Air quality modelling of smoke exposure from the Hazelwood mine fire

[摘要] The Hazelwood coal mine fire began on February 9th 2014 and burned for 45 days. A preliminary smoke tracer study, conducted by CSIRO in April 2015, showed most towns within a 30km radius of the mine fire were exposed to high concentrations of smoke. The highest concentrations of smoke from the fire occurred when the winds were blowing from a south westerly direction. The preliminary study was conducted in the absence of smoke emission rates for the fire and also was not able to resolve strong concentration gradients for locations (such as Morwell) immediately adjacent to the mine. In the current study our smoke exposure modelling has been refined through the use of quantitative estimates of hourly emission rates of various species based on parameters such as how much coal was burned using maps of the area burned drawn by the Country Fire Authority and estimated emission factors. Additionally we have used both a high-resolution near-field model and a regional scale model in order to better resolve smoke concentration gradients across the impacted areas in the Latrobe Valley – Gippsland region.The high resolution near-field concentration predictions were generated using CSIRO’s The Air Pollution Model (TAPM) in order to properly resolve the smoke plume around Morwell within a domain of 10 km × 10 km. The coal fire area was represented by a number of ground-level point sources each 100 m apart. Particulate matter with an aerodynamic diameter of 2.5 microns or less (PM2.5) and carbon monoxide (CO) from the fire were treated as tracer species (i.e. no chemistry acted upon them) within the selected domain and dispersed using TAPM predicted meteorology. Hourly PM2.5 concentrations of up to 3700 g m-3 and CO concentration up to 60 ppm were predicted in the early phases of the fire at Morwell South. These predictions for Morwell East were 2200 g m-3 and 35 ppm, respectively. Direct comparisons between the model and observations showed that TAPM predicted the correct magnitude, but did not always predict observed temporal maxima at the same time. However, when concentrations are unpaired in time TAPM simulates the concentration distribution around Morwell satisfactorily. This result demonstrates that the assumptions made in the modelling, especially in the emissions methodology, are realistic.The second modelling system conducted full chemistry simulations. The CSIRO Chemical Transport Model (CTM) coupled with the CCAM meteorological model was used to predict the exposure of residents living in towns across the Latrobe Valley, such as Sale, chosen as the control group for the long term Hazelwood health study.Regional modelling of background concentrations showed PM2.5 levels at 6 g m-3 in Morwell. This demonstrates the level of PM2.5 that would have been present in Morwell if the Hazelwood mine fire had not taken place. It includes the contribution from all anthropogenic sources and any wildfires occurring in Victoria during February and March of 2014. Likewise, background CO concentrations were 0.07 ppm. The air quality in the closest towns to the mine was returned to background levels after March 12th 2014.Inclusion of the mine fire in the CTM resulted in peak concentrations of up to 2000 g m-3 for PM2.5 at Morwell South. In the absence of measurements at this time, this predicted concentration is reasonable as fire activity was maximum during the early phases of the fire. The magnitudes of the predicted PM2.5 and CO as compared to the observations demonstrates that the emission rates calculated in the first exercise are reasonable.Overall residents of Morwell were exposed to the greatest number of breaches of the 24 hour air quality standard for PM2.5, exceeding the 25 g m-3 limit on 27 days. Outside of Morwell PM2.5 concentrations decreased rapidly. There were 3 breaches of the PM2.5 air quality standard at Churchill, and 2 breaches at Moe and Traralgon. There were no breaches of the CO air quality standard (9ppm as an 8

[发布日期] 2016-10-28 [发布机构] CSIRO

[学科分类] 地球科学(综合)