准确掌握地表2米气温（T2）的时空分布状况，是开展区域气候变化分析的重要前提。然而，基于观测的气温数据通常涵盖了气候系统自然变率与人类活动对区域水文气候状况的共同影响，尤其在大规模植被恢复区（如黄土高原），以生态工程为代表的人类活动对区域气候变化的定量贡献仍有待进一步明晰。为厘清植被恢复工程的水文气候效应，本研究基于区域气候模式WRF，以ERA-Interim再分析资料作为气象驱动数据，制备动态变化的下垫面数据和静态不变的下垫面驱动数据，在黄土高原地区分别开展了实施植被恢复（DYN）和未实施植被恢复（CTL）情景下的长时间序列高分辨率气候模拟，生成了两种情景下1999–2018年10 km空间分辨率的日平均T2数据集。经与CMFD资料T2数据对比验证，两种情景下WRF模拟结果的空间相关系数PCC大于0.94，均方根误差RMSE小于1.17°C，偏差Bias主要集中于[−1°C, 1°C]区间，表明其在黄土高原地区具有较好的适用性。本数据集可为定量评估黄土高原植被恢复的水文气候效应提供可靠的数据基础。
\n本数据以NetCDF格式（*.nc）存储，共有两个数据文件。数据文件命名格式为T2day_WRF_TYPE_Daily_19990101_20181231_LoessPlateauRegion.nc。其中，TYPE为情景名称，DYN为实施植被恢复情景，CTL为未实施植被恢复情景。数据文件中T2day为WRF模拟的日平均地表2米气温，时空分辨率分别为日、10 km，时间范围为1999-2018年，XLAT和XLONG分别为模式格点的纬度和经度。", "ds_source": "
本研究采用了多源高精度的气象数据和动态变化、静态不变的下垫面数据开展区域气候模拟研究。实施植被恢复（DYN）、未实施植被恢复（CTL）情景下气象驱动数据均使用由欧洲中期天气预报中心（ECMWF）提供的ERA-Interim再分析资料（https://gdex.ucar.edu/datasets/d627000/#），该产品系统性地融合了卫星、探空、地面观测资料，确保了全球尺度的数据覆盖与一致性，其时间分辨率为6 h，空间分辨率为0.75°，时间范围为1979年1月–2019年9月。在实施植被恢复情景下，需制备动态变化的下垫面数据。其中，土地利用数据采用由欧洲空间局提供的ESA-CCI数据（http://www.esa-landcover-cci.org/），该产品通过协同处理MERIS、AVHRR等多源卫星传感器数据，生成了高精度、一致性强和长序列的土地利用数据，其时间分辨率为逐年，空间分辨率为300 m，时间范围为1992–2020年。植被特征参数采用由中国国家地球系统科学数据中心提供的GLASS数据集（https://www.geodata.cn/thematicView/GLASS.html?guid2=52092346086099），该产品基于多源遥感数据和实测站点数据，结合反演算法获得了长序列、高质量、空间连续的遥感反演数据，其时间分辨率为8 day，空间分辨率为0.05°，时间范围为1982–2018年。未实施植被恢复的情景则使用WRF提供的静态不变下垫面数据。其中，土地利用数据、植被特征参数数据均采用WRF默认使用的静态地理数据，分别由美国地质调查局等机构发布，代表1992–1993年全球地表下垫面的状况，可反映工程实施前的区域下垫面状况（https://www2.mmm.ucar.edu/wrf/users/download/get_sources_wps_geog_V3.html）。
", "ds_process_way": "（1）利用WRF预处理系统（WPS）对ESA-CCI土地利用数据和GLASS植被特征参数数据进行预处理，制备动态变化的地表驱动数据（替换黄土高原区域内的叶面积指数、植被盖度、地表反照率和土地利用类型），使其与模型设置的分辨率一致；（2）利用WPS制备静态不变的地表驱动数据，使其与模型设置的分辨率一致；（3）利用WPS对ERA-Interim再分析资料进行预处理，制备模拟时段的气象驱动数据，使其与模型设置的分辨率一致；（4）在优选参数化方案组合的基础上，利用（1）、（2）和（3）中的数据驱动WRF模式开展有、无实施植被恢复情景下的黄土高原区域水文气候模拟；（5）输出两种情景下WRF的模拟结果，并与CMFD格点数据进行对比验证，评估模拟结果精度。
", "ds_quality": "采用空间相关系数（PCC）、均方根误差（RMSE）、偏差（Bias）三个指标，结合CMFD格点数据对WRF模拟结果进行了评估。评估结果显示，WRF的模拟结果与CMFD资料结果在空间分布上相关性强，PCC大于0.94，空间上RMSE小于1.17°C，Bias主要集中于[−1°C, 1°C]区间。此外，我们还对WRF模拟气温区域平均值的年时间序列和年内分布状况也展开了评估，结果表明WRF模拟结果与CMFD资料高度一致。上述结果表明，WRF模拟的T2在黄土高原地区具有较好的适用性，可用于黄土高原地区植被恢复工程水文气候效应的评估。
", "ds_acq_start_time": "1999-01-01 00:00:00", "ds_acq_end_time": "2018-12-31 00:00:00", "ds_acq_place": "黄土高原", "ds_acq_lon_east": 114.57, "ds_acq_lat_south": 33.69, "ds_acq_lon_west": 100.86, "ds_acq_lat_north": 41.28, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "login-access", "ds_total_size": 1477886240, "ds_files_count": 0, "ds_format": "*.nc", "ds_space_res": "10km", "ds_time_res": "日", "ds_coordinate": "WGS84", "ds_projection": "Lambert Conformal Conic Projection System", "ds_thumbnail": "6a529e85-f0ba-48fc-9fdd-51ef4434e314.jpeg", "ds_thumb_from": 0, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": "", "organization_id": "bf138922-7121-438c-8d1b-19d5f751c907", "ds_serv_man": "李红星", "ds_serv_phone": "0931-4967592", "ds_serv_mail": "ncdc@lzb.ac.cn", "doi_value": "", "subject_codes": [ "170.15" ], "quality_level": 0, "publish_time": "2026-05-21 09:37:06", "last_updated": "2026-05-21 10:48:50", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.ncdc.loess.db7329.2026", "i18n": { "en": { "title": "Daily 2-meter air temperature dataset for the Loess Plateau from WRF simulations (1999-2018)", "ds_format": "*.nc", "ds_source": "This study employed the multi-source high-precision meteorological datasets along with both dynamic and static land surface datasets to conduct regional climate simulations. For both dynamic vegetation (DYN) and control (CTL) experimental scenarios, the meteorological forcing was provided by the ERA-Interim reanalysis dataset from the European Centre for Medium-Range Weather Forecasts (ECMWF, https://gdex.ucar.edu/datasets/d627000/#). This product systematically integrates satellite retrievals, radiosonde, and surface station observations, ensuring global data coverage and internal consistency. It features a temporal resolution of 6 hours and a spatial resolution of 0.75°, covering the period from January 1979 to September 2019. For the vegetation restoration (DYN) scenario, dynamic land surface data were prepared. The land use and land cover (LULC) data were sourced from the European Space Agency’s Climate Change Initiative (ESA-CCI) dataset (http://www.esa-landcover-cci.org/). Generated through the synergistic processing of multi-source satellite sensor data (e.g., MERIS, AVHRR), this product provides a long-term, high-accuracy, and spatially consistent LULC record. It has an annual temporal resolution and a spatial resolution of 300 m, spanning from 1992 to 2020. Vegetation characteristic parameters were obtained from the Global LAnd Surface Satellite (GLASS) dataset provided by the National Earth System Science Data Center (https://www.geodata.cn/thematicView/GLASS.html?guid2=52092346086099). This product derives long-term, high-quality, and spatially continuous remote sensing retrievals by combining multi-source satellite data with ground-based measurements through advanced retrieval algorithms. It features an 8-day temporal resolution and a spatial resolution of 0.05°, covering the period from 1982 to 2018. For the control (CTL) scenario without vegetation restoration, static land surface data native to the WRF modeling system were utilized. Both the LULC data and the vegetation parameters employ the default static geographical datasets provided with WRF. These datasets, originally released by institutions such as the United States Geological Survey (USGS), represent the global land surface conditions during 1992–1993. They are thus suitable for characterizing the pre-restoration state of the regional surface in this study (https://www2.mmm.ucar.edu/wrf/users/download/get_sources_wps_geog_V3.html).", "ds_quality": "
The simulation results from the Weather Research and Forecasting (WRF) model were evaluated against the China Meteorological Forcing Dataset (CMFD) using three key metrics: the spatial correlation coefficient (PCC), root mean square error (RMSE), and bias. The evaluation reveals that the spatial distribution of WRF-simulated 2-meter air temperature (T2) strongly agrees with the CMFD reference data, as evidenced by a PCC exceeding 0.94, a spatially averaged RMSE below 1.17 °C, and biases predominantly confined to the interval [−1 °C, 1 °C]. Furthermore, an assessment of the regional-average annual time series and intra-annual distribution of simulated temperatures demonstrates a high degree of temporal consistency between the WRF outputs and the CMFD data. These collective results confirm that the WRF-simulated T2 data exhibit good applicability over the Loess Plateau region, thereby providing a reliable basis for assessing the hydroclimatic effects of vegetation restoration projects in this area.", "ds_ref_way": "", "ds_abstract": "
Accurately characterizing the spatiotemporal distribution of surface air temperature at 2 m height (T2) is a fundamental prerequisite for regional climate change analysis. However, observed T2 data inherently encompass the combined influences of natural climate variability and human activities on regional hydroclimatic conditions. This is particularly significant in regions undergoing large-scale vegetation restoration, such as the Loess Plateau, where the quantitative contribution of human interventions—primarily through ecological engineering projects—to regional climate change remains insufficiently quantified. To elucidate the hydroclimatic effects of vegetation restoration, this study employed the Weather Research and Forecasting (WRF) regional climate model. The model simulations were driven by ERA-Interim reanalysis data, with land surface conditions prescribed using two distinct datasets: one incorporating dynamically varying surface parameters and another maintaining static parameters. Long-term, high-resolution climate simulations were conducted over the Loess Plateau under two contrasting scenarios: vegetation restoration (DYN) and a control without restoration (CTL). This framework produced two corresponding datasets of daily mean T2 for the period 1999–2018 at a 10 km spatial resolution. Comparative validation with the China Meteorological Forcing Dataset (CMFD) indicated that the WRF simulations under both scenarios achieved a spatial correlation coefficient (PCC) greater than 0.94, a root mean square error (RMSE) below 1.17 °C, and biases predominantly within the range of −1 °C to 1 °C. These metrics confirm the good applicability of the simulated data over the Loess Plateau. Consequently, these datasets provide a reliable and robust foundation for the quantitative assessment of the hydroclimatic effects induced by vegetation restoration on the Loess Plateau.\r\n
The dataset is stored in NetCDF format (*.nc) and comprises two data files. The naming convention for the data files is T2day_WRF_TYPE_Daily_19990101_20181231_LoessPlateauRegion.nc. Within this format, TYPE represents the scenario names, where DYN denotes the scenario with vegetation restoration, and CTL denotes the control scenario without vegetation restoration. Each file contains the primary variable T2day, which represents the WRF-simulated daily mean air temperature at 2 meters. The temporal and spatial resolutions of the data are daily and 10 km, respectively, covering the period from 1999 to 2018. The auxiliary variables XLAT and XLONG provide the latitude and longitude coordinates, respectively, for each model grid point.", "ds_time_res": "", "ds_acq_place": "Loess Plateau", "ds_space_res": "", "ds_projection": "", "ds_process_way": "
(1) Employing the WRF Preprocessing System (WPS) to preprocess the ESA-CCI land use data and GLASS vegetation parameter data and generate dynamic land surface forcing data, involving replacing the values for leaf area index (LAI), fractional vegetation cover (FVC), surface albedo, and land use categories within the Loess Plateau region to ensure consistency with the model’s configured spatial resolution. (2) Using the WPS to prepare static land surface forcing data, ensuring it alignment with the model’s configured resolution. (3) Utilizing the WPS to preprocess the ERA-Interim reanalysis data and generate meteorological forcing data over the simulation period, which was then resampled to match the model’s spatial resolution. (4) Based on the pre-optimized combination of physical parameterization schemes, running the WRF model with the data from steps (1), (2), and (3) to conduct regional hydroclimate simulations over the Loess Plateau under scenarios with and without the implementation of vegetation restoration. (5) Exporting the simulation outputs from both scenarios and subsequently comparing them with gridded data from the China Meteorological Forcing Dataset (CMFD) to assess the accuracy of the simulation results.", "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, "ds_topic_tags": [ "WRF模拟", "气温", "长时序", "高分辨率" ], "ds_subject_tags": [ "大气科学" ], "ds_class_tags": [], "ds_locus_tags": [ "中国", "黄土高原" ], "ds_time_tags": [ 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018 ], "ds_contributors": [ { "true_name": "张宝庆", "email": "baoqzhang@lzu.edu.cn", "work_for": "兰州大学", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "张宝庆", "email": "baoqzhang@lzu.edu.cn", "work_for": "兰州大学", "country": "中国" } ], "ds_managers": [ { "true_name": "张宝庆", "email": "baoqzhang@lzu.edu.cn", "work_for": "兰州大学", "country": "中国" } ], "category": "气象" }