Continuous fossil fuel CO$_2$ emission estimates based on $^{14}$C-calibrated urban eddy-covariance fluxes of co-emitted species

Atmospheric tall-tower measurements of radiocarbon ($^{14}$CO$_2$), carbon monoxide (CO), and nitrogen oxides (NO𝑥) show great potential for determining fossil fuel (ff) CO$_2$ emissions. While $^{14}$CO$_2$ relaxed eddy accumulation (REA) enables 𝐹f fCO$_2$ estimation during individual hours, 𝐹CO and 𝐹NO𝑥 , measured directly and continuously by eddy covariance (EC), can be used as proxies for continuous 𝐹f fCO$_2$ once local 𝐹proxy /𝐹f fCO$_2$ ratios (𝑅proxy ) are known. We combined for the first time half-hourly EC 𝐹CO and 𝐹NO𝑥 , measured along 𝐹CO$_2$ at a tall tower in Munich from July 2024 to February 2025, with co-located 𝐹f fCO$_2$ from 76 hour-long $^{14}$CO$_2$ REA measurements. First, we examined the general applicability of CO- and NO$_𝑥$-based 𝐹f fCO$_2$ estimation by analyzing the $^{14}$C-based 𝑅proxy. Then, we applied the $^{14}$C-calibrated ⟨𝑅proxy ⟩ to the 𝐹proxy time series to obtain continuous f fCO$_2$ and non-fossil (nf) CO$_2$ fluxes, and analyze the 𝐹CO$_2$ composition. We demonstrate the potential of the purely observation-based partitioning approach by comparing 𝑅proxy and half-hourly 𝐹f fCO$_2$ and 𝐹nfCO$_2$ with modeled fluxes obtained from a footprint-weighted state-of-the-art Munich emission inventory and biospheric flux model. ... mehrAlthough variability in 𝑅proxy limits precision and accuracy of half-hourly, proxy-based 𝐹f fCO$_2$ , average daytime fluxes could be estimated with about 20% uncertainty. On average, 𝐹f fCO$_2$ accounted for 70% of 𝐹CO$_2$ during daytime. The absolute modeled 𝐹f fCO$_2$ was 46% higher due to increased modeled heating emissions in the cold months. Observation-based and modeled 𝐹nfCO$_2$ indicated substantial human respiration and minor biospheric flux contributions. The multi-proxy EC and REA measurements enabled unique analysis and independent evaluation of urban f fCO$_2$ and nfCO$_2$ fluxes.