Document Type : Research Paper
Authors
- Maryam Gholami 1
- Ataollah Ebrahimi 1
- Esmaeil Asadi Boroujeni 1
- Elham Ghahsareh Ardestani 1
- Hamzeh Ali Shirmardi 2
1 Department of Nature Engineering, Faculty of Natural Resources and Earth Sciences, Shahrekord University, Shahrekord, Iran
2 Agriculture and Natural Resources Research and Education Center of Chaharmahal-va-Bakhtiari Province, Shahrekord, Iran
Abstract
Detecting vegetation changes in arid and semi-arid regions, particularly for identifying dried patches, is a major challenge in remote sensing. This study was conducted to identify and analyze areas affected by dryness in the rangelands of Chaharmahal-va-Bakhtiari Province, using remote sensing data and analyzing vegetation changes in the years 2013 and 2023. After obtaining the image differencing for the years 2013 and 2023 for various indices such as NDVI, VHI, TCT, and SR, thresholding of the difference values for each index was performed to identify dried patches. To validate the results, after identifying the dried patches, field visits and sampling were carried out to assess the correspondence between the identified areas and the actual conditions. The results showed that the SR index had the best performance in identifying dried patches. The findings indicated that the SR index, with an overall accuracy of 75% and a Kappa coefficient of 0.75, had the best performance in identifying dried patches, demonstrating a strong agreement with the field data. Following SR, the VHI index—with 60% accuracy and a Kappa value of 0.40—was identified as the second most suitable index for detecting dried areas. The TCT index was also able to detect vegetation changes with an overall accuracy of 52% and a Kappa coefficient of 0.32. Among the indices, NDVI, with 40% accuracy and a Kappa coefficient of 0.28, showed lower-than-expected performance and was less effective in identifying vegetation changes in the dried areas. This study, demonstrates that the SR index, as an effective tool for detecting dried patches in rangeland areas, can significantly contribute to the early-warning, monitoring and optimal management of natural resources, i.e., rangelands.
Keywords
)