Spectral aspects for monitoring forest health in extreme season using multispectral imagery

Forest health monitoring is needed for forest management in respect to eradication of diseases and insects. Spectral indicators help to retrieve forest health at different scales. Although, indicators of forest condition from remote sensing require narrow spectral bands but these bands are not available in multispectral imagery. Therefore, a systematic approach to evaluate forest health using Sentinel 2A imagery is developed. Anthocyanin Reflectance Index (ARI1), Structure Insensitive Pigment Index (SIPI) and Normalized difference vegetation index (NDVI) were combined for forest health analysis using ENVI forest health tool model. Canopy Chlorophyll Content (CCC) was retrieved using the SNAP software approach. The overall accuracy in forest health mapping was 0.82 and 0.86 in the month of May and October respectively, validated through in-situ analysis. The excellent forest health exhibit increased by 6 % in October (after the rainy season) compared to May (summer season). The substantial proportion of forest in the area is mature which showed low changes compared to young forest. The reduced regression was found during May (R2 = 0.249) between chlorophyll content and forest health due to decrease in leaf photosynthetic pigments whereas better relationships were noticed in October (R2 = 0.58). The 38% differences in the chlorophyll content in the two seasons indicated that CCC is sensitive to stress pigments in forest.(C) 2021 National Authority for Remote Sensing and Space Sciences. Production and hosting by Elsevier B. V.& nbsp;