Airborne pollens cause pollinosis and have the potential to affect microphysics in clouds; however, the number of monitored species has been very limited due to technical difficulties for the morphotype identification. In this study, we applied an eDNA approach to the airborne pollen communities in the suburbs of the Tokyo metropolitan area in Japan, within a mixed urban, rural, and mountain landscape, revealing pollen seasonality of various taxa (a total of 78 families across the period) in the spring season (February to May). Those taxa distinctly shifted in the season, especially in the beginning of February and the middle of April. Air temperature shift was an obvious key factor to affect the airborne pollen community, while the influence of other meteorological factors, such as wind speed, humidity, and precipitation, was not clear. Taxonomic classification of major Amplicon Sequence Variants (ASVs) indicates multiple pollen sources, including natural forest, planted forest, roadside, park lands, and horticultural activities. Most major ASV belongs to Japanese cedar (Cryptomeria japonica), which is the most notable allergen that causes pollinosis in Japan, peaking in mid-February to March. Backward trajectory analysis of air masses suggests that the Japanese cedar and other Cupressaceae plantation forests in the western mountains were a significant source of airborne pollen communities detected at our sampling site. Other major plant pollen sources, including Japanese zelkova (Zelkova serrata) and ginkgo (Ginkgo biloba), emanated from the nearby parks or roadside regions. This study’s approach enables us to visualize the phenology of multiple pollen, including timing and duration. Long-term monitoring of this type would provide additional insight into understanding the role of climate change on pollen transmission and links to flowering events.