Effect Of Storage Flux Estimation Methodology On Eddy Covariance Carbon And Energy Fluxes Estimation

Accurate assessment of forest carbon and energy budgets is of great significance to the global carbon and water cycle.Eddy covariance(EC)can directly measure the matter and energy fluxes in the ecosystem,which has the advantages of large measurement spatial range and high temporal resolution,plays a huge role in monitoring forest carbon and energy cycles.With the expansion of research requirements,the application of EC technology has become more and more extensive,but it still has its own limitations,including advection problems under complex terrain,imperfect flux resolution theory,and unclosed energy balance problems and so on.These errors will cause systematic deviations on the scale of global ecosystems.Therefore,solving the problems in the practical application of EC technology is very important for the accurate assessment of carbon and energy budget on the global ecosystem scale.CO2 and energy storage flux measurement errors can lead to flux underestimation,but storage flux algorithms still lack standards.In this paper,we systematically studied storage flux measurement methods in the Maoershan deciduous broad-leaved forest flux observation site compared the differences of CO2 storage flux(Fs)and energy storage flux(SE)among different measurement methods,and evaluated the effects of SC and SE on forest CO2 net ecosystem exchange(NEE)and energy balance.In this paper,the Fs estimated by single point(Ps),30 min profile(P30 min)and 2 min profile(P2 min)methods and the subsequently calculated NEE,respiration(Re)and gross primary productivity(GPP)were compared;the three-dimensional stem temperature(Tb)of major tree species was measured.Combined with the EC system,the sub-items of SE in the ecosystem were comprehensively measured,and SE and energy balance closure(EBC,the slope of principal axis regression between turbulent energy and available energy)estimated by 30-min,15-min,5-min,1-min and 5-s time windows were compared.The results showed that:The impact of the Fs estimation method on forest carbon fluxes generally increases with the increase of the time scale,indicating that the interpolation and resolution of the flux data will further amplify the influence of the Fs calculating method.On the half-hour scale,the Fs calculated by the Ps method in the growing season was 39.5%lower than the P30 min method,while the Fs calculated by the P2 min method was 2.0%higher than the Fs of the P30 min method;the P2 min and Ps method in the non-growing season were 9.7%and 34.5%lower than the P30 min method.On the daily scale,the variables of NEE,Re and GPP of the P2 min method were 6.1%,13.4%and 5.0%higher than those of the P30 min method,respectively.The NEE and Re of the Ps method were 1.8%and 2.4%higher than those obtained by the P30 min method,respectively,while the GPP of the single-point method was 2.0%lower than that of the P30 min method.On the annual scale,the NEE based on P2 min and Ps method were 36.3%and 29.4%lower than that based on P30 min,the Re based on P2min was higher than that based on P30 min and Ps by 8.7%;the GPP based on P2min was 5.43%higher,while that based on Ps was 2.1%lower than that based on P30 min.The P30 min methodology ignored the instantaneous changes in CO2 concentration,Ps missed the changes of CO2 concentration within forest canopy,and both understated the actual Re.The approximately instantaneous profile(2-min mean profile)had higher temporal and spatial resolution and could more accurately estimate the forest carbon budget with non-flat terrain and complex canopy structure.This finding has great implications for solving the underestimation of forest Re and GPP as well as the overestimation of carbon sink on complex conditions with the EC methodology.There were significant differences in the estimation of SE in different time windows.The daily variation of Sb in a typical sunny day with the 5-s time window was higher than that with 30-min time window,and the Sb estimated with the 5-s time window was 3.6%higher than that with 30-min time window;The SE sub-items with the 5-s time window were higher than those with 30-min time window,and the total energy storage flux(SE)was 4.9%higher than that with 30-min time window.At the same time,when St was at its maximum,it contributed 54.72%to net radiation.On the half-hour time scale,the EBC calculated with 5-s time window showed no significant difference with that with 30-min time window.The branch energy storage flux(Sb)with the 5-s time window obtained by measuring Tb in three-dimensional space was significantly improved than the simple estimation;After considering the SE subitems as much as possible,on the half-hour scale,the EBC with the 5-s time window was increased by 4.1%compared to the EBC that only considers the soil heat flux(G).Comprehensively measured tiny SE,improved the existing sub-item of SE measurement method to further increase the SE,thus the energy balance had been improved;the approximate instantaneous time window reduced the underestimation of SE and provided a reference for further solving the problem in non-closure of energy balance.The improvement of calculating methodology for the storage term provides methodology and data support for accurate estimation of forest Fs and SE,which is of great significance to improve the reliability of forest carbon and energy flux measurement.