1. 数据集概述\n
本数据集面向青藏高原长江黄河源区冰冻圈变化及其水文生态影响研究，基于CMIP6多模式集合平均，提供SSP126、SSP245和SSP585三种情景下未来至21世纪末的冰川、冻土、积雪及降水相态预测数据。冰川、积雪和降水相态数据覆盖2015-2100年，冻土数据覆盖2018-2100年。各类数据均按照统一的长江黄河源研究区边界裁剪。\n
2. 数据内容与要素\n
冰川：包括冰川面积和冰川储量，单位分别为km²和km³。\n
冻土：包括活动层厚度（ALT）和年平均地温（MAGT），单位分别为m和℃。\n
积雪：提供积雪深度，单位为cm。\n
降水相态：包括降雨量（Rainfall）、降雪量（Snowfall）和雨夹雪量（Sleet），单位均为mm。\n
3. 时空分辨率与存储格式\n
冰川：空间表现形式为矢量，时间分辨率为日。冰川单元的静态边界以Shapefile格式存储；2015-2100年逐日冰川面积与冰川储量以NetCDF格式存储，数据维度为时间×冰川，并通过冰川编号（GLAC_ID）与静态矢量边界一一对应。\n
冻土：空间分辨率为1 km，时间分辨率为年，以GeoTIFF格式存储。\n
积雪：空间分辨率为0.05°，时间分辨率为日，以NetCDF格式按年存储，数据维度为时间×纬度×经度。\n
降水相态：空间分辨率为0.1°，时间分辨率为日，以NetCDF格式按年存储，数据维度为时间×纵向坐标×横向坐标。\n
4. 数据组织与命名\n
数据按冰川（Glacier）、冻土（Permafrost）、积雪深度（SnowDepth）和降水相态（PrecipitationType）分类，各类未来数据按SSP126、SSP245和SSP585情景分别组织。冻土数据进一步分为ALT和MAGT，降水相态数据进一步分为Rainfall、Snowfall和Sleet。\n
冰川数据由一套静态矢量边界和三份情景时间序列文件组成。glacier_geometry.shp为冰川单元静态边界，配套的.shp、.shx、.dbf、.prj和.cpg文件应共同保存和使用；glacier_ssp126_2015_2100.nc、glacier_ssp245_2015_2100.nc和glacier_ssp585_2015_2100.nc分别存储三种情景下2015-2100年的逐日冰川面积与冰川储量。NetCDF中的glacier_id与静态矢量属性表中的GLAC_ID一一对应。\n
积雪深度和降水相态逐日文件采用YYYYDDD命名，其中YYYY表示年份，DDD表示年内日序；冻土文件采用ALT_YYYY或MAGT_YYYY命名，其中YYYY表示年份。\n
5. 质量评估\n
冰川模型计算所得冰、雪度日因子的数值与已有研究接近，且均呈随海拔升高而增大的总体趋势，支持模型参数的合理性。\n
冻土模拟结果采用研究区多年冻土钻孔地温和活动层观测资料进行验证。年平均地温（MAGT）模拟结果的平均绝对误差为0.71 ℃，均方根误差为0.93 ℃；活动层厚度（ALT）模拟结果的平均绝对误差为0.11 m，均方根误差为0.41 m。结果表明，GIPL2模型能够较好模拟研究区多年冻土热状态及活动层厚度变化，模拟结果与观测资料具有较好一致性。\n
积雪深度经EDCDF校正后，多模式集合平均与参考数据的区域日平均雪深对比显示，平均绝对误差为0.252 cm，均方根误差为0.430 cm，相关系数为0.948，表明校正后的历史期结果与参考数据具有较好一致性。\n
降水相态采用的湿球温度方案经青藏高原75个气象站观测资料评估，模拟的非降雨量（降雪与雨夹雪合计）与观测年际变化的相关系数为0.90，整体表现优于采用固定温度阈值的雨雪分割方案。相比平均低估非降雨量约36%的CMIP6原始模拟，湿球温度方案的模拟结果更接近观测值。\n
6. 数据组织与命名\n数据按冰川（Glacier）、冻土（Permafrost）、积雪深度（SnowDepth）和降水相态（PrecipitationType）分类，各类未来数据按SSP126、SSP245和SSP585情景分别存储。冻土数据进一步分为ALT和MAGT，降水相态数据进一步分为Rainfall、Snowfall和Sleet。\n
冰川数据由一套静态矢量边界和三份情景时间序列文件组成。glacier_geometry.shp为冰川单元静态边界，配套的.shp、.shx、.dbf、.prj和.cpg文件应共同保存和使用；glacier_ssp126_2015_2100.nc、glacier_ssp245_2015_2100.nc和glacier_ssp585_2015_2100.nc分别存储三种情景下2015-2100年的逐日冰川面积与冰川储量。NetCDF中的glacier_id与静态矢量属性表中的GLAC_ID一一对应。\n
冻土数据以GeoTIFF格式按年存储，文件采用ALT_YYYY.tif或MAGT_YYYY.tif命名，其中YYYY表示年份。\n
积雪深度数据以NetCDF格式按年存储，各情景目录包含2015.nc至2100.nc。每个年度文件包含当年365个或366个逐日时间步，数据变量为snow_depth，单位为cm。\n
降水相态数据以NetCDF格式按年存储，Rainfall、Snowfall和Sleet下的各情景目录均包含2015.nc至2100.nc。每个年度文件包含当年365个或366个逐日时间步，数据变量分别为rainfall、snowfall和sleet，单位均为mm。\n
7. 应用范围\n
本数据集可用于不同气候情景下的冰川面积与储量变化分析、多年冻土热状况与活动层变化评估、积雪变化及降水相态转变研究，并为流域水文模拟、水资源变化评估及生态环境影响研究提供数据支持。", "ds_source": "", "ds_process_way": "", "ds_quality": "", "ds_acq_start_time": "2015-01-01 00:00:00", "ds_acq_end_time": "2100-12-31 00:00:00", "ds_acq_place": "长江黄河源区", "ds_acq_lon_east": 103.53027777777778, "ds_acq_lat_south": 30.8, "ds_acq_lon_west": 90.39500000000001, "ds_acq_lat_north": 36.15833333333333, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "login-access", "ds_total_size": 6471983483, "ds_files_count": 0, "ds_format": "*.nc,*.tif,*.shp", "ds_space_res": "冰川：矢量数据；冻土：1 km；积雪：0.05°；降水相态：0.1°", "ds_time_res": "冰川：日；冻土：年；积雪：日；降水相态：日", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "5fd83334-cabe-4534-97e5-9a91bbb58cab.jpg", "ds_thumb_from": 0, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "None", "ds_from_station": null, "organization_id": "52b7b79b-860c-49a5-9083-9a70cf8bed5a", "ds_serv_man": null, "ds_serv_phone": null, "ds_serv_mail": null, "doi_value": "10.12072/ncdc.db7795.2026", "subject_codes": [ "170" ], "quality_level": 0, "publish_time": "2026-09-20 09:46:35", "last_updated": "2026-09-20 09:46:35", "protected": false, "protected_to": "2028-08-30 00:00:00", "lang": "zh", "cstr": "11738.11.ncdc.db7795.2026", "i18n": { "en": { "title": "Scenario Projections of Glaciers, Permafrost, Snow, and Precipitation Phase in the Source Region of the Yangtze and Yellow Rivers Through the End of the 21st Century (2015-2100)", "ds_format": "*.nc,*.tif,*.shp", "ds_source": "", "ds_quality": "", "ds_ref_way": "", "ds_abstract": "
1. Dataset Overview\r\n
This dataset is designed to support research on cryospheric changes and their hydrological and ecological impacts in the Source Region of the Yangtze and Yellow Rivers on the Tibetan Plateau. Based on the CMIP6 multi-model ensemble mean, this dataset provides projections of glaciers, permafrost, snow cover, and precipitation phase under three future climate scenarios (SSP126, SSP245, and SSP585) through the end of the 21st century. Glacier, snow cover, and precipitation phase datasets cover the period from 2015 to 2100, while permafrost datasets cover the period from 2018 to 2100. All datasets were clipped according to the unified boundary of the Source Region of the Yangtze and Yellow Rivers.\r\n
2. Dataset Contents and Variables\r\n
Glacier: Includes glacier area and glacier volume, with units of km² and km³, respectively.\r\n
Permafrost: Includes active layer thickness (ALT) and mean annual ground temperature (MAGT), with units of m and ℃, respectively.\r\n
Snow Cover: Provides snow depth data, with a unit of cm.\r\n
Precipitation Phase: Includes rainfall (Rainfall), snowfall (Snowfall), and sleet (Sleet), with all variables expressed in mm.\r\n
3. Spatiotemporal Resolution and Storage Formats\r\n
Glaciers: spatially represented as vector data with a daily temporal resolution. The static boundaries of glacier units are stored in Shapefile format. Daily glacier area and glacier ice storage for 2015–2100 are stored in NetCDF format with dimensions of time × glacier. Each glacier in the NetCDF files is linked to the corresponding static vector feature through its glacier identifier (GLAC_ID).\r\n
Permafrost: a spatial resolution of 1 km and an annual temporal resolution, stored in GeoTIFF format.\r\n
Snow: a spatial resolution of 0.05° and a daily temporal resolution. The data are stored annually in NetCDF format with dimensions of time × latitude × longitude.\r\n
Precipitation phase: a spatial resolution of 0.1° and a daily temporal resolution. The data are stored annually in NetCDF format with dimensions of time × y × x.\r\n
4. Production Methods\r\n
Glacier data were developed based on the Sanjiangyuan Glacier Dataset from the Second Chinese Glacier Inventory. Glacier boundaries in the Source Region of the Yangtze and Yellow Rivers were extracted, and initial glacier ice storage was estimated using a glacier volume area scaling relationship. Glacier evolution was simulated using the modified FLEXG distributed degree day model incorporating shortwave radiation and surface albedo. Glacier albedo information was obtained from the Annual Albedo Dataset for Glaciers in High Mountain Asia. CMIP6 meteorological forcing data statistically downscaled and bias corrected using the EDCDF method were used to simulate snow accumulation, snowmelt, and glacier melt on a 100 m computational grid. Glacier area and ice storage were updated daily. The static vector boundaries record glacier locations and identifiers, while daily variations in glacier area and ice storage are stored as NetCDF time series.\r\n
Permafrost data were simulated using the GIPL2 model to estimate active layer thickness and mean annual ground temperature under future climate scenarios. The model was primarily driven by CMIP6 meteorological data statistically downscaled and bias corrected using the EDCDF method. Land cover, soil properties, soil moisture, snow, and other parameters were incorporated into the model inputs. Soil freezing and thawing processes were simulated based on the one dimensional transient heat conduction equation. Following historical period simulations, the model was driven by SSP126, SSP245, and SSP585 scenario data to produce annual active layer thickness and mean annual ground temperature projections for the Source Region of the Yangtze and Yellow Rivers.\r\n
Snow depth data were developed using the Tibetan Plateau 0.05° Daily Snow Depth Dataset as the reference dataset. The EDCDF method was applied to statistically downscale and bias correct the CMIP6 data, producing daily snow depth projections under the three scenarios.\r\n
Precipitation phase data were produced by statistically downscaling and bias correcting CMIP6 meteorological forcing data using the EDCDF method. Rainfall, snowfall, and sleet were identified and calculated using a wet bulb temperature based precipitation phase partitioning scheme that incorporates air temperature, humidity, atmospheric pressure, and elevation.\r\n
5. Quality Assessment\r\n
The calculated ice and snow degree-day factors from the glacier model were consistent with values reported in previous studies and showed an overall increasing trend with elevation, supporting the rationality of the model parameters.\r\n
The permafrost simulations were validated using observed ground temperature and active layer thickness measurements from permafrost boreholes within the study region. The mean absolute error (MAE) and root mean square error (RMSE) of simulated mean annual ground temperature (MAGT) were 0.71 ℃ and 0.93 ℃, respectively. For active layer thickness (ALT), the MAE and RMSE were 0.11 m and 0.41 m, respectively. These results indicate that the GIPL2 model can effectively represent permafrost thermal conditions and active layer variations in the study region, with good agreement between simulations and observations.\r\n
After EDCDF correction, the multi-model ensemble mean snow depth was compared with reference data during the historical period. The regional daily mean snow depth showed a mean absolute error (MAE) of 0.252 cm, a root mean square error (RMSE) of 0.430 cm, and a correlation coefficient of 0.948, indicating good consistency between the corrected historical simulations and reference data.\r\n
The wet-bulb temperature-based precipitation phase scheme was evaluated using observations from 75 meteorological stations across the Tibetan Plateau. The correlation coefficient between simulated non-liquid precipitation (the combined amount of snowfall and sleet) and observed interannual variations was 0.90. The method performed better than fixed temperature threshold-based precipitation phase partitioning schemes. Compared with original CMIP6 simulations, which underestimated non-liquid precipitation by approximately 36% on average, the wet-bulb temperature-based method produced results closer to observations.\r\n
6. Data Organization and Naming Convention\r\n
The dataset is organized into four categories: Glacier, Permafrost, SnowDepth, and PrecipitationType. Future projection datasets are stored separately according to the SSP126, SSP245, and SSP585 scenarios. Permafrost data are further divided into ALT and MAGT, while precipitation phase data are further divided into Rainfall, Snowfall, and Sleet.\r\n
The glacier data consist of one static vector boundary dataset and three scenario time series files. The glacier_geometry.shp file contains the static boundaries of the glacier units. Its associated .shp, .shx, .dbf, .prj, and .cpg files must be stored and used together. The glacier_ssp126_2015_2100.nc, glacier_ssp245_2015_2100.nc, and glacier_ssp585_2015_2100.nc files contain daily glacier area and glacier ice storage under the three scenarios from 2015 to 2100. The glacier_id variable in the NetCDF files corresponds directly to the GLAC_ID field in the static vector attribute table.\r\n
Permafrost data are stored annually in GeoTIFF format. The files follow the ALT_YYYY.tif or MAGT_YYYY.tif naming convention, where YYYY represents the year.\r\n
Snow depth data are stored annually in NetCDF format. Each scenario directory contains files from 2015.nc to 2100.nc. Each annual file contains 365 or 366 daily time steps. The data variable is snow_depth, with a unit of cm.\r\n
Precipitation phase data are stored annually in NetCDF format. Each scenario directory under Rainfall, Snowfall, and Sleet contains files from 2015.nc to 2100.nc. Each annual file contains 365 or 366 daily time steps. The corresponding data variables are rainfall, snowfall, and sleet, all expressed in mm.\r\n
7. Applications\r\n
This dataset can be used to analyze glacier area and volume changes under different climate scenarios, evaluate permafrost thermal conditions and active layer variations, investigate snow cover changes and precipitation phase transitions, and provide data support for watershed hydrological modeling, water resource assessment, and ecological impact studies.", "ds_time_res": "Glaciers: daily; permafrost: annual; snow: daily; precipitation phase: daily", "ds_acq_place": "Source area of Yangtze and Yellow River", "ds_space_res": "Glacier: vector data; frozen soil: 1 km; snow cover: 0.05°; precipitation phase: 0.1°", "ds_projection": "", "ds_process_way": "", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 0, "recommendation_value": 0, "license_type": "https://creativecommons.org/licenses/by/4.0/", "doi_reg_from": "reg_local", "cstr_reg_from": "reg_local", "doi_not_reg_reason": null, "cstr_not_reg_reason": null, "is_paper_in_submitting": false, "belong_to_nieer": false, "allow_update_data": false, "created_from": null, "ds_topic_tags": [ "冰川", "冻土", "积雪", "降水相态", "情景预测" ], "ds_subject_tags": [ "地球科学" ], "ds_class_tags": [], "ds_locus_tags": [ "长江黄河源区" ], "ds_time_tags": [ 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022, 2023, 2024, 2025, 2026, 2027, 2028, 2029, 2030, 2031, 2032, 2033, 2034, 2035, 2036, 2037, 2038, 2039, 2040, 2041, 2042, 2043, 2044, 2045, 2046, 2047, 2048, 2049, 2050, 2051, 2052, 2053, 2054, 2055, 2056, 2057, 2058, 2059, 2060, 2061, 2062, 2063, 2064, 2065, 2066, 2067, 2068, 2069, 2070, 2071, 2072, 2073, 2074, 2075, 2076, 2077, 2078, 2079, 2080, 2081, 2082, 2083, 2084, 2085, 2086, 2087, 2088, 2089, 2090, 2091, 2092, 2093, 2094, 2095, 2096, 2097, 2098, 2099, 2100 ], "ds_contributors": [ { "true_name": "李珅", "email": "lishen@stu.xjtu.edu.cn", "work_for": "西安交通大学", "country": "中国" }, { "true_name": "赵煜峰", "email": "zhaoyuf@stu.xjtu.edu.cn", "work_for": "西安交通大学", "country": "中国" }, { "true_name": "姚莹莹", "email": "yaoyy27@xjtu.edu.cn", "work_for": "西安交通大学", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "李珅", "email": "lishen@stu.xjtu.edu.cn", "work_for": "西安交通大学", "country": "中国" } ], "ds_managers": [ { "true_name": "李珅", "email": "lishen@stu.xjtu.edu.cn", "work_for": "西安交通大学", "country": "中国" } ], "category": "基础地理" }