With the ever-growing chemical industry, chemical storage and transportation may pose a threat to neighboring communities should there be an accidental release. In order to safeguard the workers and communities, agencies such as the National Institute for Occupational Safety and Health (NIOSH) and the Environmental Protection Agency (EPA) have made it mandatory for all industrial facilities to comply with the rules and regulations listed thereby. The Emergency Planning and Community Right-to-know Act (EPCRA) requires that the public be given access to information on chemicals at individual facilities for emergency planning in case of episodic releases at the site. A large number of state-of-the-art modeling technologies and engineering tools have been developed and are readily available to successfully predict and track the cloud or plume of a chemical during an accidental release using only information on the release scenario and the meteorological data at the time of release. These models provide an assessment of the impact of a release and are essential and extremely useful for emergency planning in order to save lives and resources in the event of a release.

After the events of September 11, 2001, the paradigm has shifted to predicting an intentional release of a toxic chemical, which has potentially high impact if the targets are public locations such as football stadiums, shopping malls, and army bases located either domestically or internationally. In such cases, identification of the source location and the rate of release becomes a big challenge for hazmat personnel tasked with responding to the chemical attack. This is in great contrast to situations involving an accidental chemical release where the source of a release is known, or, in most cases, can be quickly identified.

In an attempt to find a solution to this problem, a method is proposed for locating the source of a chemical using only the basic knowledge of meteorological information (wind speed, direction and stability) and two or more concentration measurements of the chemical released. The technique used in this approach is called the reverse corridor or back trajectory method. This method uses the horizontal dispersion information obtained from surface meteorological data at receptor locations to narrow the search domain for the release location, thus reducing the computational cost. The effects of the main parameters of dispersion, such as wind meander and fluctuations in sensor concentration on the sensitivity of the model, are discussed. This model establishes a confidence level for prediction of a release location under varying circumstances.

The data regarding prediction of the source location and its associated confidence level are valuable pieces of information for emergency responders when seconds count.