Relationship Between NDVI and ENSO/Arid Indices in Pampa Grasslands in South Brazil and Uruguay.

The occurrence of annual cycle’s variability, which the El Niño Southern Oscillation (ENSO) is the main factor, has intensified water use and the occurrence of droughts in several regions with grasslands like the Pampa in South Brazil and Uruguay. Besides recent land use change, climate variability has been a key factor in the modification of vegetation patterns. ARID (Agricultural Reference Index for Drought) has been found to be a good indicator of the effects of climatic conditions in the vegetation. In addition, the Normalized Difference Vegetation Index (NDVI) shows significant variability in annual phenological response and could be used to characterize grasslands patterns. This study describes the spatial variability of the correlations between time series images of NDVI and climatic indices in this the Pampa region. ENSO was characterized by the Multivariate ENSO Index (MEI), and ARID was calculated using daily spatial weather data from the Climate Forecast System Reanalysis. The NDVI was acquired from MODIS, 16 days composited 250 m resolution, for 2000 to 2011, and from GIMMS, biweekly composited 8-km resolution, for 1981 to 2006. The analyses conducted were based on monthly periods. Preliminary results indicate high positive correlations between NDVI and MEI, mainly on summer on grasslands. The vegetation decrease during La Niña phases, and the opposite behavior during El Niño phases corroborated with the observed trends and this patterns. The positive correlation found during winter its being investigated, we speculate that it may be caused by an increase in freeze events during La Niña and warm moist soils during El Niño. Results of correlations between NDVI and ARID indicated a threshold occurrence around 0.5 for ARID beyond which vegetation starts to be affected mainly during the summer, but it requires further investigation.