Evaluation of SNODAS
Snow Depth and SWE
D.W. Clow, L. Nanus, K. Verdin, J. Schmidt
Hydrolological Processes, 2012
doi: 10.1002/hyp.9385.
CWCB
National Snow Analyses - NOHRSC - The ultim ate source for snow inform ation
6/ 15/ 11 3:56 PM
weather.gov
National Weather Service
SNODAS


SNOw Data
ASsimilation model
NOAA - NWS



www.nohrsc.nws.gov/nsa/
Daily estimates of
snow depth, SWE, …
1 km resolution for
conterminous U.S.
National Operational Hydrologic
Remote Sensing Center
Home
News
Home
Organization
Search
Enter Search Here
National Snow Analyses
Snow Information
National Analyses
Interactive Maps
3D Visualization
Airborne Surveys
Satellite Obs
Forecasts
Data Archive
SHEF Products
Snow Reports
Model Assimilation Schedule
Snow Survey Schedule
Automated Model Discussion:
June 15, 2011
Area Covered By Snow:
2.4%
Area Covered Last Month: 4.5%
Snow Depth
Average:
1.2 in
Minimum:
0.0 in
Maximum:
973.9 in
Std. Dev.:
9.9 in
Snow Water Equivalent
Average:
0.5 in
Minimum:
0.0 in
Maximum:
508.9 in
Std. Dev.:
4.2 in
more...
Metric Units...
Click On Map for Regional Analyses
Observations near
City, ST
Go
Go
Science/Technology
NOHRSC
GIS Data Sets
Special Purpose
Imagery
Select Region and Date
About The
NOHRSC
Staff
National
Snow Water Equivalent
2011
June
15
Snow Depth
-
+
Go
Average Snowpack Temp
NOAA Links
Snow Climatology
Related Links
Help
Help and FAQ
Site Map
Contact Us
Please Send Us
Comments!
Animate: Season --- Two weeks -- Animate: Season --- Two weeks -- Animate: Season --- Two weeks -- One Day
- One Day
- One Day
SWE Change
Snow Precipitation
Snow Melt
Animate: Season --- Two weeks -- Animate: Season --- Two weeks -- Animate: Season --- Two weeks -- One Day
- One Day
- One Day
Blowing Snow Sublimation
Surface Sublimation
Non-Snow Precipitation
Animate: Season --- Two weeks -- Animate: Season --- Two weeks -- Animate: Season --- Two weeks -- One Day
- One Day
- One Day
http:/ / www.nohrsc.nws.gov/ nsa/
Page 1 of 2
Problem and Approach

Problem:


Little data left for
validation
Approach:


Snow Surveys
Water-Balance
Calculations
Snow Surveys

Snow Depths
Snow depth at
~40 points/grid

• ~50 sites
Snow Surveys

Snow Depths


Snow density in
pits
Snow Surveys

Snow Depths



SWE = Depth x
Density
Snow depths agreed well in forest
Estimates were biased in alpine zone
Estimates were biased in alpine zone
Watershed Scale

How does SNODAS perform at
watershed scale?

Do errors cancel?

Water balance approach
Water Balance Calculations

Runoff =
SWEApril 1 + precipitation - sublimation - ET ± storage
Snowmelt Period:
April – June
Runoff is from USGS gages
All other terms from SNODAS
~25 sites
Water balance results indicate
moderate agreement
R2=0.52
ET and recharge account for ~ 20 cm
20 cm
Factors influencing snow distribution
Loch Vale, Rocky Mountain National Park

Snow Survey
Conducted in 2003
(318 points)

Regression model
developed using:

Wind Direction

Slope

Aspect

Vegetation
SNODAS does
not account for
wind redistribution
of snow
Examples of Wind Drifts

Wind drifts can be caused by


topographic features (above)
or vegetation (right)
Wind Redistribution in
SNODAS grids
Deposition
215
cm
121
cm
113
cm
Wind
Scour
 Effect
varies with distance
Wind Drift near Continental Divide

Persistent patterns
Sites were categorized

GIS analysis
of terrain in
upwind direction



Scour
Deposition
No effect
Adjusted SNODAS snow depths
agree well with measurements

Objective:


Improve
SNODAS
estimates
Model Includes:


SNODAS snow
depth
Wind effect
variable
Conclusions

SNODAS performed well in
forested areas, but not in
alpine zone

Water balance results
indicated moderate agreement

SNODAS could be improved
using topographic, vegetation,
meteorological information to
account for wind redistribution
Clow, et. al., 2012, Hydrol. Processes,
doi: 10.1002/hyp.9385.
Study performed by USGS in cooperation
with Colorado Water Conservation Board