Analysis Of The Spatial Variability And Evaluation On Soil Biochemical Properties Of Gravel-mulched Land

In the arid area in central Ningxia,the agricultural production was seriously affected because of the water and heat condition.Gravel-sand mulched field is a unique conservation tillage method developed by farmers in northwestern China in the long-term labor practice.It has the functions of heat preservation and permeability improvement and soil physical and chemical properties,which can significantly improve the efficiency of precipitation and increase crop yield.The study on the spatial variability of soil properties in gravel-sand mulched field mostly explained the spatial distribution of soil moisture and salinity,and neglected the comprehensive study on the spatial variability of soil biochemical on mesoscale in the past.Field sampling and indoor analysis were conducted in grave-sand mulch field in Xiangshan area.By the methods of mathematical statistics and geostatistics,the spatial variability and distribution of soil biochemical properties were analyzed in the horizontal directions.RandomForest method,geographically weighted regression Kriging and regression Kriging were used for spatial interpolation of soil properties and compared with ordinary Kriging in this study,and the accuracy of spatial prediction was evaluated by above methods.Examined comprehensive quality index of soil biochemical properties in grave-sand mulch field.Studied on evaluation of soil biochemical properties and its spatial variability can provide scientific basis for agricultural production and ecological restoration in the arid and semi-arid regions of China.In this study.The main conclusions of this study were as follows.(1)The total potassium content have low variability,while others have moderate variability.The geostatistical statistics results showed that the soil nutrients are defined by the Gaussian model,spherical model and exponential models.The nugget to sill ratios indicated that all soil nutrients have moderate dependencies except the total potassium content.The area with high total potassium content is concentrated in the southeast.Soil total phosphorus,total potassium and available nutrients decreased with elevation.(2)The coefficients of variation of soil urease and invertase activities are more than 10%with moderate variability,the hydrogen peroxidase activities have low variability.Geostatistics showed that soil soil urease and invertase activities are best fitted by spherical model,others are best fitted by exponential model.The spatial structure characters of soil invertase were not significant compared to urease and hydrogen peroxidase activities.The soil orgsnic matter and total nitrogen can represent the main limiting factors of soil enzyme activity.(3)Descriptive statistics indicated that soil heavy metals are moderately variable,Cu and Cd were the main variables affecting soil quality in study area.Geostatistical analyses showed that the spatial autocorrelation of all heavy metals are strong except Cu.The ordinary kriging maps can provide useful information for development and application of precision agriculture in grave-sand mulch field.Study on correlation and principal component analysis showed that heavy metals are affected by natural source,traffic source and agricultural activity source.(4)Descriptive statistics showed that the soil cation exchange capacity have moderate variability.Geostatistics showed that both soil cation exchange capacity content and its residuals(including ordinary least squares residual and geographically weighted regression residual)had strong spatial autocorrelation(nugget value was less than 25%).Interpolation accuracies by different methods were estimated based on a data set of validation samples,results showed that the randomForest method significantly improved the accuracy of ordinary Kriging interpolation of soil cation exchange capacity due to their incorporation of auxiliary variables.The map interpolated using randomForest method was revealed more details of soil cation exchange capacity distribution.(5)The minimum data set suitable for use in the soil fertility quality assessment of the oasis is composed of three indices,that is,soil total nitrogen,total phosphorus and available potassium.When the Nemerow comprehensive pollution index of heavy metals was less than 0.485,the comprehensive soil quality improved as the soil fertility quality improved.The comprehensive evaluation of soil quality index based on soil fertility quality and heavy metal pollution can reflect the real characteristics of soil in this research,and the soil quality were low in the grave-sand mulch field of NingXia.