{
    "created": "2025-08-05 18:11:39",
    "updated": "2026-05-09 15:36:13",
    "id": "d77f793a-94c3-428c-a282-b7a82920cba1",
    "version": 2,
    "ds_topic": null,
    "title_cn": "新疆和中亚五国GEDTM120米地形数据",
    "title_en": "",
    "ds_abstract": "<p>基于GEDTM120米地形数据，经过掩膜处理得到的新疆、哈萨克斯坦、塔吉克斯坦、吉尔吉斯斯坦、乌兹别克斯坦和土库曼斯坦6个区域的地形数据，数据分辨率120米。</p>",
    "ds_source": "<p>该数据集是荷兰OpenGeoHub机构的何玉峰团队于2025年2月发布在Zenodo网站上的数据，自今年2月以来已更新了4个版本。该数据集简称为GEDTM30。由于30米数据集目前还未正式出版，数据论文正在Research Square平台的评审之中，数据作者仅开放了120米的DEM地形数据，30米地形数据需等评审完成后才能获取。</p>",
    "ds_process_way": "<p>论文作者提出全球-局部迁移学习框架，生成1arcsec（约30m）分辨率的无空隙全球DEM（GEDTM30），并提取坡度、曲率、汇水面积等15类地表形态与水文参数。研究整合多源异构数据（光学影像、雷达、激光雷达），通过中位数滤波与异常值剔除生成初始DEM，结合ICESat-2（20亿点）和GEDI（10亿点）激光雷达数据构建训练样本，利用GNSS站数据验证模型精度。针对数据稀疏区域（如太平洋岛屿），创新性地采用全球-局部迁移学习策略：全局模型基于10%随机样本训练随机森林（RF）捕捉共性特征，局部模型按5°×5°分块，融入高程异常与植被覆盖等本地化样本微调参数，显著提升区域适应性。地表参数化阶段，通过白盒工具（WhiteboxTools）结合Equi7投影系统实现多尺度参数高效计算（30–960m），优化水文连通性与计算效率，最终生成兼具垂直精度（森林区RMSE23.2m）与区域适应性（沿海误差降低50%）的全球地形基准数据集。</p>",
    "ds_quality": "<p>森林区RMSE23.2m，沿海误差降低50%。</p>",
    "ds_acq_start_time": null,
    "ds_acq_end_time": null,
    "ds_acq_place": "",
    "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": "在线下载",
    "ds_total_size": 39133930707,
    "ds_files_count": 32,
    "ds_format": "",
    "ds_space_res": "120米",
    "ds_time_res": "",
    "ds_coordinate": "WGS84",
    "ds_projection": "GCS_WGS84",
    "ds_thumbnail": "d77f793a-94c3-428c-a282-b7a82920cba1.png",
    "ds_thumb_from": 2,
    "ds_ref_way": "",
    "paper_ref_way": "",
    "ds_ref_instruction": "Ho, Y., & Hengl, T. (2025). Global Ensemble Digital Terrain Model 30m (GEDTM30) (v1.1) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.15689805",
    "ds_from_station": null,
    "organization_id": "a5877b42-96ea-4f13-af7e-246f355413d6",
    "doi_value": "",
    "subject_codes": [
        "170.40",
        "170.45"
    ],
    "quality_level": 1,
    "publish_time": "2025-08-06 16:35:45",
    "first_publish_time": null,
    "last_updated": "2025-08-06 16:35:45",
    "protected": false,
    "protected_to": null,
    "lang": "zh",
    "cstr": null,
    "license": null,
    "extra": null,
    "files_shape": [
        {
            "name": "GEDTM_120m.rar",
            "size": 37658708677,
            "is_dir": false
        },
        {
            "name": "v1_covered_f59649fa-39aa-4fbd-8c86-91f4966f0d34.pdf",
            "size": 25272802,
            "is_dir": false
        },
        {
            "name": "Clip",
            "size": null,
            "is_dir": true
        }
    ],
    "features": null,
    "data_level": 0,
    "ds_topic_tags": [
        "DEM",
        "DTM",
        "随机森林",
        "集成数字地形模型"
    ],
    "ds_subject_tags": [
        "地图学",
        "地理学"
    ],
    "ds_class_tags": [],
    "ds_locus_tags": [
        "全球",
        "中国新疆",
        "哈萨克斯坦",
        "塔吉克斯坦",
        "乌兹别克斯坦",
        "吉尔吉斯斯坦",
        "土库曼斯坦"
    ],
    "ds_time_tags": [
        2006,
        2007,
        2008,
        2009,
        2010,
        2011,
        2012,
        2013,
        2014,
        2015
    ],
    "ds_contributors": [
        "李锦"
    ],
    "ds_meta_authors": [
        "李锦"
    ],
    "ds_managers": [
        "李锦"
    ],
    "category": "DEM数字高程"
}