RELATION BETWEEN NOAA-AVHRR SATELLITE DATA AND STOCKING RATE OF RANGELANDS

Biomass of both wild herbivores and livestock in rangelands is correla ted with rainfall at a regional scale. Thus, rainfall may be a good pr edictor of actual stocking rates. However, rainfall data are scarce in many regions, and their spatial resolution is usually much coarser th an needed to set or to evaluate wildlife or livestock stocking rates. We here show a relationship between livestock biomass and an annual ve getation index (normalized-difference vegetation index-integrated valu e, NDVI-I) calculated from remotely sensed data on spectral properties of rangelands of Argentina. The relationship is as strong or even str onger than previously reported correlations between herbivore biomass and rainfall. This, together with the greater availability and higher spatial resolution of satellite data, makes remote sensing a potential ly valuable tool to predict stocking rates for regions, landscapes, an d different portions of a landscape. The form of the relationship betw een stocking rate and average NDVI-I was exponential, which, as previo usly shown, indicates an increasing herbivore load per unit of primary production as rainfall or productivity increases. This may be at leas t partially explained by the fact that the NDVI interannual variation and seasonality were negatively related with average NDVI-I. Thus, sto cking rate may increase exponentially because of an increasing year-to -year reliability of the forage resource and a more even distribution within the year.