Dry deposition of ozone in the earth system: controlling processes and local-to-global impacts. Figure illustrated by Simmi Sinha.

Tropospheric ozone is an air pollutant — harmful to both humans and vegetation — and a potent greenhouse gas, as well as the main source of the hydroxyl radical, which controls how long reactive greenhouse gases and toxic pollutants stay in the atmosphere. While the last thirty years of tropospheric ozone research focused on the role of sources, my research focuses on one of the main sinks, dry deposition. Dry deposition of ozone happens when turbulence transports ozone to the surface and surface-mediated chemistry removes ozone from the atmosphere. Ozone depositing through plant stomata (the small pores on leaves used for gas exchange) can be injurious to vegetation and influence local-to-global carbon and water cycling. Nonstomatal deposition, occurring to snow-covered surfaces, leaf cuticles, soil, or bodies of water, is highly uncertain, but an important fraction of the total ozone dry deposition. 

Current research priorities & relevant work

Advance understanding of dry deposition

Screen Shot 2019-03-13 at 9.56.29 AMWith one of the longest datasets of ozone eddy covariance fluxes, I show that there are strong interannual variations in ozone deposition velocity, a measure of the efficiency of the ozone removal by the surface, at Harvard Forest (Clifton et al. 2017). While stomatal uptake is often considered the driver of ozone dry deposition, I use process modeling to show that stomatal uptake does not drive the interannual variability in ozone deposition velocity at Harvard Forest (Clifton et al. 2017). Instead, I suggest that the variability is driven by increases in ozone uptake by soil when soil is dry (Clifton et al. 2019). For more information on ozone depositional processes, as well as a current synthesis of modeling and observations, see Clifton, Fiore, Massman, et al. (2020)

Constrain the influence of dry deposition on air pollution 

As illustrated by the above figure, the strong observed interannual variability in ozone deposition velocity at Harvard Forest is not simulated by a global chemical transport model using the widely used Wesely (1989) dry deposition scheme. Given that simulated ambient ozone concentrations are sensitive to ozone deposition velocities, my findings suggest that using atmospheric chemistry models that employ the Wesely scheme to interpret observed year-to-year changes in ozone pollution may lead to a model overemphasis of the role of ozone precursor emissions.

The NOAA GFDL global chemistry-climate model now has dry deposition of some reactive trace gases and aerosols simulated in the land component of the model (Paulot et al. 2018), allowing for stomatal and nonstomatal deposition to be coupled to terrestrial carbon and water cycling, land use, and vegetation dynamics. I updated the dry deposition scheme in the GFDL model for ozone to be consistent with current understanding, and recently published a paper investigating the effect of dynamic representation of ozone dry deposition on ozone pollution. In Clifton, Paulot, Fiore, et al. (2020), I show that nonstomatal ozone dry deposition is important for ozone pollution, including hemisphere-scale levels and extremes.

Evaluate the importance of meteorology-plant functioning interactions for ozone plant damage 

Meteorology and plant functioning are often overlooked in considering how ozone damages ecosystems. Here I am using (i) a multilayer canopy large eddy simulation (LES) model that uniquely resolves turbulence above and inside the forest canopy (Patton et al. 2016) and (ii) the new version of the global NOAA GFDL model as described above to investigate interactions between ozone, meteorology, and stomatal aperture as relevant for ozone plant damage.

Screen Shot 2020-10-23 at 9.22.50 AM

In terms of (i), the above figure shows horizontal variability in instantaneous ozone mixing ratio and leaf uptake in a ~2km x 2km forest under moderately unstable atmospheric boundary layer conditions as simulated by the NCAR LES. Model resolution is 2 m, with each grid cell in the canopy solving the equations of the multilayer canopy model. Structure in the variability on the scale of the boundary layer height and forest canopy is apparent in both within-canopy ozone and leaf uptake, but the variability differs between the two quantities. At heights much higher than the canopy, variability in ozone occurs at mostly scales on the order of the boundary layer height. In general, ozone variability reflects sinking structures enriched in ozone and rising structures depleted in ozone due to dry deposition in the canopy. Leaf uptake variability reflects turbulence’s influences on humidity and temperature and thus leaf energy balance, photosynthesis, stomatal aperture, and the formation of thin water films on leaves. Stay tuned for what this turbulence-induced variability in ozone and leaf uptake means for ozone dry deposition and plant damage. 

Screen Shot 2020-10-23 at 9.44.54 AM

In terms of (ii), in Clifton, Lombardozzi, et al., 2020, I use the NOAA GFDL model to show interannual variability (IAV) in stomatal conductance is key for the cumulative stomatal uptake of ozone, which indicates the amount of ozone entering the leaf over time available to cause physiological damage. IAV in stomatal conductance is more important than IAV in ozone air pollution (see above figure). My findings imply that the most ozone damage happens in years when ecosystems are most productive, challenging widely used metrics suggesting that the most ozone damage occurs in the highest ozone years.