Research On Satellite Remote Sensing Soil Moisture Correction And Inversion Algorithm

Remote sensing has the advantages of large-scale,long-period observations.Remote sensing monitoring of soil moisture can be divided into two categories: optical remote sensing and microwave remote sensing.The spatial resolution of optical remote sensing is high,but the penetrating power of this band is limited,and it is greatly affected by weather phenomena such as clouds,so it is impossible to observe all-weather.Microwave remote sensing can penetrate the cloud and observe the soil moisture all day and all weather,but its spatial resolution is low and it is greatly affected by the surface roughness.Due to the lack of in-situ data,there are still many uncertainties in the accuracy of soil moisture products,which directly affect the application of products.It is critical to evaluate product quality before using different soil moisture products.At the same time,the accuracy of remote sensing inversion of soil moisture products is uncertain,so it is necessary to carry out scientific quality correction of soil moisture products.In order to obtain accurate soil moisture information,five soil moisture products: SMAP(soil moisture active and passive detection satellite),FY-3C,ERA-Interim,NCEP R-1 and NCEP R-2 were evaluated.Based on the results of the authenticity test,the soil moisture data of the FY-3C microwave imager(MWRI)was used to correct the soil moisture product using the variational correction method,then the correction effect was evaluated.The spatial variation of soil moisture is large.Based on the advantages of high spatial resolution of optical remote sensing,the spatial distribution of two-dimensional spectra obtained by radiation in the near-infrared(NIR)and red-light(RED)bands is used to propose Vegetation Vertical Drought Index(VPDI)Act.The main work and achievements are as follows:In this paper,five soil moisture products,SMAP(Soil Moisture Active and Passive Detecting Satellite),FY-3C,ERA-Interim,NCEP R-1 and NCEP R-2 were evaluated.The in-situ data is the external field observation test data of the Tibet Plateau in 2016..Compared with other products,the SMAP soil moisture product in the Tibet Plateau has the smallest deviation from the measured value and the highest correlation coefficient.In the case of low vegetation cover,the satellite soil moisture data has better quality,but in the case of high vegetation coverage,the soil moisture data provided by the reanalysis data has less deviation from the measured data.The satellite products can better reflect the seasonal changes in the Tibet Plateau,however,the soil moisture changes provided by the reanalysis data are not obvious.In this paper,the soil moisture data of the FY-3C microwave imager(MWRI)was used to evaluate the soil moisture in the Tibet Plateau using the variational correction method.The results show that the accuracy of the product can be greatly improved after the correction.In order to further test the correction method,the soil moisture data of Shandong Agricultural Meteorological Stations was selected to evaluate the product before and after FY-3C soil moisture correction.The results showed that the error between FY-3C soil moisture product and in-situ data was reduced after correction.The correlation is significantly improved.The soil moisture changes of FY-3C soil moisture data before and after the correction of the drought process in Shandong Province from March to April 2016 were compared.After correction,the soil moisture of FY-3C more accurately reflected the drought process,indicating that the variational correction method can effectively improve the data.The soil moisture product reflect the characteristics of the drought process better.In this paper,we used the two-dimensional spectral spatial distribution obtained by the near-infrared(NIR)and red-light(RED)bands to separate the vegetation from the soil spectrum,and propose a new drought monitoring method based on vegetation coverage.Vegetation Vertical Drought Index(VPDI)method for remote sensing data.In this paper,the VPDI and PDI were calculated using the Himawari-8 data,and the VPDI and PDI were compared using the in-situ data in the Tibetan Plateau.The results showed that VPDI accuracy was significantly improved in high vegetation coverage area compared with PDI.With theincrease of vegetation cover,the correlation between PDI and in-situ soil moisture data gradually deteriorated,and the correlation coefficient between soil moisture of VPDI and in-situ data was always high.VPDI overcomes the distortion caused by the increase of vegetation coverage.VPDI is highly related to the in-situ soil moisture data and can be used for real-time dynamic drought remote sensing monitoring under different vegetation cover conditions.