Search

Mainland and maritime Southeast Asia is home to more than 655 million people, representing nearly 10 % of the global population. The dry season in this region is typically associated with intense biomass burning activity, which leads to a significant increase in surface air pollutants that are harmful to human health, including ozone (O3). Latitude-based differences in the dry season and land use distinguish two regional biomass burning regimes: (1) burning on the peninsular mainland peaking in March and (2) burning across Indonesia peaking in September. The type and amount of material burned in each regime impact the emissions of nitrogen oxides (NOx = NO + NO2) and volatile organic compounds (VOCs), which combine to produce ozone. Here, we use the nested GEOS-Chem atmospheric chemistry transport model (horizontal resolution of 0.25∘ × 0.3125∘), in combination with satellite observations from the Ozone Monitoring Instrument (OMI) and ground-based observations from Malaysia, to investigate ozone photochemistry over Southeast Asia in 2014. Seasonal cycles of tropospheric ozone columns from OMI and GEOS-Chem peak with biomass burning emissions. Compared to OMI, the model has a mean annual bias of −11 % but tends to overestimate tropospheric ozone near areas of seasonal fire activity. We find that outside these burning areas, the underlying photochemical environment is generally NOx-limited and dominated by anthropogenic NOx and biogenic non-methane VOC emissions. Pyrogenic emissions of NOx play a key role in photochemistry, shifting towards more VOC-limited ozone production and contributing about 30 % of the regional ozone formation potential during both biomass burning seasons. Using the GEOS-Chem model, we find that biomass burning activity coincides with widespread ozone exposure at levels that exceed world public health guidelines, resulting in about 260 premature deaths across Southeast Asia in March 2014 and another 160 deaths in September. Despite a positive model bias, hazardous ozone levels are confirmed by surface observations during both burning seasons.

Isoprene oxidation schemes vary greatly among gas-phase chemical mechanisms, with potentially significant ramifications for air quality modeling and interpretation of satellite observations in biogenic-rich regions. In this study, in situ observations from the 2013 SENEX mission are combined with a constrained 0-D photochemical box model to evaluate isoprene chemistry among five commonly used gas-phase chemical mechanisms: CB05, CB6r2, MCMv3.2, MCMv3.3.1, and a recent version of GEOS-Chem. Mechanisms are evaluated and inter-compared with respect to formaldehyde (HCHO), a high-yield product of isoprene oxidation. Though underestimated by all considered mechanisms, observed HCHO mixing ratios are best reproduced by MCMv3.3.1 (normalized mean bias = −15%), followed by GEOS-Chem (−17%), MCMv3.2 (−25%), CB6r2 (−32%) and CB05 (−33%). Inter-comparison of HCHO production rates reveals that major restructuring of the isoprene oxidation scheme in the Carbon Bond mechanism increases HCHO production by only ∼5% in CB6r2 relative to CB05, while further refinement of the complex isoprene scheme in the Master Chemical Mechanism increases HCHO production by ∼16% in MCMv3.3.1 relative to MCMv3.2. The GEOS-Chem mechanism provides a good approximation of the explicit isoprene chemistry in MCMv3.3.1 and generally reproduces the magnitude and source distribution of HCHO production rates. We analytically derive improvements to the isoprene scheme in CB6r2 and incorporate these changes into a new mechanism called CB6r2-UMD, which is designed to preserve computational efficiency. The CB6r2-UMD mechanism mimics production of HCHO in MCMv3.3.1 and demonstrates good agreement with observed mixing ratios from SENEX (−14%). Improved simulation of HCHO also impacts modeled ozone: at ∼0.3 ppb NO, the ozone production rate increases ∼3% between CB6r2 and CB6r2-UMD, and rises another ∼4% when HCHO is constrained to match observations.

The Framework for 0-D Atmospheric Modeling (F0AM) is a flexible and user-friendly MATLAB-based platform for simulation of atmospheric chemistry systems. The F0AM interface incorporates front-end configuration of observational constraints and model setup, making it readily adaptable to simulation of photochemical chambers, Lagrangian plumes, and steady-state or time-evolving solar cycles. Six different chemical mechanisms and three options for calculation of photolysis frequencies are currently available. Example simulations are presented to illustrate model capabilities and, more generally, highlight some of the advantages and challenges of 0-D box modeling.

The chemical link between isoprene and formaldehyde (HCHO) is a strong, nonlinear function of NOx (i.e., NO + NO2). This relationship is a linchpin for top-down isoprene emission inventory verification from orbital HCHO column observations. It is also a benchmark for overall photochemical mechanism performance with regard to VOC oxidation. Using a comprehensive suite of airborne in situ observations over the southeast US, we quantify HCHO production across the urban–rural spectrum. Analysis of isoprene and its major first-generation oxidation products allows us to define both a "prompt" yield of HCHO (molecules of HCHO produced per molecule of freshly emitted isoprene) and the background HCHO mixing ratio (from oxidation of longer-lived hydrocarbons). Over the range of observed NOx values (roughly 0.1–2 ppbv), the prompt yield increases by a factor of 3 (from 0.3 to 0.9 ppbv ppbv−1), while background HCHO increases by a factor of 2 (from 1.6 to 3.3 ppbv). We apply the same method to evaluate the performance of both a global chemical transport model (AM3) and a measurement-constrained 0-D steady-state box model. Both models reproduce the NOx dependence of the prompt HCHO yield, illustrating that models with updated isoprene oxidation mechanisms can adequately capture the link between HCHO and recent isoprene emissions. On the other hand, both models underestimate background HCHO mixing ratios, suggesting missing HCHO precursors, inadequate representation of later-generation isoprene degradation and/or underestimated hydroxyl radical concentrations. Detailed process rates from the box model simulation demonstrate a 3-fold increase in HCHO production across the range of observed NOx values, driven by a 100 % increase in OH and a 40 % increase in branching of organic peroxy radical reactions to produce HCHO.