| This study investigates the spatial distribution characteristics, autocorrelation, and influencing factors of selenium (Se) in topsoil from the cultivated lands of the Luokouxi small watershed, Fujian Province. Fifty-seven topsoil samples were collected and analyzed using traditional statistics, geostatistics, and spatial autocorrelation analysis. The results show an average soil Se content of 0.34 mg?kg-1, which falls within the sufficient-to-enriched range, with a moderate coefficient of variation (47.99%). Soil Se content was significant positively correlated with the contents of Pb, As, and Cr, as well as with the distance to water systems. Semivariogram analysis indicated moderate spatial autocorrelation (nugget coefficient=48.10%), suggesting that spatial variability was influenced by both structural and random factors. Spatial interpolation via Inverse Distance Weighting (IDW) revealed a relatively high Se content in the middle reaches of the watershed, with lower levels in the upper and lower reaches, exhibiting an overall zonal distribution pattern along the southwest-northeast direction. This pattern is closely associated with soil parent materials, hydrological processes, and agricultural activities near villages. The global spatial autocorrelation index (0.21) indicated significant positive autocorrelation with distinct directional variations, primarily affected by land use patterns and farming intensity. Local spatial autocorrelation analysis identified two primary agglomeration area types: high-high (H-H) agglomeration area, mainly distributed in the southwestern area and correlated with the enrichment of Pb, As, and Cr in soil parent materials; and low-low (L-L) agglomeration area, concentrated in the upper reaches and influenced by hydrological scouring and migration. This study provides a scientific basis for the precise management and safe utilization of local Se-enriched land resources. |
