地表土壤水分（SM）对荒漠化地区的水文过程和陆地生态系统起着至关重要的作用。土壤水分主动被动（SMAP）卫星等被动微波遥感产品已被证明能很好地监测地表土壤水分。然而，这些产品的空间分辨率较低，缺乏全面覆盖，极大地限制了它们在荒漠化地区的应用。为了克服这些局限性，我们结合多种机器学习方法，包括多元线性回归（MLR）、支持向量回归（SVR）、人工神经网络（ANN）、随机森林（RF）和极端梯度提升（XGB），对 36 千米的 SMAP SM 产品进行了降维处理，并根据植被指数和地表温度等相关地表变量生成了空间分辨率更高的 SM 数据。研究选取了对 SM 敏感的华北荒漠化地区作为研究区域，并制作了 2015 至 2020 年分辨率为 1 km 的日尺度降尺度 SM。", "ds_source": "
（1）本文使用了MODIS产品MOD09A1、MOD11A1、MOD13A2、MOD15A2H和MCD43D58（表2）。每天 1 公里的 LST 由 MOD11A1 提供，1 公里的 16 天 EVI 和 NDVI 由 MOD13A2 提供。MOD15A2H提供了空间分辨率为500 m的8d叶面积指数（LAI），MCD43D58提供了空间分辨率为30角秒（∼1000 m）的每日反照率数据。一些与土壤湿度相关的指数，包括NDWI、NSDSI和地表水指数（LSWI），是由MOD09A1制作的。\n
（2）地形因素与SM密切相关，包括海拔、坡度和坡向，从陆地过程分布式活动档案中心 （LP DAAC） 获得\n
（3）本研究使用的1000 m分辨率土壤数据，包括沙子、淤泥和粘土的比例，使用了从国家青藏高原数据中心获得的中国土壤特征数据集（CSCD）\n
（4）原位SM测量值是从Maqu监测网络和Babao监测网络提供的数据中收集的。\n
（5）日降水量和气温数据来自中国气象数据服务中心的131个气象站。", "ds_process_way": "
（1）根据选定的变量指标（主要包括地形数据、土壤数据和一些MODIS产品）和机器学习方法，构建了一个基于多种机器学习方法的SMAP SM降尺度框架，选择了目前广泛用于构建SM及其相关变量回归模型的机器学习方法；\n
（2）首先，所有数据都需要进行预处理。每日 LST 数据可能会受到云的影响，因此我们使用其质量控制 （QC） 波段对MOD11A1产品进行质量控制，并选择高质量的无云像素。所有选定的变量，包括LST、反照率、LAI、NDWI、LSWI、NSDSI、NDVI、EVI、DEM、坡度、坡向、沙子、淤泥和粘土，都以GeoTIFF格式聚合为1 km的分辨率。使用最近邻插值法进一步对这些变量进行重采样，以达到 SMAP SM 数据的空间分辨率 （36 km）；\n
（3）其次，获取有效样本并进行拆分；\n
（4）第三，基于训练集和测试集确定回归模型。考虑到样本数量对回归模型的准确性至关重要，我们只选择了样本超过 100 个的时期来构建模型；\n
（5）最后，对超参数进行转动，并选择最优模型。", "ds_quality": "
结果表明，与原地观测的 SM 数据相比，其性能良好，平均无偏均方根误差值为 0.057 m3 m-3。此外，其时间序列与降水量一致，性能优于常见的网格化 SM 产品。这些数据可用于评估土壤干旱情况，并为扭转研究地区的荒漠化提供参考。", "ds_acq_start_time": "2015-01-01 00:00:00", "ds_acq_end_time": "2020-12-31 00:00:00", "ds_acq_place": "华北荒漠区", "ds_acq_lon_east": 120.0, "ds_acq_lat_south": 30.0, "ds_acq_lon_west": 70.0, "ds_acq_lat_north": 50.0, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "open-access", "ds_total_size": 6818320970, "ds_files_count": 7, "ds_format": "tif", "ds_space_res": "1km", "ds_time_res": "年", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "7fb1dd58-ab1c-40aa-b18e-05e9d8867c9a.png", "ds_thumb_from": 0, "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": "0931-4967596", "ds_serv_mail": "ncdc@lzb.ac.cn", "doi_value": "", "subject_codes": [ "170.4510" ], "quality_level": 3, "publish_time": "2024-07-19 10:32:45", "last_updated": "2025-05-29 11:33:08", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.NCDC.FIGSHARE.DB6535.2024", "i18n": { "en": { "title": "Daily soil moisture mapping at 1 km resolution based on SMAP data for desertification areas in northern China", "ds_format": "TIFF", "ds_source": "
(1) This article uses MODIS products MOD09A1, MOD11A1, MOD13A2, MOD15A2H, and MCD43D58 (Table 2). 1 kilometer of LST per day is provided by MOD11A1, and 1 kilometer of 16 day EVI and NDVI is provided by MOD13A2. MOD15A2H provides 8d Leaf Area Index (LAI) with a spatial resolution of 500 m, while MCD43D58 provides daily albedo data with a spatial resolution of 30 arcseconds (∼ 1000 m). Some indices related to soil moisture, including NDWI, NSDSI, and Surface Water Index (LSWI), were produced by MOD09A1.\n
(2) Terrain factors are closely related to SM, including altitude, slope, and aspect, obtained from the Land Process Distributed Activity Archive Center (LP DAAC)\n
(3) The 1000 meter resolution soil data used in this study, including the proportions of sand, silt, and clay, was obtained from the China Soil Characteristics Dataset (CSCD) obtained from the National Qinghai Tibet Plateau Data Center\n
(4) The in-situ SM measurement values were collected from the data provided by the Maqu monitoring network and Babao monitoring network.\n
(5) The daily precipitation and temperature data come from 131 meteorological stations of the China Meteorological Data Service Center.", "ds_quality": "
The results indicate that compared with the SM data observed in situ, its performance is good, with an average unbiased root mean square error value of 0.057 m3m-3. In addition, its time series is consistent with precipitation, and its performance is superior to common grid based SM products. These data can be used to assess soil drought conditions and provide reference for reversing desertification in the study area.", "ds_ref_way": "", "ds_abstract": "
Surface soil moisture (SM) plays a crucial role in hydrological processes and terrestrial ecosystems in desertification areas. Passive microwave remote sensing products such as Soil Moisture Active Passive (SMAP) satellites have been proven to effectively monitor surface soil moisture. However, the low spatial resolution and lack of comprehensive coverage of these products greatly limit their application in desertification areas. To overcome these limitations, we combined various machine learning methods, including multiple linear regression (MLR), support vector regression (SVR), artificial neural network (ANN), random forest (RF), and extreme gradient boosting (XGB), to perform dimensionality reduction on the 36 kilometer SMAP SM product. We also generated higher spatial resolution SM data based on relevant surface variables such as vegetation index and surface temperature. The study selected desertification areas in North China that are sensitive to SM as the research area and produced daily downscaled SM with a resolution of 1 km from 2015 to 2020.
", "ds_time_res": "年", "ds_acq_place": "North China Desert Region", "ds_space_res": "1km", "ds_projection": "", "ds_process_way": "(1) Based on selected variable indicators (mainly including terrain data, soil data, and some MODIS products) and machine learning methods, a SMAP SM downscaling framework was constructed using multiple machine learning methods, and machine learning methods widely used to construct SM and its related variable regression models were selected;\n
(2) Firstly, all data needs to be preprocessed. The daily LST data may be affected by clouds, so we use its quality control (QC) band to perform quality control on MOD11A1 products and select high-quality cloud free pixels. All selected variables, including LST, albedo LAI、NDWI、LSWI、NSDSI、NDVI、EVI、DEM、 Slope, aspect, sand, silt, and clay are all aggregated in GeoTIFF format to a resolution of 1 km. Use nearest neighbor interpolation to further resample these variables to achieve spatial resolution of SMAP SM data (36 km);\n
(3) Secondly, obtain valid samples and split them;\n
(4) Thirdly, determine the regression model based on the training and testing sets. Considering that the sample size is crucial for the accuracy of the regression model, we only selected periods with more than 100 samples to construct the model;\n
(5) Finally, rotate the hyperparameters and select the optimal model.", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 0, "license_type": "CC BY 4.0", "doi_reg_from": "reg_outside", "cstr_reg_from": "reg_outside", "doi_not_reg_reason": null, "cstr_not_reg_reason": null, "ds_topic_tags": [ "地表土壤湿度（SM）", "多元线性回归（MLR）", "支持向量回归（SVR）", "人工神经网络（ANN）", "随机森林（RF）", "土壤湿度主动被动（SMAP）卫星" ], "ds_subject_tags": [ "自然地理学" ], "ds_class_tags": [], "ds_locus_tags": [ "华北荒漠区" ], "ds_time_tags": [ 2015, 2016, 2017, 2018, 2019, 2020 ], "ds_contributors": [ { "true_name": "王芳", "email": "657563390@qq.com", "work_for": "中国水利水电科学研究院流域水循环模拟与调控国家重点实验室", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "王芳", "email": "657563390@qq.com", "work_for": "中国水利水电科学研究院流域水循环模拟与调控国家重点实验室", "country": "中国" } ], "ds_managers": [ { "true_name": "王芳", "email": "657563390@qq.com", "work_for": "中国水利水电科学研究院流域水循环模拟与调控国家重点实验室", "country": "中国" } ], "category": "水土保持" }