Get documents about "mrm_mdm
Document Sample
Case Study of the ArcGIS Marine
Data Model: Examining Habitat
Utilization Patterns of Reef Fish
Along the West Coast of Hawaii
Photos from www.fishbase.org
Marine Resource Management
Marine resources are at risk (eg. Leslie et al.,
2003, Mumby et al., 2001, Puniwai et al., 2003)
Growing interest in Marine Protected
Areas (MPAs)
• Management tool
• Effective both ecologically and
economically (eg. Carr et al., 2003, Friedlander and Brown,
2003, Tissot et al., 2004)
2
Marine Resource
Management
How can we make
MPAs as effective
as possible?
• Identify areas with
patterns of high
habitat utilization
3
Two Main Questions
Question 1: What is the habitat utilization
of specific fish species at a coarse-scale?
4
Two Main Questions
Question 1: What is the habitat utilization
of specific fish species at a coarse-scale?
• To answer question, use ArcGIS MDM
5
Two Main Questions
Question 1: What is the habitat utilization
of specific fish species at a coarse-scale?
• To answer question, use ArcGIS MDM
• Use real data to test model itself
6
Two Main Questions
Question 1: What is the habitat utilization
of specific fish species at a coarse-scale?
• To answer question, use ArcGIS MDM
• Use real data to test model itself
Question 2: How well does the MDM meet
the needs of the marine GIS community?
7
Why use the MDM?
1. Standardized template
Implement project
Organize data
2. GIS and data management
Easy to share
Build upon
3. Querying ability
4. Object orientation
8
Testing the ArcGIS Marine Data
Model: 3 Goals
Case study tested 3 goals (Wright
et al, 2002):
Goal 1: Assembling, managing, and
querying in ArcGIS
Goal 2: Provide a template
Goal 3: Improve the understanding of
geodatabase models
9
Data Gathered
West Hawaii Aquarium Project
(WHAP) (Tissot et al., 2004)
Coral Reef Assessment and
Monitoring Program (CRAMP) (Brown et
al., 2004)
State of Hawaii’s Division of Aquatic
Resources (DAR) (DAR, 2004)
Saving Maui’s Reefs project (Brown, 1998)
Individual peer-reviewed journal
articles
• Date, sites, method, frequency, info
10
Formatting: Getting Data into the MDM
Formatting
data to fit
MDM template
Portion of MDM
MS Access
11
http://dusk.geo.orst.edu/djl/arcgis/diag.html
Importing: Personalizing the MDM
Adding fields to the
feature classes,
tables
Adding new tables
and relationships
Can add new fields
Attention to detail
required
12
Database Set Up
Table Names Survey Divemaster RunDetail Species
Location Information
(Survey Point) (Integrated
Species)
Information SurveyID Reference Transect_Co TaxonID
stored in Reference location TaxonID EBrownID
each table location Transect_Co Count TaxonName
Latitude TransectRunID Size Alpha
Longitude Transect Comments Type
Island Date Family
Site Time FamCode
Area_Desc Observer Quest
Depth Source DAR
Comments Synomyn
HawaiiName
CommonName
Trophic 13
Status
14
Querying
15
Goal 3: Improve Understanding of
Geodatabase Model
Benefits of
MDM
The MDM
Tutorial
Response
16
Results: Goals
Goals Results
Goal 1: Common structure for Met
assembling, managing, and
querying in ArcGIS
Goal 2: Provide a template Met
Goal 3: Improve the understanding of Met
the Geodatabase model
17
Habitat Utilization
1. What is the habitat utilization
pattern of selected fish species at a
coarse-scale?
2. Can fine-scale substrate info be
correlated to coarse-scale habitat
info?
3. Does depth play a role in the
location of coarse-scale habitat
types?
18
Methodology: Habitat Utilization
Select WHAP data
Done with the MDM 2.
Classify WHAP transect
points by habitat Calculate mean % of
and zone fine-scale substrates
and classify by
presence in coarse-
1. 3. scale habitat types
Run logistic Identify depth
regression analysis values of each site Run two-sample
t-test
Examine habitat Run two-sample
utilization patterns t-test Determine if certain
substrates are found
more often in RA
Determine if depth or RCB habitats
plays a role in
habitat classification
19
Fish Analysis done with MDM
Select WHAP data
Determined site
location in each
habitat type and
zone
20
Sites classified by Habitat and Zone
Reef/Aggregate Reef/Colonized Volcanic
Habitat Type Coral (RA) Rock/Boulder (RCB) No data
Location Anaehoomalu Kalahiki Beach Honokohau
Keawaiki Kamilo Gultch Hookena (Auau)
Keei Kaupulehu Makalawena
N. Keauhou Kealakekua Bay Manuka
S. Oneo Bay Keopuka Omakaa
Red Hill Kualanui Pt Wawaloli
Waiakailio Bay Lapakahi Wawaloli Beach
Papawai
Puako
Zone: All sites with data were located in the shelf zone, with the exception
of Puako which was located in the fore-reef zone.
21
Querying to Determine Fish Location
22
Logistic Regression Analysis
Percentage of species found in Percentage of species found in
Fish Species RA habitat type RCB habitat type
N (absent) Y (present) N (absent) Y (present)
Acanthurus triostegus 57.1 42.8 55.5 44.4
Ctenochaetus strigosus 14.2 85.7 0 100
Scarus dubius 14.2 85.7 0 100
Scarus psittacus 0 100 11.1 88.8
Zebrasoma flavescens 0 100 0 100
Chaetodon ornatissimus 0 100 0 100
Chaetodon quadrimaculatus** 57.1 42.8 0 100
Forcipiger spp. 0 100 0 100
Acanthurus achilles 0 100 0 100
Naso lituratus 0 100 0 100
Paracirrhites arcatus 0 100 0 100
Paracirrhites foresteri 0 100 0 100 23
Acanthurus nigrofuscus 0 100 0 100
Total % per location Total % per location
0
10
20
30
40
50
0
10
20
30
40
Boulder Boulder
Dead Coral Dead Coral
Porites Porites
Compressa Compressa
Porites
Puako
Porites
Lobata
Anaehoomalu
Lobata
Substrate
Substrate
Pocillopora
meandrina Pocillopora
meandrina
Rubble
Rubble
Sand
Sand
coarse-scale habitat info?
24
Can fine-scale substrate info be correlated to
Substrate Results
Substrate RA Mean % Cover RCB Mean % Cover P
type (SD) (SD)
Boulder 0.29 (0.59) 2.26 (3.92) 0.153
Porites 34.9 (13.6) 12.2 (11.1) 0.004*
compressa
Porites 23.26 (9.97) 33.1 (8.96) 0.059
lobata
Pocillopora 0.167 (0.44) 0.88 (1.87) 0.277
meandrina
6.73 (6.23) 7.14 (6.68) 0.897
Rubble
0.92 (1.77) 7.00 (10.9) 0.117
Sand
28.6 (10.1) 32.3 (17.6) 0.601
Dead Coral 25
Study Site Habitat Type Depth (ft)
Anaehoomalu RA 32.00 Does depth play a role
Kealakekua Bay RCB 32.00
Puako Bay RCB 32.00
in the location of
Papawai RCB 34.00 coarse-scale habitat
Keopuka RCB 35.00 types?
Kualanui Pt. RCB 36.00
S. Oneo Bay RA 37.00
Ke'ei RA 39.00 Mean depth for
Hookena (Kalahiki)
RCB 39.00
RCB habitat type
Kamilo Gultch RCB 40.00 is 37 +/- 3.65ft
Lapakahi RCB 40.00 Mean depth for RA
Ka'upulehu RCB 42.00
habitat type is
41.71 +/- 5.77ft
Waiakailio Bay RA 45.00
Keawaiki RA 46.00
N. Keauhou RA 46.00 26
Red Hill RA 47.00
Does depth play a role in the location of
coarse-scale habitat types?
Boxplot of Depth by Layer
P=0.088 48
Trend
46
44
indicates that 42
RA habitat Depth
40
occur deeper
38
36
than RCB 34
32
30
RA RCB
Layer
27
Discussion: Habitat Analysis
Future research:
Tie fine-scale 1. How do specific
substrate info be fish utilize coarse-
tied to specific scale habitats?
fish? A. WHAP A: Inconclusive
Fish Info
C. Fine-scale B. Coarse-scale
Regional Habitat
WHAP substrate NOAA habitat
Utilization Patterns?
info info
3. Does depth play a role
in the location of RA and 2. Can fine-scale substrate
RCB habitat types? info be correlated to coarse-
A: Strong trend, p=.088 scale habitat types?
A: Yes, PC (p<.05) in RA and
a strong trend with PL 28
(p=.059) in RBC
Conclusions: Management
Use information from this study to
evaluate current protected areas
For example:
• RA habitat shows trends toward being located
deeper
• P. compressa is found significantly more often
in RA habitat
• P. compressa provides and important habitat
for yellow tang (Tissot et al., 2003, Dollar,
1982)
• Thus, to protect the yellow tang, must have
protected areas that extend into deeper waters
• Consistent with study that MPA encompassing
range to depth values shown to be more
efficient for yellow tang (Tissot et al, 2004) 29
Conclusions
The MDM is a powerful tool
Information to aid managers
evaluate the West Coast MPAs
Future research
• Data to HNHP
30
The 13 reef fish analyzed
Aquarium Fish Non-aquarium fish
Acanthurus achilles Acanthurus nigrofuscus
(Achilles Tang) (Brown surgeonfish)
Chaetodon ornatissimus Acanthurus triostegus
(Ornate Butterflyfish) (Convict surgeonfish)
Chaetodon Ctenochaetus strigosus
quadrimaculatus (Striated surgeonfish)
(Foursport Butterflyfish)
Paracirrhites arcatus
Forcipiger spp (Arc-eye hawkfish)
(eg., Longnose
Butterflyfish) Paracirrhites foresteri
(Blackside hawkfish)
Naso lituratus
(Orangespine Scarus dubius
Unicornfish) (Regal parrot)
Zebrasoma flavescens Scarus psittacus
(Yellow Tang) (Common parrotfish) 31
All photos from www.fishbase.org
Statistical Analysis
Boxplot of PC by Layer
Porites compressa
50
(finger coral)
40
P=.004
30
Significantly more
PC
20
abundant in RA
10
relative to RCB
0
RA RCB
Layer
50
Boxplot of PL by Layer
Porites lobata (lobe
40
coral) p=.059
Shows strong
PL
30
trends toward
being found in RCB
20
relative to RA
10
RA RCB 32
Layer
What is the habitat utilization pattern of
selected fish species at a coarse-scale?
?
www.fishbase.org
fine vs. coarse-scale
habitat information
Importance
• Habitat utilization
• Management
33
Can fine-Scale Substrate Info be Correlated
to coarse-Scale Habitat Info?
fine-scale WHAP
substrate information ?
Importance:
• Cost
• Time
• Information
34
Study Site Habitat Type Depth (ft) Does depth play a
Anaehoomalu RA 32.00
Kealakekua Bay RCB 32.00
role in the
Puako Bay RCB 32.00 classification of
Papawai RCB 34.00
Keopuka RCB 35.00 coarse-scale habitat
Kualanui Pt.
S. Oneo Bay
RCB
RA
36.00
37.00
types?
Ke'ei RA 39.00
Hookena
(Kalahiki) RCB 39.00 Importance:
Kamilo Gultch RCB 40.00 • Management
Lapakahi RCB 40.00
Ka'upulehu RCB 42.00
Waiakailio Bay RA 45.00
Keawaiki RA 46.00
N. Keauhou RA 46.00
Red Hill RA 47.00
35
Why use the MDM Model?
Why a data model?
1. Standardized
template
Implement project
Organize data
2. GIS and data
management
Easy to share
Build upon
3. Querying ability
4. Object orientation
36
