Skip to main content
ARS Home » Pacific West Area » Tucson, Arizona » SWRC » Research » Publications at this Location » Publication #246800
Title: Assessing phenological change in China from 1982 to 2006 using AVHRR imagery

Author
item WEI, H. - University Of Arizona
item Heilman, Philip - Phil
item QI, J. - Michigan State University
item Nearing, Mark
item GU, Z. - Beijing Normal University
item ZHANG, Y. - Beijing Normal University

Submitted to: Frontiers of Earth Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/2/2012
Publication Date: 9/1/2012
Citation: Wei, H., Heilman, P., Qi, J., Nearing, M.A., Gu, Z., Zhang, Y. 2012. Assessing phenological change in China from 1982 to 2006 using AVHRR imagery. Frontiers of Earth Science. 6(3):227–236.

Interpretive Summary: Vegetation phenology parameters are indicators of seasonal biological life stages and are important for monitoring vegetation growth and agriculture production. Historical phenology parameters can be estimated from satellite remote sensing data to detect long term vegetation growth and evaluate the impacts from climate change and human activity on ecosystems. TIMESAT is a software program developed to estimate phenology parameters such as the beginning, end of growing season, and the seasonal production based on remote sensing images. In this study, we used TIMESAT to estimate phenology trends across China based on the Normalized Difference Vegetation Index (NDVI) at 8km spatial resolution from the Advanced Very High Resolution Radiometer (AVHRR) of Global Inventory Modeling and Mapping Studies (GIMMS). AVHRR NDVI images were used because they have the advantage of high temporal resolution since 1980s and are available globally. Along with the rapid social change and economic growth that have taken place in China in the last several decades, the terrestrial ecosystems and vegetation production have also experienced dramatic changes. However, the spatial patterns of growing season production across the country and the causes of the changes remain unknown. The objective of this study is to use AVHRR data and TIMESAT to visualize the spatial patterns of vegetation growth trends across China from 1982 to 2006. Linear temporal regression was used to detect the significant trends in each pixel at the 95% significance level. Eight pairs of Landsat images at finer spatial resolutions (30-60m) were classified into vegetated area and non-vegetated area for validation. Results showed great spatial variation in vegetation growth changes across the country during the 25 year study period. Significant decreases in vegetation production were detected in the grasslands of Inner Mongolia, and the industrializing regions including the Pearl River Delta, the Yangtze River Delta, and areas along the Yangtze River. Significant increases in vegetation production were found in Xinjiang, central China, and northeast China. Validation of the NDVI values and vegetated area changes was conducted using Landsat imagery and the results were consistent with the analyses from AVHRR data. Moreover, we also found that although the causes of the vegetation change may vary locally, there was a coincidence of the spatial pattern of vegetation change and the scales of national policies of China launched in the 1970s such as the opening of coastal economic zones in southern China and inland cities along the Yangtze River; and the ‘Three-North Shelter Forest Programme’, which is the largest nation-wide afforestation project covering 4 million km2 in north China. The results from this study indicate the impacts of national policies on ecosystem change and that such impacts can be detected using the method described in this paper.

Technical Abstract: Long term trends in vegetation phenology indicate ecosystem change due to the combined impacts of human activities and climate. In this study, we used 1982 to 2006 Advanced Very High Resolution Radiometer Normalized Difference Vegetation Index (AVHRR NDVI) imagery across China and the TIMESAT program to quantify annual vegetation production, and used that information to examine changes in seasonal vegetation growth. Results showed great spatial variation in vegetation growth changes across the country during the 25 year study period. Significant decreases in vegetation production were detected in the grasslands of Inner Mongolia, and in industrializing regions in southern China, including the Pearl River Delta, the Yangtze River Delta, and areas along the Yangtze River. Significant increases in vegetation production were found in Xinjiang, central China, and northeast China. Validation of the NDVI values and vegetated area changes were conducted using Landsat imagery and the results were consistent with the analysis from AVHRR data. We also found that although the causes of the vegetation change vary locally, the spatial pattern of the vegetation change and the areas of greatest impact from national policies launched in the 1970s, such as the opening of economic zones and the ‘Three-North Shelter Forest Programme’, are similar, which indicates the impact of national policies on ecosystem change and that such impacts can be detected using the method described in this paper.