利用昆明站1991 -2020年近30年逐日降水数据计算昆明雨季（5 -10月）的起止时期， 进一步确定昆明市雨季的长短。又基于云南省和昆明市的统计年鉴数据， 使用年末总人口、 城市建成区面积、 城镇化率、 人均GDP等城市发展因子确定昆明的城市发展进程， 将昆明市的城市发展进程划分为缓慢发展期（1991 -2003年）和快速发展期（2004 -2020年）， 进而分析比较两段时期中昆明市雨季长短的特征和差异， 采用统计分析、 小波分析和M-K突变检验等综合分析方法， 系统分析了昆明市雨季长短的时间变化特征， 并用灰色关联度分析方法分析了昆明市雨季长短与城市发展的关联性。结果表明， 1991 - 2020年昆明市的雨季开始日呈逐渐偏晚的趋势， 而雨季结束日呈逐渐偏早的趋势， 总体上雨季长度呈逐渐缩短的趋势； 小波系数分析结果显示， 在8年以下的时间尺度上， 昆明市雨季长短变化的周期不存在明显的规律性， 在17年时间尺度上的周期变化明显， 呈偏短-偏长-偏短-偏长-偏短的5个循环交替， 2003 - 2008年、 2014 -2017年雨季增长， 1991 -2002年、 2009 -2012年、 2018 -2020年雨季缩短， 2018 - 2020年等值线未闭合说明还有进一步缩短的趋势。通过M-K检验表明昆明市的雨季长短在1991-2020年间出现4次突变， 分别发生在2002年、 2008年、 2012年和2017年。从昆明城市发展与雨季长短的关系来看， 昆明城市发展缓慢期的雨季长短的变化趋势较为平稳， 而城市发展快速期2004年以后， 昆明市雨季长度缩短的变化明显， 并随着城市发展进程的加快其极端波动性更加明显。运用SPSS（Statistical Product and Service Solutions）软件对未来10年昆明的雨季长短进行预测， 结果显示未来10年昆明雨季长短将持续偏短的趋势。在灰色关联度分辨率为0.5时， 表征城市发展进程的4个因子对昆明雨季长短变化均产生不同程度影响， 其关联度系数都在0.70以上， 表明昆明城市发展与雨季长短显著关联性， 其中影响最大的因子是年末总人口， 最小为人均GDP， 灰色关联度分别为0.88和0.70， 属于高度关联和显著关联。对4个因子的关联系数进行排序为： 年末总人口>城镇化率>城市建成区面积>人均GDP。

The daily precipitation data of Kunming Station for nearly 30 years from 1991 to 2020 were used to calculate the beginning and ending periods of Kunming rainy season （May-October）， further determine the length of the rainy season in Kunming.Additionally， statistical yearbook data for Yunnan Province and Kunming City were used， including year-end total population， urban built-up area， urbanization rate， per capita GDP， and other urban development factors， to determine the urban development process in Kunming.This process divided Kunming's urban development into a slow period （1991 -2003） and a fast period （2004 -2020）.The characteristics and differences in the length of the rainy season in Kunming City during these two periods were then analyzed and compared.Various analytical methods， including statistical analysis， wavelet analysis， and Mann-Kendall （M-K） mutation test， were employed to systematically analyze the temporal changes in the length of the rainy season in Kunming City.Additionally， the grey correlation analysis method was used to assess the correlation between the length of the rainy season and urban development in Kunming City.The results indicate that from 1991 to 2020， the start date of Kunming City's rainy season gradually became later， while the end date gradually became earlier， resulting in an overall trend of the rainy season getting shorter.Wavelet coefficient analysis revealed that there was no obvious regularity in the variation of the rainy season's length on time scales below 8 years， but on a 17-year time scale， there was a noticeable cycle of short-long-short-long-short variations， The rainy seasons from 2003 to 2008 and from 2014 to 2017 were relatively long， while the rainy seasons from 1991 to 2002， 2009 to 2012， and 2018 to 2020 were relatively short.The unclosed contour lines from 2018 to 2020 indicate a further trend of shortening.The M-K test showed that the length of the rainy season in Kunming City experienced four abrupt changes between 1991 and 2020， Occurring in 2002， 2008， 2012 and 2017.Regarding the relationship between Kunming's urban development and the length of the rainy season， the variation trend of the rainy season length during the slow urban development period remained relatively stable.However， after 2004， during the rapid urban development period， there was a significant reduction in the length of the rainy season in Kunming City， Extreme volatility became more pronounced as the urban development process accelerated.The SPSS （Statistical Product and Service Solutions） was used to predict the duration of rainy season in the next 10 years in Kunming City， The results show that the rainy season will continue to be shorter in the next 10 years in Kunming.When the grey correlation resolution was set at 0.5， four factors representing the urban development process had varying degrees of influence on the length of the rainy season in Kunming City， with correlation coefficients all exceeding 0.70， the results show that there is a significant correlation between the urban development and the length of rainy season in Kunming.Among these factors， the most influential one was the year-end total population， while the least influential was per capita GDP， with grey correlation coefficients of 0.88 and 0.70， respectively， signifying a high and significant correlation.The order of correlation coefficients for the four factors is as follows： year-end total population > urbanization rate > urban built-up area > per capita GDP.