Modeling the spatial distribution of REEs and Li within two coal ash impoundments

Abstract

Historical and newly acquired data are used to model the spatial variability of rare earth elements (REE), Sc, Y, and Li concentrations in two coal ash impoundments. Samples were collected using a limited sampling scheme and modeled in ArcGIS Pro for total REYSc, Ce, La, Sc, Y, Nd, and Li using 3-dimensional empirical Bayesian Kriging (EBK 3D). Although a higher sample density was modeled at Site 1, the composition of the ash is more variable and randomly distributed throughout the impoundment, resulting in poorly constrained models. At Site 2, the models better represent the spatial distributions of REYSc, Ce, La, Nd, and Li within the coal ash, which vary systematically from east to west, as well as with depth. These results demonstrate that EBK 3D offers a distinct and effective approach for modeling elemental concentrations in coal ash impoundments with limited sampling. Unlike traditional interpolation methods, EBK 3D accounts for spatial uncertainty and variability in three dimensions. It is particularly useful when ash remains undisturbed, the coal source is consistent, or the coal source transitions gradually. This approach introduces a novel methodology for evaluating the resource potential of impounded coal ash as an unconventional feedstock for critical elements.