Özet:

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

Anahtar Kelimeler: