Phenology-tuned karst rocky desertification monitoring using satellite image time series

Remote sensing has been used in karst studies. The aim of this work is to investigate the use of MODIS time series to monitor and estimate karst rocky desertification (KRD). We propose a new phenology-tuned KRD model based BFAST (Breaks For Additive Seasonal and Trend) approach, Our strategy for KRD monitoring involving several steps: (a) phenology decomposition with BFAST; (b) seasonal parameters extraction; (c) phenology-tuned KRD modeling and ecological responses assessment. The strategy is applied in KRD monitoring and ecological responses assessment with the 4-year MODIS NDVI products that cover Qiubei and Yanshan counties, Yunnan province, and the results indicate that our approach is valid in ecological responses assessment of KRD.