Spatiotemporal analysis of particulate matter following the WTC disaster: Initial Results using a geostatistical approach

Particulate matter (PM10 and PM2.5) is a criteria air pollutant providing a useful indicator to assess the impact of the WTC disaster on air quality. The objective of this initial analysis is to use a geostatistical approach to estimate particulate matter over space and time in the surroundings of the WTC site in order to assess the ability of available monitoring data to reconstruct the plume of air toxics released by the collapse of the towers and the ensuing fires. In this analysis we use the Bayesian Maximum Entropy (BME) method of modern Geostatistics, which performs a composite space/time analysis and provides the capability to account for measurement errors when that information will be available. The implementation of the BME method for the WTC situation resulted in the development of the BME-WTC model presented herein. Using this model we analyzed monitoring data from the US-EPA AIRS database, which contains monitoring data for the federal criteria air pollutants measured at a fixed network in the US, as well as from monitoring stations that were deployed after the 9/11 disaster to specifically monitor the air toxics coming from the WTC site. Our initial results show that the BME-WTC model provides a useful framework for the spatiotemporal exposure mapping in the WTC situation, however the data used so far in our analysis is not sufficient to reconstruct the space/time plume. Future work is focused on getting a more complete database of the monitoring data in order to obtain a better reconstruction of the plume, and combining this reconstruction of the plume with predictions from physical air transport simulations