

   Lateral hydraulic conductivity                    Exponential decrease in lateral hydraulic
                                                       conductivity
   Maximum infiltration rate                         Capillary drive
   Surface albedo                                    Number of soil layers
   Porosity                                          Pore size distribution
   Bubbling pressure                                 Field capacity
   Wilting point                                     Bulk density
   Vertical conductivity                             Thermal conductivity
   Thermal capacity                                  Mannings n


Values for each soil type were estimated based on published literature (Bowling and
Lettenmaier, 1997; LaMarche and Lettenmaier, 1998; Freeze and Cherry, 1979; Dunne and
Leopold, 1978). Appendix A: DHSVM model configuration file lists the values used for each of
the soil types.

4.2.1.4         Streams

The DHSVM model requires spatially distributed data on the location and characteristics of
stream channel types. Stream locations and characteristics influence where and under what
conditions subsurface flow becomes surface flow, and the rate at which streamflow is routed to
downstream locations. The model uses a simple linear routing algorithm to move water through
the channel system.

It is critical to the models operation that the vector stream coverage matches exactly the
topographical low points (i.e., the valleys) in the DEM. Consequently, it was necessary to
construct a vector stream coverage from the DEM. This was done by varying the number of
upstream pixels needed for channel initiation so that the created stream vectors reasonably
approximated the stream coverage for the area. The resultant coverage closely approximated
mapped stream locations.

Parameters that need to be estimated for each stream segment include the following:

   Active channel width                              Active channel depth
   Channel slope                                     Channel roughness

Channel width and depth were estimated based on drainage area (Figure 25). Insufficient local
data were available to develop watershed-specific relationships. Consequently, regression
equations for width and depth were based on stream survey data from the North Fork Coeur




Granite Creek Assessment                       WPN                                       Page 38
d‟Alene subbasin17. Channel width showed a strong correlation with drainage area. Although
channel depth shows a much poorer relationship it is still adequate for the purposes of the model.
Channel slope was calculated within GIS. A roughness value of 0.065 was used for all stream
segments.


                        1.6                                                               25
                                                                                                             0.4888
                                                                                                   y = 0.1475x
                        1.4                                                                            2
                                                                                                      r = 0.8182
                                  y = 3E-05x + 0.2895                                     20
                        1.2            r2 = 0.2648
   Bankfull Depth (m)




                                                                     Bankfull Width (m)
                         1                                                                15
                        0.8

                        0.6                                                               10

                        0.4
                                                                                           5
                        0.2

                         0                                                                 0
                              0   5000    10000    15000   20000                               0      5000    10000   15000   20000
                                    Drainage area (ha)                                                  Drainage area (ha)



Figure 25. Relationship between bankfull width (right) and depth (left) and contributing
area for streams in the North Fork Coeur d’Alene subbasin.



4.2.1.5                           Roads and Road Drainage

Within DHSVM roads have two primary effects on modeled streamflow. First, road surfaces are
treated as near-impervious surfaces and precipitation falling on these surfaces is treated as
surface flow (although it may reenter the soil profile once it leaves the road surface). Secondly,
road drainage ditches may convey all or a portion of the ditch flow to adjacent pixels, or directly
to streams, altering the timing of streamflow at downstream locations. As with streams, road
input has a spatial component that varies with road conditions.

The USFS identifies 128 miles of system roads18 in the Granite Creek watershed, 57 miles of
which were field-surveyed (IPNF, 2008)19. An additional 3 miles of non-system roads were also
included in the surveys. A total of 169 culverts were surveyed; 51 in 2006, and the remainder
(118) in 2007 (Figure 44). Surveys focused on higher-use roads. 10.3 miles of surveyed road


17 http://www.deq.idaho.gov/about/regions/north_fork_cda_river_wag/wpn_channel_analysis_report.pdf
18 Roads that are maintained by the US Forest Service
19 See Figure 43 and Figure 44 in Section 6.3.1


Granite Creek Assessment                                           WPN                                                           Page 39
(14% of the total) were identified as having a surface connection to the stream network20 (Figure
26, Table 13). The longest length (4.5 miles) and largest percentage (33%) of surveyed roads
that are hydrologically connected are located in the North Fork subwatershed.




Figure 26. Surveyed roads having road ditches that are hydrologically connected to the
stream system.




20 Refers to road ditches where the surface runoff in the ditch shows evidence of flowing directly into a stream
channel (as opposed to flow that diffuses over the forest floor)


Granite Creek Assessment                                WPN                                                  Page 40
Table 13. Summary of surveyed roads that are hydrologically connected. Percentages are
percent of surveyed roads.

                                          Main Granite     North Fork       South Fork         Entire
                                             Creek        Granite Creek    Granite Creek     watershed
Hydrologically connected road segments      3.5 (7%)       4.5 (33%)            2.3 (20%)    10.3 (14%)
All surveyed roads                           48.4             13.9                11.2          73.5



Model parameters needed for each road segment included:

   Ditch width                     Ditch depth                          Ditch slope
   Ditch roughness                 Infiltration rate                    Outsloped/crowned/insloped
    (Manning‟s n)
   Road width                      Cutbank height                       Ditch connectivity

Infiltration rate was assumed to be 0.00003 m/s in all situations. Ditch roughness was assumed
constant with a value of 0.100. Slope was calculated from a 10-meter DEM. The following
assumptions were made for all remaining parameters:

   Ditch width and depth were assumed to be 1.0 and 0.5 feet respectively

   All ditches were assumed to be outsloped

   Road width was either available from the IPNF roads GIS layer, or assumed to be 12‟

   Cutbank height was estimated from the local hillslope (as calculated using the DEM), by
    assuming that road cut and fill were balanced in the cross-section

   Flow in all non-surveyed ditches was assumed to return to the pixel from which it originated

4.2.1.6         Meteorological Data

The DHSVM model is driven by meteorological data run at a sub-daily time step. Inputs to each
grid cell, either as precipitation or as inflow from adjacent cells, is processed for each time step,
and then is passed on to down-gradient cells. As such, model runtime is directly proportional to
the time step chosen. For this application we chose a 3-hour time step. The following input data
is needed for each time step:

       Air temperature                    Wind speed                           Relative humidity
       Shortwave radiation                Longwave radiation




Granite Creek Assessment                            WPN                                          Page 41
Two climate stations are located in the vicinity of the project area; the Bunchgrass Meadow
SNOTEL site21 located at 5,000 feet elevation in the headwaters of the South Fork subwatershed,
and the Priest Lake RAWS station22 located approximately 5 miles south of the watershed near
the Priest Lake Ranger District at 2,600 feet elevation. Unfortunately, climate records for both
stations have large periods of missing data and could not be used. Fortunately, modeled daily
data was available for a location within the Main Granite creek subwatershed23.

Air temperatures for each time step were estimated from daily min/max temperatures using a sin
curve relationship following the techniques of Waichler and Wigmosta (2003). Mean daily wind
speed values were used for all time steps on a given day. Relative humidity was estimated for
each time step using daily min/max RH values, and estimated air temperature values at the given
time step. Incoming shortwave radiation was estimated for each time step from daily min/max
air temperatures and daily precipitation using the SolarCalc hourly solar radiation estimation
program24 (Spokas and Forcella, 2006). Incoming longwave radiation values were estimated
from shortwave values using relationships from Bowling and Lettenmaier (1997). A constant
lapse rates were used for temperature (-0.0089 deg. C/meter). Precipitation was scaled to each
grid cell using gridded monthly precipitation data available from the PRISM climate mapping
system25.

4.2.2 Model Calibration

We performed limited model calibration by adjusting model parameters (Appendix A) to achieve
a better fit between observed and modeled discharge at the Granite Creek gage (Table 4)
location. Saturated hydraulic conductivity is the single parameter that the model is most
responsive to. Initial hydraulic conductivity values were increased to provide the best fit
between observed and modeled data. Soil depth was also adjusted to a uniform 2-meter depth
throughout the watershed. The timing of modeled peak flows corresponded well with observed
values; however, the magnitude of the modeled response was typically larger than what was
observed (Figure 27).    A plot of observed vs. modeled values (Figure 28) indicate that the
model does a reasonably good job at capturing the magnitude of daily flow events. Furthermore,
differences between observed and modeled flows at the gage are not necessarily a major concern,
as the subsequent analysis considers differences among modeled results; not between modeled
and observed results.




21 http://www.wcc.nrcs.usda.gov/snotel/snotel.pl?sitenum=376&state=wa
22 http://www.wrcc.dri.edu/cgi-bin/rawMAIN.pl?idIPRL
23 T he interagency Fire Program Analysis (FPA) program provides historical daily weather data for grid cell
locations at a 32-km spatial resolution across North America. This dataset consists entirely of data from the North
American Regional Reanalysis (NARR) dataset for North America The NARR is a modeled dataset from 1979-
present that incorporates data from rawinsondes, dropsondes, pibals, aircraft, selected surface stations, and
geostationary satellites. It also incorporates high-resolution data from a variety of other sources such as the NCEP /
Climate Prediction Center (CPC), Canadian, and Mexican precipitation network. More information is available at
http://www.wrcc.dri.edu/fpa/README_GRIDDED.pdf
24 http://www.ars.usda.gov/services/software/download.htm?softwareid=62
25 http://prism.oregonstate.edu/index.phtml


Granite Creek Assessment                                 WPN                                                   Page 42
Figure 27. Observed and modeled streamflow at the Granite Creek gage.




Figure 28. Observed vs. modeled mean daily discharge at Granite Creek gage.



Granite Creek Assessment                 WPN                                  Page 43
4.3    RESULTS
The DHSVM model was run using data for the period 9/1/1993 to 9/30/200626. Results were
evaluated by comparing annual peak flow events at selected locations over the modeling period
(WY 1994 – 2006; n = 13) across the range of modeled scenarios. The current vegetation model
that did NOT include roads was used as the baseline model against which all other model
iterations were compared. The DHSVM model produces continuous hydrographs at specified
locations for the entire modeling period. Only two metrics are discussed below; changes in the
magnitude of the annual peak flow event, and changes in summertime low flows.

4.3.1 Annual peak flow magnitudes

The dates for the thirteen annual peak flow events varied greatly among locations within a given
modeling scenario, and at the same locations across the three modeling scenarios. Microclimatic
conditions within the individual catchments are the most likely reason for this wide range in
dates of the annual events. Microclimatic conditions vary greatly for example from small north-
facing catchments (e.g., Athol Creek) to relatively large catchments with a lot of southern
exposure (e.g., The North Fork Granite). Date of the annual peak flow also varied between
modeling scenarios at a given site; with the largest variation seen between vegetation scenarios
(median = 19 days), and less so between the unroaded and roaded scenarios (median = 5 days).
Summaries of results for all analysis locations are provided in Table 14.




26
  The climate record used was applied to all modeling scenarios. Given that our question is how will the
watersheds respond to the same climatic conditions given varying vegetation, roads, and soil conditions. It does not
matter from what time period the climatic data is from, as long as it is representative of conditions experienced in
the watershed


Granite Creek Assessment                               WPN                                                   Page 44
Table 14. Summary of DHSVM results. Values are median, minimum and maximum for
thirteen annual peak flow events (WY 1994 – 2006) at the mouths of significant tributaries
and the watershed outlet.

                           % ∆ in annual peak flow from current % ∆ in annual peak flow from current
                             vegetation, no roads scenario to     vegetation, no roads scenario to
                              potential vegetation no roads        current vegetation with roads
                                         scenario                             scenario
      Catchment              Median        Min          Max       Median        Min         Max
          Sema                -10%        -14%          -3%         0%          -2%         2%
          Cache                -3%        -12%          9%          0%          -1%         2%
      South Fork               -1%         -6%          1%          0%          -1%         1%
          Willow               -5%        -17%          9%          1%          -4%         2%
          Tillicum             4%          -4%          14%         0%          0%          0%
      North Fork               -1%         -3%          0%          4%          3%          5%
          Fedar                2%          -6%          17%        -1%          -6%         6%
       Blacktail               -4%        -14%          6%         -2%          -6%         2%
           Athol               7%         -11%          35%         3%         -13%         14%
          Packer               -1%         -8%          3%          1%          -3%         5%
           Zero                1%          -5%          7%          0%          -2%         3%
   Watershed outlet            0%          -1%          2%          2%          1%          4%



4.3.1.1              Vegetation effects

As discussed in section 4.2.1.2 above, at the watershed scale current vegetation conditions are
similar to potential conditions, given the historic range of variability. For this modeling exercise
the current vegetation was randomly altered at the watershed scale to represent one possible
potential vegetation scenario. Consequently, the results at the watershed outlet (Table 14) show
little change in magnitude for the 13 annual peak flow events (range from –1% to 2% change)
and show no change in the median value for all 13 events.

Within the individual sub-drainages the modeled peak flow magnitudes range from as much as a
35% increase (Athol Creek outlet; Table 14) to a 17% decrease (Willow Creek outlet) for
individual events; with median values for all storms ranging from a 7% increase (Athol Creek) to
a 10% decrease (Sema Creek). These results however, should be interpreted with caution, as the
modeling only represents one possible scenario. Running several additional model runs using
alternative distributions of potential vegetation may be one way of defining the natural range of
variability in peak flow values for the various catchments.




Granite Creek Assessment                          WPN                                         Page 45
4.3.1.2         Road effects

The magnitude of modeled road drainage effects correspond to those sub catchments with the
highest densities of connected road ditches. Modeled peak flow magnitudes range from as much
as a 14% increase (Athol Creek outlet; Table 14) to a 13% decrease (also Athol Creek outlet) for
individual events; with median values for all storms ranging from a 4% increase (North Fork
outlet) to a 2% decrease (Blacktail Creek outlet).

4.3.2 Low flows

Mean low flow values for the month of September are given in Table 15. As with peak flows the
modeled flows for the lowest-flow month of the year appear to be little affected by current
vegetation conditions or roads. Minor increases are probably due to lower modeled ET losses
from existing vegetation as compared to the potential future condition, and the lack of ET losses
in roaded areas. Some of the smaller drainages show large percent increases (e.g., Athol Creek);
however, the magnitude of change is generally small.

Table 15. Modeled mean daily flow for the month of September. Values are modeled mean
September flows for WY 1994 – 2006 at the mouths of significant tributaries and the
watershed outlet.

                                                                  % ∆ from current % ∆ from current
                   Mean September Mean September Mean September vegetation, no        vegetation, no
                      daily flow:    daily flow:     daily flow:  roads scenario to roads scenario to
                       Current       Potential        Current         potential          current
                    vegetation no  vegetation no  vegetation with   vegetation no    vegetation with
  Catchment          roads (cfs)    roads (cfs)     roads (cfs)    roads scenario roads scenario
     Sema                  4.6            4.1             4.7             -13%             2%
     Cache                 1.5            1.4             1.5             -8%              0%
   South Fork            17.5            17.3            21.1             -1%             20%
     Willow                8.0            8.2             8.0             3%               0%
    Tillicum               6.7            7.2             6.7             8%               0%
   North Fork            37.9            39.2            38.6             3%               2%
     Fedar                 0.7            0.7             0.7             0%               0%
    Blacktail              2.1            2.1             2.1             -2%              0%
     Athol                 0.8            1.0             3.1             29%             307%
     Packer                1.9            1.9             1.9             -1%              0%
      Zero                 3.3            3.5             3.3             6%               1%
Watershed outlet         69.2            71.5            75.9             3%              10%




Granite Creek Assessment                         WPN                                            Page 46
4.4       INFORMATION GAPS AND MONITORING NEEDS
The assessment presented here was based solely on hydrologic modeling using the Distributed
Hydrology-Soils-Vegetation Model (DHSVM). To accurately model a given watershed the
model requires representations of spatial data (vegetation conditions, soil characteristics, stream
and road characteristics), watershed-specific meteorological records, and continuous flow
records from one or more locations within the watershed. Soil and vegetation data were
probably accurate enough for the purposes of the model. The lack of an overlapping streamflow
and climate record limited our ability to accurately calibrate the model, and probably represent
the biggest data gap affecting the assessment.

The primary monitoring need is to establish continuous flow stations at the outlet of the
watershed and some of the principal tributaries. These data will help calibrate future models,
and will help to verify modeled management-related flow impacts.

4.5       RECOMMENDATIONS
Based on the analysis presented here it appears that the hydrologic impacts of recent forest
harvest is negligible, and probably within the natural range of variability for the area. Road
drainage impacts are also relatively slight. As noted in the Sediment section of this report this is
likely due in large part to an aggressive and effective program of road decommissioning and
maintenance of existing roads. Road densities are low, and, with a few notable exceptions, roads
are located away from streams and outside of floodplain and wetland areas. Recommended
actions include:

         Decouple road drainage and stream networks in the lower North Fork subwatershed, and
          within the Athol Creek drainage. Modeled impacts from hydrologically connected road
          segments were greatest in these two areas. See GIS data (e.g., Figure 26) for specific
          locations of hydrologically connected road segments.

         Continuation of road decommissioning projects throughout the watershed. Road impacts
          on hydrology (and sedimentation; discussed in sediment section below) are relatively
          slight, due in large part to an aggressive past program of road decommissioning.
          However, there are still some problem areas (i.e., lower North Fork subwatershed, and
          within the Athol Creek drainage as discussed above) that need to be addressed.

         Keep future road densities low, and locate future roads away from sensitive floodplains,
          wetlands, and mass-wasting prone areas. Rosgen channel types (Figure 4) provide a
          good starting point to identify priority streams, with “E” channels having the most
          sensitive floodplain/wetland complexes that need protecting, followed by “F” and “C”
          channels. Rosgen “B” and “A” type channels are relatively confined and have limited
          floodplain area.




Granite Creek Assessment                        WPN                                          Page 47
                                 5.0      RIPARIAN CONDITIONS

5.1       CRITICAL QUESTIONS
The purpose of this section is to provide an assessment of the current conditions for the riparian
areas within the Granite Creek watershed, and to establish a range of potential future conditions
(~50-100 year timeframe) to determine the trajectories of primary riparian ecosystem services:
potential large wood debris (LWD) inputs and stream shade. The focus is on the forested riparian
components of the system, with specific emphasis on how the riparian zones influence fisheries
habitat.

Specifically the critical questions are:

         What are the current and potential future forest components that are available to
          potentially contribute LWD to the stream system?

         At the subwatershed and reach scales, what are the short- and long-term trends in tree
          growth and mortality?

         What reaches may be limiting in riparian shade, and how does that change through time?

         Are there adequate riparian forest resources in stream reaches currently identified as
          being limited in LWD?

5.2       METHODS
5.2.1 Riparian Zone Mapping

The riparian zone was defined as the land area 150 ft (45.7 m) immediately adjacent to the
mapped stream or stream bank (where visible) from aerial imagery27. This 150 ft buffer was
created on both sides of the stream and was further delineated using available aerial imagery in
the GIS. Delineations were made on the basis of homogeneous structure (i.e. density and
distribution of trees, shrubs, and outcrops) and compositional types (hardwood, conifer, changes
in species, etc.). These types were consolidated into functional groupings based on the imagery
to assess the potential LWD and shade factors along each stream reach.

5.2.2 Field Sampling

A total of 75 stands (~161 acres) within 24 functional groups were randomly selected for field
sampling using a modified Forest Service stand exam procedure28 (Table 16). These functional
groups represented the range of cover types that would likely contribute LWD to the stream
channel (or associated riparian zone); functional groups were a composite of observed forest

27 This buffer width was chosen to represent the area within the Forest Service Riparian Habitat Conservation Areas
(RHCA).
28 http://www.fs.fed.us/emc/nris/products/fsveg/index.shtml


Granite Creek Assessment                              WPN                                                  Page 48
cover types, with the goal to stratify the landscape for sampling forest structure. Non-forested or
young managed stands were not sampled. Approximately 3 plots were randomly established
within each stand, with care given to ensure plot locations did not allow for sampling of trees
outside of the stand, or within an adjacent plot. Variable plot radius was limited to trees 5
inches DBH (diameter at breast height, measured at 4.5 ft); fixed circular plots were established
to sample understory vegetation per common stand exam protocols. Growth reference trees were
identified and diameter growth rates were measured for approximately 9 trees per stand.

Table 16. Sampled stands within the watershed following a modified Common Stand
Exam procedure.

      Subwatershed             Number of Stands    Avg. Size (Acres)   Total Sampled Acres
    Main Granite Creek               40                   2.2                  86.2
North Fork Granite Creek             16                   2.1                  33.3
South Fork Granite Creek             19                   2.2                  41.2
           Total                     75                   2.2                 160.8



The Forest Service collected data in the field, compiled it in FSVEG and loaded to an Access
database.

5.2.3 Growth and Mortality Modeling

Measured stands were inputted into the Forest Vegetation Simulator (FVS) using the Kootenai
Variant (KT)29. This is a growth and mortality simulation model that allows for stand structure
and composition information to be “grown” at set intervals through time. This is a useful tool for
managing harvest activities on forested landscapes; in the case of predicting LWD and shade, the
model outputs provide tree mortality, height, diameter and crown cover measures into the future.

Stands were compiled and modeled for a 100-year growth scenario, in ten 10-year growth
increments beginning in growth year 2008 to the end growth year 2108. The growth scenario
relied on natural regeneration and no active management. Where available in the field dataset,
field-derived tree growth reference information was applied to obtain riparian- and site-specific
growth modeling. An “expansion set” was created using the weighted average modeled output
for each functional group. This dataset was applied to unmeasured stands of like functional
groups. Non-sampled groups were not included as part of this analysis.

Due to the considerably large size of this output dataset (>20 million lines of data), the tree data
were consolidated into respective species types (conifer/ hardwood) and into different size
classes. These size classes were chosen as size ranges that are common in forestry applications
as well as providing generalized size-height relationships of what types of forest components
would naturally recruit to the stream channel or riparian forest floor. Diameter classes and size
class boundaries are presented in Table 17.

29 http://www.fs.fed.us/fmsc/fvs/


Granite Creek Assessment                          WPN                                        Page 49
Table 17. Tree diameter size class groupings.
    DBH Class              DBH Size Class Boundaries
         1                          0 - 5 in.
         2                         5 - 12 in.
         3                         12 - 20 in.
         4                         20 - 30 in.
         5                         30- 50 in.
         6                          > 50 in.


Data including trees per acre (TPA), dead trees per acre (dead TPA), basal area, average height,
average crown width and percent were calculated for each size class within each stand, for ten
10-year periods.

For the purposes of this report, data are reported in units per acre, which is a conventional
method for forest management. Data were summarized as a weighted average by acre at the
subwatershed and stream reach scales. For direct wood debris and riparian tree availability for
active restoration, average units per acre were converted to trees per 100 m of stream length and
summarized at the reach level. The direct conversion from units per acre to units per 100 m of
stream was made using the following equation:

    Units per 100 m stream length = [reported units] per acre * 1.13 acres per 100 m

For example, a stream reach reporting 27 large trees per acre (TPA 20 inches DBH) would have
an average of 30.5 trees per 100 m of stream length.

5.2.4 LWD Recruitment Potential

Wood recruitment to the stream channel is a difficult process to model directly and has high
levels of uncertainty. Two main factors of the riparian zone are considered in this analysis: live
standing trees and tree mortality through time. Though several diameter classes can be
considered appropriate sizes for “functional” pieces of wood potentially entering the stream, this
analysis focused on conifers 20 inches DBH. Large trees were selected in this analysis for two
reasons: a large tree is more likely to remain in-place during high flow events, and this size class
represents generally higher “scores” in the measurement of “key pieces” of LWD in the stream
channel (as it relates to the R1 R4 Stream Survey Protocols). Focus on this size class provides a
conservative view of the number of pieces that could potentially be recruited; inclusion of small
size classes would only increase the potential LWD recruitment to the channel. Actual function
within the stream channel is dependent upon many factors other than size and length of the tree
bole.

The overall pool of live and dead trees was examined on the acre- and stream length basis.
Weighted averages by acre contribution were calculated and summarized on two spatial scales:
subwatershed and stream reach scales. LWD values (see fisheries section) calculated from
stream survey data (LWD pieces plus debris dams) were compared. For active restoration

Granite Creek Assessment                         WPN                                         Page 50
potential, the primary assumption was a single, live or dead conifer tree 20 inches DBH was
equivalent to a single count of a LWD piece or debris dam. With this assumption, direct
comparisons were made with the reach-level riparian structure and reach-level requirements to
meet the 75th percentile of reference reach conditions for LWD pieces and debris dams.

5.2.5 Stream Shade

Stream shading was estimated as the combined function of riparian vegetation and topography
(i.e. effective shade) for the current and potential future conditions (50- and 100- year
scenarios). Our approach in estimating future effective shade levels assumed no management
within riparian areas. Future riparian stand conditions (i.e., years 2058 and 2108) were modeled
using the Kootenai Variant (KT) of the Forest Vegetation Simulator (FVS)30 as described in
section 5.2.3 above.

Effective shade values were estimated using the HeatSource version 7.0 temperature model
(Boyd and Kasper 2003)31. Effective shade was estimated within a GIS using the ArcView 3.x
TTools extension. Input data is assembled within T-tools at sampling points located along a
given stream. Sample data was assembled at a 100-meter interval along several principal
streams in the Granite Creek watershed. Location elevation, channel gradients, and topographic
characteristics were calculated using the finest-resolution DEM available (i.e., 1/3 arc-second; ~7
meter resolution). Active channel widths were mapped within GIS from color orthophotos as
part of this assessment.

The TTools extension was also used to sample riparian stand conditions, and assemble these data
for evaluation within HeatSource. Several stand metrics are needed by HeatSource in calculating
effective shade, these include stand height, canopy density, and canopy overhang. Seventy-five
stand types were derived from field data (described in section 5.2.2) and used in this analysis.



5.3      RESULTS
5.3.1 Riparian Zone Distribution

The majority of the riparian zone follows the patterns of the watershed as a whole (Figure 17),
with approximately 90% of the land area in forested cover (Table 18). The non-forest types are
typically dense riparian hardwood meadows and areas outside the immediate riparian zone with
distribution of hardwood trees and shrubs. For purposes of this assessment and plan, the
designation of „forested‟ is biased toward species that could contribute to coniferous LWD
materials.




30
     http://www.fs.fed.us/fmsc/fvs/
31
     http://www.deq.state.or.us/wq/TMDLs/tools.htm


Granite Creek Assessment                             WPN                                    Page 51
Table 18. Riparian forest and non-forest functional groups derived from the 3,783 stands
within the Granite Creek watershed.
        Subwatershed                 Forested            Non-Forested       Total
            (SWS)               Acres     % of SWS      Acres   % of SWS    Acres
      Main Granite Creek        1,395       87%          207      13%       1,603
   North Fork Granite Creek     1,120       89%          141      11%       1,261
  South Fork Granite Creek      1,610       91%          156       9%       1,767
             Total              4,126       89%          505      11%       4,631



Data from sampled stands were expanded to representative functional groups on the landscape.
This provided a coverage of approximately 71% of the riparian zone area for the Main and NF
Granite Creek and 63% of the acres for South Fork Granite (Table 19). This was based on a
random distribution of the most common forested functional groups, with an underlying set
sampling effort by the Forest Service. At the landscape scale, a total of 2,893 stands were
represented (75% of the total landscape). It is important to recognize that not all of the landscape
was sampled, and that the reach-level averages presented in this report are dependent upon the
forested functional groups sampled. This provides a conservative estimate in available LWD, as
the sample focus was on forested environments.



Table 19. The distribution of representative acres sampled or expanded by like functional
groups for the watershed.

     Subwatershed          Represented Acres    Watershed Acres         % Sampled
   Main Granite Creek           1,188.8               1,674.1               71%
North Fork Granite Creek         933.7                1,324.3               71%
South Fork Granite Creek        1,160.2               1,851.8               63%
          Total                 3,282.7               4,850.1               68%



5.3.2 Major Vegetation Types

Stand composition is moderate to rich in diversity, with successional western hemlock (TSHE)
and western redcedar (THPL) as co-dominants in the overstory, following earlier seral species of
grand fir (ABGR), Douglas-fir, ponderosa pine, western larch, and western white pine. Cooler
sites are co-dominant Engelmann spruce, subalpine fir, lodgepole pine. Understory compositions
are somewhat simplistic, and include western oakfern (Gymnocarpium dryopteris) and ladyfern
(Athyrium filix-femina), forbs such as queencup beadlilly (Clintonia uniflora), and moist
perennial shrubs (e.g. Vaccinium membranaceum). Barren understories populated by thick wood
and duff are also common in closed canopy environments. The long-term structure and
composition of this forest type is a multi-strata forest with large, old-growth dominance of


Granite Creek Assessment                        WPN                                          Page 52
shade-tolerant species. Cold-dry sites (e.g. subalpine fir) are somewhat open, with graminoid
sedge understories. Though not measured for distribution, the invasive grass reed canarygrass
(Phalaris arundinacea) was observed in the system, particularly in the Main Granite Creek
subwatershed.

Multiple plot-level plant associations were recorded for individual stands; the data displayed in
Table 20 and Table 21 represent the observed sampled plant associations and the relative
distribution of the plant communities on the landscape. The dominant association in all cases is
a cool-wet plant association type with moist fern dominated understories.

Table 20. A summary of plant associations and sampled acres

Plant Association Code             Plant Association Description               Sampled Acres
       ABGR/CLUN                        Grand fir/ bride's bonnet                     1.5
        THPL/ATFI                     Western redcedar/ ladyfern                      46.9
       THPL/CLUN                   Western redcedar/ bride's bonnet                   16.3
       THPL/OPHO                     Western redcedar/ devils club                    8.9
       TSHE/CLUN                   Western hemlock/bride's bonnet                     15.6
       TSHE/GYDR                       Western hemlock/ oakfern                      69.973




Table 21. Major plant association groups identified as part of the field effort. Values are
expressed as percentages of sampled acres within each subwatershed.

      Subwatershed            ABGR/CLUN THPL/ATFI THPL/CLUN THPL/OPHO TSHE/CLUN TSHE/GYDR
   Main Granite Creek              2%             25%           17%             3%            14%             39%
North Fork Granite Creek                          22%            5%             19%            4%             51%
South Fork Granite Creek                          45%                                          6%             49%



5.3.3 Subwatershed Scale Growth & Mortality Projections

At subwatershed scales, there is projected to be a steady increase in available large trees (both
live and dead) 20 inches DBH. The mainstem Granite Creek subwatershed contains almost
twice the standing stock of the NF and SF subwatersheds (Figure 29). Standing stock in the
mainstem ranges between 20 and 35 TPA (25 – 40 trees per 100m) 32 in the first 50 year period.
The SF and NF available stock is approximately half, ranging between 10 and 25 TPA by 2058
(12 – 28 trees per 100m). Large tree mortality increases from approximately 1 to 2.5 TPA in the
mainstem Granite and from 0.5 to 1.5 TPA in the other subwatersheds (Figure 30). Projected

32       A conversion multiplier of 1.13 is used to convert units of trees per acre to trees per 100 m of stream
channel for the 150 ft wide (45.7 m) riparian buffer zone. See methods section.


Granite Creek Assessment                                WPN                                                   Page 53
tree mortality, expressed as the ratio of dead:live trees per acre is projected to increase from 4-
5% to a steady state of 6-8%, reached at approximately the 50 year time step (Figure 31).

In all cases at the subwatershed scale, there does not appear to be a limit in available large and
durable conifer tree material for potential recruitment to the stream system. Current large-scale
projections indicate an increase in recruitment of both live and dead trees through time without
any additional active management. In addition, a considerable amount of basal area is present,
suggesting some local thinning may provide opportunities for direct LWD recruitment (Figure
32). Actual recruitment to the stream channel is not directly predicted here, as it is dependent
upon a series of factors, including proximity to the stream, slope, aspect, and obstructions, as
well as stochastic factors (landslides, wind throw, insects, disease, and fire). The purpose of this
analysis was to predict wood available for active (placement) or passive (natural mortality)
recruitment directly from the riparian system only. Hence this analysis is conservative in
predicting potentially available wood, as added effects from stochastic factors and upslope
recruitment will likely increase probabilities and timing for recruitment.

                               50

                               45

                               40
  Trees >20 in. DBH per Acre




                               35

                               30

                               25

                               20

                               15

                               10

                               5

                               0
                               2000        2020            2040         2060             2080            2100          2120
                                                                  Modeled End Year

                                      Main Granite Creek      North Fork Granite Creek          South Fork Granite Creek




Figure 29. Trees per acre 20 inches DBH for a 100 year time period.




Granite Creek Assessment                                                                 WPN                                  Page 54
                                     4

                                    3.5
  Dead Trees >20 in. DBH per Acre




                                     3

                                    2.5


                                     2

                                    1.5

                                     1

                                    0.5

                                     0
                                     2010     2020   2030      2040     2050   2060    2070      2080     2090   2100    2110      2120
                                                                            Modeled End Year

                                          Main Granite Creek          North Fork Granite Creek          South Fork Granite Creek




Figure 30. Projected tree mortality per acre for trees 20 inches DBH.




Figure 31. The ratio of projected dead to projected live trees 20 inches DBH.




Granite Creek Assessment                                                                            WPN                                   Page 55
  Basal Area (ft2 per acre) for trees >20 in. DBH   250



                                                    200



                                                    150



                                                    100



                                                    50



                                                     0
                                                     2000          2020        2040         2060          2080         2100           2120
                                                                                      Modeled End Year

                                                          Main Granite Creek   North Fork Granite Creek    South Fork Granite Creek




Figure 32. Basal area (ft2 per acre) of trees 20 inches DBH.



5.3.4 Reach Scale Growth & Mortality Projections

The previous section demonstrated subwatershed-scale averages for standing live and projected
mortality of large trees (20 inches DBH) were predicted to steadily improve through time. A
total of 131 stream reaches were examined for the 100 year time period. The results for each
stream reach were compared directly to the average response for the subwatershed it occupies.
Hence, values were relativized through time to the responses found in Section 5.3.3 above.

As displayed in the previous section, the NF and SF watersheds had generally fewer large trees
(20 inches DBH) than the Mainstem Granite Creek subwatershed. When compared with each
subwatershed mean values (Figure 32), the current (2008) distribution of live large trees shows
very few areas within the mainstem Granite Creek and tributaries with lower-than-average large
tree densities.

Limitations (as compared with subwatershed mean values) begin to appear in the lower reaches
of the transition to the North Fork and South Fork Granite Creek. Reaches 5 and 8 in the Lower
NF Granite are approximately 35% and 67% below the average for the NF subwatershed, though
both of these sections are associated with wide floodplain meadows (and road influence). Orwig
Creek appears to be below average in large trees, though this is likely a legacy of having few
measured samples; the aerial imagery suggests large tree density is well within limits.




Granite Creek Assessment                                                                                  WPN                                Page 56
Figure 33. The degree of departure from the subwatershed average, for standing TPA 20
inches DBH in 2008.33
In the SF Granite subwatershed, reaches 1 and 2 of the SF Granite are 120% of average large tree
densities for the subwatershed. Reach 3 drops below average, primarily due to open non-
forested meadows. The Upper SF Granite Creek is mostly within average limits, excepting the
meadow-driven components of the system. Sema Creek is mostly within average limits or
within 30% of the subwatershed average, with the variations in large tree density following the
amount of meadow environment within the riparian zone. Tobasco Creek has very low levels of


33      Actual values presented in Appendix B.


Granite Creek Assessment                         WPN                                     Page 57
large trees compared with the average, primarily due to timber harvest activities and riparian
meadows.




Figure 34. The average number of dead trees 20 inches in diameter likely to be produced
in 10-year increments per 100m of stream for year 10, 50 and 100 (1st, 5th and 10th time
step), separated by reach.




Granite Creek Assessment                    WPN                                        Page 58
Trees that were modeled to die from stem exclusion and other natural mortality events at the end
of each 10 year period were summarized by stream reach and presented as large trees (20
inches DBH) per 100 m of stream (Figure 34). At the end of the first time-step (2018), most
areas along the NF and SF Granite Creek are projected to produce fewer that 1 tree per 100 m in
length. Most of the tributaries and the mainstem Granite are predicted to produce between 1 and
2 large dead trees per 100 m, with the exception of the lower reaches of Granite Creek, where
almost no mortality in the 20 inch size class was modeled to occur.

As displayed at subwatershed scales (Figure 30), tree mortality increases through time to
increased ratios of ~6-8% of the standing live trees by year 100, yielding only 3 reaches of
Granite Creek producing less than 1 tree every 200 m (i.e. 0.5 trees per 100 m): Reach 1, SF
Reach 1 and Upper SF Reach 9. All other reaches except Upper SF Granite Creek Reach 1 and
Reach 10 produced over 1 tree per 100 m. The mainstem Granite and tributaries were the most
productive through the time series.

5.3.5 Modeling Conclusions

The use of fine-scale mapping (polygons) for riparian zones is a useful tool in developing a long-
term riparian strategy. How polygons are grouped and sampled is highly dependent upon the
spatial scale of the desired assessment. For the purposes of this project, our area of interest was
at the subwatershed and stream reach scales. Hence, field sampling was organized to capture the
widest array of vegetation groups, with the goal to sample on the basis of acres represented on
the landscape.

Due to the inherent nature of riparian zone dynamics, the structure and composition is highly
diverse, is spatially heterogeneous, and is more prone to stochastic factors that alter the
composition, structure, and their function for the stream system. Capturing these diverse factors
in a field sample is a difficult and expensive task, especially over large land areas. This becomes
compounded with forward-growth projections, and uncertainty increases with the amount of time
projected.

The purpose of this modeling run was to evaluate overall subwatershed forest growth and
mortality processes and identify reach-level areas of concern where LWD recruitment may be
limiting. The following are fundamental findings regarding the current and potential future
forest and LWD dynamics for the Granite Creek Watershed:

       There appears to a steady supply of large conifer trees in the system. Trees per acre
        and basal area of large trees (20 inches dbh) is relatively high (~10 – 20 per acre and 50
        –100 ft2/acre), and recruitment into this size class increases steadily through time.
        Reach-level metrics are consistent with the subwatershed averages; areas that are below
        average are most often associated with riparian meadows.

       Available large dead trees steadily increases and levels to a steady range. As with
        standing live trees, large dead trees increases from 4-5% of the live tree ratio to



Granite Creek Assessment                       WPN                                          Page 59
        approximately 6-8%. Recruitment to large dead trees steadily increases without
        adversely affecting live tree densities.

       Ample opportunities exist to modify trees per acre/ basal area targets for forested
        sites. The standing basal area of large trees appears to be sufficient for developing site-
        specific silvicultural prescriptions to remove whole trees for in-stream LWD placement
        projects. This potential increases through time.

       Additional field sampling and model re-runs will improve estimates at finer scales.
        At targeted reach scales, it is recommended to increase focus field efforts in the near-
        stream environment to develop and evaluate effects of silvicultural treatments in the
        riparian zones. The current fine-scale riparian polygons developed for this project are
        adequate and useful for additional field sample and prescription design.

Following these assessment conclusions, areas of opportunity can be developed to refine an
Action Plan strategy.

5.3.6 Stream Shade Results

Current and potential future shade levels for the principal streams in the Granite Creek watershed
are shown in Figure 35 and summarized in Table 22. Only the mainstem Granite, the North and
south Forks, and Blacktail Creek were included in this assessment, as these are the only streams
having thermograph data (used in section 7 below). Blacktail Creek has high levels (median >
90%) of effective shade, with low variability (standard deviation < 10%), under current and
future conditions (Table 22). In contrast, the South and North Forks, and lower mainstem of
Granite, have lower shade levels (60-80% effective shade) and higher variability (standard
deviation > 20%). The primary reasons for this are the narrow canopy openings found along
small tributary streams (i.e., Blacktail Creek). Shade levels in the large streams vary
considerably longitudinally given the frequency of large non-forested stream-adjacent areas,
wide active channels, and varying aspect of the stream channels. Modeled effective shade
generally increases from the current (2008) condition to 2058, but declines slightly over the
following 50-year period in most of the streams.

Longitudinal effective shade profiles for the four-modeled stream segments are shown in Figure
36 and Figure 37. Areas where future conditions show higher effective shade values are likely
areas for protection and passive restoration. Areas where there is little or no change in future
effective shade values may represent areas that are either fully functioning (in terms of shade), or
have a natural or human-caused impediment to stand development; in either case they represent
areas where enhancement opportunities could be investigated. Areas where future values are
lower then current values represent areas where shade conditions are expected to deteriorate in
the absence of active management.




Granite Creek Assessment                       WPN                                           Page 60
Figure 35. Current effective shade (top), and potential future shade conditions in 50
(middle) and 100 (bottom) years along principal streams, Granite Creek watershed.




Granite Creek Assessment                WPN                                    Page 61
Table 22. Stream Shade Summary. Modeled average (minimum-maximum) effective
shading (%) for principal stream systems in the watershed in 2008 and projected 50 and
100 years in the future.
                                             Current     Potential future conditions:
                                           conditions:
      Location               Statistic     (7/16/2008)   7/16/2058           7/16/2108
                      Mean                    67%           72%                 71%
                      Median                  80%           88%                 89%
 South Fork Granite Standard Deviation        29%           30%                 30%
                      Minimum                  2%           2%                   2%
                      Maximum                 95%           97%                 96%
                      Mean                    69%           74%                 74%
                      Median                  78%           84%                 87%
 North Fork Granite Standard Deviation        27%           26%                 25%
                      Minimum                  7%           7%                   7%
                      Maximum                 98%           97%                 97%
                      Mean                    61%           68%                 69%
                      Median                  68%           74%                 73%
  Lower Mainstem
                      Standard Deviation      22%           23%                 22%
     Granite
                      Minimum                  1%           1%                   1%
                      Maximum                 93%           95%                 95%
                      Mean                    92%           95%                 93%
                      Median                  92%           96%                 94%
   Blacktail Creek    Standard Deviation       5%           5%                   5%
                      Minimum                 59%           62%                 61%
                      Maximum                 94%           97%                 95%




Granite Creek Assessment                      WPN                                     Page 62
Figure 36. Longitudinal effective shade profiles for South Fork Granite Creek (top) and
North Fork Granite Creek for current and modeled future conditions.




Granite Creek Assessment                 WPN                                     Page 63
Figure 37. Longitudinal effective shade profiles for Lower Mainstem Granite Creek (top)
and Blacktail Creek for current and modeled future conditions.




Granite Creek Assessment                 WPN                                     Page 64
5.4       AREAS OF OPPORTUNITY: LWD ENHANCEMENT
From the model and summary information presented above, data were examined to evaluate
restoration opportunities to increase in-stream LWD potentials, through active- and/or passive
restoration strategies. Not all stream segments are discussed, though similar data presented here
can be found in Appendix B: Digital Appendices to the Riparian Assessment.

Using the riparian model outputs, this section evaluates restoration opportunities to increase in
stream LWD potential. Current values for instream wood debris were evaluated at reach scales
(see Fisheries Section). Reference reach information was analyzed and the 75th percentile value
for combined in-stream wood and debris dams was chosen as restoration targets for the surveyed
streams. A total of 30 surveyed stream reaches did not meet the 75th percentile criteria for
instream wood and/or debris dams per 100 m of stream. This section considers:

         The number of pieces of LWD required to meet reference reach targets, and

         The residual standing stock of trees in the riparian zone should wood be sourced to meet
          targets in the next year.

The current and projected average standing large live trees per 100 m (20 inches DBH) in the
riparian zone were compared with the instream LWD values to determine if enough standing
wood was available to meet the instream LWD 75th percentile restoration target for a given
reach. Figure 38 displays the streams that were measured for instream LWD, and highlights
reaches that do not currently meet the LWD target. Within these highlighted reaches, the amount
of wood required to meet the targets are mapped in black. The colored lines represent the
residual (remaining) standing large trees per 100 m, expressed as a percentage of their current
and projected future unmanaged standing stocks.

The following sections highlight these potential restoration opportunities, organized by
subwatershed.




Granite Creek Assessment                        WPN                                        Page 65
Figure 38. Measured LWD sites. Light shading represents reaches that meet or exceed the
75th percentile as compared with reference reaches. Black lines indicate the number of
pieces and/or dams required to meet the 75th percentile. Assuming active restoration, the
colored lines indicate the remaining standing conifer trees 20 inches DBH, expressed as a
percentage of the current and projected values per 100 m. Green is a >75% retention,
orange is 50 – 75% and red is <50% retention or there are insufficient trees on site.




Granite Creek Assessment                  WPN                                      Page 66
5.4.1 Mainstem Granite Creek

Within the Mainstem Granite Creek subwatershed, a total of 15 stream reaches did not meet the
75th percentile instream LWD targets, as defined from reference reach data. All reaches of
Granite Creek and WF Packer Creek did not meet targets. Packer Creek reach 2, the mid reaches
of Fedar Creek and the upper reach of Athol Creek also failed to meet the 75th percentile for
LWD (Table 23).

Table 23. The summary of instream LWD in reaches that does not meet the 75 th percentile
target (of reference reaches), and how many pieces are required to meet those targets.
                                                                                       th
                       Current Pieces + Dams   Additional Pieces Required to Meet 75
   Stream Reach              per 100 m               Percentile Target per 100 m
   Athol Creek 3               13.6                             1.6
   Fedar Creek 4                6.4                             8.8
   Fedar Creek 5                7.5                             7.7
  Granite Creek 1               3.4                             4.7
  Granite Creek 2               2.6                             5.5
  Granite Creek 3               2.3                             5.8
  Granite Creek 4               4.1                              4
  Granite Creek 5               5.9                             2.2
  Granite Creek 6               1.3                             6.8
  Granite Creek 7               4.6                             3.5
  Granite Creek 8               4.8                             3.3
   Packer Creek 2              11.7                             3.5
Packer Creek, WF 1             14.1                             1.1
Packer Creek, WF 2              0.7                            14.5
Packer Creek, WF 3              4.3                            10.9



The number of pieces needed to achieve the in-stream LWD targets for these reaches is listed in
Table 23 above. In many cases the requirement exceeds the current value within the stream.
Figure 39 and Figure 40 display the difference between the reach-level averages for standing live
trees 20 inches DBH in the associated riparian zones and the wood required per 100 m to meet
the LWD restoration targets (i.e. “residual standing stock”, or the trees remaining in the riparian
zone should whole trees be removed and used to meet the LWD restoration target). These values
show the potential reach-level effect to the large tree stock for each of the 10, 10-year
timeframes. Residual Standing Stock is the number or percentage of trees remaining after
harvest for LWD improvement projects.




Granite Creek Assessment                       WPN                                          Page 67
  Remaining Trees >20 in. DBH per 100 m   70


                                          60


                                          50


                                          40


                                          30


                                          20


                                          10


                                          0
                                          2000     2020            2040           2060          2080         2100            2120
                                                                          Modeled End Year

                                                 Granite Creek 1     Granite Creek 2     Granite Creek 3   Granite Creek 4
                                                 Granite Creek 5     Granite Creek 6     Granite Creek 7   Granite Creek 8



Figure 39. The projected remaining standing live trees 20 inches DBH per 100 m if
instream LWD restoration projects sourced wood from the immediate riparian zone (Table
23).



To meet LWD restoration targets in Granite Creek, the residual large trees in reaches 2 and 6
would drop below 75% of the current (2008) standing stock. In 2028, it is projected that the
current LWD needs would result in a residual >75% for reach 2, and in 2068 for reach 6.
Absolute numbers of standing trees (Figure 39) is the lowest in Reach 6, and this reach is highly
constrained by a road and granite dome. All other reaches have current residual stock values
>75%, with reach 5 having a >90% residual value.

For the tributaries to Granite Creek, WF Packer Creek (reaches 2 and 3) would currently require
240% and 70% more trees than are currently available to meet LWD targets. Reach 2 is a
floodplain meadow and is not expected to meet any LWD recruitment targets in 100 years.
Reach 3 is forested and is projected to meet the current wood large wood requirement with 75%
retention at the 100-year time step. Reach 1 of WF Packer Creek and Athol Creek can currently
meet LWD targets with >90% residual standing stock. Packer Creek Reach 2 can meet targets
with 86% residual. The limiting reaches in Fedar Creek (4 and 5) would have a 68% and 62%
residual in 2008 and >75% residual in 2028.




Granite Creek Assessment                                                                 WPN                                        Page 68
                                          80.0

                                          70.0
  Remaining Trees >20 in. DBH per 100 m



                                          60.0

                                          50.0

                                          40.0

                                          30.0

                                          20.0

                                          10.0

                                            0.0
                                              2000          2020         2040            2060          2080    2100          2120
                                          -10.0

                                          -20.0
                                                                                   Modeled End Year

                                            Athol Creek 3          Fedar Creek 4            Fedar Creek 5     Packer Creek, WF 1
                                            Packer Creek, WF 2     Packer Creek, WF 3       Packer Creek 2



Figure 40. The projected remaining standing live trees 20 inches DBH per 100 m if
instream LWD restoration projects sourced wood from the immediate riparian zone (Table
23).




Granite Creek Assessment                                                                        WPN                                 Page 69
5.4.2 North Fork Granite Creek

Within the NF subwatershed, 8 reaches did not meet the instream targets for LWD, though NF
Willow (reach 1) and Lower NF Granite Creek reach 9 were within 0.1 “piece” of meeting the
75th percentile of instream wood from reference conditions (Table 24).

Table 24. The summary of instream LWD in reaches that does not meet the 75 th percentile
target (of reference reaches), and how many pieces are required to meet those targets.

                                                              Additional Pieces Required to Meet
                                                                 th
      Stream Reach          Current Pieces + Dams per 100 m    75 Percentile Target per 100 m
Granite Creek, Lower NF 2                 2.4                                5.7
Granite Creek, Lower NF 5                 5.1                                 3
Granite Creek, Lower NF 6                 3.3                                4.8
Granite Creek, Lower NF 7                 3.1                                 5
Granite Creek, Lower NF 9                 8                                  0.1
      Willow Creek 1                     12.4                                2.8
      Willow Creek 2                     10.5                                4.7
    Willow Creek, NF 1                   15.1                                0.1



As expected, both creeks requiring less than 1 piece could meet the LWD targets with >99%
retention in large tree riparian zone structure. Willow Creek (reaches 1 and 2) can meet LWD
targets with 83 and 86% of the residual standing stock. Reach 2 on the lower NF Granite can
likewise meet targets in 2008 with 76% of the residual large tree structure.

Reaches 5-7 on lower NF Granite Creek currently require 3, 4.8 and 5 pieces to meet LWD
targets. This translates to harvest with approximately 69% residual stock of large trees to meet
these targets for these reaches. Assuming these current targets are constant in 2018, the LWD
goals can be met, leaving ~75% of the standing large tree stock. All three of these reaches are
constrained or partially constrained by a road, and are associated with floodplain meadow
environments. Total trees reflects this (Figure 41), with 10 or fewer trees 20 inches DBH per
100 m of stream length.




Granite Creek Assessment                        WPN                                           Page 70
  Remaining Trees >20 in. DBH per 100 m   60


                                          50


                                          40


                                          30


                                          20


                                          10


                                          0
                                          2000          2020           2040          2060            2080          2100            2120
                                                                              Modeled End Year

                                                 Granite Creek, Lower NF 2    Granite Creek, Lower NF 5     Granite Creek, Lower NF 6
                                                 Granite Creek, Lower NF 7    Granite Creek, Lower NF 9     Willow Creek 1
                                                 Willow Creek 2               Willow Creek, NF 1



Figure 41. The projected remaining standing live trees 20 inches DBH per 100 m if
instream LWD restoration projects sourced wood from the immediate riparian zone (Table
24).
5.4.3 South Fork Granite Creek

The South Fork drainage contrasts with the other subwatersheds as 4 of the 7 stream reaches
require ~1.2 - 1.5 pieces per 100 m to meet LWD targets, and only a single reach in Tobasco
Creek exceeds 4 piece per 100 m for enhancement (Table 25).

Table 25. The summary of instream LWD in reaches that does not meet the 75 th percentile
target (of reference reaches), and how many pieces are required to meet those targets.
                                                                                                                                              th
                                                                                                    Additional Pieces Required to Meet 75
                                           Stream Reach             Current Pieces + Dams per 100 m       Percentile Target per 100 m
                           Granite Creek, SF 1                                       6.9                                            1.2
                           Granite Creek, SF 2                                       6.7                                            1.4
                           Granite Creek, SF 3                                       5.3                                            2.8
                                           Sema Creek 4                             11.5                                            3.7
                                           Sema Creek 6                              14                                             1.2
                                           Sema Creek 7                             13.7                                            1.5
                                          Tobasco Creek 2                            5.3                                            9.9



Reaches 1 and 2 of SF Granite Creek require 1.2 and 1.4 pieces per 100 m to meet targets.
Removal of large trees in these areas to meet this goal would result in ~95% residual stock large

Granite Creek Assessment                                                                    WPN                                           Page 71
trees in the riparian zone. Reach 3 requires approximately twice the LWD and has
approximately 25% of the available large trees due to the prominent meadow community (Figure
42). Sourcing trees from this reach to increase instream LWD values by 2.8 trees per 100 m
would result in standing stock residual values ranging from 65% in 2008 and 76% in 2058,
assuming constant instream LWD values.

Sema Creek reaches 6 and 7 could meet targets from standing large trees, leaving 89% and 91%
of the large trees on average for the reaches. Reach 4 of Sema Creek is associated with a very
large riparian meadow and would not meet current reference targets for instream LWD without
heavy reductions in standing stock, until 2098 (65% residual) and 2108 (72% residual). Tobasco
Creek Reach 2 is likewise associated with floodplain meadows and would have fewer than 1 tree
per 100 m surplus (Figure 42) assuming active LWD enhancement from riparian stocks in 2108
(i.e. <5% residual standing stock of projected growth).




                                          70
  Remaining Trees >20 in. DBH per 100 m




                                          60

                                          50

                                          40

                                          30

                                          20

                                          10

                                            0
                                            2000              2020       2040              2060         2080         2100          2120
                                          -10
                                                                                  Modeled End Year

                                               Granite Creek, SF 1   Granite Creek, SF 2      Granite Creek, SF 3   Sema Creek 4
                                               Sema Creek 6          Sema Creek 7             Tobasco Creek 2



Figure 42. The projected remaining standing live trees 20 inches DBH per 100 m if
instream LWD restoration projects sourced wood from the immediate riparian zone (Table
25).



5.4.4 Areas of Opportunity: LWD Enhancement Summary

This analysis was based on several assumptions about instream LWD and associated riparian
community structure. The first is the 75th percentile of combined pieces and debris dams


Granite Creek Assessment                                                                          WPN                                     Page 72
calculated from reference reaches is a relevant target to assume for instream LWD enhancement
for the Granite Creek watershed. The second assumption we made was one whole live tree 20
inches DBH found in the riparian zone (on a per 100 m stream length basis) is equivalent to one
piece or one dam measured as part of the stream surveys. Arguments for this equivalency can be
made that a whole tree includes branches and potential root wad structure that can provide the
necessary function; actual diameter class required to serve this function should be made on the
site-specific basis.

With the assumptions well understood, the overall restoration effort to actively manage riparian
stands for instream LWD can be very focused. Several conclusions can be made:

       With few exceptions, meadow-driven reaches are not meeting LWD targets nor are
        expected to carry enough trees to actively input from the stream bank. The LWD
        targets do not likely apply to these reaches, and additional criteria and reference reach
        sampling may need to be explored to ensure LWD functions are not limiting in these
        reaches.

       Staging larger LWD enhancement projects through time would defray impacts to
        forest health. Five- or ten-year incremental additions and anchoring of whole trees in
        the system would allow for gradual stream improvements and allow for younger cohort
        trees in the riparian zone to recruit to larger size classes. This is a key process to site
        design development and to the next level of sampling and management for these riparian
        stands.

       Thinning and release for instream enhancement projects could improve forest
        health. Average forest basal area should be monitored in context with insects and
        diseases to ensure the riparian forests continue to function well for the stream system as
        well to achieve desired conditions. It is probable that selected thin-from-below tactics
        could produce whole tree material smaller than 20 inches DBH to achieve the desired
        outcome for enhancing instream LWD.

       Evaluate roads, culverts and bridges. Prior to any LWD enhancement projects, the
        road and bridge networks will need to be evaluated to ensure long-term natural
        recruitment of LWD from the riparian zone can incorporate into the stream system
        without harming existing infrastructure.

This analysis also does not take into account several non-measured stream reaches for LWD. As
additional instream LWD data becomes available (from additional reference or from Granite
Creek streams), this analysis can be re-run in relative short order to identify additional areas of
interest or opportunity for direct LWD enhancement.

5.4.5 Stream Shade Conclusions and Recommendations

Discussed in section 7.5 (Water Quality).



Granite Creek Assessment                       WPN                                          Page 73
                                  6.0     SEDIMENT SOURCES
The primary sources of coarse and fine sediment inputs in the Granite Creek watershed include:

     Coarse and fine inputs from mass-wasting
     Upland erosion associated with forest harvest
     Road-related sediment inputs

Although mass-wasting events can contribute large volumes of sediment to stream channels, it is
unlikely that mass wasting is a large source of sedimentation within the Granite Creek
watershed. As noted above, mass failure potential is generally low, with the exception of steep
headwalls and incised stream canyons (Figure 13). We do not attempt to quantify sediment
inputs from mass wasting in this analysis.

Forest harvest by itself does not typically cause significant erosion; rather it is the extraction
methods (e.g., ground-based harvest methods) that may result in significant erosion (Megahan,
1986). Erosion per unit area is typically much less that for roads than for harvest areas.
However, the area of harvest is usually large relative to roads, so total erosion volumes from
harvest areas may be similar to that from roads. Harvest rates in the Granite Creek watershed
have been low in recent years, and few unvegetated harvest units exist (Figure 17). We do not
attempt to quantify sediment inputs from forest harvest in this analysis.

The following assessment is focused on quantifying road-related sediment inputs in the Granite
Creek watershed. We focus on 1) sediment production from road surfaces and delivery to
streams, and 2) the risk of point sources of sediment from culvert failures/washouts.

6.1     CRITICAL QUESTIONS
The analysis is built around answering the following questions:

      1. What are the delivered sediment loads from the existing road system, and how do these
         compare with natural background sediment loads34?
      2. Which road segments are contributing the highest delivered sediment loads?
      3. What are the risks posed by culvert failures on the existing road system?

6.2     METHODS
Two primary methodologies were used in this assessment, both of which rely on field
information collected by USFS field crews in 2006 and 2007. Road-related sedimentation was
evaluated using the Water Erosion Prediction Project (WEPP) computer model (Flanagan and
Livingston 1995), described below. Several variants of the WEPP model are available and can



34 Background sediment loads were estimated by the IPNF (2006). Values are given as part of the Landtype GIS
coverage.


Granite Creek Assessment                            WPN                                                Page 74
be run via the Internet35, for this analysis we used the WEPP Road interface. The culvert risk
assessment was conducted following methodologies developed by Flanagan et al. (1998).

6.2.1 Field Data Collection

The USFS identifies 128 miles of system roads36 in the Granite Creek watershed, 57 miles of
which were field-surveyed (IPNF, 2008; Figure 43, Figure 44). An additional 3 miles of non-
system roads were also included in the surveys. A total of 169 culverts were surveyed; 51 in
2006, and the remainder in 2007 (Figure 44). Surveys focused on higher-use roads.




Figure 43. System roads in the Granite Creek watershed by maintenance category, and
roads and culverts surveyed in 2006-2007.


35 http://forest.moscowfsl.wsu.edu/fswepp/
36 Roads that are maintained by the US Forest Service


Granite Creek Assessment                                WPN                            Page 75
Figure 44. Miles of road by subwatershed and survey status.



6.2.2 WEPP Road modeling

WEPP ROAD is a computer model that estimates sediment yield from individual road
segments37. Field data collected during the 2006-2007 season were use in the model. Field data
collected for each segment included the following:

       Road Design (insloped, bare ditch; insloped, vegetated or rocked ditch; outsloped,
        unrutted; or outsloped, rutted)

       Road surfacing (native, graveled, paved)

       Traffic level (high, low, none)

       Gradient of the road, the road fillslope, and the buffer area between the road and stream

       Length and width of road and fillslope

37 Individual segments were defined as contiguous sections of roads draining to a culvert. There were 162
individual road segments draining to culverts


Granite Creek Assessment                              WPN                                                   Page 76
       Buffer distance from the stream

Additional parameters needed for the WEPP ROAD model include the following:

       A representative climatic record for the area. A 30-year modified record was created for
        the Granite Creek watershed based on actual records from the Sandpoint Experiment
        Station (NCDC coop station #108137).

       Soil textural information. For the purposes of this assessment all soils were considered to
        be sandy loam soils

       Estimates of the rock content of road fill. No information was available on the rock
        content of roads in the project area. An assumed 20% rock content was used in all model
        runs.

       Modeling period. A 30-year period was used in each model run.

6.2.3 Culvert Risk Assessment

The culvert risk assessment was conducted following methodologies outlined by Flanagan et al.
(1998). The methodologies were modified for use on the IPNF by Forest staff. For each culvert
surveyed an Environmental Risk Score was calculated based on four sub-factors:

    Environmental Risk Score = Culvert Hazard Score + Fill Hazard Score + Consequence
                               Score + Impact Score

Culvert Hazard Score: The culvert hazard score is a rating based on the physical characteristics
of the culvert itself. It is calculated as the following:

    Culvert Hazard Score = [(Hydrologic Capacity * 2) + (Woody Debris Capacity * 2) +
                           (Sediment Transport) + (Debris Upslope) + (Inlet Plugged *
                           3) + (Inlet Crushed *3) + (Current Condition *6)]/1.65

    Where: Hydrologic Capacity      = Qualitative rating of the capacity of the culvert to
                                      convey flood flows (Great/good/fair = 1; Bad = 2,
                                      Extreme =3)
              Woody Debris Capacity = Rating of the capacity to pass woody debris:
                                      3 = Bad/Extreme (culvert diameter < 0.5 * flood
                                             prone channel width)
                                      2 = Good/Fair (culvert diameter > 0.5 and < 1.0 *
                                             flood prone channel width)
                                      1 = Great (culvert diameter > 1.0 * flood prone
                                             channel width)
              Sediment Transport    = Rating of the capacity to pass sediment:


Granite Creek Assessment                       WPN                                          Page 77
                                           3   = Bad/Extreme (Slope of Pipe/Slope of Channel
                                                  < 3)
                                           2   = Good/Fair (Slope of Pipe/Slope of Channel >
                                                 3 but < 6)
                                           1   = Great (Slope of Pipe/Slope of Channel > 6)

              Debris Upslope           = Rating of the debris upslope that might plug culvert
                                         (Great = 0; else = 1)
              Inlet Plugged            = Rating of the extent that inlet is plugged (Great = 0;
                                         Good/Fair = 1; Bad = 2; Extreme = 3)
              Inlet Crushed            = Rating of the extent that inlet is crushed (Great = 0;
                                         Good/Fair = 1; Bad = 2; Extreme = 3)
              Current Condition        = Rating of the overall current condition of the culvert
                                         (Great/Good = 0; Fair/Bad = 2; Extreme = 3)

Fill Hazard Score: The Fill Hazard Score is a rating of the propensity of the road fill itself to
failure. It is calculated as the following:

    Fill Hazard Score = (Flow Under CMP * 6 + Saturated Road Fill * 12 + Road Fill
                        Failing * 6) / 1.65

    Where: Flow Under CMP         = Qualitative rating of the extent to which water flows under
                                    the culvert barrel (Great/good = 0; Fair/Bad = 1, Extreme
                                    =2)
              Saturated Road Fill = Rating of the extent to which fill is saturated
                                    (Great/Good/Fair = 0; Bad/Extreme = 1)
              Road Fill Failing = Rating of the extent to which fill is currently failing
                                    (Great/good = 0; Fair/Bad = 1, Extreme =2)

Consequence Score: The Consequence Score is a rating of the physical consequences of failure.
It is calculated as the following:

    Consequence Score = [(Fill Volume *2) + ( Diversion *4) + (Diversion Distance *3) +
                        (Geology *2)] / 1.65

    Where: Fill Volume           = Rating based on the volume of fill available to fail (Great =
                                   1; Good = 2; Fair/Bad = 3, Extreme = 4)
              Diversion          = Rating of the likelihood for stream diversion at the crossing
                                   (Great = 0; Good/Fair = 2; Bad/Extreme = 3)
              Diversion distance = Rating of the distance the stream is likely to be diverted
                                   (Great = 0; Good/Fair = 2; Bad/Extreme = 3)
              Geology            = Rating of the geologic hazard at the site (Great = 0; Good =
                                   1; Fair = 2; Bad/Extreme = 3)



Granite Creek Assessment                       WPN                                        Page 78
Impact Score: The Impact Score is a rating of the potential impact of the culvert and fill on
sensitive resources. It is calculated as the following:

      Impact Score = [(Fisheries *4)+(Water Supply *2) + (TMDL *3)] / 1.65

      Where: Fisheries     = Rating based on the presence / absence of sensitive fisheries
                             (Present = 1; Absent = 0)
              Water Supply = Rating of the presence / absence of a downstream water supply
                             source (Present = 1; Absent = 0)
              TMDL         = Presence / Absence of a TMDL in the catchment (Present = 1;
                             Absent = 0)

6.3    RESULTS
6.3.1 WEPP Road modeling

Total road density ranges from 0.7 mi/mi2 in the South Fork Subwatershed, to 2.2 mi/mi2 in the
Main Granite subwatershed, and is 1.3 mi/mi2 overall (Table 26). There were 160 individual
road segments within the Granite Creek watershed that are connected to the stream system. The
length of surveyed road connected to the stream system was lowest in the South Fork (2.4 miles),
but the proportion of connected surveyed roads was lowest in the Main Granite (7%). Model
input and outputs for these 160 segments are given in Appendix C: WEPP Inputs and Outputs.

Table 26. Road densities and surveyed road drainage connectivity.

                           All Roads:                    Surveyed roads only:
                              Road    Road length Surveyed road Surveyed road    Number of
                            density    surveyed       length         length    connected road
                                  2
           Area             (mi/mi )     (mi)     connected (mi) connected (%)   segments
   Main Granite Creek         2.2        48.4         3.5            7%               91
North Fork Granite Creek      0.9        13.9         4.5            33%              54
South Fork Granite Creek      0.7        11.2         2.4            21%              15
      Entire watershed        1.3        73.5         10.4           14%              160



Results from the WEPP ROAD modeling are summarized in Table 27. Modeled road sediment
yields are very low, ranging from 0.2 tons/year in the South Fork to 6.1 tons/year in the Main
Granite, and the cumulative average annual sediment rate from roads is 8.6 tons/year for the
entire watershed (Table 27). These rates were normalized by watershed area and compared to
natural sediment loads (Figure 13, Table 27). The percent increase in delivered sediment from
roads over background rates is less than 1% watershed wide.




Granite Creek Assessment                        WPN                                        Page 79
Table 27. WEPP Road modeling summary.


                                  Average annual      Average annual      Area-weighted      % increase
                                 sediment leaving    sediment leaving Natural Sediment          over
           Area                    buffer (ton/yr)   buffer (ton/mi2/yr) Load (ton/mi2/yr)   background
   Main Granite Creek                   6.1                0.1726              29.9             0.6%
North Fork Granite Creek                2.3                0.0787              31.1             0.3%
South Fork Granite Creek                0.2                0.0063              27.9             0.02%
       Grand Total                      8.6                0.0867              29.5             0.3%



6.3.2 Culvert Risk Assessment

A total of 169 culverts were assessed in the Granite Creek watershed during the 2006-2007 field
season (Figure 43). The Environmental Risk score was calculated for each culvert location as
described in Section 6.2.3 above. Each location was assigned an overall Environmental Risk
Rating using the categories given in Table 28. The majority (150) of the surveyed culverts were
rated as “Good”, and only 19 surveyed culverts were rated as “Fair” (Figure 45; Figure 46).
None of the surveyed culverts fell into the “Bad” or “Extreme” categories. Culvert Risk Rating
values for the 169 culverts are given in Appendix D: Culvert Risk Assessment Data.

Table 28. Environmental Risk Scores and overall rating.
            Environmental Risk Score                                Environmental Risk Rating
                           0-4                                                Great
                        5 - 21                                                Good
                      22 - 38                                                  Fair
                      39 - 58                                                  Bad
                      59 - 75                                                Extreme




Granite Creek Assessment                             WPN                                           Page 80
Figure 45. Environmental Risk Ratings for 169 surveyed culverts.




Figure 46. Summary of Environmental Risk ratings for surveyed culverts.




Granite Creek Assessment                  WPN                             Page 81
6.4       INFORMATION GAPS AND MONITORING NEEDS
The assessment presented here was based solely on road and culvert field survey data collected
in 2006-2007. No assessment of harvest related or mass wasting related sedimentation was
performed. Mass Wasting potential is generally low throughout the watershed, however, a
review of aerial photos during periods of more intense harvest would be helpful to verify that
inputs from these other sources are negligible.

The road and culvert assessments presented here were based on field survey information that
covered approximately 60% of the system roads in the watershed. Assessments the remainder of
the system roads, as well as existing non-system roads would give a more complete view of
road-related sediment issues in the watershed.

A periodic (~five year frequency) resurvey of roads and culverts is recommended to monitor
road-related sedimentation over time.

6.5       CONCLUSIONS AND RECOMMENDATIONS
Based on the analysis presented here it appears that sediment delivery to streams from roads in
the Granite Creek Watershed is negligible. At the subwatershed scale sediment inputs are
estimated to be less than one percent over background levels. This is likely due in large part to
an aggressive and effective program of road decommissioning and maintenance of existing
roads. Road densities are low, and, with a few notable exceptions, roads are located away from
streams and outside of floodplain and wetland areas. Recommendations include the following:

         Decommission (if feasible) or modify road segments that are hydrologically connected to
          stream networks to eliminate or reduce road sediment delivery. Hydrologically
          connected road segments are identified in GIS (e.g., Figure 26). Highest priority should
          be given to those connected segments that receive highest traffic levels (Figure 43), and
          those that are in closest proximity to sensitive fisheries streams. In particular, connected
          road segments along the North Fork Granite Creek should be evaluated for removal or
          remediation.

         Future road construction should be designed to keep road densities low within fish-
          bearing tributaries, and roads should be located far enough away from sensitive
          floodplains, wetlands, and mass-wasting prone areas to avoid or minimize sediment
          delivery to streams. Rosgen channel types (Figure 4) provide a way to stratify streams
          that will need wider buffers from roads, with “E” channels having the most sensitive
          floodplain/wetland complexes and thus needing the widest buffers, followed (in order of
          decreasing buffer width) by “F”, “C”, “B” and “A” type channels.




Granite Creek Assessment                         WPN                                           Page 82
                                      7.0     WATER QUALITY

7.1       CRITICAL QUESTIONS
         What are the beneficial uses of water for streams within the Granite Creek watershed?
         What are the water quality criteria that apply to the stream reaches?
         Are the stream reaches identified as water quality limited segments on the 303(d) list by
          the state?
         Do water quality data indicate that water quality has been degraded or is limiting the
          beneficial uses?

7.2       METHODS
Information from the Idaho Department of Environmental Quality (IDEQ) and the Washington
Department of Ecology (WDOE) was reviewed to identify beneficial uses, water quality criteria,
and water quality limited status of stream segments in the Granite Creek watershed. Data from
the USFS and Kalispel Tribe were analyzed to further refine where water quality has been
degraded.

7.3       RESULTS
The IDEQ assigns water bodies one or more beneficial uses. The designated beneficial uses
within the Granite Creek There are eight beneficial uses assigned to streams within the Granite
Creek watershed38:

     Aquatic Life Use – Cold: Water quality appropriate for the protection and maintenance of a
      viable aquatic life community for coldwater species.
     Aquatic Life Use – Salmonid spawning (SS): Waters that provide or could provide a habitat
      for active self-propagating populations of salmonid fishes.
     Secondary contact recreation: applies to waters where people engage in activities where
      ingestion of water may occasionally occur, such as fishing, boating, wading, and infrequent
      swimming.
     Drinking water supply: water quality appropriate for drinking water supplies.
     Agricultural water supply: water quality appropriate for the irrigation of crops or as drinking
      water for livestock. This use applies to all surface waters of the state.
     Industrial water supply: water quality appropriate for industrial processes. This use applies to
      all surface waters of the state.
     Wildlife Habitats: The standards associated with this use are designed to protect water
      quality appropriate for wildlife habitat. This use applies to all surface waters of the state.
     Aesthetics: This use applies to all surface waters of the state.


38 Descriptions from http://www.deq.state.id.us/water/data_reports/surface_water/monitoring/beneficial_uses.cfm


Granite Creek Assessment                              WPN                                                Page 83
The IDEQ has divided streams in the Granite Creek watershed into three assessment units for the
purpose of assigning beneficial uses (IDEQ 2005; IDEQ 2008; Table 29; Figure 47). The
mainstem Granite Creek, and the North Fork Granite from Tillicum Creek down to the
confluence of the South Fork makes up the first unit. All eight beneficial uses have been
assigned to this unit. Of the eight beneficial uses only Aquatic Life – Cold and Aquatic Life –
Salmonid Spawning have been assessed. Both the 2002 Integrated Report (IDEQ 2005) and the
draft 2008 Integrated Report (IDEQ 2008) find that this assessment unit is not currently
supporting either beneficial use due to Thermal Modification (i.e., exceedance of water
temperature criteria). The remaining streams in the Granite Creek watershed are contained
within two additional assessment units (Table 29; Figure 47). Drinking water supply has not
been designated as a beneficial use in either of these units. Both the 2002 and 2008 Integrated
Reports find that the Aquatic Life – Cold, Aquatic Life – Salmonid Spawning, and Secondary
Contact Recreation uses are fully supported within these units; all remaining uses have not been
assessed.

Table 29. IDEQ Assessment Units in the Granite Creek watershed, and 2002 report status.
                                                       Status by assessment unit:
                                                        NF Granite (Willow Ck
                                                        to Tillicum Ck), lower
                               Mainstem Granite Creek, Tillicum Ck, SF Granite
                                     and NF Granite      (Sema Ck to mouth),
                                downstream of Tillicum      lower Sema Ck      All remaining tributaries
       Beneficial Uses         Ck (ID17010215PN022_04) (ID17010215PN022_03) (ID17010215PN022_02)
Aquatic Life Use -- Cold             Not supporting               Fully                   Fully
Aquatic Life Use --SS                Not supporting               Fully                   Fully
Secondary Contact Recreation         Not assessed                 Fully                   Fully
Drinking Water Supply                Not assessed                   -                       -
Agriculture Water Supply             Not assessed             Not assessed           Not assessed
Industrial Water Supply              Not assessed             Not assessed           Not assessed
Wildlife Habitats                    Not assessed             Not assessed           Not assessed
Aesthetics                           Not assessed             Not assessed           Not assessed
    Segment Size (miles)                 13.94                    10.44                  103.73




Granite Creek Assessment                         WPN                                        Page 84
Figure 47. IDEQ assessment map units within the Granite Creek watershed.



Water temperature standard applicable in the Granite Creek watershed are given in Table 30. All
streams are subject to the cold water and salmonid spawning standards. Several streams within
the watershed are subject to the Idaho Bull Trout standards, however, these are superseded by
Federal Standards in the principal streams (Figure 48).




Granite Creek Assessment                     WPN                                        Page 85
Table 30. Water temperature standards applicable in the Granite Creek watershed.
                                                                                                      Federal
                                                                                    3940                      41
                                                        State of Idaho Standards        :            Standard
                                                                Salmonid
                    Metric                      Cold Water      Spawning          Bull Trout         Bull Trout
   Max. Daily Max. Temperature (MDMT)           22 °C (72°F) 13 °C (55°F)            N/A                N/A
                                                                             13 °C (55°F) June
                                                                                   -August     10 °C (50°F)
     Maximum Weekly (7-day average)
                                                     N/A            N/A                           June -
     Maximum Temperature (MWMT)                                              9 °C (48°F) Sept – September
                                                                                     Oct
   Maximum Daily Average Temperature
                                               19 °C (66 °F) 9 °C (48°F)             N/A                N/A
                (MDAT)




39 http://www.deq.state.id.us/water/data_reports/surface_water/monitoring/temperature.cfm
40 http://www.deq.state.id.us/water/data_reports/surface_water/monitoring/wqs_epa_action.cfm#trout
41 Applies to streams specified in 40CFR 131.33 (http://ecfr.gpoaccess.gov/cgi/t/text/text-
idx?c=ecfr&rgn=div8&view=text&node=40:21.0.1.1.18.4.16.3&idno=40 )


Granite Creek Assessment                             WPN                                                 Page 86
Figure 48. Bull trout temperature standards and thermograph locations.



Data for the twenty thermograph locations was plotted (e.g., Figure 49) and summarized to show
the percentage of days in each year (June – September) that the Maximum Weekly (7-day
average) Maximum Temperature (MWMT) exceeded 10 degrees C (Figure 50). Results indicate
that most streams (even tributary and headwater locations) exceed the 10 degree C criteria most
of the time.




Granite Creek Assessment                     WPN                                        Page 87
Figure 49. Example thermograph for Granite Creek at the Road 302 bridge.




Granite Creek Assessment                WPN                                Page 88
                                                                                                                 100%

                                                                                                             90%

                                                                                                             80%

                                                                                                             70%
                                                                                                             60%
                                                                                                            50%




                                                                                                                              tag f ays
                                                                                                                        M M >10d C eg
                                                                                                                         ercen eo d
                                                                                                            40%
                                                                                                            30%
                                                                                                            20%




                                                                                                                          WT
                                                                                                            10%
                                                                                                            0%




                                                                                                                        P
                      2007
                       2006
                        2005
                         2004
                          2003




                                                                                                    ridge
                                                                                                    oad
                           2002
                                                                              raniteaboveTillicum
                                                                                                ache

                                                                                    ranite@319R
                                                                           raniteC @302B
                                                                       raniteC belowZeroC         k

                            2001
                                                                                              reek




                                                                     lacktail above1341road

                                                                                            ridge
                                                                                reek outh




                                                                               ranitebelowC
                                                                                         ridge
                                                                                          reek
                                                                                   9: ZeroC

                                                                                    edia k




                     Year
                                                                 12: Fedar C @m



                                                                    ranite@3112ndB
                                                              raniteC belowM C
                                                                           8: TillicumC
                                                                                     reek




                                                                                     reek
                                                               raniteC @638B
                                                                            rw reek




                                                               k Q tation(G A     R 1)
                                                                            acker C
                                                                       4: O igC

                                                 6: S aC belowTobasco
                                                                       thol reek




                                                                     19: S Fk. G
                                                            raniteaboveS a    em




                                                                                 k
                                                        outh em reek




                                                                 17: S Fk. G
                                                                18: N Fk. G
                                                                   1: A C




                                                                         reek




                                                                           .
                                                                       5: P




                                                                    20: G
                                                    2: M of S aC
                                                raniteB H Lake




                                                                       k




                                                                       .
                                                      raniteC W S




                                                                15: G


                                                                      .
                                                               14: B

                                                       16: S Fk. G
                                                        elow uff


                                                     em reek




                                                         13: G
                                                        11: G




                                                            .
                                            7: S ForkG


                                                10: G
                                                outh
                                   3: N ForkG
                                       orth




Figure 50. Percentage of days that the Maximum Weekly (7-day average) Maximum
Temperature (MWMT) exceeded 10 degrees C.



We developed a temperature model for the Granite Creek watershed using available thermograph
data from twenty sites located in the watershed (Figure 48, Figure 49, Table 31). Regression
analysis was used to determine the relationship between the maximum weekly (7-day average)
maximum water temperature (MWMT) and the environmental variables most likely to affect
water temperatures. The variables considered in the regression analysis were:




Granite Creek Assessment                                        WPN                                                                       Page 89
   Solar radiation. The seven-day moving average of total solar daily radiation calculated for
    the centroid of the watershed on the day of the annual MWMT. Solar radiation was
    calculated following the methods of Spokas and Forcella (2006) 42.

   Streamflow index. No data was available from the Upper Priest River gage (Table 4) for
    summer 2005-2007. Consequently mean daily flow on the day of the annual MWMT at
    USGS gage #12321500, Boundary Creek near Porthill Idaho 43 was used as an index of
    streamflow within the watershed.

   Air temperature. The seven-day moving average of daily maximum air temperature at the
    Bunchgrass Meadow SNOTEL station, located at an elevation of 5,000 feet in the headwaters
    of the South Fork of Granite Creek on the day of the annual MWMT.

   Site elevation. The elevation at the stream temperature-monitoring site as determined from
    digital elevation model data

   Distance from the watershed divide. The final variable used in the regression analysis was
    distance from watershed divide. Distance from the watershed divide provides an index of the
    time that water has been exposed to ambient air temperatures. The implication is that streams
    that have a shorter distance to the watershed divide would be expected to have lower water
    temperatures

   Effective shade. Average 2008 effective shade levels (expressed as a decimal) for the 2,000
    foot reach upstream of the temperature monitoring site. Values for effective shade calculated
    for the current condition (as discussed in the Riparian section, Stream Shading) were used in
    developing the initial regression equations.




42 The model is available at http://www.ars.usda.gov/services/software/download.htm?softwareid=62. Daily max
and min air temperatures and precipitation values are needed to run the model. Data available at
http://www.wrcc.dri.edu/fpa/gridded/ was used
43 http://waterdata.usgs.gov/id/nwis/nwisman/?site_no=12321500&agency_cd=USGS


Granite Creek Assessment                            WPN                                               Page 90
Table 31. Site location, responsible agency, and period of record for thermographs within
the Granite Creek watershed. Thermograph data and data plots are available for all
stations and years in Appendix E: Thermograph Data and Data Plots.

Map
 Id                 Thermograph Location                     Agency 2001 2002 2003 2004 2005 2006 2007
  1   Athol Creek                                             Tribe                                          a
  2   Mouth of Sema Creek                                     Tribe                                          a
  3   North Fork Granite Below Huff Lake                      Tribe                                          a
  4   Orwig Creek                                             Tribe                                          a
  5   Packer Creek                                            Tribe                                          a
  6   Sema Creek below Tobasco                                Tribe                                          a
  7   South Fork Granite above Sema                           Tribe                                          a
  8   Tillicum Creek                                          Tribe                                          a
  9   Zero Creek                                              Tribe                                          a
 10 Granite Ck WQ Station (GRA1)                              Tribe
 11 Granite Ck below Media Ck                                 USFS
 12 Fedar Creek @ mouth                                       USFS                                     b
 13 Granite Creek @ 638 Bridge (below Blacktail Ck)           USFS
 14 Blacktail above 1341 road                                 USFS
 15 Granite Ck below Zero Ck                                  USFS                         b
 16 S. Fk. Granite @ 311 2nd Bridge                           USFS
 17 S. Fk. Granite below Cache                                USFS
 18 N. Fk. Granite above Tillicum                             USFS
 19 S. Fk. Granite @ 319 Road                                 USFS
 20 Granite Creek @ 302 Bridge                               USFS
Notes:    a        Examination of the record indicates that the thermograph was installed too late in the season
          to capture the MWMT; data was not used in the analysis
          b        Examination of the record indicates that the gage was out of the water or otherwise
          malfunctioning at the time of the MWMT; data was not used in the analysis

A stepwise approach was taken to eliminate those variable from the regression equation that
were not statistically significant at the p < 0.05 level. The final form of the regression included
only the streamflow index and distance to watershed divide variables. Regression statistics for
the final equation are presented in Table 32.




Granite Creek Assessment                             WPN                                                Page 91
Table 32. Regression statistics for the MWMT prediction equation.

                      Adjusted R Square                          0.626768925
                           Standard Error                        0.784318429
                           Observations                               28
                                                                  Coefficients         P-value
                             Intercept                           15.00539519         1.43042E-20
                  Dist to drainage divide (feet)                 2.60307E-05         7.60716E-06
                     Streamflow Index (cfs)                      -0.022831918        0.000951017



7.4       INFORMATION GAPS AND MONITORING NEEDS
In the assessment presented above we attempted to build a stream temperature equation that was
specific to Granite Creek and its tributaries. The final form of the equation (Table 32) left over
1/3 of the observed variation unexplained, and did not include effective shade, the variable most
affected by management activities, as a significant variable. The primary reason for this poor
relationship was most likely due to the lack of temperature data across a wide range of stream
sizes and conditions. The majority of the data collected to date is from the mainstem Granite
Creek, and the lower portions of the North and South Forks. Data were collected in 2007 from
several tributary streams (Table 31), however, the thermographs were installed too late to capture
the peak summertime temperatures, and the data could not be used in this analysis.

Future monitoring recommendations include the following:

         Continue to collect data from established sites (Table 31). This will allow future
          modeling efforts to build on existing data sets.

         Add additional data collection sites that cover a wide range of stream sizes and effective
          shade conditions.

         Establish continuous stream gage locations on the mainstem Granite Creek (primary
          need) and tributaries. Streamflow is an important variable in explaining observed
          temperature variations (Table 32).

7.5       CONCLUSIONS AND RECOMMENDATIONS
The temperature modeling conducted as part of this analysis was inconclusive with respect to our
ability to make quantitative management recommendations. Future efforts will be needed to
refine temperature models for the watershed that can tie to management prescriptions. However,
we can several conclusions and recommendations based on a qualitative interpretation of the
results:

         Few consumptive water uses are present within the Granite Creek watershed, and upland
          watershed management appears to have little impact on summertime streamflow
          quantities (see Hydrology section)

Granite Creek Assessment                           WPN                                       Page 92
       Current effective shade levels are generally high (see Riparian section), and do not appear
        likely to change dramatically over the next ~100 year period.

       Given the above, it is unlikely that summertime water temperatures within the Granite
        Creek watershed are currently significantly different than what they were historically.
        However, additional monitoring and analysis would be necessary prior to delisting these
        streams as being water quality limited for temperature. The shades estimates generate as
        part of this analysis are adequate to use in future modeling exercises, however,
        continuous flow monitoring data will be needed to drive a temperature model, and more
        extensive thermograph and (if possible) FLIR data will be needed to calibrate the model.

Recommended future actions include:

       Eliminate human-caused activities that significantly reduce shade along fish-bearing
        streams. This will primarily involve identifying and removing road segments in riparian
        areas that limit effective shade, and restoring these areas to forested conditions.

       Avoid management activities that will decrease effective shade levels (e.g., riparian
        harvest, unless designed to promote shade development; road construction within riparian
        effective shade zones). There may be conflicts between maintaining adequate shade and
        placing trees for LWD enhancement. Development of a more robust temperature model
        (described above) would allow for case-by-case evaluation of temperature sensitivities
        associated with large wood placement. In those reaches where on-site removal of trees
        for instream use would be expected to significantly raise water temperatures (as estimated
        through temperature modeling) the instream wood should be brought in from off-site, and
        adjacent stands should be maintained.




Granite Creek Assessment                       WPN                                          Page 93
                                        8.0    FISHERIES

8.1     INTRODUCTION
The Fisheries Section of the watershed assessment will evaluate existing information and data on
aquatic habitat and fish populations with the goal of identifying limiting factors that have
contributed to the decline of native salmonid populations in the Granite Creek watershed.

Priest Lake once supported a healthy population of native salmonids that were used by Native
Americans, settlers, and recreational fisherman. The native fish community comprised of
westslope cutthroat trout, bull trout, and mountain whitefish spawned in tributaries of Priest Lake
and grew to large size in the lake. Introduction of non-native species of fish and invertebrates
combined with habitat degradation contributed to the substantial decline in the native fish
populations.

8.2     CRITICAL QUESTIONS
The Fisheries section will address the following critical questions:

      1. What is the historical fish distribution and the legacy effects of fishery management on
         the fish populations in Granite Creek.
      2. What is the fisheries habitat condition in the watershed and habitat limiting factors?
      3. What are the distribution and relative abundance of salmonid species that occur in the
         watershed?
      4. Where are the fish passage barriers in the watershed?


Section Organization

The report is organized in the order of the critical questions:

1) Information on the background of fisheries in Priest Lake,

2) Current fisheries management programs,

3) Methods used for data collection by the Kalispel Tribe and Forest Service

4) Results in order of habitat conditions, fish densities, and fish passage barriers. The data is
organized by subwatersheds – North Fork, South Fork, and Main Stem Granite Creek.

5) The last section will summarize the limiting factors by subwatershed.




Granite Creek Assessment                          WPN                                         Page 94
8.3   FISH SPECIES AND LIFE HISTORIES
Fish species present in the Priest Lake system are a combination of native and introduced
species. The species list below is based on the IDFG report by Mauser (1985), Idaho Fish
Collection (Orma J. Smith Museum, 2008), and Simpson and Wallace (1982).


                   Common Name                Scientific Name         Native/Introduced
              Westslope cutthroat trout      Salmo clarki lewisi           Native

                       Bull trout           Salvelinus confluentus         Native

                  Mountain whitefish        Prosopium williamsoni          Native

                     Brook trout             Salvelinus fontinalis       Introduced

                      Lake trout            Salvelinus namaycush         Introduced

                       Kokanee              Oncorhynchus nerka           Introduced

                    Slimy sculpin              Cottus cognatus             Native

                   Pygmy whitefish           Prosopium coulterii           Native

                Northern pikeminnow       Ptychocheilus oregonensis        Native

                   Longnose dace           Rhinichthys cataractae          Native

                   Redside shiner          Richardsonius balteatus         Native

                  Longnose sucker          Catostomus catostomus           Native

                  Largemouth bass          Micropterus salmoides         Introduced

                  Smallmouth bass           Micropterus dolomieu         Introduced

                    Yellow perch              Perca flavescens           Introduced

                       Bluegill             Lepomis macrochirus          Introduced


The predominant species found during fish surveys in Granite Creek now are westslope cutthroat
trout, brook trout, and sculpin.

Westslope Cutthroat Trout

Westslope cutthroat trout (cutthroat) were historically a dominant species in streams throughout
central and northern Idaho. They are thought to have evolved in coexistence with bull trout,
mountain whitefish, and northern pikeminnow (Rieman and Apperson 1989). Cutthroat mature
at 4 or 5 years of age and spawn entirely in streams. Cutthroat trout have both resident and
migratory life histories. Most cutthroat trout in Priest Lake are adfluvial; spawning begins in
April and generally ends by mid-June (IDFG 2007).



Granite Creek Assessment                         WPN                                      Page 95
Eggs are deposited in stream gravels where the developing embryos incubate for several weeks.
Several days after hatching from the egg, the cutthroat fry, about 2.5 cm long, emerge from the
gravel and disperse into the stream. Spawning habitat for cutthroat trout occurs in low-gradient
stream reaches that have gravel substrate ranging from 2 mm to 75 mm in diameter, water depths
near 0.2 m, and mean water velocities from 0.3 to 0.4 m/sec. Proximity to cover (e.g.,
overhanging stream banks) is an important component of spawning habitat for adult cutthroat
trout. Non-native brook trout are considered a significant competitor to cutthroat trout (USFWS
1999).

Bull Trout

Priest Lake bull trout are adfluvial, with most fish maturing at age-5 or age-6, and entering
spawning tributaries as early as May to spawn in September. Bull trout generally live in tributary
streams for two or three years before migrating to lakes, and have a life expectancy of 10 or
more years. During August and September, when surface temperatures reach 68°F, bull trout in
Upper Priest Lake and Priest Lake occupy depths of >50 ft where temperatures range from 45-
55°F. When surface temperatures are below or near 55°F in the spring and fall, bull trout can be
found closer to the surface (IDFG 2007). Historically, bull trout were common in the Priest Lake
basin and most of the major tributaries supported large spawning runs. By 1985, adfluvial bull
trout runs into tributaries of Priest Lake were essentially gone, and the only strong number of
bull trout occurred in the Upper Priest Lake basin (IDFG 2007). The decline in bull trout is
primarily attributed to predation and/or competition with the non-native lake trout.

Brook Trout

Brook trout are a char, like bull trout, but were introduced from eastern United States. Brook
trout reach sexual maturity in 2 or 3 years. Spawning takes place in the fall, usually in late
September and October. Like other trout species they seek out gravel areas in small streams or
along lake shores. The fry emerge from the gravel in April and May and move into deeper pools
or lakes. The primary food items of brook trout are aquatic insects and other small invertebrates.
Large individuals eat small fish including other brook trout (Simpson and Wallace 1978). The
impacts from historic logging activity (e.g. elevated stream temperatures and fine sediment
levels) in the Priest Lake tributaries is thought to favor brook trout since brook trout out-compete
cutthroat trout and bull trout in degraded habitat (IDFG 2007). Habitat degradation combined
with competition from brook trout is an obstacle to recovery of cutthroat trout and bull trout in
the Priest Lake system. Furthermore, brook and bull trout can hybridize thereby potentially
wasting reproductive effort of sensitive bull trout stocks.

8.4   HISTORICAL FISHERIES INFORMATION
Current fish populations in Granite Creek are the result of past management of the fishery and
landscape in Priest Lake and the watershed. Historically Priest Lake supported a highly valued
native fishery comprised of adfluvial populations of westslope cutthroat trout, bull trout, and
mountain whitefish. Introductions of lake trout, kokanee, and Mysid shrimp changed the
dynamics in the fishery and led to the collapse of the native fisheries over time. Priest Lake



Granite Creek Assessment                       WPN                                           Page 96
continues to support diminished bull and cutthroat trout populations but is dominated by the
introduced lake trout.

North of the Narrows (An anecdotal history)

The book, North of the Narrows (Simpson 1981), provides oral histories transcribed by Claude
and Catherine Simpson from long time residents of Priest Lake and provides some insight into
the fishery in Priest Lake. As the introduction to the fish chapter states, “even if half true, or
less, the many fish stories gives credence to the fact that Priest Lakes and feeder creeks were at
one time a fisherman‟s paradise”.

Cutthroat trout fishing was the attraction for vacationers and resorts in the teens and 1920‟s.
“Many times I have taken several fly fishermen to the little lake in my launch for evening
fishing. The launch had a long deck and three or four good flycasters could fish at the same
time. What a thrill it was to have three or four 16” cutthroats on the lines at the same time.”
Simpson (1981) provides several historic photos of large catches in his book.

Whitefish were once abundant in Priest Lake and tributaries. “The whitefish are gone. The
famous whitefish runs, in the first half of the 20th century, reported and witnessed by many
present day Priest Lakers, are a thing of the past”. “The demise of the whitefish that supplied
winter food for local inhabitants as well as Indians from Montana, Washington, and Idaho, is a
direct result of the white man‟s desire to interfere with nature‟s plan of survival” (Page 171,
Simpson 1981).

North of the Narrows contains little mention of fishing for bull trout. One story, described as the
tallest fish tale in the Kaniksu was taken from A.K. Klockman‟s diary. Bill Stoner and
Klockmann took a “white fly” (a stick of dynamite) to American Falls on Upper Priest River to
get a big char they had previously seen in the pool at the foot of the falls. “Stoner, fully six foot
two inches tall, put a heavy stick through the gills and carried him over his shoulder. To give
you an idea of the size of the fish, his tail was dragging on the ground behind Stoner.” “We
butchered him like an animal … and he had several large trout in his stomach.”

Fisheries Management History

The first scientific study of fisheries in Priest Lake was completed by Ted Bjornn, Idaho Fish
and Game, in 1955-1956 (Bjornn 1957) to investigate the decline in sport fishing. Numerous
studies have been completed by IDFG since then to evaluate cutthroat, kokanee, and lake trout
fisheries in Priest Lake. Specific references to Granite Creek are listed in Table 33.

The development of Priest Lake country as a recreational area began as early as 1905 with a
hotel at Coolin at the south end of the lake, and the 1930‟s the resort business began to boom. In
spite of the light fishing pressure the quality of the cutthroat trout fishing declined sometime in
the period between the late 1930‟s and 1940‟s. During the 1955 and 1956 survey, only about
5,000 cutthroat a year were caught and the size of fish declined from the legendary 14 to 15-inch
spawner to mostly immature fish averaging approximately 11 inches in length (Bjornn 1957).



Granite Creek Assessment                        WPN                                           Page 97
Through the years, six exotic species of game fish have been introduced into Priest Lake. In
1925, lake trout (mackinaw trout) was added to Priest Lake. Brook trout were introduced before
the 1920‟s. Rainbow trout were planted in the lake one or two times but no specimens were
found in the 1955-1956 survey. Introductions of kokanee to Priest Lake were made from 1942 to
1944. Kokanee became the most abundant fish in the catch (Bjornn 1957), but the kokanee
populations declined in the 1960‟s following the introduction of Mysid shrimp. Although no
record of stocking exists, largemouth bass and sun fish populations have been established in
Priest Lake.

The effect of non-native species introductions on Priest Lake up to 1983 was addressed in the
IDFG report (Mauser and Ellis 1985) as summarized below:

Cutthroat trout were the primary game species at Priest Lake until the 1950's; however, by the
late 1940's cutthroat fishing was reported to be fairly poor. The average size of fish reported in
the creel in 1956 were immature cutthroat trout 28 cm long (Bjornn 1957). The average size of
cutthroat in the catch increased slightly to 31.5 cm by 1978; however, catch rates dropped from
former levels of 0.55 fish per hour to 0.20 fish per hour. Cutthroat catches have fluctuated
around 2,500 fish annually since census programs were initiated in the mid-1950's.

In the 1960's, an extensive cutthroat trout rehabilitation program involving eyed egg plants
combined with fry and fingerling releases failed because westslope cutthroat trout were not
available. Beginning in 1982, the cutthroat trout limit was reduced to 2 fish, none under 38 cm.
Major tributary streams were closed to protect juvenile cutthroat which rear 2 to 3 years before
entering the lake at 13 to 18 cm. A stock of cutthroat trout derived from Priest Lake in the early
1940's was used for releases in the drainage beginning in 1981.

Kokanee were introduced into Priest Lake in the early 1940's and quickly became the most
abundant game fish, replacing cutthroat trout. By 1956, kokanee were the predominate forage
species for bull trout and lake trout and provided a harvest of 100,000 fish to the sport fishery
(Bjornn 1957).

Opossum shrimp (Mysis relicta) were released in Priest Lake in 1965 to provide additional
kokanee forage. By the early 1970's, mysids were fully established and large kokanee were
caught, including a United States record fish of 3.1 kg in 1975. The enhanced growth of these
fish was due to the mysis food supply (Rieman and Lukens 1979). However, only a fraction of
the kokanee population fed on mysids, and in the late 1970's the population collapsed abruptly.

In 1978 only 4,500 kokanee were harvested from Priest Lake. A combination of mysid-related
problems and depensatory predation by the enhanced lake trout population drove the kokanee
population to functional extinction. The fishery is not expected to recover without releases of
kokanee fry timed to coincide with mid-summer zooplankton pulses and in large enough
numbers (5-10 million) to provide prey for lake trout.

Lake trout were introduced in Priest Lake in 1925. Their contribution to the fishery was minimal
until 1952 when kokanee provided the forage for a trophy fishery (Bjornn 1957). The state



Granite Creek Assessment                      WPN                                          Page 98
record fish of 26 kg was caught from Priest Lake in 1971 but survival of juvenile lake trout was
low and only a few hundred fish were harvested annually.

Lake trout responded rapidly to the successful introduction of mysis shrimp. In 1978 the catch
increased to almost 6,000 fish. Though mysids provided excellent forage for smaller lake trout,
the condition of larger fish accustomed to a fish diet declined in the absence of kokanee. Lake
trout of all sizes have become better table fare due to the absence of kokanee and the
predominantly mysis diet. Previously, the larger fish were unpalatable due to extreme amounts of
body fat, and flesh quality was poorer.

The bull trout population has declined dramatically since the mid-1970's. Collapse of the
kokanee fishery shifted over half of the 80,000 to 90,000 hours of angling effort to a combined
bull trout and lake trout fishery. Bull trout were apparently over-harvested in a short period of
time. In 1982, a 51 cm minimum length limit was established for bull trout. Tributary streams
have been closed to the taking of bull trout since the early 1970's; however, adult fish are
vulnerable to illegal harvest in the streams during low water conditions. Adult bull trout enter the
spawning streams as early as May, with the majority of spawning taking place during September
(Bjornn 1957).

Other potential impacts to fish in Granite Creek during the early period was the practice of taking
eggs, illegal harvest, and removal of log jams. A fish trap was operated in Granite Creek from
1939 to 1947 to take eggs from cutthroat trout, and local fishermen believed this was a
contributing factor to the decline of cutthroat in Priest Lake. Bjornn (1957) thought that the
mortality from egg taking had little effect on number of spawners in subsequent years since only
a very small percent of the cutthroat in Priest Lakes live to spawn a second time. Bjornn thought
that illegal harvest in tributaries was a problem since fish were caught before they had a chance
to deposit their eggs. Mallet (1965) reported clearing log jams in tributaries of Priest Lake but
not specifically in Granite Creek.




Granite Creek Assessment                       WPN                                           Page 99
Table 33. Background information on fisheries in Granite Creek.
    Year                                 Information                                          Source
1916 – 1917    Cutthroat Spawners.                                                    Bjornn, 1957
               A reference to Granite Creek indicating early egg take from            page 60
               cutthroat in 1916 and 1917 in Granite Creek.
               (Note: Historical information quoted in Bjornn, 1957, but no
               citation provided.)
1939 - 1947    Fish Trap. The fish trap on Granite Creek, located 6 miles from        Bjornn, 1957
               the mouth. was operated from 1939 to 1947 during which 5
               million eggs were taken.
               During 1947, the last year spawn was collected, 1,600 spawners
               were caught at the Granite Creek trap, located six miles up from
               the lake.
1956           Temperature: A maximum-minimum thermometer placed in                   Bjornn, 1957
               Granite Creek near the mouth recorded maximum temperatures             page 13
               from 60 to 70 degrees F. and minimum temperatures between
               50 to 55 degrees F. for the month of August. By the end of
               September the maximum temperatures were near 50 degrees F.
               and the minimum temperatures varied between 40 and 45
               degrees F. Daily variation between the maximum and minimum
               temperatures was as much as 10 degrees F. on some days.
1955 - 1956    Catch Composition – High Brook Trout Numbers.                          Bjornn, 1957
               Estimated catch reported by fisherman for Indian Cr., Granite          Page 23
               Cr., and Kalispel Creek in 1955 indicates the high number of
               brook trout in the tributaries.

                                       Cutthroat Trout 1,300
                                       Brook Trout     4,900
                                      Bull Trout          100
                                       Total            6,300
1956           Granite Creek Summary. Granite Creek is the probably the best          Bjornn, 1957
               tributary for spawning which enters Priest Lake. Minimum flows         P 165.
               are adequate, and the creek is free of serious obstacles for 16
               miles. The low gradient allows for extensive gravel areas suitable
               for spawning. Cutthroat spawners were much lower in 1956 than
               when the fish trap was operated in the 1940’s.
               Cutthroat, dolly varden (bulltrout), and brook trout were
               numerous in the main stem and very abundant in the tributaries.
               Many fish in the tributaries appear to be resident fish.
               Granite Creek and its tributaries probably contribute a larger
               number of cutthroat to the lake fishery than any other creek.
1956           South Fork Granite Creek Summary.                                      Bjornn, 1957
               The South Fork and North Fork come together to form Granite            p. 165
               Creek approximately 9 miles from the mouth. The South Fork
               has a large amount of suitable gravel near the forks but the
               bottom type changes between 1 to 2 miles upstream from the
               forks to large rocks and boulders type yielding very little suitable
               spawning area. Small fish are very abundant with cutthroat more
               abundant near the forks and brook trout more abundant in the
               headwaters.
1956           North Fork Granite Creek Summary.                                      Bjornn, 1957
               The North Fork contains large areas of good spawning gravel            p. 165
               interspersed with bottoms of silt, sand, and bedrock. Small fish
               are plentiful with cutthroat being the dominant species.
1956           Blacktail Creek Summary.                                               Bjornn, 1957
               Blacktail Creek, entering Granite Creek at the fish trap, has          p. 165
               some accessible spawning area near the mouth. The upper



Granite Creek Assessment                               WPN                                             Page 100
    Year                                 Information                                          Source
               reaches (2 miles from the mouth) contain typical resident
               populations, with the fish maturing at 5 to 7 inches in length.
1956           Log Jams.                                                             Bjornn, 1957
               A small study evaluated the effect of log jams on insect              Page 172
               productivity. It’s useful to note that the log jam that was studied
               was on Granite Creek.
1971           Fish Caught by Angling.                                               Irizarry, 1972
               For 50 hours of fishing in Granite Creek, June and July 1971,
               IDFG personnel caught 14 cutthroat, 1 eastern brook trout, and
               88 rainbow trout. Cutthroat trout were dominant in other
               tributaries fished.
1971           Spawning Tributary Observations.                                      Irizarry, 1972
               1956: Granite Creek and its tributaries probably contribute a
               larger number of cutthroat to the lake fishery (Priest Lake) than
               any other creek.
               1971: No large migrant spawners observed between Blacktail
               Creek Road and Athol Creek. Only very limited observations and
               sampling on lower three miles because of access and high water
               difficulties.
               Spawning gravel appears suitable in this lower section but no
               large cutthroat observed or caught. Rainbow-cutthroat
               hybridization prominent in stream (observed, but unrecorded
               data). Contribution to lake fishery cannot be confirmed or denied.
1971           Habitat Conditions Discussion.                                        Irizarry, 1972
               General to Priest Lake tributaries, not specific to Granite Creek.
               Probably the reason that we caught few fish on the lower ends of
               these creeks was because degraded habitat precluded the
               presence of fish. In many instances, heavy deposits of sand
               have destroyed the spawning gravels, filled in pools, and altered
               the food producing potential of these streams.
               At times, I found apparently suitable spawning rubble, but a
               closer inspection showed heavy sand deposits between the
               rocks. Not only is this highly unsuitable for spawning but results
               in poor survival of eggs and fry. Also, few aquatic insects can
               survive to provide food in this type of environment.
               It appears that we now have fewer cutthroat spawners, higher
               egg mortality, fewer pools, and less food for juvenile cutthroat
               which could be hampering contribution to the lake cutthroat
               fishery.
               The primary contributors of sediments to streams in the Priest
               Lake area are: (1) The erosion from roads, bridges, and logging
               skid trails and (2) Erosion after sizable forest fires such as the
               Trapper Peak and Sundance Fires of 1967.
               ―I can see little value in building stream improvement structures
               and clearing out log jams to improve lake cutthroat spawning
               habitat when the streams contain sediment from improper land
               management. Rather, I feel that the best stream improvement
               procedures is proper land use.‖

1983 - 1987    Trout Enhancement Studies.                                            Mauser, Vogelsang,
               IDFG evaluated the fishery in Priest Lake between 1983 and            Smith. 1988
               1987.
               Bulltrout redds were counted in Granite Creek annually as
               summarized separately in the text below.

               Primary conclusions from the 1988 report: 1) Lake trout fishery
               was likely to remain fairly stable depending on kokanee forage
               availability. 2) Elimination of the consumptive fishery for
               cutthroat trout was necessary to reverse the population decline.
               3) Estimated kokanee population did not provide a fishery in



Granite Creek Assessment                               WPN                                                Page 101
   Year                                 Information                                     Source
               1986 and produced a poor outlook for the future. 4) Bull trout
               population showed no response to the closure of the fishery in
               1984.
1987           Fish Densities in Granite Creek and tributaries.                 Irving, MS Thesis. 1987
               Extensive data on fish densities in Priest Lake tributaries
               including Granite Creek,
1995           Fisheries Management Summary.                                    Davis and Horner, 1995
               Between 1989 and 1991, over 235,000 fingerling cutthroat
               trout were stocked in Priest Lake in an attempt to
               reestablish a limited consumptive cutthroat trout fishery.
               This program was discontinued in 1992 due to virtually no
               return to the creel of these fish.
               Management direction for Priest Lake changed in 1992 to
               reflect the existing fish species composition and inability to
               reestablish kokanee salmon and cutthroat trout fisheries
               by stocking. Regulations were changed on lake trout to
               increase the harvest of smaller fish while allowing some
               fish to escape into the slot limit and grow to trophy size.

Bull Trout in Granite Creek

Bull trout were abundant and supported a good fishery in Priest Lake until 1978, followed by an
abrupt population decline (Mauser and Ellis 1985). IDFG evaluated bull trout populations in
tributaries of Priest Lake from 1983 to 1986 (Mauser and Ellis 1985), and made redd counts at
index sites in the basin. The index site for Granite Creek is the N.F. Granite Creek, located from
the confluence with the South Fork, upstream to Granite Creek Falls. Redd counts at the N.F.
Granite Creek site were 14 in 1983 (Figure 51), 5 in 1984, 6 in 1985, and 10 in 1986. This site
was revisited by USFWS staff in 2008, in which 2 redds were counted (Personal
Communication, Ryan Hardy, IDFG, December 2008).

In September and October, 1983, a weir operated 150 meters above the mouth on Granite Creek
recorded upstream and downstream migration of bull trout (Mauser and Ellis 1985). Juvenile
and adult bull trout were accessing Granite Creek in 1983 as indicted by the length-frequency
figure (Figure 52) constructed by Mauser and Ellis (1985).




Granite Creek Assessment                             WPN                                              Page 102
Figure 51. Location of bull trout redds in surveyed sections of the North Fork Granite Creek,
Priest Lake, 1983 (from Mauser and Ellis 1985).




Granite Creek Assessment                    WPN                                      Page 103
Figure 52. Length-frequency distribution for bull trout at the weir in Granite Creek, mid-
July to early September, 1983 (from Mauser and Ellis 1985).



8.5   FISHERIES MANAGEMENT PROGRAMS
Fish populations in Granite Creek are influenced by management actions of Idaho Fish and
Game, US Fish and Wildlife Service programs, and US Forest Service land management
programs.

8.5.1 Idaho Department of Fish and Game

Idaho Department of Fish and Game (IDFG) management direction for Priest Lake and
tributaries is described in the Priest River Drainage section of the department‟s Fisheries
Management Plan 2000-2012 (Page 133 – 142, IDFG 2007). The Fisheries Management Plan
describes the decline in native fish populations over time and efforts to restore the fisheries.


Granite Creek Assessment                     WPN                                        Page 104
Historically, the westslope cutthroat trout fishery was very popular in Priest Lake when twenty
fish limits of 15 to 20 inch cutthroat trout were common. Cutthroat trout fishing began to
deteriorate as early as the 1930‟s and 1940‟s with the decline attributed to excessive harvest,
mining of adult spawners for hatchery take, competition with kokanee and brook trout and
degradation of spawning habitat. By the late 1980‟s, lake trout predation was believed to be the
major factor suppressing the cutthroat trout.

In addition to lake trout predation, IDFG identifies competition with brook trout in tributary
streams as an obstacle to cutthroat trout and bull trout recovery. Brook trout are known to
outcompete cutthroat trout and bull trout in lower gradient streams or streams with higher fine
sediment. Habitat degradation through reduction of large wood and increasing fine sediment
may increase brook trout‟s competitive advantage.

Department Management Plan
IDFG believes that the high cost and uncertainty of success in removing lake trout coupled with
the reduced productive capacity of tributary streams makes native fish restoration in Priest Lake
problematic and intends to focus the department‟s native fish restoration efforts on Upper Priest
Lake. The Fishery Management Plan identifies six objectives for managing the fishery. (Note
that Objective Number 3 specifically addresses habitat improvement in tributaries.)

        1. Objective: Restore native fish populations in Upper Priest Lake.
        Programs: Remove lake trout with gill nets and other means, monitor lake and bull trout
        populations, evaluate effectiveness of netting and blocking movement in the thorofare to
        remove lake trout, maintain catch-and-release fishing for cutthroat trout and bull trout,
        investigate harvest fishery on kokanee to reduce competition with cutthroat.
        2. Objective: Shift management emphasis in Priest Lake to lake trout, to provide both a
        yield and trophy fishery.
        3. Objective: Protect the cutthroat trout and bull trout fishery in Priest Lake.
        Programs: Preserve genetic integrity of wild, native cutthroat trout and bull trout by
        maintaining catch-and-release fisheries in the lake and limited harvest in the tributaries.
        Work with the Forest Service and Idaho Department of Lands to improve habitat
        conditions in tributary streams.
        4. Objective: Provide a limited consumptive harvest of kokanee in Priest Lake.
        5. Objective: Provide information and education of fisheries management objectives in
        the Priest River watershed.
        6. Objective: Reduce impacts of smallmouth bass on more desired game fishes.



8.5.2 US Fish and Wildlife Service

The distinct population segments (DPS) of bull trout in the Columbia Basin were listed as
threatened under the Endangered Species Act in 1998 (63 FR 31647 June 10, 1998). A Recovery
Plan was drafted in 2002 (USFWS 2002, 67 FR 71439), but has not been finalized. In 2004, the
USFWS published a final rule designating critical habitat for bull trout which was challenged


Granite Creek Assessment                            WPN                                               Page 105
and revised in 2005 (70 FR 56212 September 26, 2005). In April 2008, the Service released the
results of a five-year status review for bull trout, recommending retention of the species
threatened status.

The Recovery Plan (USFWS 2002) divides the distinct population unit of bull trout into
Recovery Units, Subunits, Core Areas, and Local Populations. Granite Creek occurs within
Priest Subunit of the Clark Fork Recovery Unit. The USFWS identified the causes for decline of
bull trout as the combined effects of habitat degradation and fragmentation, the blockage of
migratory corridors, poor water quality, angler harvest and poaching, entrainment into diversion
channels and dams, and competition with and displacement by nonnative fish species.

Bull trout are present in small numbers in Upper Priest Lake, and the Recovery Plan asserts that
unless bull trout are successfully restored in Upper Priest Lake it is doubtful that recovery can be
achieved in Priest Lake. Brook trout are widely distributed throughout the historic range of bull
trout in the Priest Subunit, present in nearly all known spawning and rearing streams, including
Granite Creek and its tributaries. Historic and current forestry practices and transportation
networks pose threats to bull trout, particularly in the half of the subunit where soils are
classified as highly erodible. Road densities in the Granite Creek drainage are relatively low
(less than 1.2 miles per square mile); however, the Recovery Plan states that road crossings
present passage barriers throughout the Priest Subunit, including Granite Creek and North Fork
Granite Creek. Roads in Granite Creek and elsewhere in the Subunit encroach on riparian areas,
posing threats associated with sediment input into streams and channel confinement. In addition,
residential development, mostly associated with seasonal recreational use, may increase
sedimentation and degrade water quality.

The Service‟s 2008 Five-Year Review for bull trout classifies the threat ranking for the Priest
Core Area as “substantial and imminent”, and ranked the species as being at high risk in the
Subunit overall. For the Priest Subunit, the Service objective is to have five local populations of
more than 100 bull trout, for a total of 1,000 total adults. Specific recovery actions for the Priest
Subunit include the following:

Priority 1 Actions

           Reduce sediment sources. Granite Creek is among the watersheds identified as
            priority for this action.
           Identify suitable unoccupied habitat.
           Eliminate passage barriers—reduce or remove culverts or bridges that impede
            passage in drainages including Granite Creek and South Fork Granite Creek.
           Conduct watershed problem assessments for watersheds that have not been
            evaluated to identify site-specific threats to bull trout. Quantify populations, numbers,
            and trends for several watersheds including Granite Creek.
           Evaluate and prevent overharvest and incidental angling mortality of bull trout.
           Design and implement a standardized monitoring to assess the effectiveness of
            recovery efforts affecting bull trout and their habitats.

        Priority 2 Actions

           Improve maintenance along transportation corridors.



Granite Creek Assessment                           WPN                                                  Page 106
             Improve instream habitat by increasing amounts of large woody debris in several
              drainages, including Granite Creek.
             Implement control of non-native fishes where found to be feasible and appropriate.
             Use partnerships and collaborative processes to protect and conserve bull trout and
              bull trout habitats.

        Priority 3 Actions


             Minimize recreational development in bull trout spawning and rearing habitat

8.6   METHODS
8.6.1 Data Collection

The Kalispel Tribe and Forest Service collected fisheries habitat and fish population data from
2004 to 2008 as the basis for the watershed assessment using the following methods.

Fisheries Habitat

Habitat Inventory. Habitat data was collected using the R1/R4 Fish and Fish Habitat Standard
Inventory Procedures (Overton and others 1997). A total survey of habitat conditions is
evaluated by walking up the stream, measuring habitat dimensions, patterns, and habitat quality
descriptors. The R1/R4 habitat data can be evaluated at the stream reach, subwatershed, and
watershed scale.

Rosgen Stream Classification. Stream channels were classified to Rosgen Level II
classification for each reach (Rosgen 1996) based on field collection of channel features.
Wolman Pebble Counts were sampled as described in Rosgen (1996) as part of the classification
procedure.

Large Woody Debris. Large woody debris (LWD) was evaluated using the procedure described
in Davis and others (2001). This method quantifies the number of individual pieces of wood and
the number of debris dams in a reach (Stage 1). LWD is defined as the wood (sticks to logs)
over 1 meter in length and at least 10 cm in diameter at one end. When multiple pieces
accumulate in the stream channel it is counted as a debris dam.

The method also evaluates the function of the wood (Stage 2) in influencing stream morphology,
hydrology, and organic material retention by calculating a LWD functional index or LWDi.

The functional influence of the wood is evaluated by scoring each piece and debris dam for their
relative influence on the channel.

Each LWD piece is scored from 1 to 5 with respect to seven factors: 1) Length/bankfull width, 2)
Diameter, 3) Location in the channel, 4) Type, 5) Structure, 6) Stability, and 7) Orientation.
Debris Dams are rated from 1 to 5 based on: 1) Length (% of bankfull width), 2) Height (% of
bankfull width), 3) Structure, 4) Location and 5) Stability. The scores for individual factors are


Granite Creek Assessment                           WPN                                              Page 107
summed to provide an LWDi score for individual pieces and an LWDi for debris dams. The
scores are a function of both the number of pieces (or debris dams) and functional rating. To
compare these scores between channels the results are reported as Percent of the Potential Score.

An example illustrates the LWDi calculations:

Individual Piece Count – 44.        Debris Dam Count = 9

Length/Bankfull Width Score = (1*6)+(2*6)+(3*8)+(4*13)+(5*11) = 149

Potential Score for Pieces = # of Pieces x # of Factors x Highest Score = 44 x 7 x 5 = 1,540

Percent of Potential Score for Pieces = 931/1,540 x 100 = 60%

      PIECES               1    2       3       4      5       LWDi Score
   Length/Bankfull
       Width                6   6        8     13      11         149
      Diameter              8   6       11      4      15         144
      Location             11           14     17       2         131
        Type               17           15      7       5         115
      Structure            39            5              0          54
      Stability             4           14             26         176
     Orientation            7    4       4     10      19         162
        Total              92   16      71     51      78         931
   DEBRIS DAMS
       Length              0    3       2       0      4            32
       Height              0    3       3       0      3            30
      Structure            3            1              5            31
      Location             1    0       3       0      5            35
      Stability            1            4              4            33
        Total              5    6       13      0      21         161



Spawning Habitat. Spawning habitat was evaluated using the methods described in the
Clearwater Stream Methodology (Espinosa 1988). All potential spawning habitat is estimated
using a minimum gravel size patch of 0.5 square meters and summed into number of square
meters by quality rating of good, fair, and poor.

Spawning habitat is characterized by three basic components:

    1. Clean spawning size gravels, free of excessive sand and fine sediments.

    2. The proper gravel size for resident fish ranges from 0.6-6.4 cm; optimum size will range
       from 1.3-3.8 cm (Espinosa 1988).

    3. Located within close proximity to cover and resting pools.


Granite Creek Assessment                         WPN                                     Page 108
    4. Desirable spawning habitat will include a pool area of sufficient depth (> 0.6 m) and
       cover (undercut banks, logs, boulders) to provide shelter and rest within 30.5 m up or
       downstream of the spawning gravels.

    5. Current velocities of 0.2 - 0.6 m (0.5 - 2.0 ft) per second are considered optimum and
       desirable for spawning activity and incubation requirements.

                           Table 34. Spawning Gravel Condition Rating.

                            Spawning Gravel Condition                Rating
                    All three components of spawning habitat
                                                                     Good
                            exist in optimum conditions
                    Two components of spawning habitat exist
                        or conditions for any of the three            Fair
                         components are not optimum.
                    Two components of spawning habitat exist
                                                                     Poor
                          and conditions are impaired.



Fish Population Sampling

Relative abundance data was collected using both electrofishing and snorkel counts following the
methods described in Overton and others (1997).

Fish Passage Barriers

Fish passage barriers were identified in a GIS shape file (Granite_d8_fish_barrier.shp) provided
by the US Forest Service. No further information about this data or current status of these
barriers is known.

Information on natural barriers was taken from a University of Idaho MS Thesis (Irving 1987).
The study evaluated production of westslope cutthroat trout in tributaries to Priest Lake, so the
barriers were evaluated as upstream blockages to adfluvial cutthroat trout. These barriers were
located by their description in the thesis table as distance in meters from the mouth of the stream.
Although there may be some degree of inaccuracy in mapping these distances, this information is
very useful in understanding the potential extent of spawning habitat available to adfluvial
cutthroat trout.

8.6.2 Data Evaluation

Measured habitat conditions are evaluated in relation to the habitat requirements for westslope
cutthroat trout and bull trout. Although habitat requirements are generally well understood –
clean sediments, quality pools, cover, and temperature – quantitative measurements from other
studies describing preferred habitat are either site specific to the study site or are not directly
comparable due to differences in specific monitoring methods. For this reason, we will use data
collected by the Kalispel Tribe for reference reaches within the Priest Lake system where


Granite Creek Assessment                        WPN                                         Page 109
feasible. Habitat conditions are reviewed briefly below to provide comparison to previous
studies.

Bull Trout

Bull trout distribution or abundance is influenced by channel stability, substrate, cover,
temperature, and the presence of migration corridors (Rieman and McIntyre 1993). In addition,
maintaining diversity of habitats is important to bull trout conservation. Rieman and McIntyre
noted that they could not define clear limits or thresholds for habitat conditions that directly
control the distribution and abundance of bull trout.

Other biologists have attempted to quantify habitat thresholds or desired future condition for bull
trout. The USFWS Bull Trout Matrix sorts habitat quality into three categories for determination
of project effects: 1) functioning appropriately, 2) functioning at risk, and 3) functioning at
unacceptable risk. Habitat characteristics in the matrix include water temperature, percent
surface fines in spawning areas, embeddedness in rearing areas, bank stability, LWD frequency,
off channel habitat, pool frequency, frequency of deep pools, and width/depth ratios (USFWS
1993). Dambacher and Jones (1997) developed ranges of habitat quality benchmarks (low,
moderate, high) for bull trout from Oregon Department of Fish and Wildlife Habitat surveys for
percent shade, riffle gravel, bank erosion, undercut banks, riffle fines, and LWD frequency. In
summarizing a number of studies, Dunham and Chandler (2001) noted that it is often believed
that bull trout select larger habitats (e.g., > 2 meter width) with low levels of fine sediment, with
cooler temperatures, and higher levels of shade, large wood, undercut banks, and deeper water.

Dunham and Chandler (2001) evaluated eight habitat variables and two biological variables in
Washington State to predict bull trout occurrence: bankfull width, LWD frequency, undercut
bank area, maximum depth, maximum water temperature, stream gradient, mean surface fines,
brook trout density, and rainbow/cutthroat trout density. Of the habitat variables evaluated, they
found only a strong relationship between juvenile bull trout and temperature. As water
temperature exceeds a single daily maximum of 20 C, it becomes increasingly unlikely that
juvenile bull trout will be found using a given habitat. Other variables were not useful in
(quantitatively) explaining distribution of bull trout in this study.



Westslope Cutthroat Trout

Reiman and Apperson (1989) summarized the literature for of westslope cutthroat trout habitat
characteristics, noting that specific limiting factors associated with habitat, as with bull trout,
cannot be readily defined. In general, distribution of westslope cutthroat trout tends toward
higher elevations and lower stream orders. Westslope cutthroat trout will use all of the major
habitat components (i.e. pool, run, riffle, pocket water) with distribution tending toward lower
gradients and lower velocities. Several workers found pools to be particularly important habitat
for rearing cutthroat. Cover and complex habitat are also important for cutthroat trout,
particularly juvenile fish. Small cutthroat are typically associated with some form of instream
cover, such as cobble or woody debris. In many systems, westslope cutthroat trout move


Granite Creek Assessment                        WPN                                          Page 110
extensively using different reaches and habitats between spawning, summer rearing, and
overwintering. High quality pools and gravel substrate seem to be particularly important in
winter habitat use. Clean substrates, not embedded with fines, are important to allow fish
utilization of this habitat.

Selection of Habitat Variables

Habitat characteristics, such as the R1/R4 procedure used here, have been used successfully to
detect differences in reference and managed forested watersheds (Kershner and others 2004).
The methods have also been criticized because of observer variation, inconsistent protocol
application, inconsistent training, and the difficulty in detecting change caused by management
activity (Roper and others 2002). The potential limitation of this kind of data needs to be kept in
mind when interpreting the information in relation to land management and potential restoration
actions.

Habitat variables were selected based on the habitat requirements of the native salmonids as
described above, as well as with consideration for the limitations with precision/accuracy noted
by Roper and others (2002). The selected habitat variables are listed in Table 35.

Table 35. Habitat Variables summarized at the stream reach scale.
              Habitat Variable                                    Purpose
           Reach Characteristics
                                                Characterize natural channel form/ potential
           Rosgen Channel Type                                    habitat.
         Main Channel Length (Km)
                               2
          Total Habitat Area (m )                      Measure of Habitat Quantity
        Average Width and Depth (m)                         Habitat Dimensions
         Average Width/Depth Ratio               Indicator of habitat depth (habitat quality)
Percent Fast Water and Percent Slow Water                Summary of habitat types
     Streambank Condition and Pools
           Percent Bank Stability                       Channel condition indicator
           Percent Bank Undercut                    Important habitat cover component
              Riffle/ Pool Ratio                  Indicator of balanced habitat complexity
      Average Residual Pool Depth (m)                    Habitat quantity indicator
                Pools per Km                            Critical habitat component
            Large Woody Debris
           LWD Pieces per 100 m                         Habitat forming component.
                                               Large Woody Debris Index (LWDi) – functional
LWDi Score – Percent of potential for pieces                 habitat rating.
        Debris Dam count per 100 m                      Habitat forming component.
LWDi Score – Percent of potential for debris   Large Woody Debris Index (LWDi) – functional
                 dams                                        habitat rating.
  Count of Pieces plus Debris Dams/100m            Sum of habitat formed by large wood.



Granite Creek Assessment                        WPN                                             Page 111
         Substrate Characteristics
                                              Excessive fines degrade spawning and rearing
      Percent Surface Fines in Riffles.                         habitats.
                                               Gravels provide invertebrate production and
      Percent Surface Gravels in Riffles               hiding cover for small fish.
                                                 Particle size indicators, gravel to cobble
           D50, D84 Particle Size                     generally is a desirable range.
              Spawning Habitat
                               2
            Spawning Area (m )                   Measure of available spawning habitat
  Spawning Habitat Quality – Percent Good
                 plus Fair                         Measure of spawning habitat quality



8.6.3 Reference Benchmarks for Habitat Variables

The fisheries literature is best used to understand the preferred habitat conditions for salmonid
species, and not quantitative benchmarks. For quantitative evaluation, comparison to reference
conditions accounts for the high natural variability associated with landscape and channel
features. Comparable reference streams were selected from the relatively undisturbed Upper
Priest River watershed that flows into Upper Priest Lake. This watershed continues to support
adfluvial runs of westslope cutthroat trout and bull trout. Stream reaches from this watershed
were sampled to develop the reference range of conditions. The method of selecting reaches and
developing the benchmark values are described in Appendix F: Habitat Benchmarks derived
from the Upper Priest River watershed.

The interquartile range, the range between the 25th and 75th percentile, is used to describe the
distribution of data for reference conditions. The descriptive rating, from 1 to 4, is based on the
interquartile range, with a rating of 1 -2 indicating habitat quality values outside of the
interquartile range of the reference stream, and 3 – 4 indicating habitat values similar to
reference conditions. The benchmark values are listed in Table 36.



Table 36. Habitat quality benchmarks from reference streams in the Upper Priest River
watershed.
                                                            Benchmark Value
     Habitat Variable              Channel   1 (Poor)      2 (Fair)     3 (Good)      4 (Good)
                                    Width
  Average Width/Depth         Large ABFW       > 75        55 - 74       41 - 54        < 41
         Ratio                Small ABFW       > 75        52 - 74       34 - 51        < 34
 Percent Bank Stability
                                               < 50        50 - 84       85 - 95        > 95
          (%)
 Percent Bank Undercut
                                              0 – 1.5      1.6 - 2.2    2.3 - 4.2       > 4.2
          (%)



Granite Creek Assessment                       WPN                                               Page 112
    Riffle Pool/Ratio                       0.1 - 0.29,   0.3 - 0.7,
                            Large ABFW                                  0.71 - 1.8
                                                >3         1.9 - 3
                                            0.1 - 0.39,   0.4 - 0.79,
                            Small ABFW                                  0.8 - 1.0
                                                >2          1.1 - 2
      Pools per Km          Large ABFW         <4            4-6          7 - 13     > 13
                            Small ABFW         < 13        13 - 25       26 - 39     > 39
 Average Residual Pool      Large ABFW        < 0.4       0.4 - 0.59    0.6 - 1.0    > 1.0
       Depth (m)                                                          0.46 -
                            Small ABFW        < 0.3       0.3 - 0.45                 > .54
                                                                           0.54
 LWD Piece plus Debris      Large ABFW         <3.5       3.5 – 6.4     6.5 – 8.1    > 8.1
 Dam (Count per 100 m)      Small ABFW        < 4.5       4.5 – 9.5     9.6 – 15.2   > 15.2
Percent Surface Fines in
                                               > 35        25 - 35        < 25
      Riffles (%)
Spawning Gravel Quality     Large ABFW         < 50        50 - 74       74 - 90     > 90
 (% rated Fair to Good      Small ABFW         < 40        40 - 58       59 - 85     > 85
             Notes to Table: Large ABFW = Average Bankfull Width of 20 meters.
                            Small ABFW = Average Bankfull Width, 4 – 5 meters.




Granite Creek Assessment                      WPN                                             Page 113
8.7   RESULTS
The results are divided into three sections; aquatic habitat; fish distribution and fish passage
barriers and fish relative abundance. The data are organized by subwatershed – North Fork,
South Fork and Main Stem, and then by stream reaches from upstream to downstream.

8.7.1 Aquatic Habitat Conditions

The data for the habitat variables are provided in tables in Appendix G: Tables for Habitat
Variables in Granite Creek. In this section, the focus is to provide a comparison of habitat
conditions in Granite Creek to habitat benchmarks as a means of identifying potential limiting
factors for native fish populations. As indicated in the previous section, a rating of “1” or “2”
represents poor to fair habitat, and a “3” or “4” represents good habitat conditions comparable to
reference stream reaches. Reaches rated as poor to fair are highlighted to assist in identifying
reaches with potential habitat limitations.

8.7.1.1         Channel and Pool Habitat Variables

Streambank conditions as measured by Bank Stability and Bank Undercut in the North Fork
watershed reaches are similar to reference streams (Table 37). Riffle/Pool ratios are generally
lower than reference conditions indicating a generally less desirable ratio. However, the count of
pools as measured by number of pools/Km is likely a better indicator of where pool habitat is
low and could use improvement.




Granite Creek Assessment                      WPN                                         Page 114
Table 37. North Fork Granite Subwatershed: Comparison of Channel and Pool Habitat
Variables to Reference Conditions.

                                                                            Avg
                           % Bank      % Bank     Riffle/Pool   Pools/                Avg W/D
  Stream, Reach                                                           Residual
                           Stability   Undercut      Ratio       Km                    Ratio
                                                                         Pool Depth
     Willow #2                4           2           2           3          2           4
     Willow #1                3           3           3           3          3           4
   Willow, NF #1              4           4           2           2          2           4
  Granite, UNF #4             4           4           3           3          2           3
  Granite, UNF #3             4           4           2           3          2           4
  Granite, UNF #2             4           4           3           3          2           3
  Granite, UNF #1             4           4           2           2          4           1
  Granite, LNF #9             4           4           3           3          3           4
  Granite, LNF #8             3           4           2           4          3           4
  Granite, LNF #7             3           4           2           3          3           4
  Granite, LNF #6             4           4           2           1          3           2
  Granite, LNF #5             4           4           3           3          3           4
  Granite, LNF #4             4           4           3           3          3           4
  Granite, LNF #3             1           4           ~           3          3           4
  Granite, LNF #2             4           4           2           3          3           4
  Granite, LNF #1             4           4           2           4          3           4
     Orwig #1                 4           4           3           1          1           3
    Tillicum #7               4           4           ~           ~          ~           1
    Tillicum #6               4           3           ~           ~          ~           4
    Tillicum #5               4           4           2           1          1           3
    Tillicum #4               4           4           2           1          3           4
    Tillicum #3               4           4           2           1          2           3
    Tillicum #2               4           4           3           2          3           4
    Tillicum #1               4           4           1           3          1           3
  Tillicum, NF #2             4           4           3           2          1           4
  Tillicum, NF #1             4           4           3           2          1           4



In the South Fork watershed, several stream reaches in the Sema Creek watershed, Tobasco
Creek, and SF Granite Creek have fewer pools per Km and lower residual pool depth than
reference conditions indicating a potential need for more pool forming features (Table 38).




Granite Creek Assessment                          WPN                                        Page 115
Table 38. South Fork Granite Subwatershed: Comparison of Channel and Pool Habitat
Variables to Reference Conditions.

                                                                          Avg
                      % Bank      % Bank     Riffle/Pool                            Avg W/D
 Stream, Reach                                             Pools/ Km    Residual
                      Stability   Undercut      Ratio                                Ratio
                                                                       Pool Depth
    Sema #7                3         4           ~            ~            ~           4
    Sema #6                4         4           2            1            4           4
    Sema #5                4         4           2            1            4           4
    Sema #4                4         4           2            2            4           4
    Sema #3                4         4           1            1            4           4
    Sema #2                4         4           3            2            4           4
    Sema #1                4         4           2            2            1           4
 Sema, Trib4 #1            3         4           3            2            2           4
 Sema, Trib3 #2            4         4           1            1            2           4
 Sema, Trib3 #1            4         4           1            1            2           4
   Tobasco #3              4         4           1            1            1           4
   Tobasco #2              4         4           2            1            3           4
   Tobasco #1              4         4           ~            ~            ~           4
 Sema, Trib1 #3            4         4           1            1            1           4
 Sema, Trib1 #2            4         4           2            1            3           4
 Sema, Trib1 #1            4         4           ~            ~            ~           2
 Granite, SF #3            4         4           ~            ~            ~          ~
 Granite, SF #2            4         4           2            1            2           4
 Granite, SF #1            4                     1            1            3           3
    Cache #7               ~         ~           3            4            ~           4
    Cache #6               ~         ~           2            4            ~           4
    Cache #5               ~         ~           3            4            ~           4
    Cache #4               ~         ~           3            4            ~           3
    Cache #3               ~         ~           2            4            ~           4
    Cache #2               ~         ~           3            4            ~           4
    Cache #1               ~         ~           2            3                        4
Granite, SF, Trib1
                           4         4           ~            ~            ~           3
        #1



The streambank variables indicate stable banks with undercut features that provide cover for fish
in Main Granite Creek streams. The channel variables (Riffle/pool ratio, Pools/Km and Residual
Pool Depth) however indicate fewer pools and lower quality pool habitat than reference
condition, likely associated with legacy effects of land management in these watersheds (Table
39).




Granite Creek Assessment                       WPN                                         Page 116
 Table 39. Main Granite Subwatershed: Comparison of Channel and Pool Habitat
Variables to Reference Conditions.

                                                                          Avg
                      % Bank      % Bank     Riffle/Pool                            Avg W/D
 Stream, Reach                                             Pools/ Km    Residual
                      Stability   Undercut      Ratio                                Ratio
                                                                       Pool Depth
    Granite #8             4         4           1            4            3           4
    Granite #7             4         3           3            3            2           4
    Granite #6             2         3           2            1            3           4
    Granite #5             3         4           3            3            3           4
    Granite #4             4         4           3            1            3           3
    Granite #3             4         1           3            1            3           3
    Granite #2             4         4           3            1            3           3
    Granite #1             3         4           3            3            3           4
     Zero #5               4         4           2            2            2           4
     Zero #4               4         4           1            2            3           4
     Zero #3               4         4           3            2            2           4
     Zero #2               3         4           2            2            2           4
     Zero #1               4         4           2            3            4           4
    Packer #6              4         4           1            1            1           4
    Packer #5              4         4           2            2            1           4
    Packer #4              3         4           2            2            1           4
    Packer #3              2         4           1            1            1           4
    Packer #2              4         4           1            2            1           4
    Packer #1              3         4           1            1            1           4
 Packer, WF #3             2         4           1            1            ~           4
 Packer, WF #2             1         4           2            1            ~           4
 Packer, WF #1             2         4           2            1            ~           4
 Packer, Trib1 #1          4         4           2            2            1           3
   Blacktail #4            4         4           2            3            2           4
   Blacktail #3            4         4           2            3            2           4
   Blacktail #2            4         4           3            2            2           4
   Blacktail #1            3         4           2            3            2           4
     Jost #3               4         4           2            4            1           4
     Jost #2               4         4           2            3            1           4
     Jost #1               4         4           3            3            1           4
 Athol Trib 2 #1           4         4           1            1            1           4
     Athol #3              4         4           1            1            1           4
     Athol #2              4         4           2            2            1           4
 Athol Trib 1 #1           4         4           2            2            1           4
     Athol #1              ~         ~           ~            1            4          ~



Granite Creek Assessment                       WPN                                         Page 117
                                                                          Avg
                      % Bank      % Bank     Riffle/Pool                            Avg W/D
 Stream, Reach                                             Pools/ Km    Residual
                      Stability   Undercut      Ratio                                Ratio
                                                                       Pool Depth
    Fedar #2               2         4           2            3            1           4
    Fedar #3               2         4           3            4            1           4
    Fedar #4               1         4           2            1            1           4
    Fedar #5               1         4           3            3            3           4
    Fedar #6               4         4           2            3            1           4
    Fedar #7               4         4           2            4            1           4



8.7.1.2         Large Woody Debris

Large woody debris (LWD) provides several important functions in forested streams – forming
pools, stabilizing channels, sorting gravel and providing habitat for fish and invertebrates. The
method used to evaluate LWD (Davis and others 2001) ranks both the quantity of large wood
and the structural integrity to provide these functions. Individual pieces and debris dams are
counted separately. These two variables were added together (Pieces plus Dams/100m) for
comparison to reference reaches since either type contributes these functions.

Streams were grouped by size, “Small” indicating streams with an average bankfull width
approximately 4-5 meters, and “Large” indicating an average bankfull width approximately 20
meters. The count of individual pieces and debris dams is much higher in small streams than in
large streams in the reference reaches as indicated by box plots of the variable, LWD Pieces plus
Debris Dams/100 meters (Appendix F). Therefore, different benchmarks were applied to Small
vs Large streams indicated in the table of benchmarks, Table 36.

The LWD Index (LWDi) rates the functional stability of the wood. The LWDi is evaluated by
comparing the measured score for the reach to the potential score for the reach if every factor
were scored at the highest rating. Therefore, the LWDi is summarized in the table as Percent of
the Potential Score. A 100% score would mean that every piece or debris dam evaluated scored
a 5 out of a possible 5. (Note: LWDi scores for references reaches, which are not shown in these
tables, were not notably different than the values measured in Granite Creek. The interquartile
range for Percent of Potential for references reaches was 55-64 percent for LWD Pieces, and 68-
77 for Debris Dams.)

Streams in the North Fork watershed generally have a similar amount of LWD to reference
streams, with the exception of several reaches in the Lower North Fork Granite Creek (Table
40). The LWDi scores are also similar to those observed in the reference streams.




Granite Creek Assessment                       WPN                                         Page 118
Table 40. North Fork Subwatershed: Large Woody Debris variables. Stream reaches
compared to the “Large” Stream LWD reference benchmark are indicated by bold text.

                              Percent of            Percent of               Rating of
                                           Debris                Pieces +
                      Piece    Potential             Potential              LWD Pieces
 Stream, Reach                              Dam                   Dams
                     /100 m   LWDi score            LWDi score                + Dam
                                           /100m                  /100m
                                Piece               Debris Dam              Count/100m

    Willow #2          5.6       72         4.9         96         10.5         3
    Willow #1          7.1       69         5.3         75         12.4         3
 Willow, NF #1         6.4       74         8.7         97         15.1         3
Granite, UNF #4       11.9       60         4.6         82         16.5         4
Granite, UNF #3       22.2       63         8.6         80         30.8         4
Granite, UNF #2       18.5       64         6.1         80         24.6         4
Granite, UNF #1       23.9       61          2          83         25.9         4
Granite, LNF #9        5.5       67         2.5         82          8           3
Granite, LNF #8        6.1       64         3.5         74         9.6          4
Granite, LNF #7        2.4       64         0.7         67         3.1          1
Granite, LNF #6        2.6       60         0.7         78         3.3          1
Granite, LNF #5        4.2       62         0.9         79         5.1          2
Granite, LNF #3       10.3       64         3.5         76         13.8         4
Granite, LNF #2        2.3       63         0.1         60         2.4          1
Granite, LNF #1       14.9       63         3.3         77         18.2         4
    Orwig #1          15.3       56         7.5         75         22.8         4
   Tillicum #7         26        61         0.5        100         26.5         4
   Tillicum #6        13.5       61         6.8         85         20.3         4
   Tillicum #5        10.4       63         4.9         79         15.3         4
   Tillicum #4        10.7       58         5.5         73         16.2         4
   Tillicum #3        10.3                  5.2         85         15.5         4
   Tillicum #2        19.4       59         7.3         80         26.7         4
   Tillicum #1        13.8       64         2.8         84         16.6         4
 Tillicum, NF #2      14.3       55         1.1         84         15.4         4
 Tillicum, NF #1       8.6       62         8.6         88         17.2         4



Streams in the South Fork watershed have LWD quantities similar to or higher than that
measured in reference streams as indicated by “4” ratings (Table 41). The LWDi scores are
similar to those observed in the reference streams.




Granite Creek Assessment                     WPN                                     Page 119
Table 41. South Fork Subwatershed: Large Woody Debris variables. Stream reaches
compared to the “Large” Stream LWD reference benchmark are indicated by bold text.

                                Percent of              Percent of               Rating of
                      LWD                     Debris                 Pieces +
                                 Potential               Potential              LWD Pieces
 Stream, Reach       per 100                 Dam per                  Dams
                               LWDi score              LWDi score                 + Dam
                        M                     100M                    /100m
                                  Piece                Debris Dam               Count/100m
    Sema #7            4.5         70          9.2         84         13.7          3
    Sema #6            8.6         68          5.4         84          14           3
    Sema #5            9.8         67          9.1         79         18.9          4
    Sema #4            6.5         70           5          78         11.5          3
    Sema #3           23.7         71         11.3         82          35           4
    Sema #2           18.5         72           5          86         23.5          4
    Sema #1           35.2         70          9.3         82         44.5          4
 Sema, Trib4 #1       28.6         65          3.4         88          32           4
 Sema, Trib3 #2       45.8         62          8.1         84         53.9          4
 Sema, Trib3 #1       38.5         65          7.3         82         45.8          4
   Tobasco #3         27.4         63          1.8         78         29.2          4
   Tobasco #2          4.9         63          0.4         98          5.3          2
   Tobasco #1         29.7         62          3.1         81         32.8          4
 Sema, Trib1 #5            *       63           *          75           *           ~
 Sema, Trib1 #4            *       60           *          81           *           ~
 Sema, Trib1 #3       38.8         65          2.3         89         41.1          4
 Sema, Trib1 #2       49.6         65          0.4         92          50           4
 Sema, Trib1 #1       36.2         62          8.7         82         44.9          4
 Granite, SF #3        3.3         68           2          79          5.3          2
 Granite, SF #2        3.7         68           3          84          6.7          3
 Granite, SF #1        3.8         73          3.1         84          6.9          3
    Cache #7          21.6         62          14          76         35.6          4
    Cache #6          15.5         67          6.8         78         22.3          4
    Cache #5          21.2         61          13          78         34.2          4
    Cache #4          20.4         63         12.4         76         32.8          4
    Cache #3          18.2         67         17.8         79          36           4
    Cache #2          23.4         62         15.2         80         38.6          4
    Cache #1          18.8         60          15          80         33.8          4
Granite, SF, Trib1
        #1             4.8         71         16.5         92         21.3          4
Notes to Table: * Stream channel intermittently ran subsurface. Therefore, no channel length to
calculate Counts/100 meter.



Granite Creek Assessment                       WPN                                       Page 120
LWD quantities are lower in reaches of mainstem Granite Creek, WF Packer Creek, and Fedar
Creek than measured in reference as indicated by the lower rating (Table 42).

Table 42. Main Granite Subwatershed: Large Woody Debris variables. Stream reaches
compared to the “Large” Stream LWD reference benchmark are indicated by bold text.

                               Percent of              Percent of               Rating of
                     LWD                     Debris                 Pieces +
                                Potential               Potential              LWD Pieces
 Stream, Reach      per 100                 Dam per                  Dams
                              LWDi score              LWDi score                 + Dam
                       M                     100M                    /100m
                                 Piece                Debris Dam               Count/100m
   Granite #8          3.7        58          1.1         69          4.8          2
   Granite #7          4.2        57          0.4         64          4.6          2
   Granite #6          0.9        57          0.4         76          1.3          1
   Granite #5          3.5        59          2.4         84          5.9          2
   Granite #4          3.6        53          0.5         79          4.1          2
   Granite #3          1.7        50          0.6         64          2.3          1
   Granite #2          2.1        49          0.5         75          2.6          1
   Granite #1          1.6        53          1.8         75          3.4          1
     Zero #5          19.2        66          5.6         79         24.8          4
     Zero #4          26.2        70           4          90         30.2          4
     Zero #3          30.7        67          3.7         86         34.4          4
     Zero #2          26.7        65           4          79         30.7          4
     Zero #1          18.8        64          3.8         76         22.6          4
   Packer #6          29.1        73         30.2         62         59.3          4
   Packer #5          26.2        72         11.2         62         37.4          4
   Packer #4          23.8        69         11.7         63         35.5          4
   Packer #3          24.7        71         14.2         65         38.9          4
   Packer #2           8.1        61          3.6         79         11.7          3
   Packer #1          11.9        59          6.2         80         18.1          4
 Packer, WF #3         3.5        58          0.8         93          4.3          1
 Packer, WF #2         0.7        35           0                      0.7          1
 Packer, WF #1        11.6        56          2.5         79         14.1          3
Packer, Trib1 #1      21.9        60           4          75         25.9          4
   Blacktail #4        22         62          4.9         73         26.9          4
   Blacktail #3       18.4        66          4.2         85         22.6          4
   Blacktail #2       17.2        65          6.4         84         23.6          4
   Blacktail #1       43.7        64          7.4         77         51.1          4
     Jost #3          16.5        65          6.2         76         22.7          4
     Jost #2          15.8        65          6.6         81         22.4          4
     Jost #1          11.9        62          9.4         83         21.3          4
 Athol Trib 2 #1      13.8        62          2.8         72         16.6          4
    Athol #3          11.9        62          1.7         75         13.6          3


Granite Creek Assessment                      WPN                                      Page 121
    Athol #2            15.4         56            3.5        77         18.9             4
 Athol Trib 1 #1        15.7         62            2.5        76         18.2             4
    Fedar #2               9                       15.3                  24.3             4
    Fedar #3             8.4                       26.9                  35.3             4
    Fedar #4             1.3                       5.1                    6.4             2
    Fedar #5               0                       7.5                    7.5             2
    Fedar #6             0.3                       16.2                  16.5             4
    Fedar #7             1.9                       56.9                  58.8             4



8.7.1.3         Substrate and Spawning Habitat

The measurements of fines in substrate are calculated from pebble counts; fines were defined as
those particle sizes less than or equal to 8mm. Pebble counts provide a fairly robust measure of
fines in comparison to the visual estimates of surface fines made as part of the R1/R4 habitat
survey procedure.

The rating of spawning habitat quality is evaluated, as described in the Methods section, based
on gravel size suitability, proximity to resting pools, and current velocities.

The rating of surface fines in riffles indicates a fairly high level of fines in the North Fork (Table
43). However, spawning habitat was rated in the combined Fair-Good category.

Table 43. North Fork Subwatershed: Substrate and spawning habitat comparison to
reference conditions.

                                              Riffle                       Spawning
                                                           Spawning
                   Stream, Reach           % Surface                     Habitat Rating
                                                          Habitat (m2)
                                          Fines Rating                   (Good + Fair)
                     Willow #2                 1              12                4
                     Willow #1                 3               5                4
                   Willow, NF #1               1              26                4
                Granite, UNF #4                3              16                4
                Granite, UNF #3                3               8                3
                Granite, UNF #2                2              20                3
                Granite, UNF #1                3               4                4
                   Granite, LNF #9             3              19                4
                   Granite, LNF #8             1              25                4
                   Granite, LNF #7             1               3                4
                   Granite, LNF #6             2               2                4
                   Granite, LNF #5             2               4                4
                   Granite, LNF #4             2               ~



Granite Creek Assessment                            WPN                                       Page 122
                Granite, LNF #3            2                 16                   3
                Granite, LNF #2            3                 0.0
                Granite, LNF #1            2                 54                   3
                   Orwig #1                2                  9                   3
                  Tillicum #7              1                 10                   3
                  Tillicum #6              3                 13                   4
                  Tillicum #5              3                  8                   4
                  Tillicum #4              3                 35                   3
                  Tillicum #3              2                 47                   3
                  Tillicum #2              3                  5                   1
                  Tillicum #1              3                  5                   3
                Tillicum, NF #2            1                  6                   4
                Tillicum, NF #1            2                  5                   4
                Notes: * No channel length to calculate Counts/100 meter.
                                                ~ No data.



In the South Fork, sediment fines were high and rated poor compared to the benchmark where
they were evaluated (Table 44). The data gaps occur where Pebble Counts were not taken or not
provided in the data base. The spawning habitat rating is consistent with the fines rating, and is
less than the reference benchmark throughout the Sema Creek watershed.

Table 44. South Fork Subwatershed: Substrate and spawning habitat comparison to
reference conditions.

                Stream, Reach             Riffle          Spawning            Spawning
                                                                   2
                                       % Surface         Habitat (m )       Habitat Rating
                                      Fines Rating                          (Good + Fair)
                   Sema #7                 ~                  5                   2
                   Sema #6                 ~                 0.0
                   Sema #5                 ~                  2                   2
                   Sema #4                 ~                 28                   3
                   Sema #3                 ~                  4                   2
                   Sema #2                 ~                  2                   3
                   Sema #1                 ~                 0.0
                Sema, Trib4 #1             ~                  ~                   ~
                Sema, Trib3 #2             2                  6                   2
                Sema, Trib3 #1             ~                  3
                  Tobasco #3               1                  6                   2
                  Tobasco #2               1                 0.5                  3
                  Tobasco #1               1                  2                   2
                Sema, Trib1 #5             2                  4                   3



Granite Creek Assessment                        WPN                                          Page 123
                Sema, Trib1 #4             1                  6                   3
                Sema, Trib1 #3             1                  2                   1
                Sema, Trib1 #2             1                  ~                   ~
                Sema, Trib1 #1                                6                   3
                 Granite, SF #3            3                 13                   4
                 Granite, SF #2            3                 76                   4
                 Granite, SF #1            3                 131                  4
                   Cache #7                ~                  ~                   ~
                   Cache #6                ~                  ~                   ~
                   Cache #5                ~                  ~                   ~
                   Cache #4                ~                  ~                   ~
                   Cache #3                ~                  ~                   ~
                   Cache #2                ~                  ~                   ~
                   Cache #1                ~                  ~                   ~
             Granite, SF, Trib1 #1         2                  3                   3
                Notes: * No channel length to calculate Counts/100 meter.
                                                ~ No data.



Percent surface fines are rated poor (high surface fines) in many reaches in the Main Stem
Granite Creek subwatershed including Zero, WF Packer, Jost, Athol, Packer and Fedar Creeks
(Table 45).

Table 45. Main Granite Subwatershed: Substrate and spawning habitat comparison to
reference conditions.

                                          Riffle                              Spawning
                                                          Spawning
                Stream, Reach          % Surface                   2        Habitat Rating
                                                         Habitat (m )
                                      Fines Rating                          (Good + Fair)
                   Granite #8              3                 551                  4
                   Granite #7              3                 42                   4
                   Granite #6              3                 12                   4
                   Granite #5              3                  9                   4
                   Granite #4              3                 68                   4
                   Granite #3              ~                 23                   1
                   Granite #2              1                  0                   ~
                   Granite #1              3                 120                  4
                    Zero #5                1                 15                   3
                    Zero #4                1                 65                   4
                    Zero #3                1                 31                   3
                    Zero #2                1                 20                   2
                    Zero #1                1



Granite Creek Assessment                        WPN                                          Page 124
                   Packer #6               1                 0.0
                   Packer #5               3                  3                 3
                   Packer #4               2                  9                 2
                   Packer #3               3                  4                 2
                   Packer #2               3                 13                 3
                   Packer #1               3                 97                 3
                Packer, WF #1              1                 0.0
                    Jost #4                3                  ~                 ~
                    Jost #3                1                 18                 4
                    Jost #2                1                 40                 3
                    Jost #1                1                 21                 3
                 Athol Trib 2 #1           3                  2                 2
                    Athol #3               2                 0.0
                    Athol #2               3                  7                 3
                 Athol Trib 1 #1           2                 0.0
                   Fedar #2                2                  ~                 ~
                   Fedar #3                3                  ~                 ~
                   Fedar #4                1                  ~                 ~
                   Fedar #5                3                  ~                 ~
                   Fedar #6                1                  ~                 ~
                   Fedar #7                1                  ~                 ~
                Notes: * No channel length to calculate Counts/100 meter.
                                                ~ No data.



8.7.2 Fish Distribution, Abundance, and Connectivity

Fish distribution and density data were collected by electrofishing or snorkel counts during the
low flow period, July to September, by the Kalispel Tribe or USFS crews. Streams were
sampled in either 2006 or 2007, with one data set from 2005. To provide a comparison between
all data sets, fish density is summarized as count of all fish/100 meters squared, and not further
divided into age or size class. It is important to note that no bull trout of any size were identified
from these fish collection efforts. (Note: it was noted that two bull trout redds were counted in
the North Fork Granite in 2008.)

Man-made barriers were identified from US Forest Service and Kalispel Tribe GIS files. The
current status of these structures as current barriers was not verified during this assessment, and
may or may not be accurate with current conditions. The extent of natural barriers to adfluvial
cutthroat trout was identified in the University of Idaho MS Thesis (Irving 1987). The location
of the natural barriers was identified by information from this source, and would be expected to
be in the approximate location of the natural barrier as indicated on these maps.




Granite Creek Assessment                        WPN                                           Page 125
8.7.3 Fish Density

Fish counts were completed by electrofishing or snorkeling in Granite Creek tributaries from
2005 to 2007. Surveys were completed in the Upper Priest River watershed in 2008 to provide a
comparison of fish abundance to relatively undisturbed watersheds. Fish densities from the
present study were also compared to densities collected in Granite Creek tributaries in the 1980‟s
to determine what changes have occurred over the last 20 years.

Fish Density in Reference Streams

Fish counts were completed by electrofishing in Cedar and Jackson Creeks, and by snorkeling in
Upper Priest River. Fish counts and descriptive statistics are summarized in Table 46.
Cutthroat trout are the dominant species in these streams with a median density of 4.6
fish/100m2, and interquartile range of 2.2 – 7.0 fish/100m2. Bull trout and brook trout occur at
much lower densities, 0.40 and 0.35 fish/100m2 respectively.

Table 46. Fish Density in Reference Reaches from Upper Priest River Watershed. (Cedar
and Jackson Creek by electrofishing, Upper Priest Rive by snorkel counts.)

                      Channel     Channel                 Bull Trout   Brook Trout   Westslope Ct
                                                Area
  Stream, Reach        Width      Surveyed                 Density       Density      Density
                                              sq meters
                        (m)      Length (m)                #/100m2       #/100m2      #/100m2
  Cedar Cr. #70           4.5         100          450        0             0             6.4
  Cedar Cr. #72           4.6         100          460        0            0.7            4.6
  Cedar Cr. #73            6          100          600        0            0.3            7.0
  Cedar Cr. #76           5.8         100          580        0             0             6.7
  Cedar Cr. #86           4.5         100          450        0             0             7.1
  Jackson Cr. #1           6          100          600        0             0            21.8
 Upper Priest #3         19.2         100         1920        0            0.6            2.0
 Upper Priest #8         13.9         100         1390       0.4           0.2            3.6
 Upper Priest #13        11.2         100         1120       0.2           0.4            1.3
 Upper Priest #53        11.6         100         1160       0.3           0.3            2.5
 Upper Priest #54         8.5         100          850       0.4           0.4            1.9
 Upper Priest #58         8.1         100          810       1.9           0.2            7.0
 Upper Priest #59         8.1         100          810       1.5            0             2.2
     Descriptive Statistics – when Fish are Present
                        Number                                 6             8            13
                          Mean                                0.8           0.4          8.9
                          Range                            0.2 - 1.9     0.2 - 0.7    1.3 - 21.8
                         Median                              0.40          0.35          4.6
                  Interquartile Range                      0.3 - 1.5    0.25 - 0.5    2.2 - 7.0
          Mean WCT:BKT Density Ratio = 16




Granite Creek Assessment                          WPN                                       Page 126
The relative magnitude of cutthroat trout density versus bull trout and brook trout is illustrated in
Figure 53.



                                 Density in Reference Reachs in Upper Priest River


                            20




                            15
         Density #/100 m2




                            10




                            5




                            0

                                        Bull Trout      Brook Trout   Westslope Cutthroat



Figure 53. Box plot of relative fish abundance from thirteen reference reaches in Upper
Priest River watershed. (25th to 75th percentile indicated by the box, median by a
horizontal line, and whiskers indicate 10th and 90th percentiles.)


The interquartile range from the reference streams was used to develop a relative abundance
rating for comparison to Granite Creek (Table 47). A Moderate rating indicates fish density
values within the interquartile range, that is, equivalent to reference condition, a Low rating
indicates densities below the 25th percentile, and a High rating indicates densities above the 75th
percentile.

Table 47. Density benchmarks for all age classes of trout from sampled reference reaches.
Abundance Rating                       Bull Trout Density         Brook Trout Density       Westslope Ct Density
                                                    2                          2                          2
                                            #/100m                     #/100m                     #/100m
Low                                           < 0.3                     < 0.25                      < 2.2
Moderate
                                            0.3 - 1.5                  0.25 - 0.5                 2.2 - 7.0
(Equal to reference)
High                                          > 1.5                      > 0.5                     > 7.0




Granite Creek Assessment                                    WPN                                            Page 127
The ratio of westslope cutthroat trout to brook trout (WCT:BKT ratio) provides an additional
metric for comparison to reference streams. Since a zero value cannot be used as the
denominator in calculating a ratio, the mean brook trout density of 0.4 fish/100 m2 was
substituted for zero values. With this substitution, the mean WCT:BKT ratio for reference
reaches was calculated as 16:1.

Figure 54 illustrates graphically the relative difference between the populations of cutthroat trout
in the reference stream, Cedar Creek, to the reduced population in Tillicum Creek, a tributary to
the North Fork of Granite Creek. (Note: These stream habitat areas were not equivalent). The
size distribution of cutthroat trout is similar between the two streams, but the number of fish is
higher in Cedar Creek.



        25                                                               25




        20                                                               20




        15                                                               15
Count




                                                                 Count



        10                                                               10




        5                                                                5




        0                                                                0
             0   50       100              150       200                         50          100           150         200

                              Tillicum                                                Cedar Creek - Reference Stream


Figure 54. Length frequency of cutthroat trout in Tillicum Creek, a tributary in Upper
Granite Creek, to Cedar Creek, a tributary in Upper Priest River watershed.


Fish Densities in Granite Creek

Presence/absence and fish density, number/100 m2, are summarized by reach in Table 48, Table
49 and Table 50for each 6th field subwatershed. The last column in the tables rates the density of
westslope cutthroat trout in comparison to the reference streams as summarized in Table 47:
High (H), Moderate (M), Low (L) or Absent (A).

Table 48. Fish Density in North Fork Granite Creek Subwatershed.
                                                                                                    Brook            WCT
                        Area             Westslope Ct      Brook Trout        Westslope Ct
                                                                                                    Trout           Density
 Stream, Reach        measured            Presence/         Presence/           Density
                          2                                                            2           Density         Relative to
                        (M )              Absence           Absence             #/100m                    2
                                                                                                   #/100m          Reference
 Granite, LNF #9        443                   P                P                  11.5               14.5              H
 Granite, LNF #8                           No Data          No Data
 Granite, LNF #7                           No Data          No Data
 Granite, LNF #6                           No Data          No Data



Granite Creek Assessment                                      WPN                                                            Page 128
Granite, LNF #5     432           P             P             7.9         27.1          H
Granite, LNF #4                No Data       No Data
Granite, LNF #3                No Data       No Data
Granite, LNF #2     382           P             P              8.4        41.1          H
Granite, LNF #1     994           P             P              1.0         9.3          L
    Orwig #1        123           P             P              8.1        13.0          H
   Tillicum #6      108           P             A              1.8          0           L
   Tillicum #5      161           P             P             11.2         4.4          H
   Tillicum #4      258           P             P             28.3         6.6          H
   Tillicum #3      340           A             P               0          2.3          A
   Tillicum #2      505           P             P              1.6         2.0          L
   Tillicum #1      609           P             P              1.0         1.3          L
Tillicum, NF #2     176           P             P              6.3         7.4          M
Tillicum, NF #1     22            P             P              9.2        64.3          H

Cutthroat trout and brook trout occur throughout the Lower North Fork of Granite Creek and
Tillicum Creek. The density of cutthroat trout varies throughout the stream system, from being
absent to high densities in comparison to the reference reach.

This data set did not include information on upper North Fork Granite Creek or Willow Creek.
However, other observation and sampling indicates that cutthroat trout occur above the falls on
the North Fork to just below the mouth of Willow Creek. The Kalispel Tribe looked for and did
not find any fish in Willow Creek and the North Fork above the mouth of Willow Creek. The
Tribe and the Forest Service Fisheries Biologist believe that these fishless areas have suitable
habitat to support the translocation of westslope cutthroat trout (Personal Communication, Todd
Anderson, Kalispel Tribe, March, 2009).

Table 49. Fish Density in South Fork Granite Creek Subwatershed.
                                                                          Brook       WCT
                    Area     Westslope Ct   Brook Trout   Westslope Ct
                                                                          Trout      Density
Stream, Reach     measured    Presence/      Presence/      Density
                      2                                            2     Density    Relative to
                    (M )      Absence        Absence        #/100m              2
                                                                         #/100m     Reference
  Sema #7           110           A             P              0           13.6         A
  Sema #6           150           A             P              0           19.3         A
  Sema #5           320           A             P              0            5.3         A
  Sema #4           280           A             P              0            6.8         A
  Sema #3           460           A             P              0            3.9         A
  Sema #2           430           A             P              0            3.5         A
  Sema #1           559           A             P              0            2.0         A
Sema, Trib4 #1      70            A             P              0           40.0         A
Sema, Trib3 #2      110           A             A              0             0          A
Sema, Trib3 #1      90            A             P              0           25.6         A
 Tobasco #3         140           A             P              0            8.6         A
 Tobasco #2         80            A             P              0           17.5         A
 Tobasco #1         120           A             P              0            9.2         A
Sema, Trib1 #5      90            A             P              0           10.0         A
Sema, Trib1 #4      90            A             P              0           23.3         A
Sema, Trib1 #3      130           A             P              0           13.8         A
Sema, Trib1 #2      130           A             P              0            8.5         A



Granite Creek Assessment                       WPN                                          Page 129
Sema, Trib1 #1      190          A             P              0           6.3          A
Granite, SF #3      750          P             P             0.5          4.5          L
Granite, SF #2      614          P             P             0.5          1.1          L
Granite, SF #1      825          P             P             2.3          0.5          L
 High Rock #1     No Data        A             P                                       A
   Cache #6         270          A             P              0          14.8          A
   Cache #5         320          P             P             0.3         11.9          L
   Cache #4         280          A             P             0.0          3.9          A
   Cache #3         350          P             P             1.1          5.4          L
   Cache #2         310          A             P              0          10.6          A
   Cache #1         350          A             P              0          10.3          A
 Granite, SF,
   Trib1 #1          70          P             P             5.7         27.1          M

Cutthroat trout have been replaced by brook trout in Sema Creek, Sema Creek tributaries, and
Tobasco Creek (Table 49). Population densities of cutthroat trout are low in South Fork Granite
Creek, and low to absent in Cache Creek.

Table 50. Fish Density in Main Stem Granite Creek Subwatershed
                                                                         Brook       WCT
                   Area     Westslope Ct   Brook Trout   Westslope Ct
                                                                         Trout      Density
Stream, Reach    measured    Presence/      Presence/      Density
                     2                                            2     Density    Relative to
                   (M ))     Absence        Absence        #/100m              2
                                                                        #/100m     Reference
  Granite #8        1080         P             P             0.8          8.1          L
  Granite #7        1250         P             P              0.7         2.0          L
  Granite #4        1450         P             P              0.9         0.8          L
  Granite #3        1442         P             P              1.5         1.2          L
  Granite #2        1578         P             P              0.4         0.9          L
  Granite #1        1103         P             P              0.3         0.5          L
   Zero #5           234         P             A             23.5          0           H
   Zero #4           379         P             A             12.9          0           H
   Zero #3           385         P             A              8.3          0           H
   Zero #2           401         P             P              4.5         9.2          M
   Zero #1           255         P             P              7.8         8.6          H
  Packer #6          160         P             A             10.0          0           H
  Packer #5          150         P             A             16.7          0           H
  Packer #4          190         P             A             20.0          0           H
  Packer #3          160         P             A             27.5          0           H
  Packer #2          377         P             A              5.3          0           M
  Packer #1          320         P             A              7.2          0           H
Packer, WF #3        33          A             A               0           0           A
Packer, WF #2        418         P             A              0.5          0           L
Packer, WF #1        182         P             A              3.3          0           L
 Blacktail #4        248         P             P             10.5         7.7          H
 Blacktail #3        288         P             P             12.2        22.6          H
 Blacktail #2        351         P             P              5.1        13.1          M
 Blacktail #1        375         P             P              5.3         4.0          M
   Jost #3           181         A             P               0         23.2          A
   Jost #2           256         A             P               0         23.5          A
   Jost #1           272         P             P              3.3         7.3          M



Granite Creek Assessment                      WPN                                          Page 130
Athol Trib 2 #1     120           P              P             8.3           12.5          H
   Athol #3         120           P              P             7.5           10.0          H
   Athol #2         320           P              P             4.1            5.9          M
Athol Trib 1 #1     60            A              A              0              0           A
   Fedar #2         91            P              A             3.3             0           M
   Fedar #3         69            P              A             8.7             0           H
   Fedar #5         257           P              P             0.4            7.4          L
   Fedar #6         146           P              P             9.6            6.9          H
   Fedar #7                    No Data        No Data


Cutthroat trout densities are low in main stem of Granite Creek in comparison to reference
reaches, but are comparable to reference reaches in most segments of Zero, Packer, and
Blacktail, Athol, and Fedar Creeks (Table 50).

Granite Creek Fish Densities in 1980’s

David Irving completed a Master of Science thesis in the 1980‟s on the productivity of cutthroat
trout in tributaries to Priest Lake (Irving 1987). This data on Granite Creek provides a point of
reference for comparison to the fish population data obtained during this study. Cutthroat trout
were found both above and below impassable fish barriers. Fish above barriers were assumed to
be fluvial in nature while those below barriers primarily were adfluvial. Bull trout were not
found above barriers. Brook trout were also found above barriers, which Irving assumed were
due to previous stocking efforts.

Reported cutthroat trout densities are highly variable so comparisons between studies need to be
made with care. For example, Irving (1987), reported annual variation in mean cutthroat trout
densities in surveyed tributaries of Priest Lake: 1.0 fish/ 100m2 in 1982, 2.0 in 1983, and 4.9 in
1984. This variation may be due to natural variation associated with climate, actual increases
associated with decreased harvest due to changed regulations or other factors. Irving believed
that these densities were relatively low compared to densities observed in other studies
completed at the time in Idaho and Montana.

Fish densities obtained during this study are compared to values reported by Irving (1987) for
Granite Creek tributaries in Table 51. Only the reaches from the matching main stream were
used for this comparison, data from tributaries to these streams were not included. The number
of reaches used to calculate average fish densities for this study are indicated in parentheses after
the stream name.

In the North Fork Granite and Tillicum Creek, both cutthroat and brook trout have increased in
number from the 1980‟s to the 2005-2007 period. In the South Fork tributaries, cutthroat and
brook trout densities have stayed about the same over time, but the brook trout densities are
much higher than the reference streams. In Main Granite Creek subwatershed, cutthroat trout
densities appear to have increased in several streams: Zero Creek, Packer Creek, Athol Creek,
and Blacktail Creek. In Jost Creek, cutthroat trout densities are about the same, but brook trout
numbers have increased. In Fedar Creek, both densities of cutthroat trout and brook trout have
decreased.




Granite Creek Assessment                        WPN                                            Page 131
Table 51. Average fish densities, fish/100m2, reported in 1983-1984 (Irving, 1987)
compared to fish densities for the current study.



                            Bulltrout          Cutthroat Trout               Brook Trout

                           1983 - 1984                                 1983 - 1984
                                          1983 - 1984
                             (Irving                     2005 - 2007     (Irving     2005 - 2007
                                         (Irving 1987)
                              1987)                                       1987)
 North Fork Granite
 North Fork Granite (3)        1.1           1.05            9.3           0.6          27.6
 Tillicum Creek (6)            0.5           0.15            7.3           0.9           2.8
 South Fork Granite
 South Fork Granite (3)       0.05           1.75            1.1          1.75           2.0
 Sema Creek (7)                 0             0.8            0.0          6.15           7.8
 Cache Creek (6)                0            0.85            0.2         12.25           9.5
 Main Granite Creek
 Granite Creek (6)              0             0.1            0.8            0            2.3
 Zero Creek (7)                0.1           6.05           11.4           0.5           3.6
 Packer Creek (6)              1.8           0.15           14.5          0.15           0.0
 Athol Creek (2)                0             1.9            5.8          27.5           8.0
 Blacktail Creek (2)            1            2.75            8.3            1           11.9
 Jost Creek (3)                0.1            2.8            1.1          8.55          18.0
 Fedar Creek (4)                0            15.4            5.5         19.45           3.6
 Granite Creek
 Drainage Average              0.4           1.55            5.4          2.95           6.9
  Upper Priest River
 Drainage
 Upper Priest River           0.03           0.3                            0
 Cedar Creek                    0            7.1                            0
 Hughes Creek                  3.1           7.6                          0.01
 Upper Priest River
 Drainage Average              1.8           6.0                           0.1

Distribution and Barriers

The current distribution of westslope cutthroat trout and brook trout is illustrated in the maps by
subwatershed in Appendix H recording just Presence/Absence from the 2005 – 2007
electrofishing/snorkel data. Known man-made barriers and the approximate location of natural
barriers is also shown in these maps. Irving (1987) was interested in estimating potential
productivity of all tributary streams and therefore accounted for all stream miles below natural
barriers he identified during field surveys. Westslope cutthroat trout and brook trout occur above
the identified natural barriers on a stream by stream basis.

Since some man-made barriers isolate native westslope cutthroat trout from competition from
brook trout, it will be important to develop a strategy for deciding which barriers to fish passage
should be left in place to assist in protect these native stocks from brook trout competition.


Granite Creek Assessment                        WPN                                            Page 132
8.8   SUMMARY
Habitat surveys generally indicate issues occur in Granite Creek watersheds with fine sediments
and poor pool habitats. In general, the ocular rating of streambank stability and undercut banks
are similar to reference conditions. This provides an indication that the physical integrity of
stream channels is currently stable or in a recovery trend. Observations of other key habitat
limiting factors – Pools, LWD, and Sediment – and fish populations is summarized by
subwatershed below.

North Fork Subwatershed

    Pool frequency in the North Fork of Granite Creek is similar to reference conditions in
     most reaches. The reaches where the pool counts are low provide potential opportunities
     for pool enhancement. Pool counts and residual pool depth is low throughout Tillicum
     Creek.
    LWD was similar to reference streams in most stream reaches with the exception of the
     Lower North Fork Granite Creek.
    Percent surface fines measured from pebble counts indicates a fairly high level of fines
     throughout the streams in the North Fork. However, spawning habitat was rated fair-to-
     good in most locations.
    Westslope cutthroat trout densities are comparable to reference streams in the Little North
     Fork and have increased since the 1980‟s. However, brook trout densities have also
     increased dramatically. In two reaches of Tillicum Creek, westslope cutthroat trout have
     increased from the 1980‟s and are comparable to reference streams, in the other reaches
     abundance for cutthroat are low, but brook trout numbers are also low.
    Currently fishless reaches of the upper North Fork of Granite Creek and Willow Creek
     may provide an opportunity for translocation of westslope cutthroat trout.

South Fork Subwatershed

  Pool habitats are poor in Sema Creek, Tobasco Creek, and reaches of the South Fork Granite
   Creek. Pool frequency is good in Cache Creek.
  LWD was similar to reference streams in majority of reaches. One reach in Tobasco Creek
   and one reach in South Fork Granite has less wood than reference conditions.
  Percent surface fines measurements were limited in the South Fork, however, where it was
   measured in Sema Creek and Tobasco Creek the surface fines were high. Spawning habitat
   was rated poor in many reaches.
  Fish counts indicate that brook trout have replaced cutthroat trout in most of the tributaries
   including Sema Creek, Tobasco Creek, and Cache Creek. A limited area in the lower South
   Fork of Granite Creek has a population of cutthroat similar to reference conditions.

Main Granite Creek Subwatershed




Granite Creek Assessment                      WPN                                        Page 133
  Pool habitats are generally poor in tributary streams in the subwatershed: Zero Creek,
   Packer Creek, Blacktail Creek, and Athol Creeks.
  The primary location of low LWD frequency was in main Granite Creek. Several reaches in
   W.F. Packer and Fedar Creek were also low.
  Percent surface fines measured from pebble counts indicated high level of fines (indicating
   poor conditions) in Zero Creek, Jost Creek, Athol Creek, and Fedar Creek. Spawning
   habitat was rated as good-to-fair in Granite Creek, and majority of Zero Creek.
  Westslope cutthroat trout densities are comparable to reference in many stream reaches in
   Zero Creek, Packer Creek, Blacktail Creek, Athol Creek, and Fedar Creek. Brook trout
   were not counted in Packer Creek and several reaches of Zero Creek and Fedar Creek.
   Where brook trout were present their numbers exceeded the density observed in reference
   streams.
  Man-made barriers created by culverts at road crossings should be left in place to protect the
   existing isolated cutthroat trout populations. These situations potentially occur in Fedar
   Creek, Zero Creek, Athol Creek, and Athol Tributary 2. These barriers need to be closely
   evaluated with respect to their potential value in isolating westslope cutthroat trout from
   brook trout.




Granite Creek Assessment                      WPN                                        Page 134
                                  9.0    REFERENCES

Bowling, L.C., and D.P. Lettenmaier, 1997: Evaluation of the effects of forest roads on
  streamflow in Hard and Ware Creeks, Washington, Water Resources Series, Technical
  Report 155, University of Washington, Seattle.

Bjornn, T. C. 1957. A survey of the fishery resources of Priest and Upper Priest Lakes and their
   tributaries. Completion Report F-24-R, 1955-1957. Idaho Department of Fish and Game,
   Boise, Idaho.

Dambacher, Jeffery M., Jones, Kim K. 1997. Stream habitat of juvenile bull trout populations in
  Oregon and benchmarks for habitat quality. Friends of the Bull Trout Conference
  Proceedings, May 5-7, 1994. W. C. Mackay, M. K. Brewin, M . Monita, Coeditors. The Bull
  Trout Task Force. Calgary, Alberta, Canada. 353-360.

Davis, J.A. and N. Horner. 1995. Regional fisheries management investigations. Region 1
   lowland lakes investigation. Project F-71-R-16. Job no. 1-b Priest Lake. IDFG, Boise, Idaho.

Davis, J.C., G.W. Minshall, C.T. Robinson, and P. Landres. 2001. Monitoring wilderness
   stream ecosystems. USDA Forest Service Rocky Mt. Research Station. RMRS-GTR-70.
   Ogden Utah. 137 p.

Doten, C.O. and D.P. Lettenmaier. 2004. Prediction of Sediment Erosion and Transport with
   the Distributed Hydrology-Soil-Vegetation Model, Water Resources Series, Technical Report
   178, University of Washington, Seattle.

Dunham, J.B. and G.L. Chandler. 2001. Models to predict suitable habitat for juvenile bull trout
   in Washington State, Final Report. USFS Rocky Mt Station, Boise, Idaho. Submitted to US
   FWS, Lacey, Washington.

Dunne, T. and L.B. Leopold. 1978. Water in environmental planning. W.H. Freeman &
   Company. San Francisco. 818 pages.

Espinosa, A. 1988. Clearwater Stream Survey Methodology, Clearwater N.F., Orofino, Idaho.

Flanagan, D. C., and S. J. Livingston. 1995. WEPP User Summary. NSERL Report No. 11, W.
   Lafayette, IN: National Soil Erosion Research Laboratory. 131 p. Available at
   http://topsoil.nserl.purdue.edu/fpadmin/weppmain/docs/usersum.pdf

Flanagan, S.A., M.J. Furniss, T.S. Ledwith, S.Thiesen, M. Love, K. Moore, J. Ory. 1998.
   Methods for Inventory and Environmental Risk Assessment of Road Drainage Crossings.



Granite Creek Assessment                     WPN                                        Page 135
    Six Rivers National Forest, Pacific Southwest Region San Dimas Technology and
    Development Center San Dimas, California. 46 pages.

Freeze, R.A. and J.A. Cherry. 1979. Groundwater. Prentice-Hall, Englewood Cliffs, New
   Jersey. 604 pages.

Idaho Chapter American Fisheries Society.           2008. Fishes of Idaho Online Database.
   http://www.idahoafs.org/fishes.php

Idaho Department of Environmental Quality (IDEQ). 2008. Working Principles and Policies for
   the 2008 Integrated (303[d]/305[b]) Report – DRAFT. State of Idaho Department of
   Environmental Quality, Boise, Idaho

Idaho Department of Environmental Quality (IDEQ). 2005. Principles and Policies for the 2002
   Integrated (303(d)/305(b)) Report. State of Idaho Department of Environmental Quality,
   Boise, ID

Idaho Department of Fish and Game (IDFG). 2007. Fisheries Management Plan 2007-2012.
   IDFG, Boise, Idaho.

Idaho Panhandle National Forest (IPNF). 2008. Idaho Panhandle National Forests – System
   Roads GIS data. Idaho Panhandle National Forests - Supervisor's Office, Coeur d‟Alene,
   Idaho

Idaho Panhandle National Forest (IPNF). 2007. Idaho Panhandle National Forests – Vegetation
   GIS data. Idaho Panhandle National Forests - Supervisor's Office, Coeur d‟Alene, Idaho

Idaho Panhandle National Forest (IPNF). 2006. Idaho Panhandle National Forests – Landtypes
   and Ownership GIS data. Idaho Panhandle National Forests - Supervisor's Office, Coeur
   d‟Alene, Idaho

Idaho Panhandle National Forest (IPNF). 2004. Idaho Panhandle National Forests –
   Management Areas GIS data. Idaho Panhandle National Forests - Supervisor's Office, Coeur
   d‟Alene, Idaho

Interagency Advisory Committee on Water Data. 1982. Guidelines for determining flood-flow
    frequency: Bulletin 17B. Hydrology Subcommittee, Office of Water Data Coordination, U.S.
    Geological Survey, Reston, Virginia. 183 pp.

Irizarry, R.A. 1972. Lake and reservoir investigations, Job No. 1-a Survival and growth of
    resident and stocked cutthroat trout in Priest and Upper Priest Lakes, Job Completion Report
    Project F-53-R-7March 1971 to February 1972, IDFG, Boise, Idaho.

Irving, D.B. 1987. Cutthroat trout abundance, potential yield, and interaction with brook trout in
    Priest Lake tributaries. Master of Science Thesis. University of Idaho, Moscow, Idaho.



Granite Creek Assessment                      WPN                                         Page 136
Kershner, J., B. Roper, N. Bouwes, R. Henerson and E. Archer. 2004. An Analysis of Stream
   Habitat Conditions in Reference and Managed Watersheds on Some Federal Lands within the
   Columbia River Basin. N.A. Journal Fisheries Management, 24:1363-1375.

LaMarche, J., and D.P. Lettenmaier. 2001. Effects of Forest Roads on Flood Flows in the
   Deschutes River Basin, Washington, Earth Surf. Process. Landforms, 26, 115-134.

LaMarche, J., and D.P. Lettenmaier, 1998: Forest road effects on flood flows in the Deschutes
   River basin, Washington, Water Resources Series, Technical Report 158, University of
   Washington, Seattle.

Mallet, J. 1965. Channel clearance in Priest Lake tributary streams. Project Number F48-D-2,
   IDFG, Boise, Idaho.

Mauser, G.R. and V. Ellis. 1985. Enhancement of trout in large north Idaho lakes (1983 field
  season), Job Performance Report, Project F-73-R-6, IDFG, Boise, Idaho.

Mauser, G.R., R.W. Vogelsang, C.L. Smith. 1988. Enhancement of trout in large north Idaho
  lakes (1986 field season) , Job Performance Report, Project F-73-R-9, IDFG, Boise, Idaho.

Mayhew, J. (editor). 1987. Iowa Fish and Fishing. Iowa Dept. of Natural Resources, Des
  Moines, Iowa. Online at http://www.iowadnr.com/fish/iafish/brooktro.html

Miller, F.K., R.F. Burmester, R.E. Powell1, D.M. Miller, and P.D. Derkey. 1999. Digital
   geologic map of the Sandpoint 1° x 2° quadrangle, Washington, Idaho, and Montana. US
   Geological Survey Open-File Report 99-144, Version 1.0
Orma J. Smith Museum of Natural History. 2008. Query of the fish collection database, queried
on November 13, 2008.. The College of Idaho. Caldwell, Idaho.

Overton, K., M. Radko and R. Nelson. 1993.            Fish habitat conditions using the
   Northern/Intermountain Regions‟ inventory procedures for detecting differences on two
   differently managed watersheds. General Technical Report INT-300. USFS Intermountain
   Research Station, Ogden, Utah, 14 p.

Overton, K., S. Wollrab, B. Roberts and M. Radko. 1997. R1/R4 Fish and Fish Habitat
   Standard Inventory Procedures Handbook. USDA Forest Service, General Technical Report
   INY-GTR-346, USFS Intermountain Research Station, Ogden, Utah, 69 p.

Peterson, James T., S. P. Wollrab, U.S. Dept. of Agriculture Forest Service, 1999, An Analysis
    of Potential Stream Fish Monitoring Procedures for the Inland Northwest, Annual Report
    1999 to Bonneville Power Administration, Portland, OR, Contract No. 92AI25866, Project
    No. 92-032-00, 61 electronic pages (BPA Report DOE/BP-25866-1)

Pratt, K.L. 1992. A review of bull trout life history. Pages 5 to 9 in Howell, P.J. and D.V.
    Buchanan, editors. Proceedings of the Gearhart Mountain bull trout workshop. Oregon
    Chapter of the American Fisheries Society, Corvallis.


Granite Creek Assessment                     WPN                                       Page 137
PRISM. 2006. United States average maximum and minimum air temperatures and mean
   annual precipitation, 1971-2000. The PRISM Group at Oregon State University, Corvallis,
   Oregon, USA. http://www.prism.oregonstate.edu/

Rieman, B.E. & McIntyre, J.D. 1993. Demographic and habitat requirements for conservation of
   bull trout. General Technical Report INT-302. Ogden, Utah: U.S. Forest Service,
   Intermountain Research Station, 38 pp.
Rieman, B.E. and K.A. Apperson. 1989. Westslope cutthroat trout synopsis and analysis of
      fishery information. Idaho Fish and Game, Boise, Idaho.

Roper, B., J. L. Kershner, E. Archer, R. Henderson, and N. Bouwes. 2002. An evaluation of
   physical habitat attributes used to monitor streams. Journal of the American Water Resources
   Association 38:1–10.

Rosgen, D. 1996. Applied River Morphology. Wildland Hydrology, Pagosa Springs, Colorado.
   Available at http://www.wildlandhydrology.com/html/applied.htm

Rosgen, D. 1996. Applied River Morphology. Wildland Hydrology. Pagosa Springs, Co.

Rosgen, D. L. 1994. A classification of natural rivers. Catena 22(3):169-199. Available at
   http://csmres.jmu.edu/geollab/eaton/web/eaton_files/Courses/Hydrogeology/CLASS_OF_N
   ATURAL_RIVERS_300.pdf

Salminen, E. and C. Heider. 2007. Hydrology Analysis North Fork Coeur d‟Alene River
   Subbasin. Prepared for the Idaho Department of Environmental Quality, Watershed
   Professionals Network.
   http://www.deq.state.id.us/about/regions/north_fork_cda_river_wag/wpn_hydrology_report.p
   df

Simpson, J.C. and R. L. Wallace. 1978. Fishes of Idaho. University Press of Idaho, Moscow,
   Idaho.

Spokas, K. and F. Forcella. 2006. Estimating hourly incoming solar radiation from limited
   meteorological data. Weed Science 54:182–189.

Storck, P., and D.P. Lettenmaier, 2000: Trees, snow and flooding: An investigation of forest
   canopy effects on snow accumulation and melt at the plot and watershed scales in the Pacific
   Northwest, Water Resources Series, Technical Report 161, University of Washington,
   Seattle.

Thyer, M., J. Beckers, D. Spittlehouse, Y. Alila, and R. Winkler. 2004. Diagnosing a distributed
   hydrologic model for two high-elevation forested catchments based on detailed stand- and
   basin-scale data, Water Resources Research 40, W01103, doi:10.1029/2003WR002414.




Granite Creek Assessment                     WPN                                        Page 138
Trout Unlimited 2006. Westslope cutthroat trout species summary. Trout Unlimited
   Conservation Success Index.
   http://www.tu.org/site/c.kkLRJ7MSKtH/b.4347987/k.F0D6/Westslope_Cutthroat.htm

U.S. Fish and Wildlife Service. 1998. A framework to assist in making Endangered Species Act
   determination of effect for individual or grouped actions at the bull trout subpopulation
   watershed scale. U.S. Fish and Wildlife Service, Portland, Oregon.
U.S. Fish and Wildlife Service. 1999. Status Review for westslope cutthroat trout in the United
       States. Region 1 and 6, U.S. Fish and Wildlife Service, Portland, Oregon and Denver,
       Colorado.

U.S. Fish and Wildlife Service. 2002. Draft Recovery Plan for the Klamath River and Columbia
   River Distinct Population Segments of Bull Trout (Salvelinus confluentus). U.S. Fish and
   Wildlife Service, Portland, Oregon. http://www.fws.gov/pacific/bulltrout/

Vanshaar, J., and D.P. Lettenmaier, 2001: Effects of land cover change on the hydrologic
   response of Pacific Northwest forested catchments, Water Resources Series, Technical
   Report 165, University of Washington, Seattle.

Waichler, S.R., and M.S. Wigmosta. 2003. Development of Hourly Meteorological Values From
  Daily Data and Significance to Hydrological Modeling at H.J. Andrews Experimental Forest.
  Journal of Hydrometeorology 4(2):251-263.

Williams, C.N. Jr., M.J. Menne, R.S. Vose, and D.R. Easterling. 2008. Long-Term Daily and
   Monthly Climate Records from Stations Across the Contiguous United States. National
   Climatic Data Center, National Oceanic and Atmospheric Administration.
   http://cdiac.ornl.gov/epubs/ndp/ushcn/newushcn.html

Zar, J.H.. 1974. Biostatistical analysis. Third Edition. Prentice-Hall. Upper Saddle River, NJ.




Granite Creek Assessment                       WPN                                         Page 139

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:89
posted:10/8/2010
language:English
pages:149