Research On Forest Carbon Flux Observation In Southwest Hilly Region Based On Eddy Correlation Method

Forest ecosystem is the main place for terrestrial ecosystem carbon exchange,and it is essential,to study the forest ecosystem carbon cycle.The vorticity correlation method is the most reliable method for direct measurement of carbon exchange.However this method is limited by the influence of terrain,thus flux data correction and quality control are very important.In this study,the vorticity correlation method was used to study the carbon flux of the subtropical coniferous broad-leaved forest in Chongqing Jinyun mountain forest ecosystem observation and research station.Using the flux and meteorological data from June 2016 to May 2017,the method for tilt correction of the data under complex terrain conditions in mountainous areas was discussed,and the appropriate tilt correction method was explored,and the local energy balance closure and the annual carbon flux characteristics were clarified.The main results are as follows:?1?The study found that the suitable tilt correction method for Jinyun mountain is the double coordinate axis rotation method?DR?.In mountainous terrains like Jinyun mountain,horizontal wind will have a significant vertical shear,which results in a large lateral stress,so the three-rotation method?TR?significantly underestimates the flux.The plane fitting method?PF?can not fit an appropriate plane to represent the underlying surface due to the large unevenness of the underlying surface,so it is not suitable for use in the Jinyun mountain area.After tilt correction by the three methods DR,TR and PF,the heat flux,carbon flux and frictional wind speed all had a decreasing trend.Overall,DR had the least effect on the flux value,which reduced the sensible heat H by 18%,latent heat LE decreases by 32%,carbon flux F_cdecreases by 40%,and friction wind speed u*decreases by 22%.After correction by DR and TR,the proportion of high-quality data?level 0?increased by 2.89?4.51%,and after correction by PF,the proportion of high-quality data decreased by 13.62?16.60%.Coordinate rotation reduced the energy closure rate?EBR?of Jinyun mountain.The EBR after rotation by DR and TR were 0.78 and 0.77respectively,and it was still within the reasonable range considered by the academic community.?2?The energy closure rate during the day was greater than that during the night,the EBR was 0.79during the day,and only 0.26 during the night,and the correlation coefficientvaried greatly.The=0.72 during the day,while theat night was only 0.01.The EBR changed drastically during sunrise and sunset,the night value fluctuated between-1.00 and 0.65,and the minimum value appeared at 19:00;EBR risen slowly during the day,starting at 7 o'clock and reaching the maximum value of 1.33 at 18:30.U*,Turbulent kinetic energy?TKE?,Monin–Obukhov stability parameter??ZD?/L?and average wind speed?WS?had a very significant linear correlation with EBR?p<0.001?,where u*had a positive correlation with EBR,other factors had a negative correlation with EBR,and u*had the greatest impact on EBR,The standardized coefficient was 0.22.?3?The average daily change of net ecosystem carbon exchange volume?NEE?in the coniferous and broad-leaved mixed forest ecosystem of Jinyun mountain were in the shape of"U".NEE was negative during the day,the ecosystem was a carbon sink,and NEE was positive at night,the ecosystem was a carbon source.The maximum and minimum daily average carbon sink peaks were in July and December.There were seasonal differences in daily carbon sink duration and daily carbon sequestration.The daily carbon sink duration was spring,summer,autumn,and winter in descending order,and the daily carbon sequestration was summer,autumn,spring,and winter in descending order.NEE,ecosystem respiration volume Re,and total ecosystem carbon exchange volume GEE annual total were-566.49,1196.68,and-1761.63 g C·m^-2,respectively.The storage of canopy carbon?F_s?at the half-hour scale had a greater impact on NEE,but at a longer time scale,the impact was not obvious,annual F_swas only 4%of the annual NEE.Overall,the ecosystem was a carbon sink,and only December was the carbon source throughout the year.U*,TKE,WS,and relative humidity?RH?had a very significant linear correlation with F_c?p<0.001?,where u*had a negative correlation with F_c,and other factors had a positive correlation with F_c;WS had the greatest influence on F_c,with a standardized coefficient of0.13.