Forest response to rising CO2 drives zonally asymmetric rainfall change over tropical land

Abstract

Understanding how anthropogenic CO 2 emissions will influence future precipitation is critical for sustainably managing ecosystems, particularly for drought-sensitive tropical forests. Although tropical precipitation change remains uncertain, nearly all models from the Coupled Model Intercomparison Project Phase 5 predict a strengthening zonal precipitation asymmetry by 2100, with relative increases over Asian and African tropical forests and decreases over South American forests. Here we show that the plant physiological response to increasing CO 2 is a primary mechanism responsible for this pattern. Applying a simulation design in the Community Earth System Model in which CO 2 increases are isolated over individual continents, we demonstrate that different circulation, moisture and stability changes arise over each continent due to declines in stomatal conductance and transpiration. The sum of local atmospheric responses over individual continents explains the pan-tropical precipitation asymmetry. Our analysis suggests that South American forests may be more vulnerable to rising CO 2 than Asian or African forests.

Why does Amazon precipitation decrease when tropical forests respond to increasing CO2?

“…A significant fraction of this dipole of precipitation change is attributed to the physiological response of vegetation to atmospheric CO 2 [Swann et al, 2016;Bonfils et al, 2017;Skinner et al, 2017;Kooperman et al, 2018;Richardson et al, 2018]. Stomatal resistance increases under elevated CO 2 , leading to a decrease in evapotranspiration and less water loss during photosynthesis [Field et al, 1995].…”

“…Stomatal resistance increases under elevated CO 2 , leading to a decrease in evapotranspiration and less water loss during photosynthesis [Field et al, 1995]. The immediate effect of this transpiration decrease is to dry out the boundary layer, and the feed-backs of this are strongest in tropical forest regions [Kooperman et al, 2018], where high initial leaf area limits the sensitivity of evapotranspiration to further increases in canopy biomass, and where boundary layer moisture is heavily dependent on canopy evapotranspiration. Developing a better understanding of how the Amazon rainforest will respond to increased atmospheric CO 2 , and the implications this will have for regional evapotranspiration and boundary layer moisture, is a key aspect of predicting future vegetation-climate interaction.…”

“…A subset of eight modeling centers produced these experiments under CMIP5, permitting intermodel comparisons on how the physiological effect will influence temperature, precipitation, and drought [Swann et al, 2016;Lu et al, 2016;Lemordant et al, 2018;Skinner et al, 2017Skinner et al, , 2018. Only some of these studies, however, have explored the structure of changes within the Amazon specifically [e.g., Abe et al, 2015;Skinner et al, 2017;Richardson et al, 2018;Kooperman et al, 2018]. These papers focus on multidecadal to century-long timescales and attribute the wet Andes, dry Amazon dipole to an increase in the column-integrated export of moisture or moist static energy (MSE) from the Amazon lowlands toward the Andean cordillera [Skinner et al, 2017;Kooperman et al, 2018].…”

“…Only some of these studies, however, have explored the structure of changes within the Amazon specifically [e.g., Abe et al, 2015;Skinner et al, 2017;Richardson et al, 2018;Kooperman et al, 2018]. These papers focus on multidecadal to century-long timescales and attribute the wet Andes, dry Amazon dipole to an increase in the column-integrated export of moisture or moist static energy (MSE) from the Amazon lowlands toward the Andean cordillera [Skinner et al, 2017;Kooperman et al, 2018]. To date, no study has assessed the rapid evolution of the forest's response to CO 2 as it develops in time.…”

“…However, a fast response defined this way still includes soil moisture and land-atmosphere feedbacks acting over several decades, and Richardson et al [2018] employ column-integrated budgets that do not interrogate the vertical structure of the boundary layer or the role it plays in the physiological response, which Kooperman et al [2018] show is important. This therefore leaves open questions about a faster process chain by which these precipitation anomalies initially develop-specifically the timescale over which they become evident and the land-atmosphere interactions or boundary layer feedbacks that accompany them.…”