本数据采用一种新型永久冻土测绘方法，以卫星衍生的地表融化与冻结指数作为输入，并以基于调查的次区域永久冻土图作为约束条件，于2010年为QTP创建了新的永久冻土分布图。该方法通过将经验土壤参数（其值通过空间聚类和参数优化，受基于调查的次区域永久冻土图约束）纳入模型，考虑了局部因素的影响，并进一步改进以减少参数等效性。该新地图显示，2010年青藏高原永久冻土总面积约为1.086×10⁶ km²（占QTP面积的41.2%），季节性冻土面积约为1.447×10⁶ km²（占54.9%），排除冰川和湖泊。通过基于调查的子区域永久冻土图（κ=0.74）和钻孔记录（整体准确度=0.85，κ=0.43）进行的验证表明，该图的准确度高于其他两份近期地图。对地图间存在明显差异的区域进行检查证实，本地图的永久冻土分布比Zou等（2017）地图更为真实。鉴于其卓越的准确性，本地图可作为青藏高原陆面模拟的基准地图，用于约束/验证模拟结果，并作为历史参考，用于在全球变暖背景下预测青藏高原未来永久冻土变化。", "ds_source": "
（1）次区域永久冻土图；
（2）中等分辨率的地表温度 （LST） 数据产品 Terra 和 Aqua 卫星上的成像光谱辐射计 （MODIS）；
（3）环境因素：归一化差异植被指数 （NDVI） 乘积 （MOD13A2）；地形因素，包括高程和坡度，源自 航天飞机雷达地形任务 90 m 数字高程数据库 ；STRM衍生的地形湿度 指数（TWI）以及2005—2010年年平均降水量 根据中国1 km月降水量数据集；平均总积雪分数 （FSC） 数据 同期 500 m 日积雪分数数据集；土壤纹理类型数据，源自中国地表建模土壤性质数据集；", "ds_process_way": "
采用了Hu等（2020）开发的FROSTNUM/COP制图方法，对青藏高原（QTP）的永久冻土分布进行了制图。该方法基于扩展地面表面霜冻数值（FROSTNUM）模型，该模型以卫星温度数据为输入，并需要各子区域的永久冻土分布图作为优化约束条件。该方法通过模型参数E考虑局部因素，其值通过空间聚类、参数优化和决策树的流程，对所有空间单元进行了最优确定。", "ds_quality": "
该地图与基于调查的次区域永久冻土地图具有良好的一致性，Kappa系数为0.74，远高于近期发表的两份地图（Zou等，2017；Wang等，2019c）的Kappa系数。在与2010年左右收集的72个永久冻土存在钻孔记录进行验证后得出结论，我们的地图比Zou地图和Wang地图表现更好。在一些我们发现我们的地图与Zou地图存在明显差异的地区，我们的地图被证明更可接受；这一结论得到了来自多个方面的证据支持，包括卫星图像、PZI统计数据、高程特征以及更多独立的钻孔记录。", "ds_acq_start_time": "2010-01-01 00:00:00", "ds_acq_end_time": "2010-12-31 00:00:00", "ds_acq_place": "青藏高原", "ds_acq_lon_east": 104.5, "ds_acq_lat_south": 26.0, "ds_acq_lon_west": 73.5, "ds_acq_lat_north": 40.0, "ds_acq_alt_low": 3000.0, "ds_acq_alt_high": 5000.0, "ds_share_type": "login-access", "ds_total_size": 24450225, "ds_files_count": 2, "ds_format": "tif", "ds_space_res": "1000", "ds_time_res": "年", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "78c4828a-6734-40fd-99eb-d7253aaf279c.png", "ds_thumb_from": 2, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": null, "organization_id": "a4dd5849-78f2-44c5-b0f1-3450e952b2a2", "ds_serv_man": "敏玉芳", "ds_serv_phone": "09314967596", "ds_serv_mail": "ncdc@lzb.ac.cn", "doi_value": "", "subject_codes": [ "170.45" ], "quality_level": 3, "publish_time": "2025-08-29 15:27:52", "last_updated": "2026-01-14 11:04:01", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.NCDC.FIGSHARE.DB6966.2025", "i18n": { "en": { "title": "Distribution Map of Permafrost on the Qinghai Tibet Plateau Based on Sub regional Survey Maps: Benchmark Map for Regional Permafrost Modeling (2010)", "ds_format": "tif", "ds_source": "
(1) Sub regional permafrost map;
(2) The imaging spectral radiometer (MODIS) on Terra and Aqua satellites is a medium resolution surface temperature (LST) data product;
(3) Environmental factors: Normalized Difference Vegetation Index (NDVI) product (MOD13A2); Terrain factors, including elevation and slope, are derived from the 90 meter digital elevation database of the space shuttle radar terrain mission; The terrain humidity index (TWI) derived from STRM and the average annual precipitation from 2005 to 2010 are based on the 1 km monthly precipitation dataset in China; The average total snow cover score (FSC) data is collected from a 500m daily snow cover score dataset during the same period; Soil texture type data, sourced from the Chinese surface modeling soil properties dataset;", "ds_quality": "
This map has good consistency with the survey based sub regional permafrost map, with a Kappa coefficient of 0.74, which is much higher than the two recently published maps (Zou et al., 2017); The Kappa coefficient of Wang et al. (2019c). After verifying with drilling records of 72 permafrost samples collected around 2010, it was concluded that our map performs better than the Zou and Wang maps. In some areas where we have found significant differences between our map and the Zou map, our map has been proven to be more acceptable; This conclusion is supported by evidence from multiple sources, including satellite imagery, PZI statistical data, elevation features, and more independent drilling records.", "ds_ref_way": "", "ds_abstract": "
This data adopts a new permafrost mapping method, using satellite derived surface melting and freezing indices as inputs, and a sub regional permafrost map based on surveys as constraints. In 2010, a new permafrost distribution map was created for QTP. This method incorporates empirical soil parameters (whose values are constrained by survey based sub regional permafrost maps through spatial clustering and parameter optimization) into the model, taking into account the influence of local factors and further improving to reduce parameter equivalence. The new map shows that in 2010, the total area of permafrost on the Qinghai Tibet Plateau was approximately 1.086 × 10 ⁶ km ² (accounting for 41.2% of the QTP area), and the seasonal permafrost area was approximately 1.447 × 10 ⁶ km ² (accounting for 54.9%), excluding glaciers and lakes. The validation based on the investigation of the sub regional permafrost map (kappa=0.74) and drilling records (overall accuracy=0.85, kappa=0.43) shows that the accuracy of this map is higher than the other two recent maps. By examining the areas with significant differences between maps, it was confirmed that the distribution of permafrost in this map is more realistic than in the map by Zou et al. (2017). Given its excellent accuracy, the local map can serve as a benchmark map for simulating the land surface of the Qinghai Tibet Plateau, to constrain/validate simulation results, and as a historical reference for predicting future permafrost changes on the plateau in the context of global warming.
", "ds_time_res": "年", "ds_acq_place": "Qinghai-Tibet Plateau", "ds_space_res": "1000", "ds_projection": "", "ds_process_way": "The FROSTNUM/COP mapping method developed by Hu et al. (2020) was used to map the distribution of permafrost in the Qinghai Tibet Plateau (QTP). The method is based on the extended ground surface frost numerical model (FROSTNUM), which takes satellite temperature data as input and requires the permafrost distribution map of each sub region as the optimization constraint condition. This method considers local factors through the model parameter E, whose values are optimally determined for all spatial units through the process of spatial clustering, parameter optimization, and decision tree.", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 0, "license_type": "CC BY 4.0", "ds_topic_tags": [ "青藏高原", "多年冻土", "基准图" ], "ds_subject_tags": [ "地理学" ], "ds_class_tags": [], "ds_locus_tags": [ "青藏高原" ], "ds_time_tags": [ 2010 ], "ds_contributors": [ { "true_name": "南卓铜", "email": "nanzt@shnu.edu.cn", "work_for": "上海师范大学", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "南卓铜", "email": "nanzt@shnu.edu.cn", "work_for": "上海师范大学", "country": "中国" } ], "ds_managers": [ { "true_name": "南卓铜", "email": "nanzt@shnu.edu.cn", "work_for": "上海师范大学", "country": "中国" } ], "category": "冻土" }