Satellite-Monitored Snow Cover in the Climate System - Satellite-based Applications on Climate Change - page 178

only give a potentially representative status in the given locality. Also, the snow
extent or snow-cover fraction (SCF) cannot be measured easily in situ.
With developments in remote sensing, satellite-derived snow information has
become an important alternative data source. Weekly snow mapping of the Northern
Hemisphere using National Oceanographic and Atmospheric Administration
(NOAA) and National Environmental Satellite, Data, and Information Service
(NESDIS) data began in 1966 (Robinson et al. 1993 ). However, the coarse spatial
resolution (1 over Northern Hemisphere) in the NESDIS data cannot well represent
the patchy and shallow snow cover at middle latitudes. In recent decades, the NOAA
Interactive Multisensor Snow and Ice Mapping System (IMS) increased spatial
resolution of snow maps to 24 km (Ramsay 1998 ). Besides visible images, passive
microwave SSMR-SMM/I (Chang et al. 1987 ), and AMSR-E (Chang and Rango
2000 ) also provide snow water estimation over the globe at low resolution, although
the accuracy still cannot reach the requirements for many modeling applications.
With improvements in polar-orbiting satellites, the National Aeronautics and
Space Administration (NASA) Earth Observing System (EOS) Terra satellite was
launched on December 18, 1999, with a complement of five instruments, one of
which is Moderate-Resolution Imaging Spectroradiometer (MODIS). Besides the
comprehensive observations of cloud, ocean, and earth surface characteristics
available from the Terra MODIS, a snow-cover product has been available since
February 2000. With substantially improved spatial resolution (500 m globally),
high temporal frequency (daily), enhanced capability to separate snow and clouds
(Hall et al. 2001 , 2002 ) due to more spectral bands (particularly in the shortwave
infrared), as well as a consistently applied, automated snow-mapping algorithm
(Riggs and Hall 2002 ), MODIS provides quantitative monitoring of global snow
extent, even in inaccessible regions such as the Tibetan Plateau (TP) and the
Himalayas. In particular, the single satellite platform provides excellent consis-
tency with MODIS snow data that are hard to obtain in previous satellite datasets.
The quality of MODIS snow data has been evaluated in several previous studies
(e.g., Hall et al. 2001 ; Klein and Barnett 2003 ). As determined by prototype
MODIS data, annually averaged, estimated error for Northern Hemisphere snow-
cover maps is approximately 8% in the absence of clouds (Hall et al. 2002 ). The
cloud mask, however, must be applied carefully, since there is a tendency to
overestimate cloud cover (Ackerman et al. 1998 ). In addition, confusion in
identifying cloud over snow has been observed in high-elevation regions, e.g., the
Sierra Nevada in California and Southern Alps of New Zealand (Hall et al. 2001 ).
This problem has been partially improved in the most recent MODIS data products
(Riggs and Hall 2002 ). Pu et al. ( 2007 ) evaluate the MODIS snow data over the
Tibetan Plateau, the third “polar” in the world, shown a practical good detection of
scatter patchy snow over these regions.
Figure 12.1 shows the MODIS-monitored Northern Hemisphere snow-cover
fraction (SCF, %) from January to December. Note that during winter there are
some blank areas in high-latitude region due to the polar night. From September to
February, snows accumulate in high-latitude polar regions and then gradually
extend to the south. The Tibetan Plateau, although located in the middle latitudes