Scholarlink

<p>Accurate and continuous monitoring of soil moisture (SM) is crucial for a wide range of applications in agriculture, hydrology, and climate modelling. In this study, we present a novel machine learning (ML) based framework for generating a continuously updated, multilayer global SM dataset: SMRFR (Soil Moisture via Random Forest Regression). Leveraging publicly available reanalysis and remote sensing data, SMRFR provides daily SM estimates at five soil layers (0–5, 5–10, 10–30, 30–50 and 50–100 cm) with a spatial resolution of 9 km, covering the period from 2000 to 2023. Evaluation results demonstrate that SMRFR effectively captures both spatial and temporal SM variability. It also exhibits strong generalization capacity, successfully transferring knowledge across continents and accurately capturing transient and seasonal SM dynamics following rainfall events. SMRFR achieved an unbiased root mean square error of 0.0339 m<sup>3</sup>/m<sup>3</sup> on the validation set. Our novel SM dataset offers a basis and valuable reference for agricultural, hydrological, and ecological research, enabling improved analysis and modelling of SM dynamics at regional to global scales.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

SMRFR: A global multilayer soil moisture dataset generated using Random Forest from multi-source data

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月9日

查看全文
  • Yuhan Liu,
  • Yuanyuan Zha,
  • Gulin Ran,
  • Yonggen Zhang,
  • Liangsheng Shi

摘要

Accurate and continuous monitoring of soil moisture (SM) is crucial for a wide range of applications in agriculture, hydrology, and climate modelling. In this study, we present a novel machine learning (ML) based framework for generating a continuously updated, multilayer global SM dataset: SMRFR (Soil Moisture via Random Forest Regression). Leveraging publicly available reanalysis and remote sensing data, SMRFR provides daily SM estimates at five soil layers (0–5, 5–10, 10–30, 30–50 and 50–100 cm) with a spatial resolution of 9 km, covering the period from 2000 to 2023. Evaluation results demonstrate that SMRFR effectively captures both spatial and temporal SM variability. It also exhibits strong generalization capacity, successfully transferring knowledge across continents and accurately capturing transient and seasonal SM dynamics following rainfall events. SMRFR achieved an unbiased root mean square error of 0.0339 m3/m3 on the validation set. Our novel SM dataset offers a basis and valuable reference for agricultural, hydrological, and ecological research, enabling improved analysis and modelling of SM dynamics at regional to global scales.