Developing yield prediction methods based on Sentinel2 satellite data

In the early phases of this study I downloaded the Sentinel satellite images from the SciHub website based on which NDVI values could have been calculated by applying ESRI ArcGIS software to extract the RED and NIR reflectance’s from the image files, and processing to get NDVI images. Then I created a model for each summer month during 2016 and 2017 years. Thus far, the NDVI value processing is completed. The next step of the study was using SPI-3 and SMI drought indexes to classify the drought risk in sample areas, next I assessed the agricultural drought risk by NDVI values and match the results with the other two drought indexes. This part is about the certify and prove the reliability of NDVI value for classification of agricultural drought, and in the end finish the comparison research via analyzing the results from SPI-3 and SMI index, as well as NDVI value in two different regions. At last, this thesis summarizes the advantages and disadvantages of NDVI values used for evaluating agricultural drought compares with other drought indices.MSc/MAAgricultural Environmental Management Engineerin