Fractional snow cover estimation

1
Fractional snow cover estimation in the complex
alpine-forested areas using MODIS and Landsat
research conducted under guidance of Dr. K.
Hirschboeck
Elzbieta Czyzowska Wisniewski
2
Why do we have to study winter precipitation and
snow cover in the Southwest USA
- winter precipitation and snow cover supply up
to 90 of annual precipitation
- winter precipitation and snow cover is very
sensitive to global climate change
- estimation of snow cover and SWE is needed for
better understanding of spatial and temporal
variation of these elements in the decadal and
century time scales

- estimation of snow water equivalent (SWE)
based on instrumental data remote sensing data
complement sparse ground data
3
Remote sensing introduction
4
Snow detection current methods NDSI
Ground
Satellite
30 m
30 m
5
Ikonos as a source information for snow cover
monitoring
Landsat
snow
30 m
how much of the pixel is covered by snow ?
30 m
6
Ikonos as a source information for snow cover
monitoring
Landsat
30 m
snow cover
1 Landsat pixel (30m) 900 Ikonos pixels (1m) 1
MODIS pixel (500m) 278 Landsat pixels 250 000
Ikonos
7
Landsat and MODIS fractional snow cover
8
Future plans 2005 - 2008
IIa. Comparison of fractional snow area
distribution with snow maps produced by MODIS
NDSILandsat NDSI NOAA SNOTEL
I. Neural network based fractional snow cover
estimator
DEM
MODIS Terra Aqua
Ikonos
Fractional snow cover in mountain areas
GIS
Landsat ETM
Model
IIb. Weekly and annual MODIS fractional snow
cover distribution
IV. SWE at 1st April (1972 2007)
Models
IIIa. Spatial and temporal changes of SWE based
on MODIS fractional snow cover

ANN
V. Reconstruction of SWE and snow days (1400
2007)
9
(No Transcript)