Particulate matter (or aerosols) are particles suspended in the atmosphere. Aerosols are believed to be the most important pollutant associated with increased human mortality and morbidity. Therefore, it is important to investigate the relationship between sources of aerosols (such as industry) and the concentration of harmful aerosols at ground level. Furthermore, aerosols influence the climate system by scattering and absorbing solar radiation and by influencing cloud properties. The total climate effect of aerosols is poorly understood compared to the climate effect of greenhouse gases. Therefore, climate studies also benefit from a better understanding of aerosols.
The goal of this thesis is to investigate the spatial distribution of aerosols over Europe with focus on the Netherlands. The aerosol life cycle and effects are calculated with numerical simulations. Performing numerical simulations of aerosols is very challenging, because, in contrast to gas molecules, each individual aerosol differs in size, composition and microphysical properties. Without simplifications, a model has to track each individual particle, which would take far too much computational time, even for modern supercomputers. The challenge is to design simplifications in such a way that the life cycle of aerosols and the effects of aerosols on human health and climate are still properly represented.
Many model studies are supported by measurements. Both the measurements and the models can have different purposes. Using the correct combination of different models and observations is key for studies on aerosols. A different combination of models and observations is required to accomplish the different sub goals of this thesis. These sub goals are:Investigation of the aerosol life cycle over Europe Improvement of the understanding of gas-aerosol phase transition of ammonium nitrate and aerosol optics Improvement of representation of aerosols and their effects in models
The life cycle of aerosols in Europe is investigated in chapter 3. The full life cycle of aerosols has been implemented in a global transport model. It is concluded that Europe is a net source of anthropogenic (man-made) aerosols and a net sink of natural aerosols. The most important sink of anthropogenic aerosols is removal by clouds and rain, while natural aerosols are removed predominantly by dry deposition processes. By comparing model results with observations, it is concluded that the largest uncertainties are caused by the parameterisation of wet removal processes and by missing emissions.
In the Netherlands, emissions of nitrogen oxides and ammonia are high because of the high population density and intensive agriculture. After oxidation of nitrogen oxides to nitric acid, ammonium nitrate aerosols can be formed. This aerosol is special, because it can evaporate under warm and dry conditions and condense back to the aerosol phase under cold and moist conditions. Like the case of clouds, the phase equilibrium changes with altitude as the atmospheric temperature decreases with altitude. The phase of ammonium nitrate is poorly detected by many measurement instruments, because the gas-aerosol partitioning can change inside the instrument. Partly due to the scarcity of reliable measurements, the phase transition of ammonium nitrate is poorly implemented in large-scale models.
Because ammonium nitrate aerosol and its phase transition is important for the aerosol budget of the Netherlands, this process has further been investigated in case studies. The goal of case studies is to gain detailed insight in the aerosol processes and, ultimately, to develop better parameterisations for large-scale models. These case-studies are performed with more detailed small-scale models. In these models, not the full aerosol life cycle is simulated but only the processes that are being investigated. A large advantage, however, is that these models have a higher resolution both in the spatial and the temporal domain. As a result, the important processes can be resolved more precisely.
Chapter 4 presents a case study where the interaction between ammonium nitrate phase transition and mixing in the lower atmosphere (boundary layer) is investigated for a warm day in spring. During an intensive measurement campaign near the Cabauw tower in the Netherlands, measurements of ammonium nitrate have been performed. Importantly, the gas and the aerosol phases have been separated with a special instrument so that both concentrations are measured without errors due to phase transition inside the instrument. It is shown that the observed partitioning between gas and aerosol ammonium nitrate deviates significantly from the thermodynamic equilibrium. The hypothesised explanation for this mismatch is that aerosol-rich air from higher altitudes (where the aerosol phase is preferred due to lower temperatures) is transported to the surface, increasing the aerosol-phase fraction of ammonium nitrate at the surface. This implies that the thermodynamic equilibrium is not instantaneously restored at the surface. A simulation of ammonium nitrate partitioning in the boundary layer has been performed with a simplified column model. The match between model results and observations improved drastically when applying a delay timescale up to two hours for the gas-aerosol equilibrium.
The interaction between turbulence and ammonium nitrate partitioning is further investigated in a more detailed model study (chapter 5). In this model, turbulent motions are explicitly resolved. As highlighted above, downward motions are associated with higher aerosol concentrations, because the phase equilibrium of ammonium nitrate is shifted towards the aerosol phase at higher altitudes. Therefore, turbulent motions induce a fluctuating concentration of aerosol ammonium nitrate with updrafts containing lower aerosol ammonium nitrate concentrations and subsidence motions containing enhanced aerosol ammonium nitrate concentrations. It is discussed that these fluctuations in observations may provide information about the speed of gas-aerosol partitioning, which is very difficult to measure directly.
Throughout chapters 3 to 5, several ideas for model improvements have been posed. These ideas originate both from knowledge gained in the studies and from further challenges that are discovered. One such improvement for models is a computationally efficient and adequate representation of the optical properties of aerosols. Implementation of aerosol optics has been quite challenging, because the physics of aerosol optics is very complicated. Chapter 6 presents a package that allows easy implementation of aerosol optics in atmospheric models that represent aerosols.
Aerosol modelling is a very challenging task and can be developed much further. In this thesis, important steps have been taken to improve knowledge about aerosols. Future research should proceed by unravelling remaining aerosol mysteries, such as those presented in the final chapter (7) of this thesis.
|Qualification||Doctor of Philosophy|
|Award date||22 Feb 2013|
|Place of Publication||S.l.|
|Publication status||Published - 2013|
- air pollution
- simulation models