Seasonal and interannual analysis of wetlands in South America using NOAA-AVHRR NDVI time series: the case of the Parana Delta Region
Articolo
Data di Pubblicazione:
2008
Abstract:
The use of NOAA-AVHRR NDVI time series from July 1981 to December 2000 was evaluated for the assessment of the functioning of a wetland macrosystem, the Parana River Delta. The spatial resolution of the dataset was 8 by 8 km. Spatial and temporal variations in NDVI pattern were analyzed and evidences for El Nino/South Oscillation events identified. We studied five wetland units (WUs) classified on the basis of landscape pattern and dominant hydrologic regime. Spearman rank correlations were performed among the NDVI time series of the different WUs. NDVI time series were correlated with water level in the Parana River and with records of local rainfall. In order to obtain a synthetic model of NDVI patterns, the autocorrelation functions (ACF) were estimated for each of the WUs. Results indicated that monthly mean NDVI values for all WUs showed a similar annual seasonal pattern, suggesting a control from the plant annual cycle on the NDVI signal. Besides, two general NDVI patterns were identified. The first pattern is represented by WUs under fluvial hydrologic regime. This is subjected to a significative interannual variability associated mainly to ENSO events. The second pattern corresponds to WUs with a very regular NDVI patterns. It includes wetlands which water input corresponds to tides or to rainfall. The ENSO had no significant influence on this pattern. This study suggests that NOAA-AVHRR NDVI long time series might provide valuable information about functioning of the large scale fluvial wetlands like those associated with South America basins.
Tipologia CRIS:
01.01 Articolo in rivista
Keywords:
wetlands macrosystems; ecosystem functioning; landscape mosaic; NOAA-AVHRR NDVI; time series; remote sensing; autocorrelation function; El Nino/South oscillation
Elenco autori:
Zoffoli, MARIA LAURA
Link alla scheda completa:
Pubblicato in: