Landscape, soil and meteorological influences on canopy dynamics of northern flooding Pampa grasslands.

POSSE, G.; MARTÍN OESTERHELD; DI BELLA, C.

APPLIED VEGETATION SCIENCE

WILEY-BLACKWELL PUBLISHING, INC

Año: 2005 vol. 8 p. 49 - 56

1402-2001

Question: How do meteorological variations at seasonal, interannual scales differentially affect the canopy dynamics of four contrasting landscape units within a region? Location: Flooding Pampa, Buenos Aires, Argentina. 5000 km2. Central point: 35°15´ S, 57°45´ W. Methods: We used a 19-year series of the normalized difference vegetation index (NDVI) derived from NOAA-AVHRR PAL (Pathfinder AVHRR Land) images and meteorological data provided by a nearby weather station. The NDVI was used as surrogate of canopy photosynthetic status. The relationship between annually integrated NDVI and meteorological conditions was explored by stepwise multiple regressions for each defined unit. PCA was performed to compare units and growing seasons on a multivariate basis. Results: Mean seasonal NDVI curve was similarly shaped among landscapes. However, the absolute values differed widely. There was high interannual variation so that the mean seasonal pattern was seldom observed in any particular year. Annually integrated NDVI of all landscapes was negatively associated with summer temperature and positively with previous year precipitation. It was also directly related with current year winter precipitation in two landscapes and with summer precipitation in the others. NDVI response to September and March precipitation accounted for some of the differences in interannual variation among landscapes. Conclusions: Our results revealed a strong intra-regional variation of canopy dynamics, closely linked to landscape (vegetation-soil) and water availability (mainly in summer and during the previous year). These links may be used to predict forage production rates for livestock.