Characterization and detection of intensified pasture areas using MODIS time series

Abstract

Brazil is one of the largest food producers in the world. Among the agricultural and livestock activities in Brazil cattle ranching has a prominent position, contributing to the country leadership in beef exporting, with the second largest cattle herd. Despite the impressive numbers the Brazilian cattle industry has lost ground for the production of agricultural commodities such as soybeans and sugar cane. Studies have indicated that the expansion of these activities occurred mostly on areas occupied with degraded pastures. The reduction on these areas might have led to intensification in use of the remaining pastures, thus maximizing their resources. An alternative to the intensive use is the integration of crop-livestock-forest (CLP) in the same area, in rotation, intercropping or in succession. Although this process is evidenced by data from Agricultural Statistics, more detailed analysis are still needed, covering issues such as the geographical distribution, mapping, estimates of intensified areas, systems involved and the management variables. In this perspective, the use GIS and Remote Sensing in the agricultural sector has contributed to better management of rural investments, reducing environmental impacts, as well as the monitoring and production estimates. The use of time series of vegetation indices, such as NDVI and EVI, enables the identification and mapping of agricultural crops in large areas, allowing the monitoring of crop development throughout their growing cycle. The object of this project is to characterize and identify different types of land use intensification in pastures and crop-livestock-forest systems, through MODIS time series data. (AU)