Spatial Distribution of Stony Desertification and Key Influencing Factors on Different Sampling Scales in Small Karst Watersheds.

Zhenming Zhang,Xianfei Huang,Yunchao Zhou,Shijie Wang

doi:10.3390/ijerph15040743

Zhenming Zhang, Xianfei Huang + Show 2 more

Open Access

https://doi.org/10.3390/ijerph15040743

Copy DOI

Abstract

Karst areas are typical ecologically fragile areas, and stony desertification has become the most serious ecological and economic problems in these areas worldwide as well as a source of disasters and poverty. A reasonable sampling scale is of great importance for research on soil science in karst areas. In this paper, the spatial distribution of stony desertification characteristics and its influencing factors in karst areas are studied at different sampling scales using a grid sampling method based on geographic information system (GIS) technology and geo-statistics. The rock exposure obtained through sampling over a 150 m × 150 m grid in the Houzhai River Basin was utilized as the original data, and five grid scales (300 m × 300 m, 450 m × 450 m, 600 m × 600 m, 750 m × 750 m, and 900 m × 900 m) were used as the subsample sets. The results show that the rock exposure does not vary substantially from one sampling scale to another, while the average values of the five subsamples all fluctuate around the average value of the entire set. As the sampling scale increases, the maximum value and the average value of the rock exposure gradually decrease, and there is a gradual increase in the coefficient of variability. At the scale of 150 m × 150 m, the areas of minor stony desertification, medium stony desertification, and major stony desertification in the Houzhai River Basin are 7.81 km2, 4.50 km2, and 1.87 km2, respectively. The spatial variability of stony desertification at small scales is influenced by many factors, and the variability at medium scales is jointly influenced by gradient, rock content, and rock exposure. At large scales, the spatial variability of stony desertification is mainly influenced by soil thickness and rock content.

Highlights

Soil is a continuum with uneven changes, and soil properties present obvious spatial variability [1].The foundation of soil science research is to obtain detailed and accurate spatial distribution information on soil properties [2]
Multiple stepwise regression and Pearson correlation analysis are used to explore the key influencing factors of stony desertification characteristics on non-sampling scales. These methods are aimed at revealing the determinants of the spatial distribution of stony desertification and the differences among different scales, and the results provide references for the treatment of stony desertification in karst areas
As the sampling scale increases, the maximum and average values of coverage of rock exposures (CRE) gradually decrease while the coefficient of variation increases

Summary

Introduction

Soil is a continuum with uneven changes, and soil properties present obvious spatial variability [1]. The foundation of soil science research is to obtain detailed and accurate spatial distribution information on soil properties [2]. The sampling scale has a decisive influence on the acquisition accuracy and quantitative expression of soil properties and spatial variability information [3]. The narrower the sampling scale, the lower the prediction error of an interpolation. It is difficult to guarantee interpolation accuracy [4]. Excessively high sampling density will require more manpower, material resources, financial resources and long

Methods

Results

Conclusion