Evaluation of spatial and temporal changes of soil quality based on geostatistical analysis in the hill region of subtropical China

Abstract

Studies on the effect of land use alteration on the spatial variability of soil properties are limited. This study addressed the spatial and temporal variability of soil properties and changes of soil quality in a hill region of subtropical China using geostatistical methods. Soil samples from 0- to 15-cm depth were collected within 105 locations, on a 100×100 m grid basis over a 112-ha field, in 1985 and 1997, respectively. Soil properties showed large variability, with the highest coefficient of variation being observed for available P, the lowest for soil pH. Over the 12-year period, a significant decrease of soil organic matter appeared with original land use patterns of wasteland and paddy field, whereas other properties showed no significant changes. In addition to the alteration of wasteland, fertilization in upland increased available P and K, whereas the opposite appeared in forest restoration system. Changing wasteland into paddy field was found to increase soil fertility. A geostatistical analysis showed that all the soil properties (pH, organic mater, available P and K) and their changes between 1985 and 1997 were spatially structured. The nugget-to-sill ratio indicates a strong spatial dependence for soil pH, and a moderate spatial dependence for other properties. The ranges for soil properties in 1985 were equal to or larger than the diameter of the hills. Changing the land use patterns decreased the ranges for soil chemical properties. The ranges for soil pH and available K were similar to the radius of hills. Interpolation using kriging showed a spatial similarity among the soil properties. Soil properties decreased in the southeast quadrant of the research area where the land use was paddy field, while they increased around the center where arable upland was initially from wasteland. Thus, we propose a process to evaluate soil quality using the geostatistical methods as a potential analysis tool for monitoring changes at a farm scale.