ABSTRACT
The ability to identify changes on the ground using multi-temporal earth observation data is one of the main issues in remote sensing. Mapping of species and identification of changes in their coverage are crucial for conserving and managing natural resources. We used multi-spectral and multi-temporal images to detect changes in Dalbergia sissoo in the Doon Valley, around Jakhan Rao, between the years 2018 and 2021. Using the Normalized Difference Vegetation Index, we reduced the dimensionality of multi-spectral images and then created a multi-temporal Normalized Difference Vegetation Index image database for the study period. This database was then used to extract D. sissoo using the fuzzy-based modified probabilistic c means method with individual samples as mean approach. The change detection in the extracted class was performed using a change matrix. Results showed that D. sissoo increased in 2021 and is susceptible to environmental factors of heavy rains and the flow of water in the Rao due to its growth at the edge of the river bank. The method described in this chapter has broad applications for classifying various species in both forest and agricultural areas, which can ultimately aid in quantifying the land cover and its changes in the regions.
