IER photochemical smog evaluation and forecasting of short-term ozone pollution levels with artificial neural networks

The experimental work of this paper has been conducted over a period of one year, starting in January 1997, for measurement of air pollutants and mete orological parameters in the urban atmosphere of the Khaldiya residential a rea in Kuwait. The measurements were carried out simultaneously every 5 min utes by using the Kuwait University mobile air pollution monitoring laborat ory (Chemical Engineering Department). The main emphasis of the paper has b een placed on the problem of ozone for those days that are characterized by events of photochemical smog. The first objective of this paper deals spec ifically with the use of the Integrated Empirical Rate (IER) photochemical kinetic mechanism that has been developed at the Commonwealth Scientific an d Industrial Research Organization (CSIRO) of Australia as a screening tool for photochemical smog assessment. The IER has been used to determine whet her the local photochemistry of ozone events is light-limited (VOC-limitcd) or NOchi-limited. Such information is necessary in developing an effective emission control plan and enables the decision as to whether NO, or NMHC e mission needs to be controlled. On the other hand, the available models to predict the concentrations of ozone are complex and require a number of inp ut data that are not easily acquired by environmental protection agencies o r local industries. Thus, the second objective concerns the short-term fore casting of ozone concentration based on a neural network method.