Air pollution diffusion simulation and seasonal spatial risk analysis for industrial areas

https://doi.org/10.1016/j.envres.2020.110693Get rights and content

Highlights

  • Investigate the air pollution risk from the largest stationary sources in Taiwan.

  • Combine historical data-driven and simulation for drawing the long-term hazard map.

  • Simulations used long-term historical meteorological data in different seasons.

  • Air pollution risk analysis was investigated for neighboring hospitals and schools.

  • Can be a reference for the establishment location of the new industrial area.

Abstract

The petrochemical industry produces many air pollutants during production, such as airborne particulate matters (PM10 and PM2.5), sulfur oxides, nitrogen oxides, volatile organic compounds, carbon oxides, etc. Petrochemical industrial accidents are more likely to cause major air pollution hazards in a short period. Therefore this study simulated diffusion and performed air pollution spatial risk analysis for potential air pollutants generated by the petrochemical industry using meteorological observation data from 2017 to 2019. The study targets were No. 6 Naphtha Cracker Complex Petrochemical Industrial Park (6NCC) of Formosa Petrochemical Corporation and Taichung Thermal Power Plant (TTPP) in central Taiwan. We used the industrial source complex model short term (ISCST3) air simulation model developed by the US Environmental Protection Agency to simulate pollutant diffusion under different weather conditions and seasons. Air pollution spatial risk was investigated for neighboring hospitals and schools for pollutant emission and diffusion to provide feedback to petrochemical related industry's risk management.

Emission areas (6NCC and TTPP) were all in the southwest since the main air pollution accumulation and diffusion is to the northeast during monsoon season (October through March). Air pollution April through September each year is more evenly distributed, with pollutant concentrations low in all directions, approximately half the concentration in winter. Simulated air pollutant concentrations often overlapped with high risk population clusters (schools and hospitals). 6NCC posed little impact on nearby schools throughout the year; whereas TTPP posed relatively low risks to nearby schools and hospitals in summer, with slightly higher risk for Shenren Elementary School in Shengang township, Changhua County in winter. Overall 6NCC posed higher risk for Mailiao and Taixi townships in Yunlin County; whereas the TTPP posed higher risk on Longjing District of Taichung City, Shengang and Xianxi townships in Changhua County, particularly during winter. The results of this study will help the petrochemical industry and public health authority to wider manage air pollution risks.

Introduction

The petrochemical industry has complex processes with many airborne pollutants, including airborne particulate matters (PM10 & PM2.5), sulfur oxides (SOx), nitrogen oxides (NOx), volatile organic compounds (VOCs), and carbon oxides (CO & CO2), etc. Many of these pose significant risk to environmental and/or human health. Although the petrochemical industry is relatively high risk industry worldwide, it is also a very important for industrial production. Therefore, higher standards are required for risk control, particularly considering that petrochemical industrial accidents can cause major air pollution hazards in a short period. For example, the second hydrogen refueling plant at the CPC Taoyuan Oil Refinery caused a fire 20 January 2018 due to an exploding heating furnace tube, leading to poor air quality in the vicinity and significant impact on nearby residents’ health. The US federal government generally adopts risk assessment methods to regulate substances harmful to human safety, health, and environmental quality. However, the large domestic petrochemical industry in Taiwan does not yet have clear definitions for risk identification and risk assessment, and also lacks a complete risk database.

Previous studies have shown that residents’ urinary carcinogenic metal levels in Mailiao and Taixi Townships, i.e., high exposure areas (HEs), were higher than for residents in more distant townships, i.e., low exposure areas (LEs), for 2009–2012, within 10 km radius from No. 6 Naphtha Cracker Complex Petrochemical Industrial Park (6NCC) of Formosa Petrochemical Corporation (FPCC). In particular, HEs had higher hepatitis C and cancer rates than LEs due to elevated pollution from various carcinogenic pollutants emitted from petrochemical manufacturing plants (Yuan et al., 2018). Hsu et al. (2018) showed that VOCs (vinyl acetate, chloroethene, and 1,2-dichloroethane) were significantly higher within 5 km radius from the Mailiao Industrial Complex (particularly Mailiao and Taixi townships, and Yunlin County, where 6NCC is located). They used a positive matrix fraction (PMF) model to show that industrial emissions (49.2–61.7%) were the main contributions to VOC sources. Inhalation exposure to benzene was most relevant to local industrial sources for residents living within 5 km from Mailiao Industrial Complex, producing the highest cancer risk (C.-Y. Hsu et al., 2018).

Kao et al. (2019) showed that wind speed and direction were the main determinants for air quality in Mailiao and Taixi townships. Northerly, Northeasterly, and Easterly winds were significant factors for elevated air pollution (Kao et al., 2019). Hsu et al. (2016) showed that the Mailiao industrial park major emission was PM2.5 in Yunlin County. PM2.5 concentration is likely to accumulate near the emission source under weak synoptic conditions and weak northeasterly monsoonal flow (C.-H. Hsu and Cheng, 2016).

Many air diffusion models have been employed to simulate spatial changes in air pollution concentrations. Air diffusion models are suitable for simulating PM2.5 concentrations in air since they explicitly consider transmission and pollutant changes in the environment, and provide good explanatory ability for time variations (Yang, 2010). The common diffusion models including steady-state Gaussian plume, Lagrangian based trajectory, and grid models based on grid division. Common diffusion models include AERMOD, jointly developed by the American Meteorological Society and US Environmental Protection Agency (USEPA) (Cimorelli et al., 2005; Rzeszutek et al., 2017); industrial source complex model (ISC3), developed by USEPA (Hanna et al., 2001; Wesely et al., 2002); community multiscale air quality (CMAQ) model developed by USEPA (Binkowski and Roselle, 2003); CALPUFF developed by Exponent, Inc., USA (Levy et al., 2002; Scire et al., 2000); and the ALOHA model developed by USEPA specifically for hazardous substances diffusion. These models have various advantages and disadvantages, and have been widely used to simulate of air pollutant transmission by various studies and compare with each other (Asadi et al., 2017; Butland et al., 2020). Table 1 list some commonly used air diffusion models.

The current study used 6NCCP and Taichung Thermal Power Plant (TTPP) in central Taiwan as study targets, and adopted the industrial source complex model short term (ISCST3) air simulation model developed by USEPA to simulate pollutant diffusion under different weather conditions and seasons. ISCST3 model is capable of handling multiple sources, including point, volume, area and open pit source types. It can evaluate pollutant concentrations emitted by different sources related to industrial parks. Several studies have shown that ISCST3 is a better model to simulate the dispersion of air pollutants especially from industrial sources (Bajoghli, 2019; Karuna et al., 2017; Prakash et al., 2017). Simulations used long-term historical observational meteorological data to analyze air pollution spatial distributions and diffusion from these petrochemical industry sites under prevailing weather conditions in different seasons. Therefore, the prone areas of air pollution exposure can be identified based on the simulation results. Air pollution spatial risk analysis was investigated for neighboring hospitals and schools, and we explored air pollutant emission impacts on these vulnerable receptors by simulation to provide feedback to petrochemical related industries, public health authorities, and residents for health risk management.

Section snippets

Study area

This paper used Mailiao 6NCC and TTPP in central Taiwan as the study areas, and adopted ISCST3 to analyze pollutant concentration distribution from the industrial park for different seasons and weather conditions. Meteorological observation data is collected from the Central Weather Bureau (CWB) from 2017 to 2019. The observed PM2.5 concentration data is calculated from the Taiwan Environmental Protection Administration (TWEPA) monitoring stations in central Taiwan in 2017 for the estimation of

No.6 Naphtha Cracker Complex

Simulated conditions for 6NCC were continuous emission from 30 chimneys. Fig. 4, Fig. 5 and Table 5 show monthly ISCST3 calculated concentrations for ambient air pollutants. Fig. 4 shows only the 2019 results as an example due to space constraints; Appendix 1 shows the complete monthly simulation results for 2017–2019 (Figure A1). Fig. 5 shows 2017–2019 time series monthly concentrations at P1 and P2 (6NCC and Taixi township office, respectively).

Fig. 6 shows that simulated main air pollutant

Discussion

The petrochemical industry has always been very important for economic development. However, it produces many pollutants, and air pollutants in particular, due to the various materials and production processes (Cetin et al., 2003; Ragothaman and Anderson, 2017). Many studies have shown that air pollutants produced by the petrochemical industry have significant impact on human health (Domingo et al., 2020; Marquès et al., 2020). For example, sulfide can affect human central nervous and

Conclusions

Long-term exposure to air pollution may cause health hazards, especially near industrial areas and petrochemical plant, where are the main source of stationary pollution. This study used 6NCC of Formosa Petrochemical Corporation and TTPP in central Taiwan as study targets, and adopted the industrial source complex model short term (ISCST3) air simulation model developed by USEPA to simulate pollutant diffusion under different weather conditions and seasons. The main accumulation and diffusion

Credit author statement

Yuan-Chien Lin: Conceptualization, Methodology, Supervision, Writing – original draft, Investigation, Funding acquisition. Chun-Yeh Lai: Data curation, Formal analysis, Visualization, Writing – original draft, Software. Chun-Ping Chu: Project administration, Funding acquisition, Resources.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

We are grateful for support from the Ministry of Science and Technology (Taiwan): Projects No. MOST 108-2625-M-008-002, MOST 108-2119-M-008-003, MOST 108-2636-E-008-004 (Young Scholar Fellowship Program), and MOST 108-2638-E-008-001-MY2 (Shackleton Program Grant). We are also thankful for the Python programing language and various modules, which provided powerful data analysis tools.

References (34)

  • M. Marquès et al.

    Health risks for the population living near petrochemical industrial complexes. 2. Adverse health outcomes other than cancer

    Sci. Total Environ.

    (2020)
  • T.-H. Yuan et al.

    Increased cancers among residents living in the neighborhood of a petrochemical complex: a 12-year retrospective cohort study

    Int. J. Hyg Environ. Health

    (2018)
  • M. Asadi et al.

    The comparison of Lagrangian and Gaussian models in predicting of air pollution emission using experimental study, a case study: ammonia emission

    Environ. Model. Assess.

    (2017)
  • M. Bajoghli

    Comparison of application of AERMOD and ISCST3 models for simulating the dispersion of emitted pollutant from the stack of an industrial plant in different time scales

    Archives of Occupational Health

    (2019)
  • F.S. Binkowski et al.

    Models‐3 Community Multiscale Air Quality (CMAQ) model aerosol component 1. Model description

    J. Geophys. Res.: Atmosphere

    (2003)
  • B.K. Butland et al.

    Comparing the performance of air pollution models for nitrogen dioxide and ozone in the context of a multilevel epidemiological analysis

    (2020)
  • F.-Y. Cheng et al.

    Long-term variations in PM2.5 concentrations under changing meteorological conditions in Taiwan

    Sci. Rep.

    (2019)
  • Cited by (0)

    View full text