Quantitative Relationship Model Between Soil Salinity And Soil Depth In Cotton Field In Southern Xinjiang

Soil salinization will lead to crop can not effectively absorb soil nutrients,prevent the normal growth of crops;In the long run,it will also lead to farmland degradation and restrict the sustainable development of agricultural production.Therefore,it is of great practical significance to study soil salinization and master the dynamic change of soil salt for the prevention and control of soil salinization in cotton fields in Southern Xinjiang.Therefore,how to quickly,conveniently and accurately obtain soil profile salinization information while saving material resources,and obtain the correlation between farmland soil depth and soil profile salt based on this,has gradually become a key difficulty and hotspot of current research at home and abroad.This paper takes the xinjiang cotton as the test area,southern xinjiang machine through the EM38-MK2 earth conductivity meter,collection and research related to apparent conductivity of soil and soil electrical conductivity and related data,using linear regression algorithm,dummy variable method,least squares method,kalman filtering algorithm,salt of the farmland soil profile model to study the relationship with soil depth.The research results of this paper mainly include:(1)For the collection of soil samples,the linear model of apparent conductivity constructed by collecting soil samples under the transmitting coil position of EM38-MK2 geodetic conductivity meter has high accuracy.When soil moisture content is less than 10%,the determination coefficient is 0.72,and the RESE is larger and the model accuracy is the worst.When soil moisture content is higher than 10%,the determination coefficient is more than 0.84,and the RESE is much smaller than that of soil moisture content less than 10%,indicating a good model accuracy.Apparent conductivity in different measurement modes was selected as the modeling factor according to different soil depths.ECh0.5 in horizontal mode could be selected as the modeling factor in the range of 0?40 cm.In the range of 40?80 cm,EC_v0.5in vertical mode or EC_h1.0in horizontal mode can be used as the modeling factor.In this paper,EC_v0.5in vertical mode was used as the modeling factor.At 80?100 cm,EC_v1.0in vertical mode was selected as the modeling factor.The inversion accuracy of local models with different soil depths is higher than that of global models,and the prediction ability is better.(2)In order to further improve the monitoring and prediction of soil salinity profile,a multivariate linear model can be established by taking the measured conductivity data obtained from fixed sampling points in each period as the whole and taking the influence of soil depth and the number of samples as independent variables by using dummy variable processing method and least square method.Then through the discussion of kalman filter algorithm and its variant forms,combined with the ideas and methods of other researchers,kalman filter algorithm is selected to carry out multiple corrections to the model,so that the predicted value is more likely to be close to the measured value.According to the research results,using kalman filter algorithm can improve the correlation between the predicted value and the measured value of the model,and improve the accuracy of the model.The results showed that soil salt accumulation appeared sawtooth fluctuation in cotton field.Although the model accuracy of the four periods is different,it can be seen that soil salinity gradually accumulates with the depth of soil layer.In march and October with high accuracy,it can be concluded that the soil salt content obviously accumulates and reaches the peak value within the depth of 40?80 cm.The measured soil salt content is in good agreement with the simulated value,and the simulated results can reflect the salt migration and accumulation in different soil depths.