Quantitative Estimation Of Soil Salinity And Its Composition Based On Spectra In The Yellow River Delta

The salinity in the Yellow River delta is widely distributed,and the degree is serious.The salinization of soil has caused serious damage to many ecosystems in the area.It is crucial to quantitatively and accurately to detect saline soil for the prevention of soil salinization and agricultural sustainable development.The objective of this study is to find the characteristic spectra of soil salt and its main ions,and build a quantitative analysis model of soil salinization which is suitable for the estuary area of Yellow River to improve the accuracy and stability of the quantitative analysis.Kenli County in Shandong Province was selected as the experimental area.Firstly,96 representative soil samples were collected on October 5th-9th,2014 and the hyper-spectra were measured in situ.After soil sample were dried,the contents of soil salt and ions were analyzed.Then the visible and near infrared reflectance hyper-spectra of the soil samples were measured respectively in the laboratory and in the outdoor,and smoothed and transformed to the first deviation.Secondly,the characteristic spectra which can represent soil salt and main ions(Cl^-,Na⁺,Ca²⁺)were analyzed synthetically and decided using correlation analysis.Finally,the methods of multiple linear regression（MLR）,support vector machine（SVM）and random forest（RF）were used respectively to build quantitative analysis models of soil salinity and ions contents.Then the sensitive bands were selected.Third,from the indoor-spectra,five types of salt spectral indices（SSI）were built including BSI,ratio spectra index（RSI）,normalized differential spectra index（NDSI）,DSI,and spectra index（SI）,were each built from the indoor-spectra,outdoor-spectra and situ-spectra and input into the monitoring models using the random forest（RF）method.The best model and optimal SSI were determined.Finally,the results from the indoor-spectra were applied to the outdoor-spectra and situ-spectra.The best model as well as the optimal SSI from the indoor-spectra was determined.The main conclusions can be drawn as follows:（1）The overall shape and trend of the spectra curves of soil salinity and major ions(Cl^-,Na⁺,Ca²⁺)content in study area were very similar.Soil salinity and major ions(Cl^-,Na⁺,Ca²⁺)spectra response regions were determined to be 1 320¹ 495,1 790¹ 920,2 120² 290 nm.On account of the correlation between the first derivative of the reflectance and the soil salinity and its main ions content,the sensitive spectral regions were 1 490¹ 520,1 890¹ 930 nm;final integrated spectra analysis and correlation analysis the characteristic bands were 1 493,1 801,1 911 and 2 289 nm and the significant characteristic bands were 1 493 and 1 911 nm.（2）The models’ accuracy based on the first deviation of reflectance on the significant characteristic bands matched the models’ accuracy based on the four characteristic bands,which indicates that the significant characteristic spectra were effective for quantitative analysis of soil salt and its main ions.Compared the three modeling methods,the prediction ability of the RF was the best,followed by the SVM,the MLR models’ precision was the lowest.The models using the above-mentioned three methods can used for quantitative analysis of salt,Cl^-and Na⁺,and had good stability and high precision;however,the prediction accuracy of Ca²⁺ contents was still to be improved.（3）In comprehensive comparison and analysis,among the built models,the RF models based on the first deviation of reflectance on the significant characteristic bands（1 493 and 1 911 nm）had higher accuracy and stability for the quantitative analysis of soil salt,Cl^-and Na⁺;it could be applied to the quantitative estimation of Ca²⁺.The significant characteristic spectra（the first deviation of reflectance on 1 493 and 1 911 nm）of soil salt and its main ions were selected synthetically using correlation analysis based on the spectral response.The quantitative estimation models of salt and its main ions was built using the RF regression method,which is suitable for the effective extraction of soil salinization information in the estuary area of Yellow River.（4）The analysis of correlation showed that the sensitive band selection,sensitive bands are mainly concentrated in the near infrared band,sensitive period of indoor-spectra of soil salt and ions were 1493 and 1 911nm;outdoor-spectra were 1536 and 2 037nm;the situ-spectra are 1726 and 2 032 nm spectra.（5）By selection of spectral parameters,the model based on the brightness spectra index（BSI）and difference spectra index（DSI）from the indoor-spectra had the best accuracy;the BSI and DSI was selected as the optimal SSI.（6）From the outdoor-spectra and situ-spectra,the precision was lower than the indoor-spectra,but was adequate for application.Therefore,the BSI or DSI built on the sensitive bands was input into the estimation model using the RF method,which could not only be used for monitoring soil salt and ions with the indoor-spectra but also be used with the outdoor-spectra and situ-spectra.