Spatially asynchronous changes in strength and stability of terrestrial net ecosystem productivity

Abstract

<p><strong>Abstract.</strong> Multiple lines of evidence have demonstrated the persistence of global land carbon (C) sink during the past several decades. However, both annual net ecosystem productivity (NEP) and its inter-annual variation (IAV<sub>NEP</sub>) keep varying over space. Thus, identifying local indicators for the spatially varying NEP and IAV<sub>NEP</sub> is critical for locating the major and sustainable C sinks on the land. Here, based on a machine-learning-derived database, we first showed that the variations of NEP and IAV<sub>NEP</sub> are spatially asynchronous. Then, based on daily NEP observations from eddy covariance sites, we found robust logarithmic correlation between annual NEP and ratio of total CO<sub>2</sub> exchanges during net uptake (<i>U</i>) and release (<i>R</i>) periods (i.e., <i>U/R</i>). The cross-site variation of mean annual NEP can be linearly indicated by ln(<i>U/R</i>), while the spatial distribution of IAV<sub>NEP</sub> was well indicated by the slope (i.e., <i>β</i>) of the demonstrated logarithmic correlation. Among biomes, for example, forests and croplands had the largest <i>U/R</i> ratio (1.06 ± 0.83) and <i>β</i> (473 ± 112 g C m<sup>−2</sup> yr<sup>−1</sup>), indicating the highest NEP and IAV<sub>NEP</sub> in forests and croplands, respectively. We further showed that the spatial variations of NEP and IAV<sub>NEP</sub> were both underestimated by the machine-learning-based and process-based global models. Overall, this study underscores the asynchronously changes in the strength and stability of land C sinks over space, and provides two simple local indicators for their intricate spatial variations. These indicators could be helpful for locating the persistent terrestrial C sinks and provides valuable constraints for improving the simulation of land-atmospheric C exchanges.</p>

Research challenges and opportunities for using big data in global change biology

“…This ratio is relatively invariant even with a larger sample size of flux sites (2.71 ± 1.61; Cui, Bian, et al, 2020; Ciu, Weng, et al, 2020). It is also interesting that the total annual net ecosystem CO 2 exchange at each site can be described as a logarithmic function of the ratio between total CO 2 exchange over carbon uptake and release periods (Cui, Bian, et al, 2020; Ciu, Weng, et al, 2020). These properties could be useful benchmarks for evaluating the simulated terrestrial ecosystem CO 2 exchange by Earth system models.…”

“…For example, Churkina, Schimel, Braswell, and Xiao (2005) have discovered the convergent ratio between mean daily CO 2 exchange over carbon uptake and release periods as 2.73 ± 1.08 across 28 sites (Churkina et al, 2005). This ratio is relatively invariant even with a larger sample size of flux sites (2.71 ± 1.61; Cui, Bian, et al, 2020;Ciu, Weng, et al, 2020). It is also interesting that the total annual net ecosystem CO 2 exchange at each site can be described F I G U R E 3 Using big data on the site level to improve process-based understanding of ecosystem functions.…”

“…For example, Churkina, Schimel, Braswell, and Xiao (2005) have discovered the convergent ratio between mean daily CO 2 exchange over carbon uptake and release periods as 2.73 ± 1.08 across 28 sites (Churkina et al, 2005). This ratio is relatively invariant even with a larger sample size of flux sites (2.71 ± 1.61; Cui, Bian, et al, 2020;Ciu, Weng, et al, 2020). It is also interesting that the total annual net ecosystem CO 2 exchange at each site can be described The hierarchical structure of variables aboveground and belowground including tree number, diameter breast at 1.3 m height (DBH), litter mass, soil organic carbon (SOC) in different layers, topography (i.e., elevation), soil microclimate (i.e., soil moisture and temperature) in the 20 ha Tiantongshan forest dynamics plot (29.8°N, 121.8°E).…”

“…(d) Probability density of experimental responses of net leaf photosynthesis to different global change factors (Walker et al, 2014). As shown in Figure 2, hundreds of plant species have been studied in different global change experiments (Figure 2a (Cui, Bian, et al, 2020;Ciu, Weng, et al, 2020). Another analysis on root traits from 369 plant species has revealed that root diameter has the strongest impact on root trait variation across species (Ma et al, 2018).…”

“…As shown in Figure 2d, net leaf photosynthesis of terrestrial plants is significantly enhanced by elevated CO 2 and nitrogen addition but reduced by drought. Although the response directions of leaf traits are divergent among species, a recent analysis on 515 species has shown robust leaf trait relationships across species under global environmental changes (Cui, Bian, et al, 2020; Ciu, Weng, et al, 2020). Another analysis on root traits from 369 plant species has revealed that root diameter has the strongest impact on root trait variation across species (Ma et al., 2018).…”