Soil properties and their spatial pattern in an oasis on the lower reaches of the Tarim River, northwest China

In this study, classical, geostatistical methods and a geographical information system have been used to identify soil properties including soil organic carbon (SOC), total nitrogen (total N), salt content (SC) and soil moisture (SM), and their spatial variation as well as their relationships with groundwater, land use, and soil texture. The data came from 36 soil samples from 0 to 20 cm depth, and 36 groundwater samples in the Tikanlik Oasis on the lower reaches of the Tarim River, northwest China. The objective was to provide a scientific basis for understanding the heterogeneity of the spatial distribution of soil properties on a large scale with the goal of accelerating the sustainable development of agriculture in the oasis. The results showed that the SC and SM of the surface soil varied greatly, but the SOC and total N varied very little. A possible reason for the low variability of SOC and total N was the relatively consistent human activities within the typical agricultural oasis with more than 75% land used as cropland and orchards, because there was no significant effect of land use types on soil total N and SOC. Geostatistical analysis and the kriging estimator showed that SC and SM had a strong spatial autocorrelation. One-way ANOVA and LSD suggested that land use was the main structural indicator that resulted in the spatial autocorrelation of SC and SM within the 0–20 cm soil depth through its significant differences in soil texture, irrigation and groundwater. In particular, croplands and orchards had significantly higher SM, silt and clay particle percentages and lower SC compared to other land uses due to regular irrigation. Additionally, irrigation could effectively relieve the degree of the hazard from soil salt and soil water stress in cropland.

Research highlights▶ Land use types cause the low variability in surface soil SOC and total N. ▶ The SC and SM within the 0–20 cm soil depth have a spatial autocorrelation. ▶ Land use types is the main structural indicator influencing the spatial dependence of SC at 0–20 cm soil. ▶ Spatial variability of soil texture, groundwater depth and TDS result from the land use types.