B53B-0186:
Size and frequency of forest loss and gain in China during 2000-2005

Friday, 19 December 2014
Dan-Xia Song, University of Maryland College Park, College Park, MD, United States, Chengquan Huang, University of Maryland College Park, Geographical Sciences, College Park, MD, United States, Joseph O Sexton, University of Maryland, College Park, MD, United States, Saurabh Channan, Global Land Cover Facility, College Park, MD, United States and John R Townshend, University of Maryalnd- College Park, College Park, MD, United States
Abstract:
Tightly coupled to biodiversity, terrestrial carbon, climate change, and land value, forest-cover change is among the most drastic land cover changes on Earth’s surface. The size of individual events of forest loss and gain are important for inferring natural and anthropogenic drivers of forest change and for estimating carbon flux due to disturbance and regrowth. The size-frequency distribution of forest-change patches, usually modeled using power-law functions, provides a mechanism for prescribing disturbance scenarios in carbon and ecosystem models. Using a global, Landsat-based forest-cover and -change dataset, we examine the patch-size distribution of forest-cover changes in China from 2000 to 2005. China has reported a 19,860 km2/yr increase in forest area in the 1990s and 30,000 km2/yr after 2000. However, recent results show a disagreement between remotely sensed forest-area changes and official national forest inventories, observing that forest gains were offset by losses in the same periods. Understanding the spatial pattern of gains and losses is therefore of considerable importance for evaluating and making forestry policies. We examine the hypothesis that each province has a unique relationship between size and frequency for forest losses or gains and seek explanation for this variation in the different drivers of forest change among the provinces.