Hyper-Spectral Response And Remote Sensing Estimation Model Of Soil Degradation In The Yellow River Delta

Due to the large C hinese population and the increasingly smaller appropriation of per-capita quantity of land resources,natural resources have been used unreasonably for a long time.In particular,the unreasonable use of land resources has caused serious damage to regional ecological environments,consequently causing the severity of soil degradation to increase.Statistics show that the area of soil degradation accounted for more than 40% of the land area in C hina.Soil degradation will destroy the natural landscape and the human living environment,and induce soil damage,drainage atrophy,forest decline a nd climate change in the scope of regional or global.The Yellow River Delta,located on the west coast of the Bohai Sea,is an important land resource reserve in C hina.It has a fragile ecological environment and serious salinization degradation under the impact of the dynamic systems of rivers,land,ocean,and other environmental factors.Soil degradation restricts the sustainable development of the economy and society.It is necessary for preventing and curing soil degradation to know the information of soil degradation,such as the range of soil degradation,the distribution and the degree of degradation.The study area is Kenli County.First,the soil properties were obtained through field investigation,soil sampling and laboratory analysis,and the soil degradation index(SDI)was constructed by the key indicators of soil degradation,such as soil nutrients,salt,p H,texture.Then,to analyze the soil degradation response of field soil spectrum and air-dried soil spectrum obtained by ASD FieldSpec4,which is affected by multiple soil properties;to explore the method of removing the influence of the soil moisture from field soil spectrum;and to establish the hyper-spectral estimation model of SDI based on the field soil spectrum of moisture removal.O n this basis,soil hyper-spectral resampling by taking the average in the narrow-band range which is the same as band range of Landsat8-OLI image,and the SDI estimation model for the corresponding period remote sensing image was established for inversion soil degradation.Finally,the dynamic analysis of soil degradation in Kenli County from 2007 to 2015 was carried out by the spatial interpolation of the measured data.The conclusions are as follows:(1)The SDI of soil samples in the study area was between 0.237 to 0.659.According to the result of cluster analysis,soil degradation was divided into the following 3 grades: SDI<0.40,light degradation;0.40<SDI<0.54,moderate degradation;and 0.54<SDI,heavy degradation.The samples of light degradation,moderate degradation and heavy degradation respectively accounted for 39%,50%,and 11% of the total samples.Based on the analysis of the vegetation types in different degraded grades,it was found that the vegetation cover types of soil degradation grades from severe to mild were as follows: Suaeda salsa,reed,cogongrass,rice and cotton,and wheat.(2)The field soil spectrum is greatly affected by soil moisture,so it is difficult to extract the spectral information of soil degradation.The spectral characteristics of the air-dried soil spectrum were analyzed,it was found that 370~450,500~650 nm and the wavebands of large spectral reflectance in the near-infrared range show a good response to soil degradation;When the wavelength was greater than 1975 nm,there is a stable linear relationship between SDI and reflectance spectra in this range.The spectral direct transform can effectively eliminate the influence of soil moisture on the field soil spectrum.After the removal the influence of soil moisture,the response of field soil spectrum was consistent with the air-dried soil.The evaluation index of soil organic matter and salt content,which has a great influence on the spectral reflectance.The greater the contribution of the evaluation index to the SDI value,the greater the contribution to the soil spectral reflectance,so the more severe the soil degradation,the greater the overall spectral reflectance.(3)Based on the correlation analysis and multiple stepwise linear regression analysis of the field soil spectrum and the first derivative of the field soil spectrum,the sensitive 47 bands of soil degradation were selected,and five kinds of spectral parameters were constructed.The model constructed by spectral parameter(RA+RB)/(RA-RB)is as follows: Y=-0.065-0.038X2300,2147 +0.033X2340,2120 +3.54*10-6X2300,2242-3.72*10-5X1975,1488 +0.004X900,760 +1.9*10-5X2431,1442 +1.91*10-5X2211,1434(R2=0.824),which is the optimum hyper-spectral estimation model of SDI in this study.Based on the fitting of hyper-spectral data and remote sensing image,the remote sensing inversion model of soil degradation was established by spectral parameter RA+RB,which is as follows: Y=0.235+3.755 XGreen,Red +3.808 XCoast al,NIR+0.585 XNIR,SWIR2,where R2 is 0.738.(4)The remote sensing inversion of soil degradation in Kenli County in 2015 shows: the soil area of light degradation,moderate degradation and heavy degradation respectively accounted for 10.16%,48.28%,and 41.56% of the study area;the characteristics o f spatial distribution of each grade were consistent with spatial interpolation of the measured data,which shows that the degree of soil degradation in southwest inland region is relatively light,and the degree of soil degradation is more serious in the northeast coastal region.The fitting R2 of the inversion’s SDI with the rea l SDI is 0.725,the precision of remote sensing inversion of sample points is 89.83%,and the Kappa coefficient of the error confusion matrix is 0.823,which shows that the results of remote sensing inversion are approximately consistent with the actual situation.(5)The dynamic analysis of soil degradation from 2007 to 2015 showed that the soil area of light degradation decreased by 4.44%,and the soil area of moderate degradation and heavy degradation respectively increased by 0.80% and 3.64%;the degree of soil degradation in the local area of northeast and southwest has been reduced,and the degree of soil degradation in the central region of Kenli County has increased;the average SDI of point sits and study area were 0.429 and 0.452 in 2015,which were slightly higher than the average SDI in 2007,so the degree of the soil degradation in 2015 was higher than that of the year of 2007,but the degree of aggravation was not obvious.