In order to evaluate the health impact of the Yunlin offshore industrial park on the local community, this project culminated relevant literatures and data regarding spatial temporal distribution of multimedia pollutants in the Yunlin vicinity. Collated observational datasets is divided into meteorological data, pollutant concentration observation and pollutant toxicity records. Meteorological data sourced from the Central Weather Bureau, Taiwan EPA, and various meteorological data source. Observation data sources include the special industrial zone air monitoring program, Taiwan EPA monitoring station(photochemical/air quality), mobile petrochemical monitoring vehicles and other multi-media observation data in the relevant literatures. Toxicological data such as the health effects of various chemicals used in relevant exposure regulation,biological testing, toxicological dynamics and human activity pattern studies. This project compares the hazardous chemicals in the industrial zone IARC-2B and above through the literature, and compares the relevant environmental monitoring list. 16 VOC species were selected for the health risk assessment including vinyl chloride, 1,3- butadiene, and 1,2-dichloropropane, benzene, trichloroethylene, dichloromethane, tetrachloroethylene, chloroform, ethylbenzene, cumene, isoprene, styrene, 1,2-dibromoethane, 1,2-dichloroethane, 1,4-dichlorobenzene and carbon tetrachloride. This project developed a CMAQ-SMOKE-WRF-CalTOX model to estimate multi-media exposure concentration for exposure estimation in the health risk assessment. The modelling processes include assimilation of relevant observation data to approximate the real health exposure to reconstruct the spatio- temporal distribution of the exposure concentration of 2016 in Yunlin. The consequent multi- media health risk assessment assesses the quantitative health risks in Yunlin. The health risks are characterised as cancer risk and non-carcinogenic risk(acute and chronic) for sensitive groups defined. This project assess the acute risks with 24-hour exposure, 8-hour exposure or 1-hour exposure while chronic risk are estimated with mean concentration. The estimated cancer risk(median) for the 16 targeted species ranging from 2.71x10^-11to 3.02x10^-6 and the lifetime cancer risk(95th percentile) ranging from 2.65x10^-9 to 4.13x10^-4 generally have slightly higher risk in Taixi Township than Maiqi Township. Although, this difference is not statistically significant. In the non-carcinogenic risk component, the hazard factor was evaluated. The median number of 16 chemical non-oncogenic HI assessments ranged from 2.04x10^-7 to 2.28x10＆-2 , while the 95th percentile range was Between 1.77x10^-5 and 2.66x10^0, short-term hazard indicators are estimated at hourly or daily exposure concentrations for acute non-cancer risk, resulting in a HIst range of 16.20x10^-4 for 16 chemical substances in Mailio and Taixi. In the health risk part of sensitive population, the health risk assessment of vinyl chloride and benzene exposure is mainly targeted at the small area of Taixi Township. Among them, in the risk of cancer, the average cancer risk of small vinyl chloride and benzene in Qiaotou is 3x10^-6 and 4.99x10^-7, the 95th percentile were 5.19x10^-6 and 7.26x10^-7 respectively; the average risk of carcinogenicity of vinyl chloride and benzene at Xufu Campus was 2.92x10^-6 and 5.07x10^-7 , respectively. The 95th percentiles are 5.11x10^-6 and 7.43x10^-7, respectively. In the chronic non-carcinogenic risk, the average HI of vinyl chloride and benzene in Qiaotou is 1.59x10^-2 and 9.94x10^-3, respectively. The 95th percentile is 2.76x10^-2 and 1.45x10^-2 respectively. The average HI of vinyl chloride and benzene at Xufu is 1.55x10^-2 and 1.01x10^-2 , respectively. The 95th percentile is respectively For 2.70x10^-2 and 1.47x10^-2, in the acute non-carcinogenic risk part, the highest concentrations of vinyl chloride and benzene in Qiaotou are 52.7 and 66.0, respectively, and HIst are 0.1 and 6.2, respectively; The highest concentrations were 46.9 and 32.0, respectively, and HIst were 0.1 and 3.0, respectively. There were no significant differences in the health risks of children with different ethnic groups in the study area.

health risk assessment, database construction, air dispersion model, multimedia model