融雪径流模拟是干旱区水文水资源研究的热点问题，对干旱区春季融雪洪水风险评估和流域水资源管理至关重要。结合文献查询及资料分析，重点讨论了不同类型融雪径流模型的特征和发展情况，比较了不同融雪径流模型在干旱区一些典型河流的应用情况，并对其功能及优缺点进行了评估。结合对目前流域融雪径流模拟研究中存在的问题的分析，提出未来融雪径流模拟要注重提高数据分辨率的观点。借助多源遥感数据驱动获取更为精确的输入数据，在数据获取难度减小、精度提高的基础上山区融雪径流模拟将更多地以基于能量平衡的物理性模拟为主。模型的构建要充分考虑由气候变暖所带来的其它参数的变化，精确描述山区融雪过程，以提高对融雪径流的模拟精度。

为研究气候变化下澜沧江流域内径流演变趋势，构建了云南省内澜沧江流域SWAT（Soil and Water Assessment Tool）模型。根据气候变化特征及未来趋势设置不同气候变化情景，分析了气温、降水对径流和蒸散的影响。结果表明：SWAT模型在澜沧江流域具有很好的适用性，其模拟径流量判定系数及纳什系数的率定验证结果均高于0.8，澜沧江年径流量空间分布特征呈现下游、干流高的特征，径流受气温和降水的共同影响，与降水成正比，与气温成反比，径流变化由降水变化主导；澜沧江潜在蒸散量与降水成反比，与气温成正比，潜在蒸散量由气温变化主导。选取云南省内澜沧江上中下游各代表点分析气候变化特征，并由第六次国际耦合模式比较计划（CMIP6）和神经网络模型综合预测未来澜沧江流域内气候变化趋势。经Mann-Kendall趋势检验分析过去四十年间气温主要呈显著性上升趋势，降水时空波动变化较为复杂总体呈显著性下降趋势，相邻年间变化幅度分别在1 ℃和20%以内。未来气温将呈约1 ℃/20 a的上升趋势，降水将呈约10%/20 a的上升趋势。未来气温和降水变化将使径流及潜在蒸散增加，径流增加集中于雨季最大可增加约38%，蒸散集中于干季最大增加约6%，干雨季旱涝风险增加，更加需要云南省内退耕还林策略涵养水源和梯级水库的调度防控防治旱涝灾害。

In order to study the evolution trend of runoff in the Lancang River Basin under climate changes， the SWAT model of the Lancang River Basin in Yunnan Province is constructed. Different climate change scenarios are set up according to the characteristics of climate change and future trends， and the impact of temperature and precipitation on runoff and evapotranspiration is analyzed. The results show that the SWAT model is highly applicable in the Lancang River Basin. Its simulated runoff correlation coefficient and Nash coefficient calibration and verification results are all more than 0.8. The annual runoff spatial distribution characteristics of the Lancang River present the characteristics of high in downstreams and main streams. Runoff is affected by both temperature and precipitation， and is proportional to precipitation and inversely proportional to temperature. Runoff changes are dominated by precipitation changes. The potential evapotranspiration of the Lancang River is inversely proportional to precipitation and directly proportional to temperature. Potential evapotranspiration is dominated by temperature changes. The upper， middle and lower reaches of the Lancang River in Yunnan Province are selected to analyze the characteristics of climate change， and the Coupled Model Intercomparison Project Phase 6 （CMIP6） and neural network model comprehensive forecast of the climate change trend in the future are compared. After the Mann-Kendall trend test， the temperature in the past forty years mainly has a rising trend. The time and space fluctuations in the precipitation are more complicated， mainly showing a tendency of significant decline， and the variation amplitude between adjacent years is 1 ℃ and 20% respectively.In the future， the temperature will be about 1 ℃/20 a， and the precipitation will be about 10%/20 a rise trend.In the future， temperature and precipitation change will increase the runoff and potential evaporation， and the increase in runoff is concentrated on the rainy season. It can increase by about 38%， and the evaporation is concentrated on the dry season. The maximum increase of about 6%， and the risk of drought and flood in the dry and rainy seasons increases. The strategy of returning farmland to forest in Yunnan Province is more needed to conserve water sources and the dispatch of cascade reservoirs to prevent and control droughts and floods.

针对全球变化背景下极端升温、暴雪和暖湿化现象以及中国新疆地区融雪洪水灾害风险增大问题，概述了新疆不同类型洪水灾害特征，重点阐述了近年发生频率增加、致灾性强、灾害风险增大，但在新疆未引起重视的融雪洪水的研究进展，对比分析了不同类型融雪径流模型特点和研究现状。综合目前融雪径流模型已有进展和面临的挑战，提出新疆未来研究需考虑融雪径流模型的物理机制和融雪消融过程，以提高预报预警精度。回顾了融雪洪水在新疆的预报预警技术，指出构建高精度预报预警融雪洪水模型所面临的风吹雪、冻土表层雪和雪面雨等关键问题，并提出提升新疆洪水模拟、预报预警、应对突发洪水的综合能力的关键技术，为提升新疆融水洪水预报预警技术提供思路与建议。

This study examines various flood disaster types in Xinjiang, China, encompassing extreme temperature fluctuations, snowstorms, and warm, humid conditions attributed to global climate change. This study focuses on the underexplored issue of snowmelt floods, addressing their escalating frequency, severity, and associated disaster risk in Xinjiang. Using VOSviewer software, this study analyzes the research keywords and collaboration networks among different authors investigating snowmelt floods in Xinjiang. In addition, it scrutinizes the research priorities of scholars at various stages. A comparative analysis of the characteristics and research status of diverse snowmelt runoff models is presented. The study advocates for future research in Xinjiang to delve into the physical mechanisms and snowmelt processes integral to snowmelt runoff models, aiming to enhance prediction and warning accuracy. An assessment of the technologies employed for predicting and warning snowmelt floods in Xinjiang is conducted, highlighting pertinent issues such as wind-blown snow, frozen ground surface snow, and rain-on-snow. Furthermore, the study recommends key technologies to augment the overall capability of flood simulation, prediction, warning, and response to abrupt floods. It also explores strategies for optimizing the use of flood resources. In conclusion, this study furnishes recommendations for refining the prediction and warning technology pertinent to snowmelt floods in Xinjiang.

采用改进的双层积雪模型，以澜沧江上游积雪区为例进行了模型的应用，并从模型原理、模型适用条件、模拟效果等方面与传统的度日模型进行了对比分析。结果表明：以能量平衡和质量平衡为基础的双层积雪模型，在融雪物理机制上更加完善；模型不仅能模拟积雪区融雪产流量，还能模拟融雪过程中冠层对降雪的截留量、蒸发中的汽化和升华损失量、冠层对地面的二次降水量、积雪层中融雪水再结冰量等各个子物理过程；通过对双层积雪模型的改进，使融雪产流模拟结果更符合实际情况，可为寒区冰雪水文过程的研究提供借鉴。

澜沧江作为亚洲最重要的跨国水系,是“亚洲水塔”的重要组成部分。分析澜沧江上游河川径流的演变规律与变化趋势,对澜沧江水资源利用与区域绿色能源发展具有重要的战略意义。基于乌弄龙断面径流资料与上游气温、降水资料,采用线性倾向估计、Mann-Kendall 趋势检验和Pearson相关系数等方法对近30 a澜沧江上游径流演变特征进行分析。结果表明,自有径流资料以来,澜沧江上游径流在年代际尺度上变化不大,但在季节上表现为冬季径流增加,夏季径流减少的特点,其中8月份径流在20世纪70年代初存在突变,并以4.8%/(10 a)的显著趋势下降,折合水量2.0亿m³/(10 a),且近30 a下降速率还在进一步加快,上游中部地区降水量减少与径流减少的趋势一致。研究还发现,近30 a以来,随着春季气温升高,上游径流对5月份气温的响应有所减弱,而4月份两者的关系变得密切。在全球变暖的大背景下,澜沧江上游融雪(冰)期逐渐提前,需要引起高度重视,并采取积极应对措施。