Measurement and modelling of Eddy-covariance fluxes of CO2 in the city of Zurich

The city of Zurich, Switzerland, aims to achieve net-zero greenhouse gas emissions by the year 2040. To support the city in monitoring its path towards this ambitious goal, an emission monitoring program has been established with two complementary approaches. The first involves a network of CO₂ mid- and low-cost sensors in combination with atmospheric transport inverse modelling. The second, presented here, combines CO₂ flux measurements from an Eddy-covariance system installed on a 17 m mast on top of a 95 m tall building in the city center with flux footprint modeling and a high-resolution emission inventory.

Here we present a detailed comparison between hourly simulated and observed CO₂ fluxes for a period of two years (August 2022 – August 2024) to evaluate the inventory and its partitioning into source sectors. The simulated fluxes were obtained by multiplying the footprints with the sectorially resolved emissions from the inventory, all available on a 10 m x 10 m grid. The sectorial emissions were scaled by temporal factors describing diurnal, day-of-week and seasonal variability. Traffic emissions, for example, were scaled using actual traffic counts from 182 counters and heating emissions were scaled with a heating-degree-day factor based on outdoor temperatures. In addition to anthropogenic emissions, biospheric CO₂ fluxes from trees, lawns and cropland were simulated at 10 m x 10 m resolution with the Vegetation Photosynthesis and Respiration Model (VPRM), driven by local temperature and radiation measurements and Sentinel-2 satellite observations.

The simulated hourly fluxes, which change in time due to the varying footprints and temporal scaling factors, were found to be strongly correlated with the observed fluxes but were, on average, higher, suggesting that the inventory overestimates the actual emissions from the city. The comparison also allowed us to improve the temporal scaling factors of certain sectors, for example, to better represent the reduced emissions during holidays or the heating demand during the transition periods between winter and summer. Accurately representing the temporal variability is important, as it allows disentangling source sectors that follow different temporal profiles. The results demonstrate the capability of tracking the CO₂ emissions of a central part of Zurich with a single, well-placed flux tower with an accuracy that is suitable for evaluating the expected emission reductions in the coming decades.