Understanding and quantifying foliar temperature acclimation for Earth System Models

Abstract:

Photosynthesis and respiration on land are the two largest carbon fluxes between the atmosphere and Earth’s surface. The parameterization of these processes represent major uncertainties in the terrestrial component of the Earth System Models used to project future climate change. Research has shown that much of this uncertainty is due to the parameterization of the temperature responses of leaf photosynthesis and autotrophic respiration, which are typically based on short-term empirical responses. Here, we show that including longer-term responses to temperature, such as temperature acclimation, can help to reduce this uncertainty and improve model performance, leading to drastic changes in future land-atmosphere carbon feedbacks across multiple models. However, these acclimation formulations have many flaws, including an underrepresentation of many important global flora. In addition, these parameterizations were done using multiple studies that employed differing methodology. As such, we used a consistent methodology to quantify the short- and long-term temperature responses of maximum Rubisco carboxylation (V_cmax), maximum rate of Ribulos-1,5-bisphosphate regeneration (J_max), and dark respiration (R_d) in multiple species representing each of the plant functional types used in global-scale land surface models. Short-term temperature responses of each process were measured in individuals acclimated for 7 days at one of 5 temperatures (15-35°C). The comparison of short-term curves in plants acclimated to different temperatures were used to evaluate long-term responses. Our analyses indicated that the instantaneous response of each parameter was highly sensitive to the temperature at which they were acclimated. However, we found that this sensitivity was larger in species whose leaves typically experience a greater range of temperatures over the course of their lifespan. These data indicate that models using previous acclimation formulations are likely incorrectly simulating leaf carbon exchange responses to future warming. Therefore, our data, if used to parameterize large-scale models, are likely to provide an even greater improvement in model performance, resulting in more reliable projections of future carbon-clime feedbacks.