青藏高原多年冻土近地表土壤冻融循环会影响土壤和大气之间的水分与能量交换，研究其时空变化特征及对气候变化的响应对理解高原气候变化机制具有重要意义。本文基于通用陆面模式（Commu‐nity Land Model 5. 0，CLM5. 0）计算1980-2017年高原多年冻土区近地表冻融参量，即土壤冻结开始时间、冻结结束时间、融化持续时间和冻结持续时间，并分析其时空变化及与近地表温度、降水量、积雪厚度和植被指数的相关性。结果表明：（1）高原多年冻土近地表土壤冻结开始时间集中于9月到10月中下旬，结束时间集中于2-5月。半湿润区土壤融化时间最长而半干旱区最短，平均相差15 d。高原多年冻土土壤冻融状态变化显著，除喀喇昆仑山脉附近外，大部分多年冻土地区显示冻结、融化持续时间分别具有缩短和增长趋势。高原平均土壤融化持续时间增长速率为2 d·（10a）-1，其中半湿润区增长趋势最为显著，达 4 d·（10a）-1。（2）高原多年冻土冻融参量与地理因子具有联系。在 29°N-36°N 和 82. 5°E-103°E融化持续时间呈增长趋势，但速率分别降低和增加。随着海拔升高，融化持续时间增长率降低。（3）高原多年冻土融化持续时间与积雪厚度、近地表温度、降水量以及植被指数均具有相关性且不同气候区域相关性不同。近地表温度在所有区域正相关性显著，是影响土壤冻融变化的主要因素。降水量与积雪厚度分别呈现正相关和负相关且均在半湿润区显著相关。植被指数全区基本呈现正相关，且半干旱区最显著，具有较强相关性。（4）高原多年冻土融化持续时间与不同季节气候因子相关性不同。近地表温度在季节尺度同样显著影响土壤冻融过程，其中春季最为显著。降水量在夏季为显著正相关，而冬季为负相关。积雪厚度和植被指数均在春季的半干旱和半湿润区存在显著相关性，分别呈负相关和正相关。（5）近地表温度在干湿季对高原多年冻土区域土壤冻融均有影响，而积雪深度、降水量和植被指数仅在湿季有较大影响。

The freeze-thaw cycle of near-surface soil in the perennial permafrost region of the Qinghai-Xizang （Tibetan）Plateau plays a crucial role in regulating water and energy exchange between the soil and the atmosphere. Investigating its spatiotemporal characteristics and response to climate change is essential for understanding the mechanisms driving climate change on the plateau. In this study，we calculated near-surface freeze-thaw parameters-including the start and end times of soil freezing，thawing duration，and freezing duration-across the perennial permafrost region of the plateau from 1980 to 2017 using the Common Land Model 5. 0（CLM5. 0）. We further analyzed their spatiotemporal variations and correlations with temperature，precipitation，snow depth，and vegetation index. The results show that：（1）The onset of near-surface soil freezing in the plateau’s permafrost region occurs between September and mid-to-late October，while the thawing period ends between February and May. Semi-humid regions have the longest thawing duration，whereas semi-arid regions have the shortest，with an average difference of 15 days. The freeze-thaw status of permafrost soil on the plateau exhibits significant changes. Except for areas near the Karakoram Mountains，most permafrost regions show a decreasing trend in freezing duration and an increasing trend in thawing duration. The average growth rate of soil thawing duration across the plateau is 2 d·（10a）⁻¹，with the most significant increase observed in semi-humid regions， reaching 4 d·（10a）⁻¹.（2）The freeze-thaw parameters of the plateau's permafrost are associated with geographical factors. In the latitude range of 29°N-36°N and longitude range of 82. 5°E-103°E，the thawing duration shows an increasing trend；however，the rate of change decreases in some areas while increasing in others. Additionally，as elevation increases，the growth rate of thawing duration declines.（3）The duration of permafrost thawing is significantly correlated with snow depth，near-surface temperature，precipitation，and vegetation index，though these relationships vary across different climatic regions. Near-surface temperature exhibits a strong positive correlation across all regions，making it the primary driver of freeze-thaw changes. Precipitation and snow depth show positive and negative correlations，respectively，with particularly strong correlations in semihumid areas. The vegetation index is positively correlated with thaw duration in all regions，with the strongest correlation observed in semi-arid areas.（4）The relationship between thawing duration and seasonal climatic fac‐ tors varies. Near-surface air temperature exerts a significant influence on the freeze-thaw process at seasonal scales，with the most pronounced impact occurring in spring. Precipitation is positively correlated in summer but negatively correlated in winter. Both snow depth and vegetation index are significantly correlated with thawing duration in semi-arid and semi-humid regions during spring，exhibiting negative and positive correlations，respectively.（5）Near-surface temperature influences the freeze-thaw cycle in the plateau’s perennial permafrost region during both dry and wet seasons. However，the effects of snow depth，precipitation，and vegetation index are more pronounced during the wet season.