Investigation of atmospheric transport and chemistry of semivolatile organic pollutants using earth system models

The global atmospheric cycling of persistent organic pollutants is complex because of partitioning among phases of the aerosol and revolatilization. Many of the substances are detrimental to human health and the environment. Global dynamical multicompartmental chemistry and transport models are needed to investigate their fate and distributions. The first study investigates climate change influences on the meridional transports of dichlorodiphenyltrichloroethane (DDT) and polychlorinated biphenyls (PCBs) to and from the Arctic by application of the MPI-MCTM model. The objectives are to determine major transport gates along the Arctic Circle, the trends in import and export fluxes, and the relationships between transports and two selected patterns of climate variability, the Arctic Oscillation (AO) and the North Atlantic Oscillation (NAO), under present-day (1970−1999) and future (2070−2099) climate. The pollutants enter the Arctic by passing through the Alaska−Northwest Territories regions, Greenland, the Norwegian Sea−Northwestern Russia, and the Urals−Siberian; whereas they leave the Arctic via the Canadian Arctic and Eastern Russia. DDT import fluxes to the Arctic show a decreasing trend during the present climate, but the trend is expected to change to increasing import fluxes. In contrast, PCB153 export from the Arctic is expected to be increasing in the future climate. The zonal mean meridional fluxes across the Arctic Circle are positively correlated with AO/NAO in winter, corresponding to high net imports when the frequency of positive AO/NAO increases. Under the future climate, there will be an increasing significance of the correlations for DDT while the correlations for PCB153 are expected to weaken. It is concluded that the long-term accumulation trends of other persistent pollutants in the Arctic need to be studied specifically. In the second study, the new module SVOC was developed and coupled to the ECHAM/MESSy Atmospheric Chemistry (EMAC) model to facilitate a continuous development of modeling semivolatile organic compounds through a modular framework. Parameterizations of air−surface mass exchange in the EMAC-SVOC model are similar to those in MPI-MCTM. Other physics parameterizations were improved in the following ways: The gas−particle partitioning is described using poly-parameter linear free energy relationships; and aerosol particle size is discretized into a series of log-normally distributed modes. Through a sensitivity analysis with factor separation technique, the study examined the effects of four factors. These include the aforementioned parameterizations, as well as volatilization and temporal resolution of emissions. The focus here is set on four polycyclic aromatic hydrocarbons (PAHs), i.e., phenanthrene (PHE), pyrene (PYR), fluoranthene (FLT), and benzo(a)pyrene (BaP). The results indicate that seasonal emissions show dominant effects on PHE concentrations, notwithstanding the non-negligible effects from revolatilization. Other species are more sensitive to the change of internal model features. For all PAHs, the degree of model response is more determined by the interactions among factors with their contributions overall being stronger than individual factor contributions. Predicted near-surface concentrations using optimum model configuration were compared against observations. The model underestimates PHE concentrations in the Arctic and tropics but overestimates in the mid-latitudes. FLT and PYR tend to be overestimated over the Arctic and mid-latitudes, and underestimated over the tropics. There is a consistent underestimation of BaP in all regions, with bias increasing from mid-latitudes to the Arctic. The systematic underestimation of BaP concentrations is related to a too fast particulate-phase oxidation by ozone. This issue is addressed in the third study through a better description of BaP multiphase degradation. A new kinetic scheme was developed by considering the dependence of BaP reaction rate on two environmental parameters, that is, temperature (T) and relative humidity (RH). These parameters influence not only the phase state and diffusivity of organic aerosol coating but also the chemical reactivity of BaP. The significance of the new scheme (ROI-T) for distributions and fate was quantitatively assessed by regional (WRF-Chem-BaP) and global scale (EMAC-SVOC) modeling. In comparison to laboratory-based degradation schemes, the ROI-T scheme consistently shows better predictions and improved bias against observations at near-source sites, mid-latitude sites and most substantially at Arctic sites. The new scheme reasonably simulates the effect of low T and RH conditions to increase BaP atmospheric lifetime, leading to a more efficient transport at high altitudes or in a cold season/regions. The scheme can be adopted for modeling the multiphase degradation of other semivolatile organics