No abstract

No abstract

No abstract

We present data from a new mobile system intended to aid in the design of upcoming urban CO 2 -monitoring networks. Our collected data include GPS probe data, video-derived traffic density, and accurate CO 2 concentration measurements. The method described here is economical, scalable, and self-contained, allowing for potential future deployment in locations without existing traffic infrastructure or vehicle fleet information. Using a test data set collected on California Highway 24 over a 2 week period, we observe that on-road CO 2 concentrations are elevated by a factor of 2 in congestion compared to free-flow conditions. This result is found to be consistent with a model including vehicle-induced turbulence and standard engine physics. In contrast to surface concentrations, surface emissions are found to be relatively insensitive to congestion. We next use our model for CO 2 concentration together with our data to independently derive vehicle emission rate parameters. Parameters scaling the leading four emission rate terms are found to be within 25% of those expected for a typical passenger car fleet, enabling us to derive instantaneous emission rates directly from our data that compare generally favorably to predictive models presented in the literature. The present results highlight the importance of high spatial and temporal resolution traffic data for interpreting on-and near-road concentration measurements. Future work will focus on transport and the integration of mobile platforms into existing stationary network designs.