On Hot Summer Days, Munich's Street Trees Absorb More CO2 Than Its Traffic Produces

City planners have long argued for more trees. The evidence for cooling, drainage, and mental health benefits is solid. The carbon math has been fuzzier - partly because the models used to calculate urban CO2 flows couldn't actually see most of the trees. Satellite-based biogenic flux models typically operate at 500-meter resolution, which means a row of lindens along a boulevard is invisible. Individual trees, patches of park, green rooftops - all lost in the pixel.

A team at the Technical University of Munich has built something finer. Their new model runs at 10-meter resolution, fine enough to map individual urban vegetation, and the results have a striking headline: on some summer days in Munich, the city's trees absorb more CO2 than its urban traffic emits.

What ten meters of resolution actually reveals

Jia Chen, a professor of environmental sensing and modeling at TUM's School of Computation, Information and Technology, and her doctoral student Junwei Li validated the model with field measurements taken in Munich's urban parks between April 2024 and February 2025 - data collected the old way, by going outside with instruments. The combination of high-resolution modeling and ground truth from direct biospheric flux measurements is what gives the results credibility.

The model distinguishes between different types of urban vegetation, and the contrast between trees and grass is particularly interesting. Trees, it turns out, are doing most of the heavy lifting. On peak summer days their photosynthesis is strong enough to at least match - and sometimes exceed - the carbon being emitted by the city's vehicle fleet.

Grass tells a different story. Because soil respiration continues through seasons when photosynthesis slows, grassy areas emit more CO2 over the course of a full year than they absorb. On an annual basis, lawns and meadows are net carbon sources, not sinks. That finding does not mean grass is bad; it means the carbon accounting for urban greenery is more nuanced than a simple green-is-good shorthand.

The heterogeneity problem

"The current study shows that the urban vegetation landscape is very heterogeneous," Chen said. "Our high-resolution analysis reveals which areas actually have an impact on the climate."

That heterogeneity matters practically. If city planners are trying to maximize carbon sequestration per square meter of available space, the model's resolution allows them to compare specific interventions - a row of mature trees versus a ground-level park, say - in ways that 500-meter pixels simply cannot support. It also means carbon credits or offset calculations based on coarser satellite data are likely underestimating what urban trees actually do.

Chen is careful to situate the finding in its proper context. Trees do much more than absorb carbon: they lower urban temperatures through evapotranspiration, manage stormwater, and improve the quality of life in neighborhoods that might otherwise be entirely sealed surfaces. The carbon result is one layer of a larger picture.

Next cities on the list

The methodology was developed and tested in Munich and Zurich, in collaboration with the University of Basel, the Swiss materials science institute EMPA, and the German Aerospace Center (DLR), with support from the EU project ICOS Cities. The team plans to apply it to additional cities, which will eventually reveal whether the summer-absorption effect holds up across different urban morphologies, climates, and tree species mixes - or whether Munich and Zurich are in some way exceptional.

The broader implication is that the standard tools used to audit city-level carbon emissions have a systematic blind spot for urban biology. Closing that gap, one city at a time, is likely to produce a more accurate accounting of where cities actually stand on their climate commitments.