本数据集针对北极陆地自然灾害风险开展系统评估，填补了该区域多灾种综合风险数据的空白。研究基于1980–2020年多源数据，包括CPC全球降水与温度数据、NOAA再分析风速与积雪数据、LandScan人口密度数据以及基础地理信息数据，利用ArcGIS、ENVI、Matlab等工具进行统计分析与归一化处理，构建了北极地区雪灾风险、热喀斯特湖泊易发性及高温热浪危险性三类单灾种数据集，并进一步通过空间叠加形成1980–2020年北极陆地综合灾害风险数据集。其中，雪灾风险涵盖降雪频率、积雪深度、风速等多类致灾因子与承灾体信息；热喀斯特湖泊易发性基于随机森林模型构建，模型准确率达0.83，AUC为0.90；高温热浪危险性基于概率分布函数提取温度阈值，生成频率、日数、强度与振幅四类指标。本数据集可为北极地区自然灾害风险评估、生态安全研究和气候变化响应提供重要数据支撑。", "ds_source": "
MODIS数据来源于美国国家航空航天局（NASA）对地观测系统（EOS）计划中的中分辨率成像光谱仪（MODIS）传感器，可通过MODIS官方网站（https://modis.gsfc.nasa.gov/）获取。该数据产品由NASA的地球数据云平台（Earthdata Cloud）和陆地过程分布式主动存档中心（LP DAAC）负责分发与维护，涵盖大气、陆地、海洋等多学科数据集，时间覆盖范围自2000年至今，部分产品时间分辨率可达每日，空间分辨率包括250 m、500 m及1 km等多种级别。MODIS数据经过辐射定标、大气校正及几何精校正等处理，广泛应用于全球环境变化、灾害监测和地表过程研究。用户可通过Direct Broadcast提供商或官方数据门户订购和下载相关产品。", "ds_process_way": "
（1）利用ArcGIS、ENVI和Matlab等软件对CPC全球降水、温度数据，NOAA再分析积雪和风速数据等基础气象数据进行统计分析及线性归一化处理，生成包括降雪频率、最大积雪深度、积雪日数、平均降雪量和最大风速等指标的北极地区灾害因子数据集；\n
（2）基于随机森林模型，对高程、坡度、坡向、平面曲率、剖面曲率、地形湿度指数、年平均地温、活动层厚度、降水、植被指数和土壤含量等多类环境因子进行建模，评估热喀斯特湖泊在当前（1980–2016）和未来（2050s）时期的易发性，模型准确率为0.83，AUC为0.90，并以0.2为步长对概率图进行重分类，生成五级易发性分布结果；\n
（3）采用基于概率分布函数的温度阈值方法，利用CPC全球每日最高温度数据，以滑动窗口前后30天提取夏季（7–8月）日最高温度，构建温度分布曲线，取90%分位数作为阈值，生成北极地区热浪温度阈值的空间分布，进而计算得到高温热浪频率、持续日数、强度及振幅等指标，并通过加权处理合成高温热浪危险性数据集；\n
（4）综合冻融灾害、道路雪灾和高温热浪三种单灾种风险数据，通过空间叠加分析生成北极陆地地区1980–2020年综合灾害风险空间分布图。", "ds_quality": "
本数据集通过随机森林模型精度指标（准确率、AUC）及空间验证方法进行质量评估。其中，热喀斯特湖泊易发性模型的准确率达0.83，AUC为0.90，表明模型具有较高的预测可靠性。高温热浪危险性数据采用概率分布函数法计算温度阈值，基于滑动窗口统计方法有效提取热浪特征参数。\n
雪灾风险数据在线性归一化处理后，关键指标（积雪深度、风速等）与实地观测数据的一致性较高。综合风险数据通过多源灾害数据空间叠加生成，与历史灾害记录的空间匹配度超过90%。\n
整体而言，本数据集在北极大部分地区具有较好的适用性，尤其在中低纬度北极区域可靠性较高，可用于长期环境变化研究及灾害风险评估。", "ds_acq_start_time": "1980-01-01 00:00:00", "ds_acq_end_time": "2015-12-31 00:00:00", "ds_acq_place": "北极地区", "ds_acq_lon_east": 180.0, "ds_acq_lat_south": 66.33999999999999, "ds_acq_lon_west": null, "ds_acq_lat_north": 90.0, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "login-access", "ds_total_size": 22790139, "ds_files_count": 2, "ds_format": "*.tif", "ds_space_res": "0.5°×0.5°；1 km; 32 km", "ds_time_res": "年", "ds_coordinate": "WGS84", "ds_projection": "North_Pole_Lambert_Azimuthal_Equal_Area", "ds_thumbnail": "d49ceb0d-3c8e-44d4-aed0-9a3e7f00ad29.png", "ds_thumb_from": 2, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": null, "organization_id": "53943799-d453-4bf2-a141-56c205c1355b", "ds_serv_man": "李红星", "ds_serv_phone": "0931-4967592", "ds_serv_mail": "ncdc@lzb.ac.cn", "doi_value": "", "subject_codes": [ "170.4510" ], "quality_level": 3, "publish_time": "2026-05-07 16:44:09", "last_updated": "2026-05-07 16:44:18", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.NCDC.ARCTIC-CHANGE.DB7129.2026", "i18n": { "en": { "title": "The Arctic Land Natural Disaster Risk Dataset from 1980 to 2020", "ds_format": "*.tif", "ds_source": "
The MODIS data is sourced from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor in NASA's Earth Observation System (EOS) program and can be accessed through the MODIS official website（ https://modis.gsfc.nasa.gov/ ）Get it. This data product is distributed and maintained by NASA's Earthdata Cloud platform and the Land Process Distributed Active Archive Center (LP DAAC), covering multidisciplinary datasets such as atmosphere, land, and ocean. The time coverage range is from 2000 to the present, and some products have daily time resolution and spatial resolution levels including 250m, 500m, and 1km. MODIS data has been processed through radiometric calibration, atmospheric correction, and geometric precision correction, and is widely used in global environmental change, disaster monitoring, and surface process research. Users can order and download related products through Direct Broadcast providers or official data portals.", "ds_quality": "
This dataset is evaluated for quality using random forest model accuracy metrics (accuracy, AUC) and spatial validation methods. Among them, the accuracy of the susceptibility model for hot karst lakes reached 0.83, with an AUC of 0.90, indicating that the model has high predictive reliability. The probability distribution function method is used to calculate the temperature threshold for high-temperature heat wave hazard data, and the sliding window statistical method is used to effectively extract heat wave characteristic parameters.\r\n
After linear normalization, the snow disaster risk data shows high consistency between key indicators such as snow depth and wind speed and field observation data. The comprehensive risk data is generated by spatially overlaying multi-source disaster data, with a spatial matching degree of over 90% with historical disaster records.\r\n
Overall, this dataset has good applicability in most areas of the Arctic, especially in the mid to low latitude Arctic region where it has high reliability and can be used for long-term environmental change research and disaster risk assessment.", "ds_ref_way": "", "ds_abstract": "This dataset conducts a systematic assessment of natural disaster risks on Arctic land, filling the gap in comprehensive risk data for multiple disasters in the region. The study is based on multi-source data from 1980 to 2020, including CPC global precipitation and temperature data, NOAA reanalysis wind speed and snow data, LandScan population density data, and basic geographic information data. Statistical analysis and normalization were performed using tools such as ArcGIS, ENVI, Matlab, etc., to construct three single disaster datasets for snow disaster risk, susceptibility of hot karst lakes, and high temperature heat wave risk in the Arctic region. Furthermore, a comprehensive Arctic land disaster risk dataset from 1980 to 2020 was formed through spatial superposition. Among them, the risk of snow disasters covers multiple types of disaster causing factors and information on disaster bearing bodies, such as snowfall frequency, snow depth, and wind speed; The susceptibility of hot karst lakes is based on a random forest model, with an accuracy of 0.83 and an AUC of 0.90; The risk of high-temperature heat waves is based on probability distribution functions to extract temperature thresholds and generate four indicators: frequency, number of days, intensity, and amplitude. This dataset can provide important data support for natural disaster risk assessment, ecological security research, and climate change response in the Arctic region.", "ds_time_res": "", "ds_acq_place": "Arctic region", "ds_space_res": "", "ds_projection": "", "ds_process_way": "
(1) Using software such as ArcGIS, ENVI, and Matlab, statistical analysis and linear normalization were performed on CPC global precipitation and temperature data, NOAA reanalysis snow cover and wind speed data, and other basic meteorological data to generate a disaster factor dataset for the Arctic region, including indicators such as snowfall frequency, maximum snow depth, snow days, average snowfall, and maximum wind speed;\r\n
(2) Based on the random forest model, multiple environmental factors such as elevation, slope, aspect, plane curvature, profile curvature, terrain humidity index, annual average ground temperature, active layer thickness, precipitation, vegetation index, and soil content were modeled to evaluate the susceptibility of hot karst lakes in the current (1980-2016) and future (2050s) periods. The model accuracy was 0.83, AUC was 0.90, and the probability map was reclassified with a step size of 0.2 to generate a five level susceptibility distribution result;\r\n
(3) Using a temperature threshold method based on probability distribution function, the CPC global daily highest temperature data is utilized to extract the highest temperature in summer (July August) within 30 days before and after the sliding window. A temperature distribution curve is constructed, and the 90th percentile is taken as the threshold to generate the spatial distribution of the temperature threshold for heatwaves in the Arctic region. Subsequently, indicators such as frequency, duration, intensity, and amplitude of heatwaves are calculated, and a heatwave risk dataset is synthesized through weighted processing;\r\n
(4) Based on the comprehensive risk data of freeze-thaw disasters, road snow disasters, and high-temperature heat waves, a spatial distribution map of comprehensive disaster risks in the Arctic land region from 1980 to 2020 was generated through spatial superposition analysis.", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 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, "ds_topic_tags": [ "冰冻圈灾害", "自然灾害", "雪灾" ], "ds_subject_tags": [ "自然地理学" ], "ds_class_tags": [], "ds_locus_tags": [ "北极地区" ], "ds_time_tags": [ 1980, 1981, 1982, 1983, 1984, 1985, 1986, 1987, 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015 ], "ds_contributors": [ { "true_name": "刘吉夫", "email": "liujifu@bnu.edu.cn", "work_for": "北京师范大学", "country": "中国" }, { "true_name": "朱文泉", "email": "zhuwq75@bnu.edu.cn", "work_for": "北京师范大学", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "刘吉夫", "email": "liujifu@bnu.edu.cn", "work_for": "北京师范大学", "country": "中国" } ], "ds_managers": [ { "true_name": "刘吉夫", "email": "liujifu@bnu.edu.cn", "work_for": "北京师范大学", "country": "中国" } ], "category": "极地" }