Sampling Design Based On Sample Representaticve Grade And Road Network Information

Soil sampling is an important part in the investigation of soil properties,and it is of great significance to the study of detailed digital soil mapping.Therefore,how to develop soil sampling work efficiently and at low cost has become the focus of academic research.In the past studies,the design of the soil sampling program was mainly to obtain high-precision survey results,lacking in consideration of sampling efficiency and sampling cost,which often led to the occurrence of problems such as an excessive number of sampling sites and difficult sampling work.In recent years,more and more researchers have begun to reduce the number of sample points and optimize the layout of sample points,which to resolve the conflict between the input of sampling resources and the precision requirement of survey.Speculating spatial distribution of soil attributes based on individual sample representativeness.Optimizing the layout of sample points which can take advantage of the road network.In view of this,this study designs a soil sampling scheme with few sample sites,high representativeness and excellent spatial accessibility based on two aspects of sample representativeness and road network optimization,which will provide new opportunities for reducing the sampling difficulty and cost.Based on soil-landscape model theory,clustering analysis theory,and landscape fragmentation theory,this study analyzed the determination of the central study area,the design of the soil sampling scheme,and the representative verification of the samples.Taking the central area of Zhongxiang City,Hubei Province as the research area,extracting appropriate environmental factor data through terrain analysis,this study designed a sampling scheme based on the sample representative level and road network information,then evaluted the representativeness of sample design scheme.The conclusions are as follows:(1)The road network has a low impact on landscape fragmentation in the study area and sample design can be carried out within a reasonable range of roads.This study selected six environmental factors such as elevation,slope,aspect,curvature along the profile,contour curvature,and topographic moisture index as source data,which can be extracted through SoLIM software;introduced Partition coefficient and Normalized entropy determine the optimal number of clusters,based on MATLAB software cluster analysis to obtain the central area of the study area,then complete the preliminary sample design in the center area;using patch density,average plaque area and plaque aggregation degree as indicators,through overlaying the road network quantitative analysis of the impact of the road on the landscape fragmentation of the study area,built a reasonable buffer for sample placement and completed the final sample design.(2)The sampling scheme based on sample representative level and road network has good spatial accessibility.This study first calculated the average distance from the sample point to the road through ArcGIS software,obtained the road sampling cost by analyzing the best sampling path,and then compare the spatial accessibility between the new sampling point and the historical sampling point.The results show that the average distance from the new design sample points to the road is 277 m,and the average distance from the historic sampling point to the road is 369m;the road passing cost when sampling the newly designed sampling point is 78296 m,and the road passing cost when sampling the historical sampling point is 95349 m.Compared with the historical sampling points,the new design samples have obvious advantages in spatial accessibility.The layout optimization of samples has important value for improving the sampling efficiency.(3)The sampling point has a good representation for the spatial distribution of organic matter in the study area.This study first analyzes the representativeness of the newly designed sampling points for the global organic matter distribution in the study area.The results show that the mean,standard deviation,and coefficient of variation of the organic matter are very similar between the old and new sampling points.Afterwards,this study explored the representativeness of the newly designed sampling points in the organic matter content by subdividing the terrain.The results showed that the five representative terrain units in the study area,such as dry land,wet flat,flat peak,slope foot,and depression land,are equally representative.The sampling design scheme based on sample representative level and roadnetwork information proposed in this study effectively solves the problem of excessive sample number,and overcomes the difficulty of low sampler hiking sampling efficiency to a certain extent.It is of guiding significance to improving the sampling efficiency and reducing the sampling difficulty,and it also has important value to reduce costs and expenses for soil science research in the future.