Remote sensing based wheat acreage and spectral-trend- agrometeorological Yield Forecasting : Factor Analysis Approach

Forecasting of crop production is one of the most important aspects of agricultural statistics system. Crop production forecasting comprises crop identification, area estimation and predicting the yield of the crop. Crop acreages are estimated using current season’s Indian Remote Sensing Satellite data. Stratified Random Sampling approach and supervised classification of Satellite digital data has been adopted for district-level acreage estimation. In this procedure, using ground based observations collected synchronous to satellite passes, the various crops along with other vegetation classes are identified on the satellite data and spectral signatures are generated for supervised classification. The sample segments are classified using these spectral signatures and crop acreages in the district are estimated using standard statistical aggregation procedures. The Quality Assessment (QA) of district-level acreage estimates is evaluated based on the ground truth data collected in the sample segments. To estimate wheat yields in the study districts, zonal spectral–trend-agrometeorological (agromet) model has been generated using the statistical method of factor analysis. The study has been conducted for zone-II comprising of Karnal, Kaithal, Jind, Panipat, Sonipat and Rohtak districts of Haryana state. Remote sensing based wheat acreage and model predicted yields have been compared with Department of Agriculture (DOA) estimates by computing percent relative deviation. Zonal model developed on the basis of time series data from 1978-79 to 2000-01 has been used to predict the wheat yields for the period 2001-02 to 2007-08. The overall results indicate that the integration of remote sensing data with trend based yield and weather variables provides an immense scope to improve the efficiency and reliability of wheat yield forecasts. Zonal yield models provided