Fusion of pan-tropical biomass maps using weighted averaging and regional calibration data
Ge Y.
2014
关键词Biomass maps Accuracy Fusion model Spatial variability remotely-sensed imagery land-cover datasets multisensor data forest biomass carbon-dioxide sensing data soil maps classification deforestation validation
英文摘要Biomass is a key environmental variable that influences many biosphere-atmosphere interactions. Recently, a number of biomass maps at national, regional and global scales have been produced using different approaches with a variety of input data, such as from field observations, remotely sensed imagery and other spatial datasets. However, the accuracy of these maps varies regionally and is largely unknown. This research proposes a fusion method to increase the accuracy of regional biomass estimates by using higher-quality calibration data. In this fusion method, the biases in the source maps were first adjusted to correct for over- and underestimation by comparison with the calibration data. Next, the biomass maps were combined linearly using weights derived from the variance-covariance matrix associated with the accuracies of the source maps. Because each map may have different biases and accuracies for different land use types, the biases and fusion weights were computed for each of the main land cover types separately. The conceptual arguments are substantiated by a case study conducted in East Africa. Evaluation analysis shows that fusing multiple source biomass maps may produce a more accurate map than when only one biomass map or unweighted averaging is used. (C) 2014 Elsevier B.V. All rights reserved.
出处International Journal of Applied Earth Observation and Geoinformation
31
13-24
收录类别SCI
语种英语
ISSN号0303-2434
内容类型SCI/SSCI论文
源URL[http://ir.igsnrr.ac.cn/handle/311030/29865]  
专题地理科学与资源研究所_历年回溯文献
推荐引用方式
GB/T 7714
Ge Y.. Fusion of pan-tropical biomass maps using weighted averaging and regional calibration data. 2014.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace