1. Ecology and Evolutionary Biology, Cornell University, Ithaca, New York 14853, USA
2. Institute of Ecosystem Studies, Millbrook, New York 12545, USA
3. Present addresses: US EPA Western Ecology Division, Corvallis, Oregon 97333, USA (J.W.G.); Department of Integrative Biology, University of California, Berkeley, California 94720, USA (T.E.D.).

Correspondence to: Jillian W. Gregg^1,3 Email: jwg1@cornell.edu

Plants in urban ecosystems are exposed to many pollutants and higher temperatures, CO₂ and nitrogen deposition than plants in rural areas^{1, 2, 3, 4, 5}. Although each factor has a detrimental or beneficial influence on plant growth⁶, the net effect of all factors and the key driving variables are unknown. We grew the same cottonwood clone in urban and rural sites and found that urban plant biomass was double that of rural sites. Using soil transplants, nutrient budgets, chamber experiments and multiple regression analyses, we show that soils, temperature, CO₂, nutrient deposition, urban air pollutants and microclimatic variables could not account for increased growth in the city. Rather, higher rural ozone (O₃) exposures reduced growth at rural sites. Urban precursors fuel the reactions of O₃ formation, but NO_x scavenging reactions⁷ resulted in lower cumulative urban O₃ exposures compared to agricultural and forested sites throughout the northeastern USA. Our study shows the overriding effect of O₃ despite a diversity of altered environmental factors, reveals 'footprints' of lower cumulative urban O₃ exposures amidst a background of higher regional exposures, and shows a greater adverse effect of urban pollutant emissions beyond the urban core.