Masters Thesis Abstract: Mikhail Titov
Application of an international atmospheric mesoscale modelling system to anaysis of air pollution dispersion in Christchurch and surrounding areas.
Christchurch and surrounding small towns continue to experience elevated levels of air pollution during calm winter nights due to emissions from domestic fires and local meteorology. Complex orography and ocean breezes make the local atmospheric circulation much more complex to model. The ability of coupled systems of recent versions of Mesoscale Model Generation 5 (MM5) and Comprehensive Air quality Model with extensions (CAMx) to simulate the local circulations and predict 3-dimensional air pollution for the Canterbury region during winter nocturnal stagnation situations in complex orography was evaluated.
A number of experiments for various time intervals during winter 2000 were undertaken to tune MM5 using different numbers of nested grids and to adjust it to winter circulation over the Christchurch region under synoptic conditions of stagnant air with weak winds. The data from the Global Dataset of National Centre for Environmental Prediction were used as initial fields for MM5 runs over the east coast of South Island. Observations from Christchurch Air Pollution Study 2000 were used to compare modelled results with observations for a more accurate and sophisticated selection of model parameters to reproduce the local circulation systems, especially in the atmospheric boundary layer. Model Absolute Error, Pearson Correlation Coefficient and Index of Agreement were calculated for near-surface main meteorology: wind, temperature and relative humidity.
The Comprehensive Air quality Model was utilized and adjusted to the Canterbury region. This included creating South Island land use files to allow for dry deposition and utilization of initial grided emission fields for particulate material. The chemistry model used input data from MM5 output (including wind fields) to study dispersion and concentration of fine Particular Matter (PM10 and PM2.5), non-organic gases and Volatile Organic Components (VOC) over Christchurch during wintertime. The results were provided at a fine spatial resolution (1km x 1km) for the near-surface part (15.5–20.5m) of the atmospheric boundary layer.
Several experiments were run for winter 2000 to investigate the ability of the coupled system (MM5- CAMx) to predict spatial dispersion and time trends of PM, especially of elemental carbon and organic acids. A comparison was made of the CAMx4 output pollutants PM2.5 – PM10 with the observed near-surface pollution data measured at 3 sites (Coles Place, Christchurch Polytech and Packe Street) during the winter Christchurch Air Pollution Study 2000 by Environmental Canterbury. Special filters were applied to PM input grided emission data and special interaction procedures were used to adjust input emission fields to the real PM emission picture obtained from the Coles Place measurements. A few experiments were made to research CAMx possibilities to predict pollutant gases (NO, NO2, CO, SO2, VOC).