Abstract: Abstract:Understanding the changes in field straw-burning at national scale over a comparatively long period is essential to improving straw management and helping its administrative supervision. Using the Stasee-Fire data from 2014 to 2018 in China, we constructed a dataset of field straw-burning in China after data conversion, extraction, cleaning and other processing. We then analyzed its spatiotemporal changes at different scales over the past five years using the central gravity-moving model, spatial autocorrelation model and the K-mean clustering model. The results show that the mandate of banning field straw-burning has achieved its objectives, and that the number of straw-burning had fallen at an average annual rate of 22.43%. The monthly straw-burning in February-April and October-November accounted for 85.71% of the annual burning. Most of straw-burning were in northeast, the north plain and Heilongjiang, with 56.31% of the burnings occurring in Liaoning and Jilin provinces. Taking the Hu-Huanyong line as a boundary line, the straw-burning in the regions to the southeast and southwest of the line was 78.73 : 21.27. It was found from the central gravity-moving model that the gravitational center of the straw-burning was located in Liaoning and Hebei, with a tendency of shifting to southwest due to the decline in straw-burning in northeast China. The center of the monthly straw-burning showed a noticeable return-glyph trajectory. We found that the spatial autocorrelation of the straw-burning was between 0.33 and 0.49 at national scale but has been falling, and that the straw-burning was no longer concentrated in some regions and it instead occurred sparsely across the country. At provincial scale, straw-burning can be classified as autumn - winter, spring, and autumn-burning type, with specific types in each area depending on its cultivation, crop varieties and climate. Therefore, improving management of straw-burning should take climate change and crop varieties into account.