Abstract
Abstract. Biosphere–atmosphere interactions strongly influence the chemical composition of the atmosphere. Simulating these interactions at a detailed process-based level has traditionally been computationally intensive and resource prohibitive, commonly due to complexities in calculating radiation and light at the leaf level within plant canopies. Here we describe a surrogate canopy physics model based on the MEGAN3 detailed canopy model parameterized using a statistical learning technique. This surrogate canopy model is specifically designed to rapidly calculate leaf-level temperature and photosynthetically active radiative (PAR) for use in large-scale chemical transport models (CTMs). Our surrogate model can reproduce the dominant spatiotemporal variability of the more detailed MEGAN3 canopy model to within 10 % across the globe. Implementation of this surrogate model into the GEOS-Chem CTM leads to small local changes in ozone dry deposition velocities of less than 5 % and larger local changes in isoprene emissions of up to ∼40 %, though annual global isoprene emissions remain largely consistent (within 5 %). These changes to surface–atmosphere exchange lead to small changes in surface ozone concentrations of ±1 ppbv, modestly reducing the northern hemispheric ozone bias, which is common to many CTMs, here from 8 to 7 ppbv. The use of this computationally efficient surrogate canopy model drives emissions of isoprene and concentrations of surface ozone closer to observationally constrained values. Additionally, this surrogate model allows for the further development and implementation of leaf-level emission factors in the calculation of biogenic emissions in the GEOS-Chem CTM. Though not the focus of this work, this ultimately enables a complete implementation of the MEGAN3 emissions framework within GEOS-Chem, which produces 570 Tg yr−1 of isoprene for 2012.
Highlights
The biosphere plays an important role in modulating the abundance and variability of trace gases and aerosol in the atmosphere
We describe a novel method for simulating canopy physics relevant to atmospheric chemistry at very low computational cost
Our surrogate canopy model is based on the detailed canopy model in the MEGAN3 code base and simplified using a statistical learning technique for the determination of variable importance
Summary
The biosphere plays an important role in modulating the abundance and variability of trace gases and aerosol in the atmosphere. Simulating biosphere–atmosphere interactions necessitates a detailed representation of physical, chemical, and biological processes that occur at the scale of an individual plant This is typically achieved by integrating a set of energy and radiative balance equations vertically throughout a canopy (e.g., Ashworth et al, 2015, 2016; Goudriaan and Laar, 1994). This sort of physical model of the canopy calculates the environmental parameters that drive the biological and chemical processes, which impact the atmospheric fluxes of trace gases and aerosol (Guenther et al, 2012; Lamb et al, 1996). Our reduced model reproduces the output of the more detailed vegetation model well, without the large computational overhead
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
