Study On Detecting Drought-induced Disturbance Based On Multi-temporal Landsat 8 In Western Lin’an County

Hickory plantation is one of major industries in Lin An county. The high-temperature induced drought in 2013 influences Hickory production, and further influences the sustainable management and economy. Therefore, timely evaluation of drought-induced disturbances of hickory plantations is valuable for local governments to make proper decisions and for local people to adopt proper measures to reduce the drought-induced loss.Response to the drought event in 2013, this research selected western Lin’An County, Zhejiang province, as a case study to develop a new approach for rapidly mapping hickory plantation distribution and detecting its disturbance. Landsat 8 OLI images in July, October, and December 2013 were used to produce separately three fractions(i.e., green vegetation – GV, shade, and soil) using the linear spectral mixture model(LSMM). Two indices called Normalized Multi-Fractions Index(NMFI) and ratio of GV to Soil Fractions(GSI) were proposed to mapping hickory plantation distribution and detecting drought-induced disturbance respectively. The main conclusions are summarized as follows:1. Soil fraction in hickory plantation is higher than other deciduous-broadleaved vegetation, but GV and shade fractions are opposite. The developed NMFI for mapping hickory plantation distribution provided an overall accuracy of 88.67% and kappa coefficient of 0.76;2. The developed fractions can effectively reflect changes before and after drought event. The GV fraction in the post-drought plantation has lower values than before-drought plantation, but soil fraction is opposite. The GSI can be effectively used to detect different stress levels due to the drought event. Four-level drought(severe, moderate, light, and non-drought) distribution is thus classified using this index.