Felix Wagner stood at the intersection of climate science and urban design, holding 10 million data points that would fundamentally reshape how cities approach carbon emissions. His research, just published in Environmental Research Letters, analyzed mobility patterns across six major cities—Berlin, Boston, Los Angeles, the San Francisco Bay Area, Rio de Janeiro, and Bogotá—revealing that where people live matters far more than how densely they live overall.
The finding cuts against decades of urban planning orthodoxy. While city-wide population density and transit networks are important, the study shows that what truly drives down car commutes and emissions is a single, elegant principle: homes should be located close to city centers and workplaces. This means the most effective climate strategy isn't building everywhere denser—it's building smarter, in the right places.
Using GPS data, travel patterns, and artificial intelligence, Wagner's team at the Potsdam Institute for Climate Impact Research, working with partners at UC Berkeley, the University of Sussex, and others, uncovered unexpected relationships between urban factors. They discovered that population density and road connectivity are tightly coupled, while income influences driving behavior mainly through where people choose to live, not whether they can afford to drive. This kind of causal analysis, rather than simple correlation, changes everything about how planners can responsibly recommend development.
The spatial specificity is striking. In Berlin, the same densification measure produces wildly different results depending on the neighborhood—emissions could shift anywhere from minus 0.8 to plus 2.9 kilograms of CO₂ per journey relative to the city average. Two kilometers can mean the difference between a climate win and wasted effort. This granular insight is new. "Urban planning experts often discuss densification as a one-size-fits-all policy," explains PIK researcher Felix Creutzig, "but our data show that a single measure can significantly shorten commuting distances in one neighborhood, yet have little effect two kilometers away."
The study identifies a surprisingly elegant solution in monocentric cities like Berlin and Boston: a ring-shaped corridor of neighborhoods around the city center. In Boston, the sweet spot for infill development lies 10 to 21 kilometers from downtown, where land is less built-up yet the center remains accessible. Rio de Janeiro's optimal ring extends to 40 kilometers. In polycentric cities like Los Angeles and the San Francisco Bay Area, emissions fall by densifying areas with high concentrations of jobs—a different geometry for a different city shape.
But urban planning alone isn't enough everywhere. For neighborhoods far from employment centers, supplementary strategies become essential: transit-oriented development, restrictions on sprawling greenfield development, carpooling, and flexible work arrangements. The findings suggest a more realistic, locally-grounded approach to urban climate action.
Wagner's methodology is being shared openly through GitHub, inviting other researchers to build on the work. The CircEUlar project that supported this research represents a shift toward evidence-based urbanism—replacing hunches and ideological battles over density with precise maps of where intervention actually works. For cities grappling with climate targets, this is the difference between scattering efforts and focusing them where they matter most.