Wheat Yield Forecasting for the Tisza River Catchment Using Landsat 8 NDVI and SAVI Time Series and Reported Crop Statistics

Due to the increasing global demand of food grain, early and reliable information on crop production is important in decision making agricultural production. Remote sensing (RS)-based forecast models developed from vegetation indices have potential give quantitative timely crops for larger regions or even at farm scale. Different are being used this purpose, however, their efficiency estimating yield certainly needs be tested. In study, wheat was derived by linear regressing reported values against a time series six different peak-seasons (2013–2018) using Landsat 8-derived Normalized Difference Vegetation Index (NDVI) Soil Adjusted (SAVI). NDVI- SAVI-based forecasting were validated based 2018–2019 datasets compared evaluate most appropriate index that performs better Tisza river basin. Nash-Sutcliffe positive with E1 = 0.716 model NDVI SAVI 0.909, which means method performed good efficiency. The best prediction 8-SAVI found beginning full biomass period 138th 167th day year (18 May 16 June; BBCH scale: 41–71) high regression coefficients between yield. RMSE NDVI-based 0.357 t/ha (NRMSE: 7.33%). 0.191 (NRMSE 3.86%). validation results revealed provided more accurate forecasts NDVI. Overall, probable amount possible predict far before harvest (six weeks earlier) 8 generating simple thresholds forecasting, loss can mapped.