Answers are grounded in this paper and cite their sources.
This analysis compares polynomial regression and spatial interpolation methods in estimating porosity values, highlighting differences in accuracy.
Two reservoir porosity datasets are mapped using the Polynomial Regression (PR; zonal estimation), Inverse Distance Weighting (IDW) and Ordinary Kriging (OK; both are interpolations). Data were collected from sandstone reservoirs “K” (19 samples) and “L” (25 samples), of the Lower Pontian age, and located in the Sava Depression, Croatia. Maps were compared via cross-validation (mean square error) and visual inspection of key features. Datasets are considered as small (“K” reservoir) and large (“L” reservoir). In the “K” reservoir, the MSE values are 0.000119 for IDW vs. 0.05401 for PR. In the “L” reservoir the MSE are 0.000676 (OK) vs. 0.040117 (PR). Zonal estimation obviously did not prove as primary mapping in the sets with about 20 datasets. A linear interpolator like the IDW or OK are much better choices, especially if a spatial model can be reliably modelled, like in this case for porosity characterised with normal distribution, which favoured the OK. However, zonal estimation can be a useful addon in interpretation, especially in zones where interpolation is considering less reliable, like in transitional areas.
No takes yet. Share an insight, caveat, or question.
Malvić et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: