Research On The Time And Spatial Interpolation Of AOD Based On The Peterson Model Of Beijing-Tianjin-Hebei Region

Atmospheric aerosol is a multiply system composed of solid or liquid and gas carriers suspended in the atmosphere.Through the absorption and scattering of solar radiation and surface radiation,the influence of atmospheric transfer has an important role in the earth’s atmospheric radiation balance and global climate change, and also reflects the atmospheric environment.Although the ground monitoring stations can monitor the aerosol concentration,the ground site distribution is uneven, the amount is not much, which has great limitations in the temporal and spatial variation of all-round monitoring of aerosol. MODIS has obvious advantages in monitoring the aerosol.The aerosol optical depth(AOD) is one of the most basic optical characteristics of aerosol extinction coefficient, generally which refers to the integration of the whole layer of the aerosol in the vertical direction and is described to weaken the role of aerosol optical physics quantity of light.The aerosol optical depth is determined the climatic effects of aerosols is an important factor affecting the changes in characteristics and regional atmospheric environmental monitoring application of aerosol widely.Due to the limitation of the algorithm, the seasonal variation of meteorological conditions or factors such as poor,unable to obtain the required images or images of poor quality, resulting in a large area on the AOD sequence of the missing value, resulting in the use of AOD to do research area,which can not guarantee the continuity of data and reduce the accuracy.Firstly the AOD were time series interpolated: choose 2009-2013 years in Beijing-Tianjin-Hebei region with 89 meteorological stations, where the MODIS atmospheric secondary standard product data mod04 atmospheric AOD(aerosol optical depth daily data and meteorological visibility degree day data. Based on the Peterson model the MODIS AOD values and V to do the calculation, it will get the site of 365 years h daily mean values for all site years h, mean and regional years of H mean Gaussian fitting,which finds H of averagedaily values with Gaussian Distribution, so as to establish a single site h fitting function and regional H fitting function, using H and V calculated AOD value prediction, and proves that the MODIS AOD prediction value is practical, which selects the 2014 level meteorological visibility and randomly selected seven sites, the establishment of function fitting forecast2014 H, value and AOD, value and error analysis of the value of prediction of AOD and MODIS AOD 、 RMSE and MRE in the range. Then the AOD were sequence space interpolation, respectively, two kinds of interpolation validation using four kinds of GIS basic interpolation method: first Class is not increased to simulate site, the value of the error of the interpolation model analysis using the image of the original site of MODIS AOD; the second is increase fitting site fitting AOD as valid data and MODIS AOD values at the same time as the interpolation model validation data, and by means of error analysis, which are used to evaluate the accuracy of the four basic interpolation methods. Select a representative data of207, 209 to do interpolation analysis, to explore in the inversion value number and different interpolation methods of different precision, and,accord to the 2009-2014 October weather stations in Grid The number of statistics, the interpolation method of grading standards are formulated. Finally, take the high accuracy AOD time and spatial interpolation methods, for2009-2014 October AOD inversion of remote sensing image because of lots of missing values to realize comprehensive interpolation and calculation of AOD.According to the full image analysis,the variations of the image of AOD also can be used to obtain changesin atmospheric environment in recent years.