Title :
Use of MODIS enhanced vegetation index to detect seasonal patterns of leaf phenology in central Amazon várzea forest
Author :
Hess, Laura ; Ratana, Piyachat ; Huete, Alfredo ; Potter, Chris ; Melack, John
Author_Institution :
Inst. for Comput. Earth Syst. Sci., Univ. of California Santa Barbara, Santa Barbara, CA, USA
Abstract :
MODIS 16-day composite EVI, NDVI, and VI Quality Analysis values for 2000-2005 were extracted for 21 varzea forest sites along the Solimoes-Amazon floodplain west of Manaus, Brazil. VI values were filtered to exclude dates with VI-QA values greater than 3, and time series of median values of the remaining pixels were examined in conjunction with river stage levels recorded at the Manacapuru gauge. All sites showed a regular seasonal variation in EVI, ranging from a mean low for all sites of 0.41 to a mean high of 0.61. The amplitude of variability in NDVI was about 50% that of EVI. Minimum EVI, corresponding to minimum leaf area, occurred in late May, about 40 days preceding maximum river stage; EVI peaked in mid-October, about 30 days before lowest river levels. These temporal patterns are in general agreement with field observations of leaf phenology at va¿rzea stands near Manaus.
Keywords :
geophysical techniques; pattern recognition; remote sensing; vegetation; AD 2000 to 2005; Brazil; MODIS enhanced vegetation index; Manacapuru gauge; Manaus; Solimoes-Amazon floodplain; central Amazon varzea forest; leaf phenology; pattern detection; Floods; Geoscience; Image resolution; Lakes; MODIS; NASA; Production; Rivers; Soil; Vegetation mapping; EVI; MODIS; Várzea; flooded forest; phenology;
Conference_Titel :
Geoscience and Remote Sensing Symposium,2009 IEEE International,IGARSS 2009
Conference_Location :
Cape Town
Print_ISBN :
978-1-4244-3394-0
Electronic_ISBN :
978-1-4244-3395-7
DOI :
10.1109/IGARSS.2009.5417550