Scholarlink

<p>Properly calibrated hydrological models are crucial for regional runoff simulation and flood forecasting. However, the limited availability of basic hydrometeorological observation data in many catchments makes it difficult to achieve a reliable and robust calibration process. Thus, a parameter calibration method for calibrating the weather research and forecasting (WRF)-Hydro model using satellite-based soil moisture (SM) and multisource meteorological forcing data is proposed. In this method, the optimal parameter schemes are determined by combining SM and model calibration decision-making basis and three forcing scenarios are established to drive the WRF-Hydro model for runoff simulation. The corresponding station data over the period of 2020–2023 in the Yuehe catchment of China are selected to verify the proposed method. Statistical results show the good performance of group combining China Meteorological Administration Land Data Assimilation System (CLDAS) precipitation and WRF three-dimensional variational (3DVar) assimilation results for model calibration with Kling–Gupta efficiency (KGE) mean improvement rates of 53.32% and 81.61%, respectively, in comparison with groups using CLDAS meteorological variables alone or Multi-Sources Weighted Ensemble Precipitation (MSWEP) products. Meanwhile, the improvement effect of meteorological data is greatly affected by flood peak levels and discharge magnitudes, and the higher the level or magnitude, the better the effect of reducing the error. In addition, the proposed calibration method is applied to Group 2 (precipitation and remaining meteorological data derived from CLDAS) with more flood events. Moreover, the evaluation indicators of floods in the validation period are basically qualified with the mean values of the Nash–Sutcliffe efficiency (NSE, 0.54), KGE (0.61) and determination coefficient (R<sup>2</sup>, 0.77), further verifying its applicability of simulating runoff in ungauged catchments. These results indicate the good application prospect of the parameter calibration method proposed in this study for forecasting floods and managing water resources in ungauged areas.</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

Novel Parameter Calibration Method of WRF-Hydro in Ungauged Areas Combining Satellite-Based Soil Moisture and Multisource Meteorological Data

  • 期刊论文

  • 发表日期：

    2025年9月4日

查看全文
  • Qingzhi Zhao,
  • Yatong Li,
  • Hongwu Guo,
  • Zufeng Li,
  • Yibin Yao,
  • Mingxian Hu,
  • Pengfei Geng,
  • Yuan Zhai,
  • Xiaohua Fu,
  • Qiong Wu

摘要

Properly calibrated hydrological models are crucial for regional runoff simulation and flood forecasting. However, the limited availability of basic hydrometeorological observation data in many catchments makes it difficult to achieve a reliable and robust calibration process. Thus, a parameter calibration method for calibrating the weather research and forecasting (WRF)-Hydro model using satellite-based soil moisture (SM) and multisource meteorological forcing data is proposed. In this method, the optimal parameter schemes are determined by combining SM and model calibration decision-making basis and three forcing scenarios are established to drive the WRF-Hydro model for runoff simulation. The corresponding station data over the period of 2020–2023 in the Yuehe catchment of China are selected to verify the proposed method. Statistical results show the good performance of group combining China Meteorological Administration Land Data Assimilation System (CLDAS) precipitation and WRF three-dimensional variational (3DVar) assimilation results for model calibration with Kling–Gupta efficiency (KGE) mean improvement rates of 53.32% and 81.61%, respectively, in comparison with groups using CLDAS meteorological variables alone or Multi-Sources Weighted Ensemble Precipitation (MSWEP) products. Meanwhile, the improvement effect of meteorological data is greatly affected by flood peak levels and discharge magnitudes, and the higher the level or magnitude, the better the effect of reducing the error. In addition, the proposed calibration method is applied to Group 2 (precipitation and remaining meteorological data derived from CLDAS) with more flood events. Moreover, the evaluation indicators of floods in the validation period are basically qualified with the mean values of the Nash–Sutcliffe efficiency (NSE, 0.54), KGE (0.61) and determination coefficient (R2, 0.77), further verifying its applicability of simulating runoff in ungauged catchments. These results indicate the good application prospect of the parameter calibration method proposed in this study for forecasting floods and managing water resources in ungauged areas.