Study On Inversion Of Soil Alkalization Degree In Qitai Oasis Based On Measured Spectral Data

Soil salinization is four kinds of ecological environment the main problem of present agricultural production worldwide in arid and semi arid regions face?soil erosion,soil erosion and desertification,soil water,soil salinization?of.Soil salinization will not only lead to the development of agriculture is limited,crop production,but also serious damage to the environment and ecosystems,ecological imbalance.Over 30% of the world's land has been plagued by the problem,with an area of more than 1 billion hectares.Therefore,the problem of soil salinization has become one of the common problems faced by all countries in the world.China is one of the influence degree of soil salinization heavier by countries worldwide,there are different degrees of soil salinization more than 20 million hectares,and the soil is mainly distributed in the agricultural irrigation area around the arid and semi arid oasis and oasis,affects the development of local agriculture and animal husbandry.Xinjiang is a national base for grain production and supply,is the largest area in arid and semi arid region,but the development of the agricultural economy has been restricted by soil salinization,therefore,to master the Xinjiang saline soil physicochemical properties,for the extent of salinization and the extent of accurate judgment is imperative,and further reveal the space-time changes of saline soil in Xinjiang area,and on this basis,the establishment of soil salinization improvement mechanisms and policies,so as to better formulate prevention measures.It also has important practical significance to the sustainable and healthy development of Xinjiang's agricultural economy and the development of Xinjiang's ecological environment system.Investigation of soil alkalization condition compared to the traditional sense of the shortcomings of time-consuming laborious,survey results are not accurate and reveal the spatial and temporal distribution of alkaline soil more difficult,the research of monitoring the properties of spectrum and image data acquired by remote sensingand field soil samples collected data on alkali soil,can achieve the ideal effect.The oasis in Xinjiang area of Qitai County as the study area,the soil ion effects on soil alkalization of the angle,using the measured spectral data,image extraction of spectral reflectance data and field sampling data,correlation analysis,extract the sensitive band of ion characteristic bands and soil alkalization index in the visible light range from the measured spectra the reflection characteristic curve,using multiple linear regression modeling method to establish ion characteristic band inversion model and test based on the index of alkalization,by using multiple linear regression,stepwise regression and partial least squares regression two method of three methods,the establishment of alkalizing index inversion model of visible light sensitive bands based on the comparison of different methods for precision influence the inversion model,and the model test,and then select the best alkalinization index inversion model,as the base Monitoring of soil to provide different new ideas.And through the sampling points the reflectance values of SPOT6 high resolution remote sensing image,through 6 kinds of transform?original,reciprocal,logarithmic and logarithmic reciprocal,square,square root?,establish the inversion model of alkalization index,to test the inversion model using different measured reflectance transformation,so as to realize on alkaline soil the scope of monitoring and forecasting using remote sensing technology.The main results and conclusions of this study are as follows:?1?The spectral characteristic curves of the soil in the oasis of Qitai county were generally wave type.The content of each ion was significantly correlated with the alkali index,and the correlation between the measured spectral reflectance curve and Mg²⁺,Ca²⁺ two ions content were not obvious,but the correlation with CO₃^2-was the best.Compared the correlation between pH values and the measured spectral reflectance,the correlation of total alkalinity and TA measured spectral reflectance is better,and the ion content(Mg²⁺,Ca²⁺,and alkalization indicators were excluded)correlation between measured spectral reflectance curve in the red spectral range?near 679nm?reached a peak value.?2?Based on the inversion model of the alkalinity index of the ion characteristic band,the model of the total alkalinity TA has the highest precision,and the R² and R² of the model are respectively 0.703 and 0.851,and the precision of the model is about higher than that of the pH value.The results indicated that the pH value as the alkalinity index was not suitable to reflect the degree of alkalization.?3?Normal distribution index alkalization of the sampling points in the study area,and the coefficient of variation from the weak variation of both the distribution of specific Xeon,which makes the model more universal.Three different methods to establish index alkalization inversion model of optimal accuracy of the model established by multiple linear regression method,the total alkalinity TA modeling precision and accuracy can reach 0.722 and 0.816 respectively,and the relative error of RPD is greater than 2,prove the prediction ability and effect model are better.?4?The correlation between the reciprocal and logarithm of the reflectance and the square root transformation of the SPOT6 images and the measured reflectance of the two types of images is better than that of the original reflectance of the two types,which is improved by about 7%.Compared with the original image correlation between reflectivity and alkalization index,reflectance and the reciprocal of logarithm of reciprocal transformation respectively with the correlation coefficient with the indexes of SAR,RSC and alkaline pH value increased,while the correlation between image logarithm and square root transformation and total alkalinity TA correlation and image original reflectance and TA is almost the same,the rest of the it is obvious decline.The correlation between the logarithm,square root and reciprocal transformation of the measured reflectance and the alkalization indexes is higher than that of the original reflectance and the alkali index?except the correlation between the reflectance reciprocal transformation and TA?.?5?In addition to alkalization index SAR,reflectance inversion model of the other three types of alkalizing index with the results of modeling results,namely the imageto extract the reflectivity index of alkaline TA inversion model established logarithmic transformation,inversion model using reciprocal reflectance transformation established index RSC and pH value of alkalization best in similar alkalization the index,which pH value of the R² model and R² test respectively reached 0.600 and0.577,the inversion is better than the other alkali index,but also reflects the image to reflect the better effect of soil pH value degree of pH.