Remote Sensing Technology and Application, Volume. 40, Issue 3, 671(2025)

Downscaling of Groundwater Storage in the Yellow River Basin based on Random Forest

Mengran LIU1, Yanping CAO2,3、*, Shaokun WANG2, and Yingjun PANG4
Author Affiliations
  • 1College of Mathematics and Statistics, Henan University, Kaifeng475004, China
  • 2Faculty of Geographical Science and Engineering, Henan University, Zhengzhou450046, China
  • 3Key Laboratory of Geospatial Technology of the Middle and Lower Yellow River Region, Ministry of Education, Henan University, Kaifeng475004,China
  • 4Institute of Ecological Conservation and Restoration, Chinese Academy of Forestry, Beijing100091, China
  • show less
    Figures & Tables(10)
    Location and elevation distribution map of Yellow River Basin
    VIP values of characteristic factor for typical months
    Time series of 0.25° GRACE monthly groundwater
    Box plot of difference between GRACE groundwater and simulating groundwater with downscaling groundwater
    Spatial distribution of 0.25°GRACE groundwater
    Correlation coefficients between measured groundwater, downscaling groundwater and GRACE groundwater, respectively
    Differences between the correlation coefficients between measured groundwater and those derived from downscaling techniques and GRACE-observed
    • Table 1. Data source and resolution

      View table
      View in Article

      Table 1. Data source and resolution

      数据来源空间分辨率/km
      降水国家青藏高原科学数据中心1
      气温国家青藏高原科学数据中心1
      蒸散发国家青藏高原科学数据中心1
      土壤湿度国家青藏高原科学数据中心1
      地表温度MOD11A21
      归一化植被指数MOD13A31
    • Table 2. Important parameters of random forest regression model

      View table
      View in Article

      Table 2. Important parameters of random forest regression model

      参数参数调整方向及其意义
      n_estimators随机森林中树的个数,默认为100
      min_samples_leaf

      在叶节点处需要的最小样本数,默认为2(最高复杂度);

      样本数越多,模型越简单

      min_samples_split

      拆分内部节点所需要的最少样本数,默认为1(最高复杂度);

      样本数越多,模型越简单

      max_depth

      树的最大深度,默认为None(最复杂);

      深度越低,模型越简单

      random_state随机种子,控制随机性
    • Table 3. Subsets of characteristic factors from January to December

      View table
      View in Article

      Table 3. Subsets of characteristic factors from January to December

      月份特征因子集
      1月蒸散发、地表温度、降水、归一化植被指数、土壤湿度
      2月蒸散发、地表温度、土壤湿度、降水、归一化植被指数
      3月蒸散发、地表温度、降水、土壤湿度、归一化植被指数
      4月蒸散发、地表温度、气温、土壤湿度、降水
      5月蒸散发、地表温度、气温、归一化植被指数、土壤湿度
      6月地表温度、蒸散发、降水、气温、土壤湿度
      7月地表温度、蒸散发、土壤湿度、气温、降水

      8月

      9月

      10月

      11月

      12月

      蒸散发、地表温度、降水、气温、归一化植被指数

      蒸散发、地表温度、土壤湿度、降水、归一化植被指数

      蒸散发、地表温度、土壤湿度、归一化植被指数、降水

      蒸散发、地表温度、气温、降水、归一化植被指数

      蒸散发、地表温度、降水、土壤湿度、归一化植被指数

    Tools

    Get Citation

    Copy Citation Text

    Mengran LIU, Yanping CAO, Shaokun WANG, Yingjun PANG. Downscaling of Groundwater Storage in the Yellow River Basin based on Random Forest[J]. Remote Sensing Technology and Application, 2025, 40(3): 671

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Feb. 16, 2024

    Accepted: --

    Published Online: Sep. 28, 2025

    The Author Email: Yanping CAO (caoyp@henu.edu.cn)

    DOI:10.11873/j.issn.1004-0323.2025.3.0671

    Topics