Estimation of forest carbon stock using the remote sensing technique

Chi Chuan Cheng, Su-Fen Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution


This study focuses on applying remote sensing to estimate the carbon stock of Nanzhuang National Forest in Taiwan based on SPOT vegetation indices. The research processes include the calculation of vegetation indices (i.e., Normalized Difference Vegetation Index and Simple Ratio Vegetation Index) from SPOT images, the establishment of a regression model based on vegetation indices and forest stocks of field permanent plots, and finally the estimation of the forest carbon stock according to the forest stock calculated by the established regression model. Meanwhile, the image shadow and seasonal image (dry and wet seasons) effects on the estimation of forest carbon stock are also investigated when applying SPOT vegetation indices. The result is as follows. Under three shadow processes (i.e., no shadow process, shadow removal, and shadow linear-correction) on dry-season images, carbon stock per hectare is 124.05 ton, 129.34 ton, and 127.05 ton, respectively. As for wet-season images, carbon stock per hectare is 128.80 ton, 125.89 ton, and 128.98 ton. When compared with the forest carbon stock of field permanent plots (i.e., 129.20 tons per hectare), the result obtained from SPOT vegetation indices seems acceptable though a slight difference exists. However, the results vary for shadow process and seasonal images. From the result, it can be concluded that the remote sensing technique is a timely, effective, feasible, and large scale approach to estimating forest carbon stock. However, the effects of image shadow and seasonal images should be taken into consideration when applying SPOT vegetation indices to estimate forest carbon stock.

Original languageEnglish
Title of host publication34th Asian Conference on Remote Sensing 2013, ACRS 2013
PublisherAsian Association on Remote Sensing
Number of pages8
ISBN (Print)9781629939100
Publication statusPublished - 2013 Jan 1
Event34th Asian Conference on Remote Sensing 2013, ACRS 2013 - Bali, Indonesia
Duration: 2013 Oct 202013 Oct 24


Other34th Asian Conference on Remote Sensing 2013, ACRS 2013

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Estimation of forest carbon stock using the remote sensing technique'. Together they form a unique fingerprint.

Cite this