Applied Research Of Particle Swarm Optimization In MAX-DOAS Profile Retrieval

In recent years,environmental problems have become increasingly prominent with the rapid development of human society.MAX-DOAS is more and more widely used in the monitoring of atmospheric composition.Currently,there is a limitation of low retrieval height in the main algorithms of MAX-DOAS.It is necessary to propose a new method to retrieve the trace gas at high altitude.This paper applies the particle swarm algorithm to the MAX-DOAS profile retrieval for the first time.A profile retrieval algorithm based on improved particle swarm algorithm has been constructed and been used to retrieve the ozone profiles of Shanghai on December 20,2017,December 25,2017,and the ozone profiles of Kunming on December 3,2017,and December 22,2017,the result was compared and verified with the Ozone Monitoring Instrument（OMI）Level 2 ozone profile data of the two places and the Electrochemical Concentration Cell（ECC）ozone sounding data of Kunming.First of all,this paper optimizes the standard particle swarm algorithm by limiting the maximum speed of particles and introducing a particle mutation mechanism,which improves the algorithm’s search ability and convergence speed.Based on this,the MAX-DOAS particle swarm retrieval algorithm is constructed,and a simulation verification of the constructed particle swarm retrieval algorithm has been completed.It can be seen from the results that the retrieval algorithm can complete the retrieval of the ozone profile under ideal conditions.The retrieval result successfully raised the profile retrieval height to 60 km,and the relative deviation at each height was basically below 20%.The feasibility of the particle swarm retrieval algorithm is verified.Subsequently,the particle swarm retrieval algorithm was used to retrieve the ozone profile from the MAX-DOAS observation results in Shanghai and Kunming.The results show that for higher altitudes,it is basically identical between the relative deviations of retrieval above 19.5km in the two places.For lower heights,the relative deviation of tropospheric retrieval in Shanghai is better than that in Kunming.The retrieval results show that the particle swarm retrieval algorithm can increase the retrieval height to 60 km when retrieving the MAX-DOAS ozone profile,and the general trend of the profile could also be retrieved.A good retrieval effect was obtained in the range of 19.5km and above.However,the retrieval effect of tropospheric ozone is not ideal,and it is difficult to retrieve the trend of convective profile well.The retrieval results are prone to large deviations,and the retrieval effects in different regions are quite different.