Evaluation of a year-long dispersion modelling of PM10 using the mesoscale model TAPM for Christchurch, New Zealand
Introduction
Internationally, New Zealand has a reputation for having a pristine environment with plenty of green spaces and lots of fresh, unpolluted air. However, in reality—at least as far as clean air is concerned—air pollution can be a serious problem in urbanized regions, especially during austral winter months. The coastal city of Christchurch, situated about 70 km east of the Southern Alps (172°37′ W–43°31′ S) and just north of a caldera (eroded volcanic crater) known as Banks Peninsula (Fig. 1), has a population of 300,000; occupies an area of about 140 km2, and usually experiences smog events for about 30 days each winter season when the daily-averaged concentration of PM10 exceeds the air quality guideline of 50 μgm−3 (Aberkane, 2000). The area of Banks Peninsula just south of the urban area is known locally as the Port Hills.
The modelling work presented in this paper is part of an ongoing pilot programme to assess and evaluate air pollution exposure for selected urban areas in New Zealand (see www.hapinz.org.nz). A key component of this assessment is the construction of spatially detailed, annually-averaged air pollution exposure maps using an existing dispersion model. The exposure maps are needed in order, for example, to ascertain inequality in exposure to air pollutants between different social groups (Pearce et al., accepted for publication). Hence, there was a need for a computationally efficient meteorological model with an air pollution module. To this end, The Air Pollution Model (TAPM; Hurley, 2002) has been chosen. TAPM is a PC-based mesoscale prognostic numerical model with meteorological and air pollution components. The meteorological module of TAPM predicts the local-scale circulations, such as sea breezes and slope flows, in conjunction with larger scale synoptic scale meteorological fields.
Section 2 of this manuscript will provide an overview of meteorology of Christchurch, with particular emphasis on winter time smog episodes. In section 3, the derivation of pollutant emissions inventories used as input for TAPM is described in detail, while section 4 offers information on the TAPM setup and results.
Section snippets
General
Christchurch is located in the mid-latitudes and its wind climate is largely controlled by eastward propagating high and low pressure systems and the city's geographic location relative to the Southern Alps (Sturman and Tapper, 1996). Over the Canterbury Plains, the synoptic scale wind is strongly modified by dynamic and thermal effects caused by the land–sea discontinuity, the Southern Alps and Banks Peninsula (McKendry, 1983). As shown in Fig. 2, synoptic scale westerly winds flow over and
TAPM grid setup and configuration
TAPM is a three-dimensional incompressible, non-hydrostatic, primitive equations model, which uses a terrain-following coordinate system (Hurley, 2002). For computational efficiency, it can be used in a telescoping nested configuration where higher resolution grids are successively placed inside coarser resolution grids. In addition, model solution for each grid is one way interacting—information is passed from the coarse grid downwards. The meteorological component of the model is supplied
Derivation of PM10 emissions data
Emission inventory data has been collected in Christchurch on a regular basis to monitor trends over time and to determine changes in the relative contribution of sources to emissions (NIWA, 1998, Wilton, 2001, Scott and Gunatilaka, 2003). These present emissions to the air for a “typical winter's day” for the area within the Christchurch territorial boundary and comprised the main portion of the Christchurch airshed, as defined by Sturman and Zawar-Reza (2002). The inventory contains raw
Results and discussion
Statistical measures, such as Root Mean Square Error (RMSE) and Index of Agreement (IOA) are used to evaluate TAPM's performance (, and ; where the Predicted (P) values by the model are compared against Observed (O) data (Willmott, 1981). The IOA is a measure of the skill of the model in predicting variations about the observed mean; a value above 0.5 is considered to be good. The modelled data, at 10 m above the ground for wind and at screen
Acknowledgements
We would like to thank the Geography support staff at the University of Canterbury for providing assistance for this project, especially James Sturman, Paul Bealing, and Matthew Faulk. The research is supported by the HRC Health and Air Pollution in New Zealand (HAPiNZ) project research grant no: 03/470.
References (33)
- et al.
A statistical procedure for determining the best performing air quality simulation
Atmos Environ
(1990) - et al.
Year-long, high-resolution, urban airshed modeling: verification of TAPM predictions of smog and particles in Melbourne, Australia
Atmos Environ
(2003) - et al.
Application of back-trajectory techniques to the delimitation of urban clean air zones
Atmos Environ
(2002) - Aberkane T. Annual air quality monitoring report. Canterbury Report No. 00/78. 2000. 32...
Particulate and gaseous emissions from wood burning fireplaces
Environ Sci Technol
(1982)Air quality model evaluation and uncertainty
J Air Pollut Control Assoc
(1988)- Hurley, P. The Air Pollution Model (TAPM) Version 2. Part 1: Technical description. CSIRO Atmospheric Research...
- Johnstone, SJ. The interaction of inversion development and katabatic winds: implications for air pollution in...
- et al.
Modification of the thermodynamic structure of the lower troposphere by the evaporation of precipitation: a GEWEX cloud system study
Mon Weather Rev
(1997) - et al.
The surface wind field during winter smog nights in Christchurch and coastal Canterbury, New Zealand
Int J Climatol
(2004)
Christchurch home heating survey: a survey of domestic home heating methods and fuels in the Christchurch metropolitan area
Spatial and temporal aspects of the surface wind regime on the Canterbury Plains, New Zealand
J Climatol
A study of interacting multi-scale wind systems, Canterbury Plains, New Zealand
Meteorol Atmos Phys
The Canterbury Plains north-easterly
Weather Clim
A numerical study of the airflow over irregular terrain
Beitr Phys Atmos
Cited by (61)
Sulfur dioxide emissions in Portugal: Prediction, estimation and air quality regulation using machine learning
2021, Journal of Cleaner ProductionAir pollution terrain nexus: A review considering energy generation and consumption
2019, Renewable and Sustainable Energy ReviewsSpatial and temporal variations of airborne dust fallout in Khorasan Razavi Province, Northeastern Iran
2018, GeodermaCitation Excerpt :Direction and speed of these two winds and their relative dominance can have significant effects on the sediment accumulation in different areas of the province over different times of the year. In the late spring and summer, winds over the plains in Turkmenistan and much of Iran are northeasterly (Rashki, 2012; Zawar-Reza et al., 2005). This wind affects the morphology of sand dunes in the eastern, western and southern portions of Khorasan Razavi (Karimi et al., 2017; Mashhadi et al., 2007) and is responsible for the loess plain, covered by aeolian dust 1–10 m thick in the northeast corner of the province (Fig. 1) (Karimi et al., 2011).
The phenomena of spreading ambient ozone at the west coast air basin of Taiwan
2016, Atmospheric Pollution Research