autumn phenology events in northern Scandinavia by applying double logistic
functions (Fischer 1994) to MODIS NDVI time series data.
MODIS VIs have been used to characterize the phenology patterns of varying
physiognomic vegetation classes in the Brazilian cerrado biome (Ratana et al.
2005). Kawamura et al. (2005) monitored short-term phenological changes in
rangeland forage conditions with EVI in the semi-arid Xilingol steppe in Inner
Mongolia. They were able to estimate forage quantity and derive seasonal changes
in live biomass and standing crude protein amounts over areas with different
grazing intensities, useful in providing timing information for hay cutting based on
nutritive value to range managers. NDVI data from the MODIS sensor has provided
timely monitoring of locust outbreaks in East China based on its accurate assessments of vegetation conditions (Zha et al. 2005). Xiao et al. (2006) developed a
MODIS-based VI phenology algorithm using NDVI, EVI, and LSWI for mapping
paddy rice distributions in support of irrigation, food security, trace gas emission
estimates, and risk assessment of avian flu over South and Southeast Asia.
Fig. 1.13 a Three-dimensional phenological transition dates of MODIS EVI along an east to
west transect in which seasonally dry rainforest grades to a perhumid rainforest. The eastern
portion is an area of forest clearing and disturbance; and b tower flux measurements of gross
primary productivity (GPP) compared with MODIS EVI at the seasonally dry Tapajos forest site
and adjacent forest disturbance area. Adapted from Huete et al. (2006), Copyright (2006).
Reprinted with permission from John Wiley and Sons
22
A. Huete et al.
Précédent

- 33/236

Suivant