Divergent historical GPP trends among state-of-the-art multi-model simulations and satellite-based products

Understanding historical changes in gross primary productivity (GPP) is essential for better predicting the future global carbon cycle. However, the historical trends of terrestrial GPP, owing to the CO2 fertilization effect, climate, and land-use change, remain largely uncertain. Using long-term satellite-based near-infrared radiance of vegetation (NIRv), a proxy for GPP, and multiple GPP datasets derived from satellite-based products, Dynamic Global Vegetation Model (DGVM) simulations, and machine learning techniques, here we comprehensively investigated their trends and analyzed the causes for any discrepancies during 1982–2015. Although spatial patterns of climatological annual GPP from all products and NIRv are highly correlated (r > 0.84), the spatial correlation coefficients of trends between DGVM GPP and NIRv significantly decreased (with the ensemble mean of r = 0.49) and even the spatial correlation coefficients of trends between other GPP products and NIRv became negative. By separating the global land into the tropics plus extra-tropical southern hemisphere (Trop+SH) and extra-tropical northern hemisphere (NH), we found that, during 1982–2015, simulated GPP from most of the models showed a stronger increasing trend over Trop+SH than NH. In contrast, the satellite-based GPP products indicated a substantial increase over NH. Mechanistically, model sensitivity experiments indicated that the increase of annual GPP was dominated by the CO2 fertilization effect (Global: 83.9 %), albeit a large uncertainty in magnitude among individual simulations. However, the spatial distribution of inter-model spreads of GPP trends resulted mainly from climate and land-use change rather than CO2 fertilization effect. Trends after 2000 were different from the full time-series, showing that satellite-based GPP products suggested weakened rising trends over NH and even significantly decreasing trends over Trop+SH, while the trends from DGVMs kept increasing. The inconsistencies are very likely caused by the contrasting performances between satellite-derived and DGVM simulated vegetation structure parameter (leaf area index, LAI). Therefore, the uncertainty in satellite-based GPP products induced by highly uncertain LAI data in the tropics undermines their roles in assessing the performance of DGVM simulations and understanding the changes of global carbon sinks.

Thirty-eight years of CO₂ fertilization have outpaced growing aridity to drive greening of Australian woody ecosystems

Climate change is projected to increase the imbalance between the supply (precipitation) and atmospheric demand for water (i.e. increased potential evapotranspiration), stressing plants in water-limited environments. Plants may be able to offset increasing aridity because rising CO2 increases water-use-efficiency. CO2 fertilization has also been cited as one of the drivers of the widespread ‘greening’ phenomenon. However, attributing the size of this CO2 fertilization effect is complicated, due in part to a lack of long-term vegetation monitoring and interannual to decadal-scale climate variability. In this study we asked the question, how much has CO2 contributed towards greening? We focused our analysis on a broad aridity gradient spanning eastern Australia’s woody ecosystems. Next we analysed 38-years of satellite remote sensing estimates of vegetation greenness (normalized difference vegetation index, NDVI) to examine the role of CO2 in ameliorating climate change impacts. Multiple statistical techniques were applied to separate the CO2-attributable effects on greening from the changes in water supply and atmospheric aridity. Widespread vegetation greening occurred despite a warming climate, increases in vapor pressure deficit, and repeated record-breaking droughts and heatwaves. Between 1982–2019 we found that NDVI increased (median 11.3 %) across 90.5 % of the woody regions. After masking disturbance effects (e.g. fire), we statistically estimated an 11.7 % increase in NDVI attributable to CO2, broadly consistent with a hypothesized theoretical expectation of an 8.6 % increase in water-use-efficiency due to rising CO2. In contrast to reports of a weakening CO2 fertilization effect, we found no consistent temporal change in the CO2 effect. We conclude rising CO2 has mitigated the effects of increasing aridity, repeated record-breaking droughts, and record-breaking heat waves in eastern Australia. However, we were unable to determine whether trees or grasses were the primary beneficiary of the CO2 induced change in water-use-efficiency, which has implications for projecting future ecosystem resilience. A more complete understanding of how CO2 induced changes in water-use-efficiency affect trees and non-tree vegetation is needed.

The quasi-equilibrium framework revisited: analyzing long-term CO₂ enrichment responses in plant–soil models

Elevated carbon dioxide (CO2) can increase plant growth, but the magnitude of this CO2 fertilization effect is modified by soil nutrient availability. Predicting how nutrient availability affects plant responses to elevated CO2 is a key consideration for ecosystem models, and many modeling groups have moved to, or are moving towards, incorporating nutrient limitation in their models. The choice of assumptions to represent nutrient cycling processes has a major impact on model predictions, but it can be difficult to attribute outcomes to specific assumptions in complex ecosystem simulation models. Here we revisit the quasi-equilibrium analytical framework introduced by Comins and McMurtrie (1993) and explore the consequences of specific model assumptions for ecosystem net primary productivity (NPP). We review the literature applying this framework to plant–soil models and then analyze the effect of several new assumptions on predicted plant responses to elevated CO2. Examination of alternative assumptions for plant nitrogen uptake showed that a linear function of the mineral nitrogen pool or a linear function of the mineral nitrogen pool with an additional saturating function of root biomass yield similar CO2 responses at longer timescales (>5 years), suggesting that the added complexity may not be needed when these are the timescales of interest. In contrast, a saturating function of the mineral nitrogen pool with linear dependency on root biomass yields no soil nutrient feedback on the very-long-term (>500 years), near-equilibrium timescale, meaning that one should expect the model to predict a full CO2 fertilization effect on production. Secondly, we show that incorporating a priming effect on slow soil organic matter decomposition attenuates the nutrient feedback effect on production, leading to a strong medium-term (5–50 years) CO2 response. Models incorporating this priming effect should thus predict a strong and persistent CO2 fertilization effect over time. Thirdly, we demonstrate that using a “potential NPP” approach to represent nutrient limitation of growth yields a relatively small CO2 fertilization effect across all timescales. Overall, our results highlight the fact that the quasi-equilibrium analytical framework is effective for evaluating both the consequences and mechanisms through which different model assumptions affect predictions. To help constrain predictions of the future terrestrial carbon sink, we recommend the use of this framework to analyze likely outcomes of new model assumptions before introducing them to complex model structures.