Abstract: We analyzed the variation trend of growing season length (GSL) of different periods in provinces (regions) of China and the corresponding movement velocity of GSL isolines at 150, 200, 250, 300 and 350 days, based on daily mean temperature data of 822 meteorological stations from 1951 to 2017. In this study, the definition of GSL given by the world meteorological organization was adopted, together with Slope, Hurst and Mann-Kendall indices. The results showed that the GSL in northern China changed significantly during 1951-2017. The extension of GSL was faster in the north than the south, and faster in high-altitude areas than low-altitude ones. The trend of future GSL change in most regions of China converged with the current extension trend. The extension of GSL in northern provinces (regions) was generally 0.1-0.2 d·a^-1, of which the fastest was Tibet with a speed of 0.44 d·a^-1. The period 1981-2000 was the most changeable time of GSL in Chinese provinces (regions). The growing season start (GSS) of all provinces (regions) contributed more to the GSL extension, except for Xinjiang, whose GSL extension was dominated by the growing season end (GSE). In the high-latitude or high-altitude provinces, GSL was more sensitive to the change of mean annual temperature. The higher the mean annual temperature, the longer the GSL. Since 1951, China's GSL isolines of 150, 200, 250, 300 and 350 days showed notable variations. The fastest movement velocity was the 200 days isoline in Northeast China with an average northward movement velocity of 6.11 km·a^-1. The general principle of the movement of China's GSL isoline was that the higher the value of the isoline, the slower the northward movement, with even a southward shift in part of the 350 days isoline. The extension of GSL in China would result in the northward shift of crop planting boundary and the extension of natural vegetation growth period. However, the specific impacts of this change on the quality, crop yield, and ecosystem carbon sequestration need further research.

Key words: growing season length, temperature, trend analysis, velocity, isoline, spatiotemporal dynamics.