积雪深度是表征陆地水资源储量与评估雪-水文过程的关键参数，而年度最大积雪深度作为其极值指标，不仅反映区域积雪累积能力与峰值水储量状态，也是融雪径流模拟、极端雪灾风险评估的重要基础数据，对冻土稳定性及生态系统过程具有重要影响。然而，受限于现有雪深产品空间分辨率较低且反演不确定性较大，难以刻画复杂地形与多样下垫面条件下积雪分布的空间异质性。\n
本研究基于 AlphaEarth Foundations（AEF）地理空间基础模型，结合地面气象站实测雪深数据，采用梯度提升树(Gradient Boosted Decision Trees , GBDT) 机器学习算法，构建高分辨率年度最大积雪深度反演模型，并首次生成了2017—2023年中国典型积雪区10 m分辨率年度最大积雪深度数据集，数据格式为 GeoTIFF。基于独立站点验证结果表明，该数据集具有良好的精度与空间一致性，相关系数（R）为0.71，均方根误差（RMSE）为6.71 cm，平均绝对误差（MAE）为4.51 cm。该数据集可为区域水资源评估、雪灾评估和气候变化等提供高精度基础数据支撑。", "ds_source": "", "ds_process_way": "", "ds_quality": "", "ds_acq_start_time": "2017-01-01 00:00:00", "ds_acq_end_time": "2023-12-31 00:00:00", "ds_acq_place": "中国典型积雪区", "ds_acq_lon_east": null, "ds_acq_lat_south": null, "ds_acq_lon_west": null, "ds_acq_lat_north": null, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "open-access", "ds_total_size": 786856876661, "ds_files_count": 470, "ds_format": "*.tif", "ds_space_res": "10m", "ds_time_res": "年", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "7fda555c-ca26-4b83-b46b-b825e9b2adc2.png", "ds_thumb_from": 0, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": null, "organization_id": "952adb3f-3ede-4a94-942a-7de772f1bfc5", "ds_serv_man": null, "ds_serv_phone": null, "ds_serv_mail": null, "doi_value": "", "subject_codes": [ "170.15" ], "quality_level": 0, "publish_time": "2026-08-25 08:36:24", "last_updated": "2026-08-26 10:38:41", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.ncdc.snow.db7725.2026", "i18n": { "en": { "title": "A dataset of annual maximum snow depth with 10-meter resolution in typical snow regions of China during 2017–2023", "ds_format": "*.tif", "ds_source": "", "ds_quality": "", "ds_ref_way": "", "ds_abstract": "
Snow depth is a key parameter for quantifying terrestrial water storage and characterizing snow hydrological processes. As an extreme snow metric, Annual Maximum Snow Depth (SDmax) reflects the peak seasonal snow accumulation and maximum snow water storage, providing essential information for snowmelt runoff simulation, snow disaster risk assessment, and studies of permafrost stability and ecosystem dynamics. However, existing snow depth products are generally limited by coarse spatial resolution and considerable retrieval uncertainties, making it difficult to accurately characterize the spatial heterogeneity of snow distribution across complex terrain and diverse land surface conditions. \r\n
In this study, we developed a high-resolution SDmax retrieval model by integrating the AlphaEarth Foundations (AEF) geospatial foundation model with in situ snow depth observations from meteorological stations using the Gradient Boosted Decision Trees (GBDT) model. Based on this framework, we produced the first 10 m Annual Maximum Snow Depth dataset for the major seasonal snow-covered regions of China spanning 2017–2023, with all products provided in GeoTIFF format. Independent validation against ground observations demonstrates that the dataset achieves good accuracy and spatial consistency, with a correlation coefficient (R) of 0.71, a root mean square error (RMSE) of 6.71 cm, and a mean absolute error (MAE) of 4.51 cm. This dataset can provide high-precision foundational data support for regional water resource assessment, snow disaster evaluation, and climate change research.", "ds_time_res": "year", "ds_acq_place": "Typical snow cover areas in China", "ds_space_res": "10m", "ds_projection": "", "ds_share_type": "open-access", "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": "fair", "ds_topic_tags": [ "中国", "2017 - 2023", "AlphaEarth", "最大雪深" ], "ds_subject_tags": [ "大气科学" ], "ds_class_tags": [], "ds_locus_tags": [ "中国典型积雪区" ], "ds_time_tags": [ 2017, 2018, 2019, 2020, 2021, 2022, 2023 ], "ds_contributors": [ { "true_name": "石英杰", "email": "shiyingjie@iga.ac.cn", "work_for": "中国科学院 东北地理与农业生态研究所", "country": "中国" }, { "true_name": "李晓峰", "email": "lixiaofeng@iga.ac.cn", "work_for": "中国科学院东北地理与农业生态研究所", "country": "中国" }, { "true_name": "卫颜霖", "email": "", "work_for": "中国科学院东北地理与农业生态研究所", "country": "中国" }, { "true_name": "潘欣", "email": "panxinpc@163.com", "work_for": "长春工程学院 计算机技术与工程学院", "country": "中国" }, { "true_name": "郑照军", "email": "zhengzj@cma.gov.cn", "work_for": "国家卫星气象中心", "country": "中国" }, { "true_name": "李怡静", "email": "liyijing@iga.ac.cn", "work_for": "中国科学院 东北地理与农业生态研究所", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "石英杰", "email": "shiyingjie@iga.ac.cn", "work_for": "中国科学院 东北地理与农业生态研究所", "country": "中国" }, { "true_name": "李晓峰", "email": "lixiaofeng@iga.ac.cn", "work_for": "中国科学院东北地理与农业生态研究所", "country": "中国" }, { "true_name": "卫颜霖", "email": "", "work_for": "中国科学院东北地理与农业生态研究所", "country": "中国" }, { "true_name": "潘欣", "email": "panxinpc@163.com", "work_for": "长春工程学院 计算机技术与工程学院", "country": "中国" }, { "true_name": "郑照军", "email": "zhengzj@cma.gov.cn", "work_for": "国家卫星气象中心", "country": "中国" }, { "true_name": "李怡静", "email": "liyijing@iga.ac.cn", "work_for": "中国科学院 东北地理与农业生态研究所", "country": "中国" } ], "ds_managers": [ { "true_name": "石英杰", "email": "shiyingjie@iga.ac.cn", "work_for": "中国科学院 东北地理与农业生态研究所", "country": "中国" } ], "category": "积雪" }