Simulating the effects of climate change and emissions of pollutants on air quality

Abstract

This dissertation is motivated by the challenge of improving air quality and the essential need to enhance our knowledge regarding the potential effects that emissions and climate change could have on the concentration levels of different air pollutants (e.g. particulate matter, ozone, Hg). It is well known that particulate matter (PM), Hg, ozone and other pollutants are subjected to a complex series of common emissions, meteorological processes and photochemical production pathways. Therefore in order to design effective mitigation strategies for improving air quality, it is essential to develop useful quantitative tools, which they can link emissions and meteorology to ambient PM, Hg, etc., concentrations through detailed descriptions of the physics and chemistry of the atmosphere. Chemical Transport models (CTMs) are well suited tools for this purpose.In the first part of this thesis, a three-dimensional CTM, PMCAMx-2008, is applied over Europe to quantify the changes in fine particle (PM2.5) concentration in response to different emission reductions as well as to temperature increase. A summer and a winter simulation period are used, to investigate the seasonal dependence of the PM2.5 response to 50% reductions of sulfur dioxide (SO2), ammonia (NH3), nitrogen oxides (NOx), anthropogenic volatile organic compounds (VOCs) and anthropogenic primary organic aerosol (POA) emissions and also to temperature increases of 2.5 and 5 K. Reduction of NH3 emissions seems to be the most effective control strategy for reducing PM2.5, in both periods, resulting in a decrease of PM2.5 up to 5.1 μg m-3 and 1.8 μg m-3 (5.5% and 4% on average) during summer and winter respectively, mainly due to reduction of ammonium nitrate (NH4NO3) (20% on average in both periods). The reduction of SO2 emissions decreases PM2.5 in both periods, having a significant effect over the Balkans (up to 1.6 μg m-3) during the modeled summer period, mainly due to decrease of sulfate (34% on average over the Balkans). The anthropogenic POA control strategy reduces total OA by 15% during the modeled winter period and 8% in the summer period. The reduction of total OA is higher in urban areas close to its emissions sources. A slight decrease of OA (8% in the modeled summer period and 4% in the modeled winter period) is also predicted after a 50% reduction of VOCs emissions due to the decrease of anthropogenic SOA. The reduction of NOx emissions reduces PM2.5 (up to 3.4 μg m-3) during the summer period, due to a decrease of NH4NO3, causing although an increase of ozone concentration in major urban areas and over Western Europe. Additionally, the NOx control strategy actually increases PM2.5 levels during the winter period, due to more oxidants becoming available to react with SO2 and VOCs. The increase of temperature results in a decrease of PM2.5 in both periods over Central Europe, mainly due to a decrease of NH4NO3 during summer (18%) and fresh POA during wintertime (35%). Significant increases of OA are predicted during the summer due mainly to the increase of biogenic VOC emissions. On the contrary, OA is predicted to decrease in the modeled winter period due to the dominance of fresh POA reduction and the small biogenic SOA contribution to OA. The resulting increase of oxidant levels from the temperature rise lead to an increase of sulfate levels in both periods, mainly over North Europe and the Atlantic Ocean.Our results regarding the strong sensitivity of PM2.5 concentrations to temperature changes support the findings from earlier modeling studies which have concluded that concentrations of PM2.5 are strongly influenced by meteorology. However most of these studies have focused on the overall effect of future climate change, and have not tried to quantify the potential effects of changes of individual meteorological parameters and processes. To address the above issues, we apply PMCAMx-2008 over Europe, trying to examine the effects of various meteorological parameters such as temperature, wind speed, absolute humidity, precipitation, and mixing height on PM2.5 concentrations. Our simulations cover three periods, representative of different seasons (summer, winter, and fall). PM2.5 appears to be more sensitive to temperature changes compared to the rest meteorological parameters in all seasons. PM2.5 generally decreases as temperature increases by 2 K, although the predicted responses vary significantly in space and time, ranging from -700 ng m-3 K-1 (-8% K-1) to 300 ng m-3 K-1 (7% K-1) due to the competing effects on the different PM2.5 species. The predicted decreases of PM2.5 are mainly due to evaporation of ammonium nitrate (decreases by 15% on average) while the higher biogenic emissions and the accelerated gas-phase reaction rates increase the production of organic aerosol (OA) and sulfate having the opposite effect on PM2.5. The predicted responses of PM2.5 to absolute humidity are also quite variable, ranging from -130 ng m-3 %-1 (-1.6% %-1) to 160 ng m-3 %-1 (1.6% %-1) dominated mainly by changes in inorganic PM2.5 species. An increase in absolute humidity favors the partitioning of nitrate to the aerosol phase and increases average PM2.5 during summer and fall. Decreases in sulfate and sea salt levels govern the average PM2.5 response to humidity during winter. A decrease of wind speed (keeping constant the emissions), increases all PM2.5 species due to changes in dispersion and dry deposition. In all periods, average PM2.5 increases by approximately 40 ng m-3 %-1, while the effects are stronger over the polluted areas of the domain. In addition, the wind speed effects only on sea salt emissions could be significant for PM2.5 concentrations over water and in coastal areas (decreases up to 200 ng m-3 %-1 in the winter). Increases in precipitation have a negative effect on PM2.5 (decreases up to 110 ng m-3 %-1) in all periods due to increases in wet deposition of PM2.5 species and their gas precursors. Changes in mixing height have the smallest effects (up to 35 ng m-3 %-1) on PM2.5. Regarding the relative importance of each of the meteorological parameters in a changed future climate, the projected changes in precipitation are expected to have the largest impact on PM2.5 levels during all periods (changes up to 2 μg m-3 in the fall). The expected effects in future PM2.5 levels due to wind speed changes are similar in all seasons and quite close to those resulting from future precipitation changes (up to 1.4 μg m-3). Absolute humidity could potentially lead to large changes in PM2.5 levels mainly in the fall (increases up to 2 μg m-3) due to the dominance of the increased particulate nitrate levels. In the other two periods the expected PM2.5 changes are smaller. Temperature is expected to have a lower impact on future PM2.5 levels compared to the rest meteorological parameters in all seasons, while the effects of mixing height are relatively small.In order to study the overall effect of future climate change on the concentrations of PM2.5 and ozone, we use PMCAMx-2008 as part of a global-regional climate-air pollution modeling system (GRE-CAPS). GRE-CAPS consists of three models spanning the global to the regional scale and is applied over Europe, focusing on Greece. Summertime periods are simulated both for the present (2000s) and the future (2050s) assuming constant anthropogenic pollutant emissions. The future time period investigated is the 2050s, using the IPCC (Intergovernmental Panel on Climate Change) A1B scenario which describes a future world of rapid economic growth, rapid introduction of new and more efficient technologies and balance between fossil fuels and other energy sources. Climate change leads to a decrease of average PM2.5 concentrations over Greece by 1.1 μg m-3 (5%), however the predicted concentration changes are spatially variable and range from -20% to 20% depending on the area. PM2.5 is predicted to decrease in Central Greece, by 1.2 μg m-3 (5.5%) on average, due to the increase in wind speed and the lower absolute humidity, as well as in North Greece and the northern parts of the Aegean. On the contrary, in Crete and the Peloponnese, predicted increases in temperature and absolute humidity lead to higher future PM2.5. Decreased precipitation and wind speed also contribute to the changes in these areas. Ozone concentrations over Greece are increased in the future under A1B scenario by 4.5% on average. Highest changes are predicted in Central Greece (8% on average) while in most areas the predicted increases range between 2-3 ppb. The predicted changes in the daily peak O3 concentrations (1-hr maximum and maximum 8-hr average) are higher. The higher future temperatures determine to a large extent the predicted O3 response. Over Europe, climate change significantly affects PM2.5 levels, with changes range from -25% to 25%. Changes in several meteorological parameters such as precipitation, temperature, wind speed, and absolute humidity may drive the PM2.5 response, having appreciable (and sometimes competing) effects on their concentration levels. Regarding ozone, the model predicts an increase on its concentration in the South Europe (up to 17%) and also in the Balkans, and a decrease in North Europe (up to 10%), with changes in temperature being the dominant factor.The final objective of this research is to provide useful information regarding the response of atmospheric mercury to future climate change. For the purpose of this study the GRE-CAPS modeling system is applied over the eastern United States in order to study the impact of climate change on the concentration and deposition of atmospheric mercury. Summer and winter periods (300 days for each) are simulated, and the present-day model predictions (2000s) are compared to the future ones (2050s) assuming constant emissions. The future climate period is based on the IPCC A2 scenario which describes a very heterogeneous world, with continuously increasing population, self-reliance and preservation of local identities. Climate change affects Hg2+ concentrations in both periods. On average, atmospheric Hg2+ levels are predicted to increase in the future by 3% in the summer and 5% in winter respectively, due to accelerated formation and slower removal. However, the predicted concentration change of Hg2+ is found to vary significantly in space due to regional-scale changes in precipitation, ranging from -30% to 30% during summer and -20% to 40% during winter. Particulate mercury, Hg(p) has a similar spatial response to climate change as Hg2+, while Hg0 levels are not predicted to change significantly. In both periods, the response of mercury deposition to climate change varies spatially with an average predicted increase of 6% during summer and 4% during winter. During summer, deposition increases are predicted mostly in the western parts of the domain while mercury deposition is predicted to decrease in the Northeast and also in many areas in the Midwest and Southeast. During winter mercury deposition is predicted to change from -30% to 50% mainly due to the changes in rainfall and the corresponding changes in wet deposition.