Multi-temporal terrestrial laser scanning for modeling tree biomass change

Above ground biomass (AGB) is a crucial ecological variable and has to be accurately estimated to understand potential changes of the climate system and to reduce uncertainties in the estimates of forest carbon budget. The overall goal of this research is to estimate tree level AGB change using multi-temporal terrestrial laser scanning (TLS) datasets for trees in East Texas. Specific objectives are to (1) develop models using TLS parameters to estimate tree level AGB; and (2) investigate different conceptual approaches for estimating AGB change. Since majority of the AGB estimation models are developed only using diameter at breast height (DBH), we investigated the potential of TLS by extracting various geometric and statistical parameters for tree level AGB estimation. National and regional level AGB estimation models were developed for loblolly pines. To estimate the change in AGB, three different approaches were followed. The best AGB estimation model for loblolly pines had DBH, height variance, and interquartile distance as independent variables. The best AGB estimation model for hardwoods included volume and crown width as independent variables. For AGB change of loblolly pines, direct modeling of AGB change with TLS data available for 2009 and 2012 provided the best results. An extensive literature review reveals that this is the first study to model the change in AGB using different innovative and conceptual approaches with multi-temporal TLS data. The results of our study indicate the capability of TLS to model the change in tree level AGB, with potential for reducing the amount of field work when using multi-temporal terrestrial TLS datasets. We believe that the results of this study will benefit forest management and planners for prudent decision making, and other remote sensing studies from airborne and spaceborne platforms, for map upscaling, data fusion, or calibration purposes. (C) 2014 Elsevier B.V. All rights reserved.