Biome-scale temperature sensitivity of ecosystem respiration revealed by atmospheric CO2 observations

Abstract

The temperature sensitivity of ecosystem respiration regulates how the terrestrial carbon sink responds to a warming climate but has been difficult to constrain observationally beyond the plot scale. Here we use observations of atmospheric CO2 concentrations from a network of towers together with carbon flux estimates from state-of-the-art terrestrial biosphere models to characterize the temperature sensitivity of ecosystem respiration, as represented by the Arrhenius activation energy, over various North American biomes. We infer activation energies of 0.43 eV for North America and 0.38 eV to 0.53 eV for major biomes therein, which are substantially below those reported for plot-scale studies (approximately 0.65 eV). This discrepancy suggests that sparse plot-scale observations do not capture the spatial-scale dependence and biome specificity of the temperature sensitivity. We further show that adjusting the apparent temperature sensitivity in model estimates markedly improves their ability to represent observed atmospheric CO2 variability. This study provides observationally constrained estimates of the temperature sensitivity of ecosystem respiration directly at the biome scale and reveals that temperature sensitivities at this scale are lower than those based on earlier plot-scale studies. These findings call for additional work to assess the resilience of large-scale carbon sinks to warming.

Mountain Peatlands and Drought: Carbon Cycling in the Pyrenees Amidst Global Climate Change

“…Mean annual temperatures in the peatland increased by 1°since the early 1960ies and maximum temperatures by 8°C for instance. Knowing the sensitivity of ecosystem respiration to temperature (Ben & Thompson, 2010;Sun et al, 2023), it is likely that the ER increased in the last 50 years. It has been observed that drier conditions in summer were associated with lower carbon uptake in Canadian boreal peatlands (McDonald et al, 2023).…”

Monitoring and modelling European carbon fluxes using atmospheric observations

“…Although we lack the observations to verify this, we hypothesise the GPP response is dominated by the atmospheric stress through VPD, while the soil moisture limitation that would also affect TER is smaller. Biosphere models vary strongly in simulations of such a TER response (Clark et al, 2011;Haynes et al, 2020;Peters et al, 2018) and recent work suggests a general overestimate of temperature sensitivity (Sun et al, 2023). The good correspondence to atmospheric data nevertheless indicates a good NEE drought response in SiB4 which, together with a reasonable response in GPP, also suggests a reasonable TER response of SiB4.…”

“…The GPP impact on atmospheric CO 2 is typically larger and more variable than that of TER in summer (Piao et al, 2020;Tolk et al, 2009).Our study nevertheless suggests also a role of TER and soil moisture in the drought response, as we find a strong reduction of GPP which, according to the atmospheric CO 2 mole fraction constraints, does not fully balance observed NEE reductions without also considering a TER reduction. However, contrary to GPP, which can be estimated from satellite products, TER cannot currently be quantified on a large scale (although recently advances in the quantification of the temperature sensitivity of TER have been made (Sun et al, 2023))). Locally, our EC observations indeed show reduced TER and reduced GPP due to the drought in the 4 4 99…”

“…Although GPP and TER are separate processes with independent responses, they often co-vary as they are sensitive to similar drivers and reduced GPP nearly always coincides with a reduction in ecosystem respiration (Baldocchi, 2008;Buttlar et al, 2018;Ciais et al, 2005;Lasslop et al, 2010;Smith et al, 2020a;Sun et al, 2023). These countering effects reduce the impact of the drought on the net ecosystem exchange.…”