A multi-model approach to monitor emissions of CO2 and CO in an urban-industrial complex
Other literature type
Denier van der Gon, Hugo A. C.
Molen, Michiel K.
Sterk, Hendrika A. M.
(issn: 1680-7324, eissn: 1680-7324)
Monitoring urban-industrial emissions is often challenging, because observations are scarce and regional atmospheric transport models are too coarse to represent the high spatiotemporal variability in the resulting concentrations. In this paper we present a new combination of a Eulerian model (WRF-Chem with an urban parameterisation) and a Lagrangian transport-deposition model (OPS), demonstrating that a plume model strongly improves our ability to capture urban plume transport. This follows from a comparison to observed CO<sub>2</sub> and CO mole fractions at four sites along a transect from an urban-industrial complex (Rotterdam, Netherlands) towards rural conditions. At the urban measurement site we find strong enhancements of up to 33.1 ppm CO<sub>2</sub> and 84 ppb CO over the rural background concentrations. These signals are highly variable due to the presence of distinct source areas dominated by road traffic/residential heating emissions or industrial activities. This causes different emission signatures that are observed in the CO : CO<sub>2</sub> ratios and can be well-reproduced with our framework, suggesting that top-down emission monitoring within this urban-industrial complex is feasible. Further downwind from the city, the urban plume is less frequently observed and its concentration becomes smaller and less variable, making these locations more suited for an integrated emission estimate over the whole study area. We find that WRF-Chem, although able to represent mesoscale patterns, lacks spatiotemporal detail to reproduce the timing, magnitude and variability of urban plumes at the regional background and urban sites. The implementation of the OPS plume model improves the simulation of the CO<sub>2</sub> and CO enhancements. The bias for extreme CO<sub>2</sub> pollution events is reduced with almost 80 % from 15.4 to 3.4 ppm, while the reproducible fraction of observed variability over 750 measurements more than doubles (to 38 %) with the use of a plume model. Therefore, we argue that a plume model with detailed and accurate dispersion parameters is crucial for top-down monitoring of greenhouse gas emissions in urban environments. Future research could benefit from assimilating observed wind fields to improve the plume representation at urban scales.