Limitations of models for characterising indoor particle concentrations from cigarette smoking in an office environment.

Mathematical models have been used by various researchers to provide both a fundamental understanding of indoor air quality dynamics and a platform for estimation of IAQ constituents in lieu of experimental measurements. Due to the diverse nature of these applications the complexity and hence applicability and accuracy of the models varies tremendously. Some models have been specifically developed for evaluation of the impact of a broad range of environmental conditions on IAQ constituents. Cigarette smoking has been included as a source in many of these models since the contribution of smoking to particle and gas concentrations is measurable in many locations. Part one of this paper uses results from tests that accurately measured particle and gas concentrations during smoking in a real world office environment reported by Curl1 at IAQ95. The test was conducted in a 511 square meter office space with ventilation and smoking controlled and measured. The smoking rate was held constant throughout the test under both integrated and segregated smoking conditions. Ventilation of the office space was controlled to two rates; no mechanically delivered outside air and full economizer. Part two of this paper explores a more detailed and controlled test of particle concentration from cigarette smoking that was conducted in a single office setting. This test measured and recorded activities on a real time basis. The information gathered from this realtime single office observation is much more detailed than that from the large office space. Single compartment models are evaluated using the experimental data from the large office space and from the more detailed single office. The site factors associated with both the comfort cooling system and the air exchange variation were expected to be challenges to the performance of the available models. The theoretical foundation for each model is similar but each model contains slightly different mathematical terms and treatments for ventilation effectiveness, deposition, and filtration. The computed values were compared to the actual data measured in the test. Each model's performance is evaluated using the criteria suggested in ASTM D5157-91 Standard Guide for Statistical Evaluation of Indoor Air Quality Models. All of the models have limited use in a real world environment where conditions are not represented by perfect mixing and uniform distribution.