Many studies have shown that urban forests function as important carbon (C) sinks by sequestering carbon dioxide (CO₂), and there are a great deal of scientific data on their potential to fix CO₂ through photosynthesis, including variations among vegetation types and temporal dynamics. However, although vehicle traffic is one of the main anthropogenic sources of CO₂, the relationship between these emissions and sequestration by vegetation is unclear. Here, we use the eddy covariance technique to directly measure the net CO₂ flux to: (1) quantitatively validate C emissions from transportation and C sequestration by vegetation at the spatio-temporal resolution of 1 km and 30 min and (2) select tree species that best sequester C by photosynthesis and identify the major controlling factors. During the daytime monitoring period (7:00-17:00) of the plant-growing season (May-October), the net photosynthetic C sequestration per tree was used to measure the C fixation capacity of different tree species, and the order was Mono maple > Amur cork tree > Goldenrain > Chinese ash > Chinese pine > Ginkgo. In the study, C sequestration by trees was primarily controlled by photosynthetically active radiation, traffic volume and relative humidity, which together explained 92.3% of the total C sequestration by trees during the monitoring period in the growing season, and C sequestration was positively correlated with photosynthetically active radiation and traffic flow, but negatively correlated with relative humidity. Furthermore, vehicle CO₂ emissions significantly increased the amount of photosynthetic C sequestration due to a fertilization effect. Therefore, it is necessary to improve the accuracy of micro-scale regional C flux measurements in order to more accurately determine CO₂ sources and sinks and inform the selection of vegetation that can maximize the sequestration of traffic CO₂ emissions.