A major source of radical species in the atmosphere is through the oxidation reactions of biogenic volatile organic compounds that are emitted from vegetation. Tropical rainforests are responsible for over half of such biogenic species that are emitted into the atmosphere and the local, regional and global impacts of their subsequent oxidation mechanisms are currently not well understood. Further, with tropical forests being removed to make way for new land uses (such as oil palm plantations), the subsequent change in the quantity and type of biogenic emissions into the atmosphere could have far-reaching impacts.
The Oxidant, Particle and Photochemical Processes (OP3) field campaign conducted at Bukit Atur in Malaysian Borneo in 2008, aimed to address some of the uncertainties that currently exist surrounding the impact of forested regions on atmospheric chemistry. In particular, this project aims to predict the concentrations of OH, HO2 and RO2 radicals at Bukit Atur in Borneo during April and July of 2008, using a near-explicit photochemical box model with 15,000 reactions and 7,200 species. The model is constrained using observations made during the two experimental campaigns, and used to compare with radical measurements.
In agreement with previous studies involving tropical forests, it was found that the standard model based on the Master Chemical Mechanism (MCM v3.1) underestimates the observed concentrations of OH by a factor of 0.5 on average and overestimates HO2 concentrations by a factor of 2. The results for RO2 were mixed with some days over-predicted and some under-predicted. The implementation of some new theoretically derived reaction pathways without the isoprene degradation scheme improved the predicted OH concentration (modelled:measured ratio improved
to 0.3), but did not improve the HO2 estimation (modelled: measured ratio changed to 2.5). It was found that the modelled: measured discrepancy was better on days when the VOC:NOX ratio was lower, suggesting that even with the updated isoprene scheme, days with high biogenic concentrations are not well represented in the model. A rate of production analysis also confirmed that days where modelled OH agreed best with measurements were dominated more by NOX reactions, and less by for instance, Criegee biradical reactions, the latter an indication of biogenic influence.
It seems likely from the results from this study (and others) that the suggested alterations to the isoprene chemistry scheme are incomplete, as they do not completely remove the model discrepancy in the predicted OH and HO2 concentrations. This work provides a useful contribution to the understanding of radical species production in tropical forests and provides more data in this area of research. However, this project also identifies that more research is required, particularly in the elucidation of isoprene degradation in the atmosphere, but also with issues such as the dry deposition rates of key intermediates in the model mechanism.
