Joint Session J8.2 On the diurnal evolution of isoprene and hydroxyl radical over tropical forests

Thursday, 5 August 2010: 1:45 PM
Red Cloud Peak (Keystone Resort)
Jordi Vilà-Guerau de Arellano, Wageningen University, Wageningen, Netherlands; and E. G. Patton, T. G. Karl, K. vandenDries, M. C. Barth, and J. J. Orlando

Presentation PDF (518.2 kB)

We investigate diurnal variability of isoprene and related chemical species in the Amazonian region. The dynamics and chemistry of an atmospheric boundary layer are reproduced by two large-eddy simulation codes and a mixed-layer model. These numerical experiments are constrained and evaluated with atmospheric dynamics and chemistry observations available for the same area. The main features of isoprene and methyl-vinyl-ketone are reproduced well, but their evolution raises questions regarding the role of different physical and chemical processes contributing to daily variability. We systematically examine the role of: a) the exchange of thermodynamic variables between the free troposphere and the atmospheric boundary layer (entrainment), b) surface isoprene and nitric oxide emissions, and c) recently proposed modifications to the chemical degradation scheme of isoprene resulting in more efficient recycling of HOx.

The entrainment flux of isoprene and a lumped chemical species (methyl-vinyl-ketone+methacrolein) is found to be equally important as surface isoprene emissions. Varying the initial relationship between the initial isoprene mixing ratio in the boundary layer and that in the overlying free troposphere in the early morning results in a 50% increase/decrease of daytime isoprene mixing ratio within the atmospheric boundary layer. An OH concentration maximum is found during the morning transition, where the magnitude of this peak depends on: a) the onset of surface isoprene emission, b) OH photolysis reaction rates and, c) dilution of boundary layer concentrations through entrainment. The introduction of an OH recycling path increases midday OH concentration variation. Our findings suggest that atmospheric dynamics and chemistry are equally important for interpreting reactant observations and for local and region-scale modeling efforts where turbulence is parameterized.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner