Research On Downscaling Soil Moisture Product Fusion Based On Passive Microwave Remote Sensing

Soil moisture is an important environmental factor that controls the water-heat exchange balance between the earth's land surface and the atmosphere and maintains global climate stability.It is one of the core input parameters of various land surface processes,ecology,hydrology and agricultural crop models.Changes in soil moisture will have a profound impact on the surface water and heat cycle,soil physical and chemical properties and climate change.Therefore,obtaining accurate and quantitative information on changes in soil moisture content in global or local areas is of great significance for maintaining the balance of the earth 's ecosystem and responding to climate change.Inversion of soil moisture based on satellite remote sensing technology is one of the most convenient and efficient means to obtain soil moisture information in large-scale areas.Soil moisture retrieval based on satellite remote sensing technology is one of the most convenient and efficient methods to obtain soil moisture information on large-scale regions.Passive microwave remote sensing is more sensitive to changes in surface soil moisture than optical remote sensing.At the same time,passive microwave estimates are higher than active microwaves.However,the spatial resolution of the passive microwave remote sensing pixels is generally low,and when the proportion of water in the pixels is high,the accuracy of soil moisture content will be very low.Although soil moisture images with higher spatial resolution can be obtained by a downscaling method based on a combination of passive microwave and optical remote sensing,due to the limited life of satellite sensors,acquiring long-term dynamic soil moisture data requires the aggregation of multi-source sensor microwaves.Data,eliminating observation errors between multi-source data is an important part of constructing long-term data sets.In this paper,using the matching method,based on AMSR-E level 3 data,all passive microwave soil moisture data will be uniformly corrected to the same time and the same detection depth.Combined with ground station data,large areas of missing pixels and invalid pixels are restored,thereby ensuring that the entire data set is complete in China.Then,based on the negative correlation between TVDI and soil moisture,a TVDI-based spatial weight decomposition model was established to reduce the spatial resolution of microwave remote sensing soil moisture data after consistency correction from AMSR-E,SMOS,and AMSR2 from 25 km to 5.6 km.The main conclusions are as follows:1)The new high-resolution soil moisture data set overcomes the problem of time-matching of multi-source data between optical and microwave data sources,eliminates the difference between the observation errors of different sensors,and has a higher spatial resolution.In addition,it can be used to study changes in soil moisture and drought during the day and night.Validation analysis shows that the new data set is highly consistent with the in situ observations(on monthly,seasonal,and annual scales,the correlation coefficients r: 0.826,0.882 and 0.901).2)Using the downscale soil moisture data set,the spatial and temporal pattern of soil moisture in China from 2002 to 2018 was analyzed.Over the past 18 years,China 's soil moisture has shown periodic fluctuations and downward trends(slope:-0.167,R: 0.750,P = 0.05),the Jiang-huai area of the North China and South China monsoon regions,the Yangtze River Delta region and The Bohai Rim region shows a rapid downward trend;however,the southern part of the northwest arid region in the northern part of the Qinghai-Tibet alpine region has a significant upward trend,which can be summarized as "wet in the south,dry in the north,increase in the west,decrease in the east".From different seasons,the soil moisture has changed significantly from spring to winter,and the seasonal change of soil moisture is mainly affected by the precipitation of the earth;while the increasing precipitation in the northwest arid area makes the soil moisture in this area show a certain upward trend It will effectively alleviate the drought disasters in the arid regions of Northwest China.This study shows that the soil moisture effect obtained by the passive microwave downscaling method of MODIS LST / NDVI products is better,which can meet the needs of the study of large-scale parameter changes in China,and can obtain more detailed local details,which can provide parameters for the corresponding scientific research.However,the accuracy of the reduced scale soil moisture and the measured data is slightly lower than that of the original soil moisture products.It is necessary to improve the quality of the reduced scale passive microwave soil moisture data.One of the problems that should be considered is the selection of some parameters and the quality evaluation.