Crop Yield Forecast Modeling

This text is mainly focused to highlight the improvement needs of the current system of statistical sample based crop yield forecasting in developing country like Pakistan with latest state of the art, Remote Sensing & GIS technologies along with agro-meteorological science. Prime emphasis is to incorporate the synoptic variability in vegetation surface on earth through satellite technology instead of localized crop yield sampling method. Secondly, there is also a comparative study for linear (Regression), non-linear (Neural Network) and Crop Simulation model to conclude basis for sugarcane crop yield forecasting system. Remote sensing is mainly based on MODIS sensor data and GIS include mainly the geo-located crop cuttings at harvesting of Sugarcane. Crop Evapo-transpiration, crop water productivity, Nutrient availability, irrigation water, crop water stress index and Vegetation Index (NDVI) are prime parameters of all crop yield models. Benefits of such forecasting system are rational policy for food security and timely availability of agricultural statistics with reliability.