基于Crocus模型的雪深模拟研究  

Research on snow depth simulation of Crocus model

作　　者：陶星宇 柳锦宝[2] 拉珍 巴桑[4] 高瑜莲 赵爽 汪菀蔚 TAO Xingyu;LIU Jinbao;Lazhen;Basang;GAO Yulian;ZHAO Shuang;WANG Wanwei(Pingshan Meteorological Service,Pingshan 645350,Sichuan,China;School of Resources and Environment,Chengdu University of Information Technology,Chengdu 610225,China;Climate Center of Tibet Autonomous Region,Lhasa 850001,China;Xizang Institute of Plateau Atmospheric and Environmental Science Research,Lhasa 850001,China;Department of Geographic Sciences,Beijing Normal University,Beijing 100875,China)

机构地区：[1]屏山县气象局,四川宜宾645350 [2]成都信息工程大学资源环境学院,四川成都610200 [3]西藏自治区气候中心,西藏拉萨850001 [4]西藏高原大气环境科学研究所,西藏拉萨850001 [5]北京师范大学地理科学学部环境变化与自然灾害教育部重点实验室,北京100875

出　　处：《冰川冻土》2024年第4期1184-1196,共13页Journal of Glaciology and Geocryology

基　　金：国家自然科学基金项目(42065008);四川省科技厅重点研发计划项目(2021YFS0328,2020YFG0146);中国气象局创新发展专项(CXFZ2021J055)资助。

摘　　要：雪深的准确模拟对于寒区的水文过程、气候变化、生态环境具有重要意义。积雪过程模型可以定量模拟物理环境对积雪参数的影响过程,从物理意义上定量模拟区域雪深。但是,目前积雪过程模型模拟存在参数设置复杂、不确定性大等问题。因此,有必要通过参数率定找到最适应区域的参数组合,从而完成模型参数的本地化。本文应用Crocus模型,在西藏自治区的聂拉木站、普兰站和帕里站进行3 a(2019—2021年)逐日雪深模拟,利用UQ-Pyl进行模型参数的敏感性分析及率定,构建以上站点的Crocus模型,并通过相关系数(R)、标准差比(SDC)、纳什效率系数(NSE)、平均差值(BIAS)和均方根误差(RMSE)综合评估Crocus模型在聂拉木站、普兰站和帕里站的适用性。结果表明:聂拉木站、普兰站、帕里站的Crocus模型主要敏感参数分别有20、15、13个;Crocus模型在聂拉木站的雪深模拟效果最佳(R=0.989、SDC=0.990、NSE=0.978、BIAS=0.276 cm、RMSE=4.280 cm),普兰站和帕里站次之;其中,聂拉木站的积雪沉降、积累和融化过程得到了很好的模拟,普兰站在积雪较厚时模拟效果更优,帕里站在每年3—4月时模型高估了雪深,但变化趋势基本一致,3个站点的日雪深模拟结果基本反映了雪深日变化过程。本文对Crocus模型的本地化是对积雪过程模拟研究的有益补充,为理解和揭示西藏自治区的积雪过程可提供科学依据和信息,同时,也为提高区域气候预测水平、水资源有效管理、灾害预防和应对提供有力支撑。In the context of global warming,the cryosphere is the second largest climate system in the world,and its changing characteristics and impacts are attracting more and more attention.In the past decade or more,the cryosphere has been shrinking all over the world.As a major part of the cryosphere,the change of snow cover will have a significant impact on climate change,ecological environment,agriculture and animal husbandry development and water resources balance.It will also affect the ecological environment,the development of agriculture and animal husbandry and the balance of water resources.The snow resources in the southern border area are extremely rich.Global warming accelerates the melting of snow and glaciers in Xizang,resulting in frequent snow and ice disasters such as avalanches,glacial lake outburst and glacial debris flow.It has a great impact on local production and residents'lives.Therefore,accurate simulation of snow depth is of great significance for hydrological processes,climate change,and ecological environment in cold regions.At present,the methods for simulating and monitoring snow depth mainly include ground snow depth observation,inversion of snow depth based on microwave remote sensing,measurement of snow depth based on satellite altimetry technology,and simulation of snow depth using snow accumulation process models.The snow accumulation process model is based on strict physical meaning and can quantitatively simulate the influence of physical environment on snow parameters,quantitatively simulating regional snow depth from a physical sense.The snow accumulation process model mainly includes single-layer snow accumulation model,medium complexity snow accumulation model,and detailed snow accumulation model.The model used in this paper is a detailed snow cover model,which is based on the layering of snow physical properties and can describe the evolution of snow microstructure over time.However,the current snow accumulation process model simulation has problems such as complex parameter set

关 键 词：积雪深度 Crocus积雪模型 雪深模拟 UQ-Pyl 敏感性分析 参数率定 

分 类 号：P426.635[天文地球—大气科学及气象学]