Atmospheric transport and dispersion modeling systems are often used in assessing human exposures to chemical hazards. Models validated through quantitative and qualitative evaluation can be applied to epidemiologic study. Here, we modeled the 2005 Graniteville, South Carolina, USA railcar release of chlorine using dense gas plume dispersion models including the Hazard Prediction and Assessment Capability (HPAC) and Areal Locations of Hazardous Atmospheres (ALOHA). The release volume (54,915 kg) and rate was estimated by an engineering analysis combining semi-quantitative observations and fundamental physical principles. The use of regional meteorological conditions was validated by statistically (correlation, mean bias, root mean square deviation) comparing 1,024 HPAC concentration and surface dosage point estimates generated by two source-location weather data sets. An improved HPAC model was then statistically (correlation, root mean square deviation) compared to the earlier HPAC model using up to 9,446 surface dosage sampling points paired in time and space. The older HPAC model consistently overpredicted compared to the newer, refined model. When compared to HPAC, the ALOHA model significantly overpredicted downwind, centerline concentrations (up to 55 times that of HPAC). The refined HPAC model was then evaluated against post-incident environmental indicators of exposure such as phytotoxicity, corrosion events, deposition benchmarks, casualty data and exposed animal health outcome. A further sub-analysis was performed by comparing observed dog health outcome-derived exposure estimations versus model-predicted exposure. This statistical sub-analysis showed good agreement between observed and estimated, particularly when a sub-cohort of indoor dogs was excluded to determine the impact of structural shielding. Although the model was favorably evaluated based on literature-established standards, further assessment should be performed before the model can be fully validated and applied in human epidemiologic study to estimate acute exposures. Language: English / 1 / Dev D. Jani
| Identifer | oai:union.ndltd.org:TULANE/oai:http://digitallibrary.tulane.edu/:tulane_58006 |
| Date | January 2016 |
| Contributors | Jani, Dev D. (author), Rando, Roy (Thesis advisor), School of Public Health & Tropical Medicine Environmental Health Sciences (Degree granting institution) |
| Publisher | Tulane University Digital Library |
| Source Sets | Tulane University |
| Language | English |
| Detected Language | English |
| Type | Text |
| Format | electronic |
| Rights | Embargo |
Page generated in 0.0019 seconds