Using Moran's I and geostatistics to identify spatial patterns of soil nutrients in two different long‐term phosphorus‐application plots

Abstract

The spatial variation of soil nutrients especially the soil test phosphorus (STP) in grassland soils is becoming important because of the use of soil‐nutrients information as a basis for policies such as the recently EU‐introduced Nitrates Directive. Up to now, the small‐scale spatial variation of soil nutrients in grassland has not been studied. The main aim of this study was to investigate the spatial patterns of soil nutrients in two grazed grassland plots with a long‐term (38 y) P‐application experiment, in order to better understand the spatial variation of soil nutrients and the correlation among soil nutrients in grasslands. Two small areas (one from a high‐P background and the other from a medium‐P background) were selected. Soil samples (304 per study area) were collected based on a 1 m × 1 m grid system. The samples were analyzed for STP, Mg, K, pH, and lime requirement (LR). The results were analyzed using conventional statistics, Moran's I, geostatistics, and a GIS.Based on the global Moran's I values, significant positive spatial autocorrelations were found for STP, Mg, pH, and LR in both study areas. Spatial clusters and spatial outliers were detected using the local Moran's I index. Clear linear‐shaped high‐high or low‐low value clusters of the studied variables except K were observed in the study areas due to long‐term usage of machine spreader or other agricultural‐management methods in the past. The corresponding linear patterns were further found in the spatial‐distribution maps. Small spatial patches were found for soil K revealing that it had a random spatial distribution related to the relatively uniform K fertilizer in the study areas. The spatial clusters revealed by local Moran's I were in line with the spatial patterns in the distribution maps.

The spatial distribution pattern of heavy metals and risk assessment of moso bamboo forest soil around lead–zinc mine in Southeastern China

“…Moran's I is a commonly used method to study the overall spatial autocorrelation (Waldhör, 1996;Fu et al, 2011), while local indicators of spatial association (LISA) are applied to identify the degree of spatial autocorrelation at a particular location using local Moran's I (Anselin, 1995). It is also useful to identify local spatial clusters by generating cluster maps (Longley and Tobón, 2004;Harries, 2006) and spatial outliers (Zhang and McGrath, 2004).…”

“…Owing to its ability of explicit and accurate identification of spatial outliers (Anselin, 1995;Zhang et al, 2008;Sugumaran et al, 2009), Moran's I seems to be a popular method for spatial cluster analysis including soil pollution and water quality research (Tu and Xia, 2008). The local Moran's I index examines the individual locations, enabling the identification of pollution hotspots based on a comparison with the neighboring samples (Overmars et al, 2003;Zhang and McGrath, 2004;Fu et al, 2011). The present study involved a combination of geostatistical methods and Moran's I analysis to ensure the accuracy of results and evidence to characterize the heavy metal contamination (Li et al, 2014a,b).…”

Using Moran's I and GIS to study the spatial pattern of forest litter carbon density in a subtropical region of southeastern China

“…On the other hand, the low-high outliers were mainly located closely to the high-high spatial-cluster area. It should be noticed that the local Moran's I index is sensitive to outliers (Fu et al, 2011). A total of 33 spatial outliers were detected.…”

Spatial Variation, Pollution Assessment and Source Identification of Major Nutrients in Surface Sediments of Nansi Lake, China

“…and coefficient of variation (C.V.). A statistical test of the Kolmogorov-Smirnov (K-S) method was applied to evaluate the normality of data sets [13]. A normal distribution is desirable in conventional statistics and linear geo-statistics [14,15], as the high skewness and outliers can endanger the spatial continuity of the variogram function.…”