Spatial Distribution And Simulation Of Soil Organic Carbon And Total Nitrogen In The West River Basin,Chongzhou

Sequestration and emission of soil organic carbon,nitrogen has important influence on greenhouse gas content and global climate change.Traditional field method was difficult to meet the real-time requirement of soil property content.Rapid,real-time access to information has become necessary for soil nutrient management and environmental protection.Based on 121 sample points in Chongzhou West River region and field investigation together with laboratory analysis,this study carried out analysis of surface soil organic carbon(SOC)and total nitrogen(TN)content under different terrain conditions,combined 3S technology with geostatistics.In this research,remotely sensed data from Landsat8 OLI,SOC,TN content and the related ground parameters(terrain,vegetation index)were integrated to construct models for spatial prediction,using the environmental inversion method of multiple linear regression models.Evaluation test would take the way of index inspection and grid computing,in order to obtain different simulation effect of the spatial distribution for SOC and TN under different terrain conditions.Finally,characteristics and differences of SOC,TN under different influence factors would be discussed based on ArcGIS10.0 spatial analysis.The main results as flow:The average content of SOC and TN content in the region were 17.71 g kg-1 and 1.18 g kg-1,with moderate variation.Under different terrain conditions,they all decreased with increasing altitude.And its content data showed normal distribution,with Gaussian the best fitting model on a regional scale.From the point of C0/C0+C result,except strong variability in hilly areas weak variability in mountain areas for SOC,SOC and TN content showed a moderate degree of variability.Its content in some extent has been influenced by the combined effect of natural factors and human factors.Interpolation by kriging indicated that regional SOC and TN content approached to the highest regions in the central Huaiyuan and southeastern Yuantong,and annularly reduced outward,to reach the lowest regions in northern Jiezi-Sanlang mountainous and southern top hills on the whole.Overall,the middle to lower level is widely distributed both for the SOC and TN content.Distribution of TN content showed similar pattern with the SOC content under different terrain conditions.Correlation analysis under different terrain conditions between SOC,TN content and remotely sensed data,ground factor showed that,there is little changes in types and number of related significant or extremely significant for various environmental variables and SOC,TN content,with the rise in altitude.However,they have focused on their respective suffered in different terrain conditions.Fitting model of SOC and TN content for each terrain zone has been established on this basis,including regression models.The results showed that,regression models introduced remote sensing spectral alone or terrain parameters performed better,except TN content in mountain areas.Prediction effect of regression models introduced combination of remote sensing spectral and terrain parameters,vegetation index showed better performance than regression models of single factors adopted,except TN content in the plains.This explained that,remotely sensed data or ground parameters can be used for modeling prediction of soil property,and a combination of both have some potential in improving model prediction.Spatial simulation of SOC and TN reflects different advantages for geostatistics and environmental inversion methods.Different landforms lead to differences in the method of error.Spatial simulation statistics showed that modeling effect of SOC behaves better under different terrain conditions compared ordinary kriging to environmental inversion.For TN,the 2 methods showed no significance in the plain,while the environment inversion method is manifested better in mountain areas.Distribution pattern of soil elements can be better reflected under 3 different terrain conditions based on the method of spatial distribution.Influence factors analysis showed that differences of SOC and TN content reached significant level under different soil types.SOC and TN content in paddy soil were significantly higher than that of yellow soil and purple soil,whose differences reflects not significant.Under the condition of different soil parent material,SOC content performed Violet ash sediment,Minjiang gray sediments,purple sediments,sand shale residual sediments and second-amassment yellow soil in descending order,while it performed Violet ash sediment,second-amassment yellow soil,Minjiang gray sediments,sand shale residual sediments and purple sediments in descending order for TN content.SOC and TN content were significantly different under different agricultural land use patterns.Among them,average of SOC and TN content for the rice-wheat rotation was significantly higher than that of other land use types,followed by rice-rapeseed rotation mode.SOC and TN content for horticultural crops,vegetables and corn-rape rotation is insignificant,while the minimum content is woodland soil.Buffer analysis showed that,SOC and TN content both showed a trend of decrease after increase with increasing distance from the township.SOC content showed a trend of decreasing,while TN content increased after decreased with increase of main road buffer distance.SOC content also showed a trend of decreasing,while TN content in the stabilized with increase of West river buffer distance.Therefore,crop cultivation and fertilization running in daily management for vicinal towns,roads,rivers should be adapted to local conditions.