Detection and study of wheat stress using satellite images (Case study: Dasht-e Moghan)

In this study, the aim is to investigate the ability of remote sensing and satellite images to study the spectral changes of wheat and diagnose diseases and its stress in the Moghan plain area in Parsabad city. Achieving such a capability can be useful in prediction of diseases and pests, as well as preventing its spread and timely use of fungicides and pesticides. In order to achieve this goal, the spectral properties of wheat and other plants are studied using IRS and Landsat 8 satellites. Wheat plant has different species in which Attila and Koohi wheat are mostly cultivated in this region, and in this study, their spectral behaviors are considered. In this study, 146 ground control points were used, the highest percentage of the region in terms of the degree of stress to grade 2 and the lowest to the healthy class. After applying atmospheric and geometric corrections on satellite images; Bond ratios were used to identify the best band composition and separability of healthy classes and stress levels one, two, and three. To achieve this goal, different vegetation indices were used. Among the indicators, the GNDVI index was the most efficient and was able to accurately estimate 81% of the areas. Also, GVI index has the highest value of kappa coefficient and overall accuracy with 0.94 and 95.3, respectively, which indicates the high use of this index in grading wheat stress. This index also gave the most area to the degree of stress one.