Spatial autocorrelation and hotspot analysis of natural radionuclides to study sediment transport

Abstract

Tracking sediment movement is typically done with artificial radionuclides. However, this can be environmentally harmful and does not allow for sediment classification. Naturally occurring radionuclides are consequently offered as an alternative. In this study, a mobile Delta Underwater Gamma System (DUGS) capable of measuring low levels of natural radionuclides in sediment was deployed in an estuary, and a radiometric map of the sediment was constructed. Spatial autocorrelation using the Moran's I statistic was used to investigate the spatial distribution patterns of natural radionuclides in the sediments. Hotspot analysis using Getis-Ord* was used to validate and map areas that had been identified as clustered by the Moran's I statistic. The Moran's I analysis indicated that 40K displayed a positive spatial autocorrelation with a value of 0.4 and a standardized Z score of 16, thus indicating that the clustering was significant. 238U and 232Th displayed a low Moran's I value but a strong positive correlation, hence indicating some spots of clustering in the river channel. Further analysis of hotspots confirmed that the identified clusters were areas with relatively high radionuclide concentrations. This proved that the hotspot areas identified have a high deposition of sediment. In situ radiometric measurements of sediment, as well as spatial analysis, are consequently useful tools to model and study spatial structure and sediment.