Exploring the potential of aggregated traffic models for estimating network-wide emissions

This paper explores the potential of aggregated traffic models based on the Macroscopic Fundamental Diagram for building a network-wide monitoring system of travel emissions. Such a system consists of two layers. In a bottom layer, an aggregated traffic model predicts the network dynamics. In a top layer, an emission model estimates the total exhaust emissions. This paper discusses how to properly calibrate the aggregated traffic models of the bottom layer, to then estimate the total network emissions. We focus on the calibration of travel distances and the network partition definition. We propose a methodology that utilizes the concept of the detour ratio as a proxy to model real travel distances within the aggregated traffic models. This methodology increases the effectiveness of aggregated traffic models for predicting network-wide emissions in realistic scenarios. We also show that the definition of the network partitioning can significantly influence the total network emissions estimation.