Diagnosis of Vegetation Recovery in Mountainous Regions After the Wenchuan Earthquake
Wang, M. a,b; Yang, W. a,b; Shi, P. a,b; Xu, C. c; Liu, L. a,b
a State Key Laboratory of Earth Surface Processes and Resource Ecology, Beijing Normal University, Beijing 100875, China;
b Academy of Disaster Reduction and Emergency Management, Beijing Normal University, 100875 Beijing, China
c Key Laboratory of Active Tectonics and Volcano, Institute of Geology, China Earthquake Administration, 100029 Beijing, China.
Abstract: The Ms 8.0 Wenchuan Earthquake in 2008 resulted in numerous landslides disturbing vast areas of vegetation. However, almost 5 years following the catastrophic event, vegetation recovery in the affected regions experienced dynamic changes. In this paper, a new method is proposed to detect poor vegetation recovery areas (PVRAs) where vegetation recovery has experienced significant difficulty. By using 8-day interval Moderate-Resolution Imaging Spectroradiometer (MODIS) images from 2000 to 2012, Normalized Difference Vegetation Index (NDVI) values after the Wenchuan earthquake are compared with past NDVI distributions before the earthquake. This method fully considers the dynamic processes of vegetation recovery, including both seasonal and annual variation in NDVI values, and identifies landslide disturbances from natural fluctuations. Then, the method has been applied in the study area in Pingwu County to detect vegetation recovery and its validity has been checked by field works and manual interpretation of different periods of SPOT5 imageries. By comparing unrecovered vegetation sites, the optimal threshold to define PVRA is determined. This research has singled out four types of PVRA (namely “deposition,” “back scar,” “mixed large,” and “land use change” types) in the study area, with each type of PVRA linked to different reasons for poor vegetation recovery. Finally, the proposed method is applied to the mostly affected counties, where Modified Mercalli Intensity (MMI) is greater than VIII. Spatial patterns of PVRA indicate that vegetation has experienced significantly poor process over relatively widespread regions.
Keywords: Dynamic process, earthquake, landslides, Moderate-Resolution Imaging Spectroradiometer (MODIS) Normalized Difference Vegetation Index (NDVI), spatial pattern, vegetation recovery.
Published in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2014, 7(7):3029-3037.
