Research On Time Series Data Processing And Prediction Model Of Atmospheric CO₂

The greenhouse effect and climate change caused by the continuous increase of atmospheric CO₂ concentration have attracted widespread human attention.In this paper,CO₂ concentration data is used as the research object,and the GOSAT satellite remote sensing data set and airs s3c2m data set are used as data sources to analyze the temporal and spatial distribution characteristics of CO₂ concentration at near-ground height and mid-tropospheric height in the global and chinese regions.On this basis,the CO₂concentration in Chinese near-surface regions retrieved by gosat is selected for analysis,and the arima model and the lstm long and short-term memory artificial neural network model are established respectively to predict the future trend of changes and scientifically predict the future trend of CO₂.,is of great significance to the realization of CO₂ emission reduction and the exploration of the carbon cycle law.The main contents of this paper are as follows:（1）Four global atmospheric background stations were used:china’s wariguan global atmospheric background station,awaii’s mouna loa atmospheric background station,algeria’s assekrem atmospheric background station and nowit ridge atmospheric local station to test gosat verification with the airs remote sensing data set,the verification results show that the atmospheric CO₂ observed by remote sensing has good consistency with the ground observation data,indicating that the satellite remote sensing data can be used as a basis for studying the characteristics of CO₂ temporal and spatial distribution.（2）The temporal and spatial distribution of global atmospheric CO₂:both remote sensing data products show that the CO₂ concentration in the northern hemisphere is higher than that in the southern hemisphere,and the CO₂ concentration on land is significantly higher than that in the ocean,and several high-value centers appear in tropical asia and temperate north america.Northeast,central africa and other places.The distribution law shows that the higher the latitude zone,the higher the atmospheric CO₂concentration,and the faster the average annual growth rate.From the time dimension,the two types of satellite remote sensing data both show a seasonal upward trend in global CO₂ concentration,with the highest in spring and lowest in summer,while the northern and southern hemispheres show opposite seasonal distributions.（3）The spatial and temporal distribution of atmospheric CO₂ in china.The concentration of CO₂ near the ground in my country is also showing an increasing trend.The spatial distribution presents the characteristics of high in the south and low in the north,high in the east and low in the west,forming several high in central china,eastern coastal areas and northern china.Value center.The overall trend of CO₂ concentration in the middle troposphere in my country is also increasing seasonally,and the spatial distribution presents a characteristic of high in the north and low in the south,which is contrary to the characteristics of near-surface spatial distribution.（4）Select the GOSAT remote sensing data set as the research object,and use the ARIMA model and the LSTM neural network model to predict the CO₂ concentration change trend in Chinese near-surface area.The prediction results show that my country’s near-surface CO₂ concentration will continue to increase seasonally in the future trend.From the fitting results,the arima prediction model has a slight advantage in fitting and predicting the gosat remote sensing data set.Figure 42 table 10 reference 62...