It has been observed that low temperature, rainfall, snowfall, frost have never occurred over the past 50 years in the southern China, and weather in this area is very complex, so the monitoring equipments are few. Op...It has been observed that low temperature, rainfall, snowfall, frost have never occurred over the past 50 years in the southern China, and weather in this area is very complex, so the monitoring equipments are few. Optical and thermal infrared remote sensing is influenced much by clouds, so the passive microwave Advanced Microwave Scanning Radiometer-Earth Observing System （AMSR-E） data are the best choice to monitor and analyze the development of disaster. In order to improve estimation accuracy, the dynamic learn- ing neural network was used to retrieve snow depth. The difference of brightness temperatures of TB18.7v and TB36.sv, TBI8.7H and TB36.sH, TB23,sv and TB89v, TBz3.8H and TB89H are made as four main input nodes and the snow depth is the only one output node of neural network. The mean and the standard deviation of retrieval errors are about 4.8 cm and 6.7 cm relative to the test data of ground measurements. The application analysis indicated that the neural network can be utilized to monitor the change of snow intensity distribution through passive microwave data in the complex weather of the southern China.展开更多
The broadband emissivity is an important parameter for estimating the energy balance of the Earth. This study focuses on estimating the window （8 -12 μm） emissivity from the MODIS （mod- erate-resolution imaging sp...The broadband emissivity is an important parameter for estimating the energy balance of the Earth. This study focuses on estimating the window （8 -12 μm） emissivity from the MODIS （mod- erate-resolution imaging spectroradiometer） data, and two methods are built. The regression method obtains the broadband emissivity from MODllB1 - 5KM product, whose coefficient is developed by using 128 spectra, and the standard deviation of error is about 0.0118 and the mean error is about O. 0084. Although the estimation accuracy is very high while the broadband emissivity is estimated from the emissivity of bands 29, 31 and 32 obtained from MOD11B1 _ 5KM product, the standard deviations of errors of single emissivity in bands 29, 31, 32 are about 0.009 for MOD11B1 5KM product, so the total error is about O. 02 and resolution is about 5km × 5km. A combined radiative transfer model with dynamic learning neural network method is used to estimate the broadband emis- sivity from MODIS 1B data. The standard deviation of error is about 0.016, the mean error is about 0.01, and the resolution is about 1 km x 1 km. The validation and application analysis indicates that the regression is simpler and more practical, and estimation accuracy of the dynamic learning neural network method is higher. Considering the needs for accuracy and practicalities in application, one of them can be chosen to estimate the broadband emissivity from MODIS data.展开更多
基金Under the auspices of National Program on Key Basic Research Project(No.2010CB951503)National Key Technology R&D Program of China(No.2013BAC03B00)National High Technology Research and Development Program of China(No.2012AA120905)
文摘It has been observed that low temperature, rainfall, snowfall, frost have never occurred over the past 50 years in the southern China, and weather in this area is very complex, so the monitoring equipments are few. Optical and thermal infrared remote sensing is influenced much by clouds, so the passive microwave Advanced Microwave Scanning Radiometer-Earth Observing System （AMSR-E） data are the best choice to monitor and analyze the development of disaster. In order to improve estimation accuracy, the dynamic learn- ing neural network was used to retrieve snow depth. The difference of brightness temperatures of TB18.7v and TB36.sv, TBI8.7H and TB36.sH, TB23,sv and TB89v, TBz3.8H and TB89H are made as four main input nodes and the snow depth is the only one output node of neural network. The mean and the standard deviation of retrieval errors are about 4.8 cm and 6.7 cm relative to the test data of ground measurements. The application analysis indicated that the neural network can be utilized to monitor the change of snow intensity distribution through passive microwave data in the complex weather of the southern China.
基金Supported by the National Program on Key Basic Research Project(No.2010CB951503,2013BAC03B00,2012AA120905)
文摘The broadband emissivity is an important parameter for estimating the energy balance of the Earth. This study focuses on estimating the window （8 -12 μm） emissivity from the MODIS （mod- erate-resolution imaging spectroradiometer） data, and two methods are built. The regression method obtains the broadband emissivity from MODllB1 - 5KM product, whose coefficient is developed by using 128 spectra, and the standard deviation of error is about 0.0118 and the mean error is about O. 0084. Although the estimation accuracy is very high while the broadband emissivity is estimated from the emissivity of bands 29, 31 and 32 obtained from MOD11B1 _ 5KM product, the standard deviations of errors of single emissivity in bands 29, 31, 32 are about 0.009 for MOD11B1 5KM product, so the total error is about O. 02 and resolution is about 5km × 5km. A combined radiative transfer model with dynamic learning neural network method is used to estimate the broadband emis- sivity from MODIS 1B data. The standard deviation of error is about 0.016, the mean error is about 0.01, and the resolution is about 1 km x 1 km. The validation and application analysis indicates that the regression is simpler and more practical, and estimation accuracy of the dynamic learning neural network method is higher. Considering the needs for accuracy and practicalities in application, one of them can be chosen to estimate the broadband emissivity from MODIS data.
