Tree Aboveground Carbon Mapping in an Indian Tropical Moist Deciduous Forest Using Object-Based Image Analysis and Very High Resolution Satellite Imagery

• Published in: Journal of the Indian Society of Remote Sensing Vol. 52; no. 4; pp. 723 - 734
• Main Authors: Singh, Neha, Nandy, Subrata, van Leeuwen, L. M.
• Format: Journal Article
• Language: English
• Published: New Delhi Springer India 01.04.2024; Springer Nature B.V
• Subjects: Area; canopy; Carbon; Carbon sequestration; carbon sinks; Classification; Climate change; Deciduous forests; Earth and Environmental Science; Earth Sciences; Forests; High pass filters; High resolution; Image analysis; Image resolution; Image segmentation; India; inventories; multispectral imagery; remote sensing; Remote Sensing/Photogrammetry; Research Article; Satellite imagery; Shorea robusta; Spatial resolution; tree and stand measurements; Trees; Uttarakhand India; Nepal; India; Carbon stock; WorldView-2; Multi-resolution image segmentation; Pan-sharpening; Canopy projection area
• ISSN: 0255-660X; 0974-3006
• DOI: 10.1007/s12524-023-01791-0

Forests’ capability to sequester and store a large amount of carbon makes it imperative to assess the carbon stocked in them. The present study aimed to map the tree aboveground carbon stock of sal ( Shorea robusta ) forests of Doon valley, India using object-based image analysis (OBIA) of WorldView-2, a very high resolution satellite imagery (VHRS). The study evaluated different pan-sharpening techniques for improving the spatial resolution of WorldView-2 multispectral imagery and found that the high pass filter resolution merge technique was better compared to others. OBIA was used for image segmentation and classification. It enabled the delineation of tree crowns and canopy projection area (CPA) calculation. The overall accuracy of image segmentation and classification were found to be 72.12% and 84.82% respectively. The study unveiled that there exists a strong relationship between diameter at breast height and the CPA of trees as well as CPA and tree carbon. The average forest carbon density in the study area was found to be 108 Mg ha −1 . The study highlighted that OBIA of VHRS imagery coupled with field inventory can be efficiently used to quantify and map the tree carbon stock.