Get documents about "hfcourse_snowppt - Slide 1
Document Sample
MODELING OF COLD SEASON
PROCESSES
• Snow Ablation and Accumulation
• Frozen Ground Processes
Snow Ablation and Accumulation
• Snow physics
• Commonly Used Parameterizations
• How well do we model snow cover
• NOAh and SNOW-17 Models
• Input Data Uncertainties
• Model Structure/Parameter Uncertainties
• Effect of Weather Conditions
• Summary
• Where to go in Snow Modeling?
Snow Physics
Sources of Energy Exchange
• Net Radiation, Qn
– Solar Incoming – Reflected
– Incoming from Atmosphere – Outgoing from Snow
• Turbulent Exchange
– Sensible Heat: Based on the Temperature Gradient, Qh
– Latent Heat: Based on the Moisture Gradient, Qe
• Mass Transfer: Heat from Precipitation, Qm
• Heat Exchange with the Underlying Soil, Qg
Qn + Qh + Qe + Qm + Qg = ΔQ
Snow Physics
Problem Complexities
• Net Solar Radiation is not easy observable and it
varies significantly with weather and snow conditions
(sky, snow reflectance (albedo), forest, topography)
Snow Physics:
Problem Complexities (cont.)
• Turbulent Transfer is a very non-linear Function of
the Wind Speed. A Turbulent Transfer Coefficient of
this Function is defined under a Number of
Assumptions.
• Large Uncertainties in Internal Snow Cover
Processes: Density Changes in Time and Depth,
Liquid Water Refreezing, Liquid Water Retention and
Transmission, Rain Effect.
• Spatial Variability and Redistribution due to Wind
Commonly Used Parameterizations:
Energy Based vs. Temperature Index
• Energy Based
– Snowpack Layers
– Albedo Definition
– Snowpack Property Dynamics
– Wind Function Approximation
• Temperature Index
– Fully Index Based or Combination
– Snow Melt Rate Dynamics
– Wind Function Approximation
– Snowpack Property Dynamics
How well do We Model Snow Cover
How well do We Model Snow Cover
NOAh Energy Based Model
• One Layer snowpack
• Variable Snow Properties
• Multi-Layer Soil Profile
Snow
Surface
Soil surface runoff
Frost Infiltration
Unfrozen
NOAh Energy Based Model
Snow ablation and accumulation
net latent sensible
radiation heat heat
Snow density Varies depending on SWE
Thermal Conductivity and snow temperature
[Anderson-Yosida’s snow
compaction model].
Thermal conductivity varies
with snow density
Spatial variability Non-linear function of SWE
100% coverage Varies depending on vegetation
threshold
Albedo Weighted snow and bare soil
albedos
ground
heat
SNOW-17: Temperature Index Model
• Conceptual model
• Simplified Heat Balance During Rainfall Events
• Degree-Day Melt Factor During Non-Rain Events
• Input Variables: Air Temperature & Precipitation
• Output: Melt, WE, Depth, Areal Extend
• Snow Cover Processes
• Snow accumulation
• Surface Energy Exchange
• Snow Compaction
• Liquid Water Transmission
• Watershed /Areal Application
• Areal Extent of Snow Cover
SNOW-17: Surface Energy Exchange
Ta > 0oC – Snowmelt
• If Rainfall > 0.25 mm/hr
- Simplified Heat Balance Equation:
- no Solar Radiation,
- Atmosphere & Snow is a Black Body,
- Relative Humidity = 90%,
- Rainwater at an Air Temperature
• Else
- Degree-day Melt Factor (Seasonal Variable)
Ta < 0oC – No Snowmelt
- Estimate Change in Snow Heat Deficit
SNOW-17: Model Parameters
Major Minor
• Accumulation SCF (1-1.2) PXTEMP (0.6-0.2)
• Surface Melt MFMAX (1.7-2.0) MBASE (0.0)
MFMIN (0.2-0.6)
UADJ (0.002*U)
• Heat Storage & NMF (0.15)
Water Retention TIPM (0.05)
PLWHC (0.02-0.05)
• Areal Coverage SI ( > Wmax )
Depletion Curve
• Ground Melt DAYGM
SWE (Plot 3) and Depth (Plot 4) from NOAh (red lines) &
SNOW-17 (white lines), Swiss Alps Site, 1992-1993
Input Data Uncertainties
Required Data for Energy Budget Models
• Incoming Solar Radiation
• Reflected Solar Radiation or Albedo
• Incoming Long-Wave Radiation
• Snow Surface Temperature
• Wind Speed
• Air Temperature
• Dew-Point
• Precipitation and Phase
• Wet Bulb Temperature
• Snow Cover Profile (Density, Temperature, etc.)
• Soil Temperature
Input Data Uncertainties
1000 700
(a) (b)
600
800
500
600
SWE, mm
SWE, mm
400
300
400
200
200
100
0 0
60 90 120 150 180 210 240 270 300 330 0 30 60 90 120 150 180 210 240
Days from A ugus t 1992 Days from Oc tober 1981
NOA H-LS M S NOW -17 Meas ured NOA H-LS M S NOW -17
Swiss Alps Site, 1992-1993 Goose Bay Site, Canada, 1981-1982
Input Data Uncertainties
Effect of Net Radiation
350 1000
300
800
Snow water equivalent, mm
Snow water equivalent, mm
250
600
200
150
400
100
200
50
0 0
0 30 60 90 120 150 180 70 100 130 160 190 220 250 280 310 340
Days from Novem ber 1996 Days from A ugus t 1992
Sleepers River Site, USA, 1996-1997 Swiss Alps Site, 1992-1993
Observed fluxes (solid lines), Incoming solar radiation with a 20 Wm-2 error
added (dotted lines), and the constant albedo of 0.7 (dashed lines).
Input Data Uncertainties
Effect of Wind Speed: V=1m/s (white), V=15m/s (colored)
Goose Bay Site, Canada
Input Data Uncertainties
Effect of Precipitation Phase: with use (red), w/o use (blue)
150 150
300 300
Danville, Vermont, USA
125 200 125 200
Snow water equiwalent (WE), mm
Snow water equiwalent (WE), mm
100 100
100 100
Snow depth, cm
Snow depth, cm
0 0
75 75
-100 -100
-200 -200
50 50
-300 -300
25 25
-400 -400
0 -500 0 -500
225 240 255 270 285 300 315 330 345 360 375 24 39 194 209 224 239 254 269 284 299 314 329 344 360 375 24 39
Time, day s Time, day s
Hobs Hsim, w/o P-type Hsim, w P-type Hobs Hsim, w/o P-type Hsim, w P-type
175 Wobs Ws im, w/o P-type Ws im,w- Ptype 400 150 Wobs Ws im, w/o P-type Ws im,w- Ptype
200
300
150
125
100
200
Snow water equiwalent (WE), mm
Snow water equiwalent (WE), mm
125
100 100 0
Snow depth, cm
100 0 Snow depth, cm -100
75
75 -100
-200
-200 50
50
-300
-300
25
25 -400
-400
0 -500 0 -500
255 270 285 300 315 330 346 361 376 25 40 55 70 255 270 285 300 315 330 346 361 376 25 40
Time, day s Time, day s
Hobs Hsim, w/o P-type Hsim, w P-type Hobs Hsim, w/o P-type Hsim, w P-type
Wobs Ws im, w/o P-type Ws im,w- Ptype Wobs Ws im, w/o P-type Ws im,w- Ptype
Model Structure/Parameter Uncertainties
Effect of Weather Conditions
SWE and Depth from NOAh (red) & SNOW-17 (white), Col de Porte Site, France
Effect of Weather Conditions
Simulated snowmelt variables from
NOAH-LSM (dashed lines)
& SNOW-17 (solid lines)
Col de Porte, France
February 10-22, 1998.
Weather:
Very low wind;
High diurnal amplitude
of air temperature
Effect:
(a) Much faster snowmelt
from SNOW-17;
(b) Significant effect of
net radiation and liquid
water refreezing
Effect of Weather Conditions
Weissfluhjoch Site
Switzerland, July 11-16, 1993
Weather:
Negative Tair
High wind speed
Effect:
Significant snowmelt
from NOAH,
Accumulation from
SNOW-17
Summary
• Model Complexity does not Guaranty Accuracy
• Temperature Index Models Provide Practically
Reasonable Results, however They are
Sensitive to Weather Conditions, Specifically
Wind and Solar Radiation Conditions
• Energy Based Models are Sensitive to Input
Data Errors, Specifically Wind, Solar Radiation
and Albedo Treatment
• Some Calibration/Tunning is Needed to get
Better Results from Both Simple and Complex
Models
Where to go in Snow Modeling?
• Energy and Temperature Based Models may
Coexist for a Long Time, Specifically in River
Runoff Prediction
• Improvements to Temperature Based Models
can be Achieved by
– Incorporating Wind and Humidity Data
– Regionalization of the Most Critical Parameters to run
in a Distributed Mode
• Energy Based Model Improvements
– Define the Most Reliable Data Sources Using
Sensitivity Tests
– Better Parameterizations of Albedo and Wind
Function
– Improvement in Prediction of Weather Variables
Appendix 1
Appendix 2
Appendix 3
Appendix 4
