Scholarlink

<p>Groundwater level monitoring data is essential for the sustainable management and planning of aquifers. However, the limited number of monitoring stations leads to data gaps in groundwater level observations. Traditional interpolation methods, such as Kriging and Cokriging, typically focus on analyzing only the spatial or temporal dimensions when filling data gaps, without fully considering the potential impacts of other related variables on groundwater dynamics, resulting in the loss of valuable information. To address this issue, this study employs the Spatio-temporal CoKriging interpolation method, making full use of known sample point information and considering the spatial, temporal, and spatio-temporal cross-variogram structures within and between different datasets. This allows for the generation of continuous predicted trend surfaces that reflect the spatio-temporal distribution characteristics of the primary variable, achieving accurate estimation from “point” to “surface.” The study selects the Datong Basin in Shanxi as a case area, using groundwater level observation data from 73 monitoring wells (2018–2022) as primary variable data, with precipitation data from the same spatio-temporal domain as a co-variable. The spatio-temporal distribution characteristics of groundwater levels in the study area were interpolated and estimated. Cross-validation results showed that the mean squared error of Spatio-temporal CoKriging was reduced by 21.11% compared to Spatio-temporal Kriging, by 35.06% compared to Cokriging, and by 44.56% compared to Kriging. Furthermore, the correlation coefficient between the estimated and observed values is 11.22% higher than that of spatio-temporal Kriging, 22.58% higher than that of Cokriging, and 30.06% higher than that of Kriging. Accuracy assessments and validation demonstrate that Spatio-temporal CoKriging effectively fills the data gaps caused by insufficient monitoring stations, showing significant advantages and reliability in predicting the spatio-temporal distribution of groundwater levels.</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

Spatio-temporal CoKriging approach for groundwater level interpolation estimation — a case study of the Datong Basin, Shanxi province

  • 期刊论文

  • 发表日期：

    2025年7月17日

查看全文
  • Hongyue Zhang,
  • Xiaoping Rui,
  • Xiting Zhao,
  • Wen Sun,
  • Yiheng Xie,
  • Yingchao Ren

摘要

Groundwater level monitoring data is essential for the sustainable management and planning of aquifers. However, the limited number of monitoring stations leads to data gaps in groundwater level observations. Traditional interpolation methods, such as Kriging and Cokriging, typically focus on analyzing only the spatial or temporal dimensions when filling data gaps, without fully considering the potential impacts of other related variables on groundwater dynamics, resulting in the loss of valuable information. To address this issue, this study employs the Spatio-temporal CoKriging interpolation method, making full use of known sample point information and considering the spatial, temporal, and spatio-temporal cross-variogram structures within and between different datasets. This allows for the generation of continuous predicted trend surfaces that reflect the spatio-temporal distribution characteristics of the primary variable, achieving accurate estimation from “point” to “surface.” The study selects the Datong Basin in Shanxi as a case area, using groundwater level observation data from 73 monitoring wells (2018–2022) as primary variable data, with precipitation data from the same spatio-temporal domain as a co-variable. The spatio-temporal distribution characteristics of groundwater levels in the study area were interpolated and estimated. Cross-validation results showed that the mean squared error of Spatio-temporal CoKriging was reduced by 21.11% compared to Spatio-temporal Kriging, by 35.06% compared to Cokriging, and by 44.56% compared to Kriging. Furthermore, the correlation coefficient between the estimated and observed values is 11.22% higher than that of spatio-temporal Kriging, 22.58% higher than that of Cokriging, and 30.06% higher than that of Kriging. Accuracy assessments and validation demonstrate that Spatio-temporal CoKriging effectively fills the data gaps caused by insufficient monitoring stations, showing significant advantages and reliability in predicting the spatio-temporal distribution of groundwater levels.