Detecting the degraded Cultivated land of Yellow River Delta during the last two decades based on time series of MODIS- derived NDVI by coupling DTW and BFAST method

The Yellow River Delta (YRD) is the most important reserved cultivated land resource in China, for it has the highest rate of natural land growth in the world. However, under the influence of soil salinization and human activities, cultivated land is prone to degradation. In this study, based on the MODIS NDVI time series dataset from 2001 to 2021, we used Breaks For Additive Seasonal and Trend (BFAST) coupled with Dynamic Time Warping (DTW) method to extract the information of degraded cultivated land in the YRD. The results show that the user accuracy, producer accuracy and overall accuracy of the extracted degraded cultivated land information by the method were 88.68%, 90.38% and 83.33%, respectively, and the accuracy of cultivated land degradation time extraction was 75.76%. From 2001 to 2021, a total of 212.25 km 2 of cultivated land in the YRD has degraded into salt pans, grasslands, artificial surfaces, and water. The area of degraded cultivated land showed an increasing trend before 2014, but the growth trend slowed down after a brief improvement in 2015. The spatial distribution of degraded cultivated land in the YRD shows a pattern of more degradation in the northern areas and less in the southern areas, and more in the eastern areas and less in the western areas, gradually decreasing from the coastal areas to the inland, and the degree of cultivated land degradation exhibits a pattern of gradually decreasing degradation from the coastal areas to the inland. The overall extraction effect is good, which can provide reference for regional degraded cultivated land monitoring.