S M Shiva Nagendra

This paper describes the development of artificial neural network (ANN) based carbon monoxide (CO) persistence (ANNCOP) models to forecast 8-h average CO concentration using 1-h maximum predicted CO data for the critical (winter) period (November-March). The models have been developed for three 8-h groupings of 10 p.m. to 6 a.m., 6 a.m. to 2 p.m. and 2-10 p.m., at two air quality control regions (AQCRs) in Delhi city, representing an urban intersection and an arterial road consisting heterogeneous traffic flows. The result indicates that time grouping of 2-10 pm is dominantly affected by inversion conditions and peak traffic flow. The ANNCOP model corresponding to this grouping predicts the 8-h average CO concentrations within the accuracy range of 68-71%. The CO persistence values derived from ANNCOP model are comparable with the persistence values as suggested by the Environmental Protection Agency (EPA), USA. This work demonstrates that ANN based model is capable of describing winter period CO persistence phenomena. © Springer Science+Business Media B.V. 2007.

Air quality indices are used for local and regional air quality management in many metro cities of the world. In the present study, air quality indices have been calculated using the US Environmental Protection Agency procedure to assess the status of ambient air quality near busy traffic intersections in Bangalore, India. The measured 24 h average criteria pollutants such as sulfur dioxide, oxides of nitrogen, respirable suspended particulate matter and suspended particulate matter for the period from 1997 to 2005 at three air quality monitoring stations are used for the development of AQIs. The result indicated that the air pollution at all the three air quality monitoring stations can be characterized as 'good' and 'moderate' for SO2 and NOx concentrations for all days from 1997 to 2004. Analysis of air quality indices values for both forms of suspended matter concentrations during 1999-2005 indicates 91% and 94% of the times days are in category 'good' and 'moderate'. The yearly average air quality indices values of respirable suspended particulate matter and suspended particulate matter concentrations indicated decreasing trend and are coming under the category of 'good' and 'moderate' form the category of 'poor' and 'very poor'. © 2007 Elsevier Ltd. All rights reserved.

This paper presents a comparative evaluation of deterministic, stochastic and artificial neural network (ANN) based line source models in predicting carbon monoxide (CO) concentrations near an urban roadway/intersection. The observed concentration data of CO for the critical winter period (21 st - 31st December. 1999). at two-air quality control regions- a traffic intersection and an arterial road in the Delhi city, have been compared with model predictions. A range of statistical indicators has been used for model performance evaluation. The results show that ANN based line source models are comparatively more accurate in predicting the CO concentration near urban roadway/intersection than the deterministic and univariate statistical models. Copyright © 2009 by IICAI.