Estimating spatiotemporal patterns of aboveground biomass using Landsat TM and MODIS images in the Mu Us Sandy Land, China

Aboveground biomass (AGB) in areas of desertification cannot only represent the status of vegetation but can also provide evidence to evaluate the effects of ecological restoration and help land managers realize sustainable development of desert ecosystems. Current research estimating AGB by remote sensing has mainly focused on forest, grasslands and crops, and has infrequently been applied to desert ecosystems. We used Landsat Thematic Mapper (TM) images and Moderate Resolution Imaging Spectro-Radiometer (MODIS) data to estimate AGB and its spatiotemporal patterns from 2000 to 2012 in the Mu Us Sandy Land of China. Results showed that: (1) AGB varied from 2000 to 2012 and total AGB showed an increasing trend of 0.1743 Tg per year. The lowest total AGB was observed in 2000 and 2001 and the highest in 2012, with slightly less in 2007. (2) AGB spatial extent (percent of ground covered) had a decreasing trend of 5.37% during the study period and AGB was mainly in the southwestern and eastern parts of the study area. AGB had no change in 2.23% of this area, and areas of no change were mainly in the northwestern and southwestern parts. There was an increasing AGB trend in 92.40% of the area, which was mainly in large areas of the middle, northeastern, and southern parts of the Mu Us Sandy Land. (3) In the sandy land from 2000 to 2012, areas with mild and moderate fluctuations and increasing AGB made up the largest part of the study area. Those two types of fluctuations accounted for 74.60% of the total area, and were widely distributed in the northeastern, eastern, central, and southern portions of the sandy land. Areas with severe and extremely severe fluctuations and decreasing AGB were relatively small. These two types represented 0.86% of the total area and were scattered in the northwestern and western parts of the sandy land. (4) With the increase of temperature and precipitation, total AGB tended to increase from 2000 to 2012, somewhat in agreement with precipitation (r = 0.595). However, precipitation was not the only factor affecting AGB. Human factors such as population, livestock, and particularly positive policies also impacted the spatiotemporal patterns of AGB. (C) 2014 Elsevier B.V. All rights reserved.