Spatial modeling of surface water contamination is an important component of environmental monitoring and enables the identification of areas where pollutant concentrations exceed established threshold values. Kriging is one of the most widely used geostatistical methods for constructing continuous spatial models from discrete observations. However, its results are highly dependent on model and spatial search parameters, and inappropriate parameter selection may lead to overestimation or underestimation of the areas where threshold concentrations are exceeded.
The study proposes an approach to calibrating Ordinary Kriging parameters based on the empirical distribution of the input data. Statistical analysis is used to determine the empirical proportion of observations exceeding the established threshold value, which serves as an independent reference indicator. Calibration is performed by comparing this indicator with the proportion of the exceedance area derived from spatial interpolation using different model parameters and by minimizing the deviation between them.
Application of the proposed approach to hydrochemical datasets with different spatial sampling densities demonstrated a significant dependence of the estimated exceedance areas on spatial search parameters. It was shown that parameters can be identified for which the spatial modeling results are most consistent with the empirical characteristics of the input dataset. The proposed approach provides a formalized procedure for Kriging calibration, reduces subjectivity in parameter selection, and improves the reliability of mapping threshold concentration exceedance zones in surface water environmental monitoring.
Dash, P. K., Miller, B. A., Panigrahi, N., & Mishra, A. (2024). Exploring the effect of sampling density on spatial prediction with spatial interpolation of multiple soil nutrients at a regional scale. Land, 13(10), 1615. https://doi.org/10.3390/land13101615
Hodson, T. O. (2022). Root-mean-square error (RMSE) or mean absolute error (MAE): When to use them or not. Geoscientific Model Development, 15(14), 5481–5487. https://doi.org/10.5194/gmd-15-5481-2022
Klypa, A. V., & Karpinskyi, Y. O. (2025). Methodology for geoinformation modeling of microelement distribution in surface waters: Case study of the Poltava Region (Ukraine). Reports on Geodesy and Geoinformatics, 120, 25–30. https://doi.org/10.2478/rgg-2025-0013
Klypa, A., Karpinskyi, Y., & Onyshchuk, V. (2025). Application of the TIN method for GIS-based modeling of barium concentrations in surface waters. International Conference of Young Professionals «GeoTerrace-2025», 1–5. https://doi.org/10.3997/2214-4609.202552017
Rafiee, M., Eslami, A., Saeedi, R., Abtahi, M., & Jahangirirad, M. (2019). Multivariate and geostatistical analysis of spatial distribution and potential sources of heavy metals in surface waters. Caspian Journal of Environmental Sciences, 17(1), 23–41. https://doi.org/10.22124/CJES.2019.3341
Wen, L., Zhang, L., Bai, J., Wang, Y., Wei, Z., & Liu, H. (2022). Optimizing spatial interpolation method and sampling number for predicting cadmium distribution in the largest shallow lake of North China. Chemosphere, 309, 136789. https://doi.org/10.1016/j.chemosphere.2022.136789
Zaresefat, M., Derakhshani, R., & Griffioen, J. (2024). Empirical Bayesian Kriging, a robust method for spatial data interpolation of a large groundwater quality dataset from the Western Netherlands. Water, 16(18), 2581. https://doi.org/10.3390/w16182581
Zhang, H., Lu, L., Liu, Y., & Liu, W. (2015). Spatial sampling strategies for the effect of interpolation accuracy. ISPRS International Journal of Geo-Information, 4(4), 2742–2768. https://doi.org/10.3390/ijgi4042742