Carbon Flux Optimization Simulation Of Pinus Tabulaeformis Forest In Helan Mountain And Its Effect Factors Analysis

The forest ecosystem is an essential component of the terrestrial ecosystem on Earth,playing a crucial role in mitigating global warming,sequestering carbon and releasing oxygen,and maintaining the water cycle of land surface and atmosphere.The forest ecosystem sequesters atmospheric CO2 through photosynthesis,making it Earth’s most critical terrestrial carbon sink.However,there are significant differences in their carbon sequestration functions in forest ecosystems formed by different geographical and climatic regions and dominant communities.This is also an important reason for the uncertainty in the assessment of carbon sink capacity in current terrestrial ecosystems.To assess the carbon sink capacity of a specific forest ecosystem,it is necessary to master the carbon exchange law between the ecosystem and the atmosphere in detail.Through the carbon flux monitoring,simulation and driving analysis of the ecosystem,the cycle law of carbon driven by radiation energy is revealed,and the environmental factors affecting the carbon cycle of the forest ecosystem are discussed,which is the scientific basis for improving the carbon sink service function of the forest ecosystem.Therefore,this study took the Pinus tabulaeformis forest in Helan Mountain of Ningxia,located in the arid transition zone of northwest China,as the research object.On the basis of analyzing the characteristics of radiation energy interception,transmission and utilization in the canopy layer of Pinus tabulaeformis forest in Helan Mountain,with the help of remote sensing multi temporal phase space data,carbon flux observation data and environmental factor data,the Biome BGC model was driven,Using the Parameter ESTIMATION(PEST)optimization algorithm and the 4-Dimensional Variational Data Assimilation(4Dvar)algorithm,the carbon flux simulation process of the model was further optimized to achieve carbon flux simulation of Pinus tabulaeformis forest in Helan Mountain at different time scales.The impact of meteorological environmental factors on carbon flux of Pinus tabulaeformis forest in Helan Mountain was analyzed.The research mainly draws conclusions from the following aspects:(1)The radiation energy characteristics of the canopy of Pinus labulaeformis forest in Helan Mountain.The solar shortwave radiation shows a "single peak" pattern during the day and has significant seasonal changes.It is the main energy source driving photosynthesis in the Chinese pine canopy,but is significantly affected by weather conditions.Cloudy weather can reduce nearly half of the total solar radiation reaching the canopy,and downward longwave radiation(atmospheric reverse radiation)can also bring heat to the canopy,but the upward longwave energy radiated from the canopy to the sky is more,The surface long wave radiation budget ratio is between 0.73 and 0.80;After the total solar radiation reaches the canopy of Pinus tabulaeformis,it undergoes processes such as reflection,transmission,absorption,and transformation.7.8%to 8.8%of the shortwave radiation is reflected back into the sky,and 32.2%to 53.9%of the PAR reaches below the canopy through the canopy.The shortwave energy absorbed by the surface is not only used to drive vegetation physiological and ecological processes such as carbon water cycle,but also partially converted into longwave radiation;The seasonal variation of solar shortwave radiation drives the seasonal fluctuations of PAR and all longwave radiation on the canopy.Due to the complex interaction between energy interception and transmission in the canopy of Chinese pine forests,the reflection and transmission characteristics of the canopy are related to biological factors such as canopy closure and growth period,and exhibit significant differences in different weather conditions and seasons.(2)Two algorithms,optimization and model assimilation,were implemented to optimize the Biome-BGC model.Firstly,the PEST method is used to initialize the Biome-BGC model and optimize its state parameters.A comprehensive sensitivity analysis is conducted on the parameters of the Biome-BGC model.Through the dynamic file splitting method,PEST is used to optimize cross file model parameters and achieve parameter localization.Secondly,the 4Dvar assimilation method is used to assimilate the Biome-BGC model.Calibrate the simulated trajectory of the model and derive the cost function.The dynamic model Biome-BGC model uses automatic differentiation technology to obtain its accompanying model code at the computer code level,and adjusts the model simulation trajectory using observation data,significantly improving simulation accuracy.(3)Based on an optimization model,the carbon flux of Pinus tabulaeformis forest in Helan Mountain was simulated.Based on PEST optimized Biome-BGC model parameters,the Gross Primary Productivity(GPP)and Ecosystem Respiration(RE)of Pinus tabulaeformis forest in Helan Mountain in 2021 were simulated.Based on the daily carbon flux observation data of the growth season of the flux station,the model optimization results were evaluated.The results showed that the R2 values of GPP and RE before model optimization were 0.31 and 0.41,respectively,and the RMSE values were 0.95 gC·m-2·d-1 and 0.54 gC·m-2·d-1,respectively.The R2 values of GPP and RE after PEST model parameter optimization were 0.44 and 0.51,respectively,and the RMSE values were 0.58 gC·m-2·d-1 and 0.35 gC·m-2·d-1,respectively,Parameter optimization has improved the accuracy of the Biome-BGC model’s carbon flux.The BiomeBGC model was assimilated using the 4DVar data assimilation method.The results showed that the R2 values of GPP and RE before model assimilation were 0.31 and 0.41,respectively,and the RMSE values were 0.95 gC·m-2·d-1 and 0.54 gC·m-2·d-1,respectively.The R2 values of GPP and RE after 4Dvar assimilation were 0.52 and 0.63,respectively,and the RMSE values were 0.53 gC·m-2·d-1 and 0.31 gC·m-22·d-1,respectively.The 4DVar data assimilation method can significantly improve the simulation accuracy of the Biome-BGC model.(4)Carbon flux characteristics of Pinus tabulaeformis forest in Helan Mountain.During the growth season from May to October 2021,the GPP of the Pinus tabulaeformis forest in Helan Mountain was mainly between 0-2.01 gC·m-2·d-1,and the RE was mainly between 0.28-1.56 gC·m-2·d-1.Both GPP and RE showed obvious seasonal variation characteristics.During a growing season,the maximum GPP value appears in June,while the maximum RE value appears in May;From April to June,the GPP and RE values showed an upward trend,with high and relatively stable values.From July to August,there was a concave valley,which began to decline after the end of the growth season in October.The soil respiration rate during the growing season showed a strong diurnal dynamics.After sunrise in the morning,the soil respiration rate gradually increased,reaching the highest at around 12:00,with an average of 0.71 μmol·m-2·s-1,the rate of soil respiration decreased in the afternoon and dropped to 0.40 at 16:00 μmol·m2·s-1,and then the soil respiration rate remained stable.The soil respiration rate showed strong seasonal variation characteristics in the year.Since late May,the soil respiration rate has changed from 0.84 μmol·m-2·s-1 gradually increased,reaching a peak of 1.27 in late September within the year μmol·m-2·s-1,and then began to decrease.(5)Factors affecting carbon flux of Pinus tabulaeformis forest in Helan Mountain.During the growth season of 2021,the air temperature(Ta)and soil temperature(Ts)of Chinese pine forests have a similar trend of change.The daily average ranges of Ta and Ts are 1.33 to 26.33℃ and 0.87 to 21.81℃,respectively.The daily highest values of Ta and Ts appear in mid July,and the daily lowest values appear in early April;The photosynthetic effective radiation(PAR)within the canopy of Pinus tabulaeformis forest exhibits a unimodal change throughout the year,with the maximum value appearing on June 3rd at 534.51 μ mol·m-2·s-1;The total rainfall is 175.9 mm,and the maximum daily saturated vapor pressure difference(VPD)appeared on April 4th,reaching 2.77 kPa,but the daily average VPD was the highest in June.During the growing season,Ta and VPD are the meteorological factors that have the greatest impact on GPP,respectively;Ts and PAR take second place.Ta is the main meteorological factor affecting RE.