Satellite retrievals of columnar nitrogen dioxide(NO_(2))are essential for the characterization of nitrogen oxides(NO_(x))processes and impacts.The requirements of modeled a priori profiles present an outstanding bott...Satellite retrievals of columnar nitrogen dioxide(NO_(2))are essential for the characterization of nitrogen oxides(NO_(x))processes and impacts.The requirements of modeled a priori profiles present an outstanding bottleneck in operational satellite NO_(2)retrievals.In this work,we instead use neural network(NN)models trained from over 360,000 radiative transfer(RT)simulations to translate satellite radiances across 390-495nm to total NO_(2)vertical column(NO_(2)C).Despite the wide variability of the many input parameters in the RT simulations,only a small number of key variables were found essential to the accurate prediction of NO_(2)C,including observing angles,surface reflectivity and altitude,and several key principal component scores of the radiances.In addition to the NO_(2)C,the NN training and cross-validation experiments show that the wider retrieval window allows some information about the vertical distribution to be retrieved(e.g.,extending the rightmost wavelength from 465 to 495 nm decreases the root-mean-square-error by 0.75%)under high-NO_(2)C conditions.Applying to four months of TROPOMI data,the trained NN model shows strong ability to reproduce the NO_(2)C observed by the ground-based Pandonia Global Network.The coefficient of determination(R2,0.75)and normalized mean bias(NMB,-33%)are competitive with the level 2 operational TROPOMI product(R^(2)=0:77,NMB=−29%)over clear(geometric cloud fraction<0:2)and polluted(NO_(2)C≥7:5×10^(15)molecules/cm2)regions.The NN retrieval approach is~12 times faster than predictions using high spatial resolution(~3 km)a priori profiles from chemical transport modeling,which is especially attractive to the handling of large volume satellite data.展开更多
基金This work was supported by the Postdoctoral Program in Environmental Chemistry of the Camille and Henry Dreyfus Foundation,the National Aeronautics and Space Administration(grant no.80NSSC19K0945)the Smithsonian Institution(grant no.SV383019)+1 种基金J.Wang’s participation is made possible by the in-kind(James E.Ashton Professorship)support from The University of Iowa.J.Jin was partially supported by the National Nature Science Foundation of China under the project no.41805027the Ministry of Science and Technology of China under the project no.2017YFC1501802.
文摘Satellite retrievals of columnar nitrogen dioxide(NO_(2))are essential for the characterization of nitrogen oxides(NO_(x))processes and impacts.The requirements of modeled a priori profiles present an outstanding bottleneck in operational satellite NO_(2)retrievals.In this work,we instead use neural network(NN)models trained from over 360,000 radiative transfer(RT)simulations to translate satellite radiances across 390-495nm to total NO_(2)vertical column(NO_(2)C).Despite the wide variability of the many input parameters in the RT simulations,only a small number of key variables were found essential to the accurate prediction of NO_(2)C,including observing angles,surface reflectivity and altitude,and several key principal component scores of the radiances.In addition to the NO_(2)C,the NN training and cross-validation experiments show that the wider retrieval window allows some information about the vertical distribution to be retrieved(e.g.,extending the rightmost wavelength from 465 to 495 nm decreases the root-mean-square-error by 0.75%)under high-NO_(2)C conditions.Applying to four months of TROPOMI data,the trained NN model shows strong ability to reproduce the NO_(2)C observed by the ground-based Pandonia Global Network.The coefficient of determination(R2,0.75)and normalized mean bias(NMB,-33%)are competitive with the level 2 operational TROPOMI product(R^(2)=0:77,NMB=−29%)over clear(geometric cloud fraction<0:2)and polluted(NO_(2)C≥7:5×10^(15)molecules/cm2)regions.The NN retrieval approach is~12 times faster than predictions using high spatial resolution(~3 km)a priori profiles from chemical transport modeling,which is especially attractive to the handling of large volume satellite data.
