| Περίληψη: | Secondary inorganic aerosol components (sulfate, nitrate, and ammonium) constitute a significant part of atmospheric PM mass and impact aerosol acidity. Aerosol nitrate and ammonium are mainly formed through gas-to-particle conversion processes of nitric acid (HNO3) and ammonia (NH3), while for sulfate the multiphase oxidation of sulfur dioxide (SO2) is mainly responsible. Chemical transport models (CTMs) have the tendency to overpredict fine nitrate aerosol levels in both U.S and Europe. This in turn can also affect fine ammonium aerosol predictions since in fine PM fraction nitrate partitions to the aerosol phase together with ammonium. Responsible for these errors in CTMs predictions may be the effects of non-volatile cations (NVCs) on fine PM levels and composition, which usually are not quantified correctly due to urban dust levels underestimation, but also the errors in the prediction of aerosol acidity. In addition, several Earth System Models (ESMs) usually neglect inorganic aerosol thermodynamics due to the additional computational burden. In this work, we use the PMCAMx CTM to quantify the effects of NVCs on fine PM levels and composition and to gain a better understanding of aerosol acidity and its various dependences. Also, ISORROPIA-lite, a simplified and lean version of the widely used ISORROPIA-II inorganic aerosol thermodynamics model is presented. Compared to its parent model, ISORROPIA-lite can simulate the effects of secondary organic aerosol water on aerosol thermodynamics. These effects are also examined.
The first part of this thesis tests the hypothesis that errors in PM predictions by CTMs, such as PMCAMx, occur at least partially due to the urban dust emissions underestimation. The simulations suggest that the corresponding emissions are underestimated in the official pan-European reported emissions by a factor of ten. This hypothesis leads to improved PM10 predictions in all sites in Europe and especially in urban areas reducing the PM10 bias by 23% and the error by 13%. Simulations with the improved urban dust emissions indicate that PM1 nitrate, sulfate and ammonium levels can decrease on average within 20% over the modeling domain, while at the same time coarse levels can increase on average within 15% due to the higher levels of urban dust.
In the second part of this thesis, aerosol acidity was simulated depending on particle size, location and altitude over Europe during summer using the hybrid version of PMCAMx for the simulation of inorganic aerosol formation. Simulations indicate that pH changes more with particle size in northern and southern Europe with differences up to 1−4 pH units between sub- and super-micron particles, while the average pH of PM1-2.5 can be as much as 1 unit higher than that of PM1. PM1 has the most water over the continental region of Europe, while coarse particles have the most water content in the marine and coastal areas due to the relatively higher levels of sea salt. Particles acidity increases with altitude (0.5-2.5 units pH decrease over 2.5 km) due to the decrease in aerosol liquid water content. Aerosol pH affects inorganic nitrate with the highest average nitrate levels predicted for the PM1-5 range and over locations where the pH exceeds 3. Dust increases aerosol pH for all particle sizes and nitrate concentrations for supermicron range particles. This effect of dust depends on calcium content.
Τhe hybrid version of aerosol dynamics in PMCAMx is also used in the third part of this thesis to quantify aerosol acidity over the U.S during a wintertime and a summertime period as a function of particle size and altitude. Average PM1 pH can be higher up to 2 units during winter than summer due to the higher aerosol water levels in the cold periods. For the supermicron range, pH values are predicted to be higher during summer due to the higher concentrations of alkaline dust. Sub-micron aerosol is more acidic than supermicron for both seasons with pH differences of up to 1−4 units. Acidity is predicted to increase with altitude by up to 1−1.5 units for PM1, and 2−2.5 units for PM1−10 in the first two kilometers due to the decrease of liquid water content with height.
In the fourth part, ISORROPIA-lite, an accelerated and simplified version of ISORROPIA-II aerosol thermodynamics model is presented and evaluated. ISORROPIA-lite assumes that the aerosol exists in liquid form even at low relative humidities (metastable state) and treats the aerosol thermodynamics using binary activity coefficients from precalculated look-up tables. These assumptions speed up the thermodynamic calculations by 35%. Application of ISORROPIA-lite in the PMCAMx CTM accelerates the simulations by about 10% with changes in the concentrations of the major aerosol components of less than 10% over Europe. Compared to ISORROPIA-II, ISORROPIA-lite also simulates the effects of organic water on aerosol thermodynamics. Simulation of these effects indicates an increase of fine nitrate and ammonium concentrations within 1 μg m−3 in places where the organic aerosol and RH levels are high.
In the fifth part, the effects of secondary organic aerosol water (SOAW) on inorganic aerosol thermodynamics are studied using ISORROPIA-lite in PMCAMx for a full year over United States. SOAW can increase annual average fine aerosol water levels up to a factor of two when secondary organic aerosol (SOA) is a major PM1 component. Total dry PM1 can increase up to 2 μg m−3 due to increased partitioning of nitrate and ammonium (nitrate levels increase up to 200%) because of the additional SOAW mass when RH levels and PM1 components concentrations are high.
|
|---|
