ObjectivesGeostatistics combined with geographical information system (GIS) technique is thought well for evaluation of the spatial structure of data and the impacting factors causing the variation. In this paper, this method was used to analyze the spatial variability and main impacting factors of available soil Fe, Mn, Cu and Zn in the permanently cultivated cropland in Tonglu County, in order to provide base for the efficient application of micronutrient fertilizer.

MethodsSoil types, land-use types, organic matter, pH, available N, available P, available K and altitude were chosen as impacting factors in the research. There were totall 386 soil samples collected from paddy fields, dry lands, tea plantations, and orchard fields after crop harvest. Semivariance analysis and Kriging interpolation were performed by GS+7.0 and ArcGIS 10.1.

ResultsThe variation coefficient of four elements was between 58.37% and 90.22%. The nugget effect value was between 10.9% and 12.5%. The spatial distribution of Fe, Mn, Cu and Zn contents was patchy and their space structure had little similarity. The contents of Fe and Zn were significantly different among different soil types, and the availabale Fe, Mn, Cu and Zn contents were significantly different among the land-use types. Fe content has extremely significant and negative correlation with soil organic matter content and available K content, and had extremely significant positive correlation with Cu content; Cu content had extremely significant positive correlation with available P, and had significant negative correlation with available K and significant positive correlation with Mn; Zn content had highly significant negative correlation with available N, highly significant positive correlation with available P and K, had significant positive correlation with Mn and had significant negative correlation with altitude. Stepwise regression analysis indicated that soil type, land-use types, organic matter content, pH, available N, available P, available K and elevation together could explain 19.1%, 2.2%, 12.2% and 12.1% of the variability of soil available Fe, Mn, Cu and Zn in the study area. The land-use types alone could explain 3.1%-13.5% of the variability.

ConclusionsThe available Fe, Mn, Cu and Zn contents were generally at the rich level, and belong to medium spatial variability in the studied area. Structural factors played a dominating role and human activity factors played a secondary role. The contributions of soil type, land-use type, organic matter content, pH, available N, available P, available K and altitude were significant but only accounted for a small proportion of the variation. Among these factors land-use types play a major role, although not the dominating factor.