Using observed traffic volumes to improve fine-grained regional emissions estimation

When translating travel demand model output to photochemical model input, p eriod-based network assignment volumes must be converted to gridded-hourly vehicle emissions. A post-processor, such as the California Direct Travel I mpact Model (DTIM2), is frequently used to disaggregate the period-based tr avel demand assignments to the fine grained spatial and temporal resolution required by the photochemical models. A recent theoretical enhancement pro posed refining the temporal and spatial resolutions of travel demand model predictions using observed count data. This method provides a technique for disaggregating the period-based travel demand model assignments (e.g., AM peak, PM peak) into the hourly summaries required by most photochemical mod el (Lin and Niemeier, 1997). In this study we present a methodological fram ework for applying the new theory and discuss the results of a large-scale application empirical comparison between the standard and proposed methods for estimating regional mobile emissions in Sacramento, California. The sta ndard method produced slightly higher estimates of daily emissions (about 1 %) when compared to the emissions estimated using observed count data. Howe ver, the two approaches produced hourly emissions estimates that differed b y as much as 15% in some hours. (C) 1999 Elsevier Science Ltd. All rights r eserved.