A linear approach for wheat yield prediction by using different spectral vegetation indices

Yield prediction before harvest is one of the important issues in terms managing agricultural policies and making right decisions for future. Using remote sensing techniques yield estimation studies steps many countries to reach their 21st-century targets. The aim this study develop a wheat model using Landsat-8 Sentinel-2 satellite data. In study, development stages winter were examined with help images obtained between years 2015-2018 selected region Sanliurfa, Turkey, it was aimed predict yields other by establishing model. established Normalized Difference Vegetation Index (NDVI), Soil-adjusted (SAVI), Green (GNDVI) Modified (MSAVI) from images. Linear regression analysis calculated NDVI, SAVI, GNDVI, MSAVI indices, actual values on pre-flowering, flowering stage, post-flowering stage. As result highest correlation coefficient found stage vegetation indices values. coefficients are 0.82, 0.80, 0.86, 0.87, respectively. With model, 2019 tried be estimated three different fields. correlations seen pre-flowering NDVI SAVI. This clearly shows that can used remarkable