Study On Temporal And Spatial Distribution Prediction And Influencing Factors Of Tropospheric Formaldehyde In Beijing Tianjin Hebei Region Based On OMI Data

The concentration of formaldehyde is relatively high in each urban area in China,and the concentration of formaldehyde is often higher in summer.As a special aldehyde ketone compound,the content of formaldehyde in the atmosphere is generally low,and the average life span is only about 3.5 hours.Formaldehyde is not only an important intermediate product of VOC photochemical reaction,but also a necessary prerequisite for the formation of free radicals,oxidants and ozone.Traditional ground monitoring methods are often difficult to systematically and continuously monitor formaldehyde in a large area because of the lack of data and large error,and remote sensing technology weakens the disadvantages of traditional ground monitoring.Firstly,based on the aura/OMI satellite formaldehyde products from 2010 to2019,this paper analyzes the temporal and spatial change law of Beijing Tianjin Hebei region,analyzes the spatial change trend of formaldehyde concentration in the past 10 years,and forecasts the change trend in the future.Then,the influencing factors affecting the distribution of formaldehyde concentration are studied,and the natural influencing factors and human influencing factors are analyzed respectively.Finally,through the combination of Omi daily average formaldehyde concentration data and CMAQ simulated formaldehyde data,the daily average formaldehyde concentration data are predicted and analyzed based on multiple regression model,time series（ARIMA）model and long-term and short-term memory machine（LSTM）neural network model,which further improves the feasibility of formaldehyde pollution prediction.This paper analyzes the temporal and spatial variation law and spatial variation trend of tropospheric formaldehyde column concentration in Beijing Tianjin Hebei region from 2010 to 2019.It is found that the tropospheric formaldehyde column concentration in Beijing Tianjin Hebei region increases first,then decreases and then increases slowly.The overall seasonal variation trend of tropospheric formaldehyde concentration in Beijing Tianjin Hebei region is basically the same as the annual distribution.The average concentration of formaldehyde column in the four seasons is summer > autumn > spring > winter.Over the past decade,the spatial distribution area of formaldehyde concentration in Beijing Tianjin Hebei region has shown a decreasing trend as a whole.In the future,the area where the formaldehyde column concentration will continue to decrease will account for about 32.51%,and the area where the formaldehyde column concentration will continue to increase will account for 35.07%.On the whole,the formaldehyde concentration in Beijing,Tianjin and Hebei will show a slow upward trend in the future.In the study of the influencing factors affecting the distribution of formaldehyde concentration in Beijing,Tianjin and Hebei,it is found that human activities and natural factors have an impact on the spatial distribution of formaldehyde concentration.Finally,combined with the formaldehyde data simulated by CMAQ and the daily average formaldehyde concentration data of Omi satellite,multiple linear regression model,time series（ARIMA）model and long-term and short-term memory machine（LSTM）neural network model are used to predict the formaldehyde concentration of typical cities in Beijing,Tianjin and Hebei.It is found that LSTM neural network model has more advantages in predicting the daily average formaldehyde concentration,The prediction results in each city are better than the linear prediction model.