The Spatiotemporal Variability Of The Relationship Between Global Forest Age And Vegetation Productivity

Vegetation productivity can vary with forest age,but the nature of the trend and the underlying mechanisms are not well understood.At the same time,vegetation productivity is an important component that regulates the feedback between terrestrial ecosystems and climate.Therefore,understanding the dynamic changes of vegetation productivity with forest age is vital for assessing the function of forest ecosystems in carbon sink.In recent years,with the continuous development of simulation methods and monitoring technologies,numerous studies have begun to explore the spatiotemporal variability of vegetation productivity.However,most of these studies focus on the impact of climate and other environmental factors but ignore the role of forest age in regulating vegetation productivity.To address this issue,this study used multi-source observation data including global forest age data,satellite-based data,eddy flux observations,in situ observation data,tree ring data,etc.,to explore the relationship between global forest age and the spatiotemporal variation of vegetation productivity.First,we analyzed the spatial distribution of the global forest age.At the same time,based on the satellite products and flux observation data,this study analyzed the relationship between forest age and vegetation productivity.Then we used the long-term records of tree rings to explore the temporal stability of vegetation productivity at different forest ages.Finally,to evaluate the relationship between forest age and vegetation carbon storage simulated by the Earth system models,a database was constructed based on in situ observation data.The main results are as follows:(1)Based on the global forest age data sets,this study analyzed the spatial distribution characteristics of global forest age over 2000-2010.Results showed that forests older than 140 years were concentrated in South America,accounting for 65.2%.Young forests(with an age class of 1-30 years)were mainly distributed in temperate forest regions of eastern North America,subtropical forest regions of eastern Asia,and tropical forest regions south of the Congo Basin in Africa.More than 90%of the forests older than 130 years were evergreen broad-leaved forests.(2)By analyzing the spatiotemporal variation of global forest productivity,it is found that the effect of forest age on vegetation productivity is nonlinear.Based on global vegetation productivity data from satellite remote sensing products,this study further quantified the relative importance of environmental factors(temperature,precipitation,radiation,nitrogen deposition)and forest age on the spatiotemporal variation of vegetation productivity using a random forest algorithm.The results showed that the effect of forest age on spatial variation of vegetation productivity was only weaker than solar radiation and nitrogen deposition(%IncMSE=109.9%),and the effect of forest age on interannual variability of vegetation productivity was only weaker than solar radiation(%IncMSE=50.7%).In tropical forests,the effect of forest age on spatial variation of vegetation productivity was only weaker than nitrogen deposition(%IncMSE=136.1%),and the effect of forest age on interannual variation of vegetation productivity was greater than that of other environmental factors(%IncMSE=77.3%).At the same time,based on 41 forest eddy-flux stations,it was found that the spatial distribution of stand age and vegetation productivity showed a nonlinear relationship,which first increased and then decreased.The relationship between stand age and interannual variation of vegetation productivity showed a bimodal curve pattern.By extracting the parameters from the seasonal variation curve of vegetation productivity,results showed that the regulation effect of stand age on the spatiotemporal variation of vegetation productivity was mainly dominated by physiological parameters(peak productivity and mean productivity)rather than phenological parameters(carbon uptake period).With the increase of stand age,peak productivity and mean productivity tended to balance,and the ratio of them converged to 0.62.(3)In extremely dry soils,vegetation productivity of different forest ages responded to asymmetric day and night warming in the same way.Based on long-term records of tree ring data from 2,294 sites,we calculated the temporal stability of wood productivity at different forest ages.Results showed that the change rate of temporal stability during 1960-1990 was less than that during 1990-2013.From 1990 to 2013,the change rates of the temporal stability of wood productivity with the age classes of 1-100,100-200,200-300 and>300 years were-2.02 year-1,-1.38 year-1,-1.39 year-1 and-0.56 year-1,respectively.Combined with the meteorological data,it was found that when soil moisture dries to a threshold,diurnal asymmetric warming has a negative impact on vegetation productivity.Further analysis showed that the allocation of biomass from wood to leaf was an important mechanism that caused daytime warming to limit wood productivity.(4)The current earth system model cannot capture the non-linear increase in vegetation carbon storage with forest age.Based on biomass measurements from 1145 in situ forest sites,this study found that both vegetation carbon storage(Cveg)and biomass allocation were non-linearly regulated by forest age.The Cveg and woody biomass fraction(fw)show nonlinear relationships with forest age.The observations constrained an exponential relationship between Cveg and fw(i.e.,Cveg=bea·fw).Such exponential relationship also existed in the current 15 Earth system models,but the coefficient b was systematically lower in the Earth system models(0.08 ± 0.11)than the observations(0.28).The exponential relationship based on observations provided one useful benchmark for earth system models when they implement the age-dependent processes in the future.Overall,this study explored the influence of forest age on the spatiotemporal variability of vegetation productivity based on multi-source observational data and evaluated the simulation of Cveg from the current earth system model without considering the stand age.The above results reveal the nonlinear regulation pattern of forest age on vegetation productivity and carbon storage,which is helpful to deepen the understanding of the relationship between vegetation productivity and stand age and provide theoretical support for improving forest dynamic prediction and scientific forest management.This study emphasizes the importance of stand age in predicting vegetation carbon storage and provides a theoretical basis for the improvement of the process and parameters of the terrestrial carbon cycle model.