| Songnen Plain is one of the major grain producing areas in China and the production base of commodity grain.The western part of China is located in the ecotone between agriculture and animal husbandry.In recent years,with the development of economy and the progress of science and technology,human activities are more and more frequent.The problem of desertification and salinization in the western region is serious.Therefore,to the high spatial resolution of the county as a necessary evolution analysis of soil organic carbon content unit and its spatial heterogeneity,the results can not only provide the basis for land resources and improve soil texture,reasonable use of the county,but also can improve and maintain and increase soil organic carbon content in grain yield with a powerful reference value.The geographical weighted regression model（Geographically Weighted Regression Model,GWR）compared with the traditional regression model,which can geographical things local parameter estimation,can better show the spatial variation,by constructing the operation and inspection of the GWR model,to the counties as a unit to analyze the soil organic carbon content changes and space the variation of driving factors.To better understand and analysis in the Western Songnen Plain farming process and mechanism of changes in soil organic carbon content,which will counties for improving soil conditions,provides a feasible basis for improving soil organic carbon content.In this paper,desertification problems in Western Songnen Plain as the key research area,using the traditional analysis,statistical analysis and GWR model ArcGIS statistics,comparative analysis of 1990 and 2015 of 14 counties in the Western Songnen Plain Farmland Soil Organic Carbon（Soil Organic Carbon,SOC）of the temporal and spatial variability,and to explore the soil pH value.Electrical conductivity,bulk density,ammonium nitrogen,nitrate nitrogen,grain size,average annual temperature,average annual precipitation of organic carbon in the spatial distribution of the dominant influence.The main results are as follows:（1）The distribution of soil organic carbon（SOC）content and its influencing factors in the Western Songnen Plain in 1990:the trend of decreasing gradually from the central part of the city to the South and the increasing trend in the north.The content of 39.94 g kg-1 and 71.71 g kg-1is highest in Longjiang county and Qigihar County in the northwest of China.The content of 43.11-74.59g kg-1 is the lowest in the southeast of Changling County,and the content is 0.92-4.52 g kg-1.The content of soil organic carbon density in1990 was negatively correlated（p<0.01）,a negative trend of the two counties in Qigihar city and Zhaoyuan county was most significant;the content of soil organic carbon and clay and silt fractions showed significant positive correlation（p<0.01）,and sand a significant negative correlation（p<0.01）;organic carbon content and the average annual temperature was negatively related（p<0.01）showed that the higher the temperature,the organic carbon content of less.（2）Correlation between soil organic carbon content distribution in the Western Songnen Plain and its influencing factors in 2015:roughly from central Zhenlai County,Da An City and Qian’an County gradually reduced to the southwest and southeast of the trend,and gradually increase to the northwest.The content in the northwest of Longjiang is the highest,the content of which is 39.94-71.79 g kg-1;the content of kg-1in southwest Tongyu is the lowest,the content is 2.3-3.7 g kg-1.There was a significant negative correlation between soil organic carbon content and pH value（p<0.01）.Through the comparative analysis of the northern pH value trend for neutral and acidic Tailai County,Dumeng County,Qigihar city and Longjiang County of SOC and pH,SOC and pH showed a positive correlation trend,soil acidity environment can make soil microbial activity decreased,reduce the decomposition and migration of soil organic matter and nutrients the elements,so that the soil organic carbon turnover decline,is conducive to the accumulation of SOC;density was negatively correlated（p<0.01）;the silt and sand component with SOC significant correlation among silt fractions showed significant positive correlation（p<0.01）,sand components showed significant negative correlation（p<0.01）;and the ammonium nitrogen and nitrate nitrogen were extremely significant positive correlation（p<0.01）;organic carbon content and the average annual temperature（T）and annual precipitation（P）was negatively related（p<0.01）,table The higher the temperature of organic carbon content less;more organic carbon content less precipitation,organic carbon loss,but smaller than the annual mean temperature and organic carbon to relationship between the number of annual precipitation,annual mean temperature shows that the effects of the Western Songnen Plain on organic carbon than annual precipitation.（3）Using principal component analysis（PCA）and GWR model to analyze the influencing factors of soil organic carbon（SOC）,the results show that 0.05-2 mm sand,T,fertilizer application are the main factors.The spatial autocorrelation test of organic carbon content showed an increasing trend compared with the Moran’s I system in 2015and 1990.The results showed that the spatial autocorrelation of soil organic carbon content was enhanced.The spatial correlation of soil organic carbon content is stronger,which is conducive to the construction of the model.The GWR model is tested to prove its feasibility and advantage.（4）According to the results of GWR model,it is found that the content of sand is negatively correlated with the content of organic carbon.In 1990 is generally enhanced gradually from the southwest to the northeast and northwest of the influence degree,influence degree of clay content is from the 2015 southeast to the northwest,gradually decreased;the influence degree of the annual average temperature mainly showed an increasing trend from the south to the north and has negative correlation with organic carbon content.This is one reason why Longjiang county and Qigihar City,higher organic carbon content;the impact of the changes in 1990 and 2015 fertilization is similar,mainly from the southeast to the northwest,the trend of decline,fertilization amount was positively correlated with organic carbon content;through the use of GWR model to analyze the temporal and spatial changes of farming factors influence on soil organic carbon in Songnen Plain in the west,it was found that the GWR model can well estimate the Western Songnen Plain tillage soil organic carbon content influence factors of regional characteristics. |
PDF Full Text Request