Abstract
Spatial distributions of micronutrients in soils of Shouguang were evaluated using semivariogram and Moran's index (Moran‘s I) techniques to compare difference and veracity of these two spatial analysis methods. A total of 601 topsoil (0–20 cm) and 155 deep subsoil (150–200 cm) samples were collected on a symmetrical grid in the regional geochemical survey of soils in Shandong Province, and copper (Cu), iron (Fe), manganese (Mn), and zinc (Zn) concentrations were analyzed and compared. The results showed significant spatial correlations of micronutrients in Shouguang soils, and the spatial correlation degree was greater in topsoil than in deep subsoil. In topsoil and deep subsoil, the spatial correlation distance for each element obtained using the semivariogram technique was 20–60 km, whereas with Moran's I technique, the positive autocorrelation distance was 20–25 km and the negative autocorrelation distance was 25–55 km. The spatial autocorrelation degree was significant (P ≤ 0.05) for every micronutrient except deep subsoil Zn. Moran's I technique was able to distinguish between positive and negative autocorrelations and the results of semivariogram analysis gave the sum of the positive and negative autocorrelations. This study shows that Moran's I is more accurate and meaningful than semivariogram analysis for spatial autocorrelation of some soil attributes. These results provide the theoretical foundation for the application of spatial analysis methods, and Moran's I in particular, in environmental research.
Acknowledgments
This work was supported by Shandong Province Higher Educational Science and Technology Program (J10LC62), the Key Laboratory of Coastal Zone Environmental Processes and Ecological Remediation, YICCAS (No. 201201), and the Ecological Geochemical Survey of Shandong Province (No. 200314200023).