Numerical models to forecast the sugarcane production in regional scale based on time series of NDVI/AVHRR images

The use of time series of meteorological satellite images, such as the AVHRR/NOAA, and agrometeorological data can be very useful in developing monitoring and forecasting methods for sugarcane crops because they are based on detection changes of space-time behavior. The knowledge about different sugarcane producing areas and climate in a given region is information required to develop models that can be applied simultaneously to several producing municipalities of sugarcane in order to assess the relation between NDVI and WRSI, the estimated productivity and the detection of similarity between the municipalities through distance functions. Thus, the main goal of this paper is to propose numerical models applied to monitor the sugarcane production based on time series of NDVI/AVHRR images and agrometeorological data. The regression method analyzes the relation between a single dependent variable (sugarcane production) and several independent variables (planted area, NDVI, WRSI), that is, use the independent variables whose values are known to predict the values of the selected dependent variable. The models proposed to estimate the sugarcane production using the variables planted area, NDVI and WRSI presented correlation coefficients (R²) around 0.9 and are able to estimate the sugarcane production for the state of São Paulo in Brazil.