Abstract:VOCs (volatile organic compounds) are believed to have close relationships with the chemical composition and physical characteristics of the atmosphere. Biogenic VOCs (BVOCs) are more reactive than anthropogenic VOCs (AVOCs), can contribute to tropospheric ozone and secondary particle formation, and have indirect effects on climate change. Estimates of regional BVOC emissions are thus crucial input parameters of air quality models. However, most of the BVOC inventory work has concentrated on natural habitats in Europe, North America and some tropical regions. Although an increasing number of studies are being carried out in tropical and mid-latitude regions throughout Asia in recent years, few studies are concerned with humid subtropical areas, especially in urban areas. Here, a field survey of vegetation composition and distribution was conducted in a subtropical urban-rural complex, Greater Taizhou Area (28°01'-29°20'N,120°17'-121°56'E), Zhejiang Province. Different models were developed and used to estimate leaf biomass of the main tree species. BVOC were divided into isoprene, monoterpenes and other VOCs (OVOCs) and different algorithms were used calculate them separately. The isoprene emission rate data of the main tree species in this area was identified using a plant enclosure approach followed by a GC-PID analysis, while other emission data was compiled from measurements available in China. Tree distribution information, leaf biomass, plant emissions, and meteorological data were combined to investigate the BVOC emission intensity, temporal and spatial patterns and species features in Greater Taizhou Area. Results showed that the annual BVOC emissions of Taizhou in 2009 is 4.6×1010 g C. Isoprene, total monoterpenes and OVOCs emissions account for 93.8%, 3.5% and 2.7%, respectively. In rural area, the emission intensity of bamboo forest (133.8 t C·km-2·a-1) is two orders of magnitude higher than those of Pinus massoniana forest, Cunninghamia lanceolata forest and evergreen broad-leaved forests (0.9, 0.8, 0.6 t C·km-2·a-1, respectively). So from the perspective of improving air quality, expansion of bamboo forests should be inhibited while increasing the proportion of evergreen broad-leaf forests. Within Taizhou city, the BVOC emission intensity of street and riparian trees is 2.4 t C·km-2·a-1, while the emission intensity of trees in parks and residential areas is 1.6 t C·km-2·a-1, both of which are higher than that of any rural forest types other than bamboo forest. Among the primary tree species within the built-up area, Salix babylonica, Albizia julibrissin, Sophora japonica and Liquidambar formosana exhibited high BVOC emission potential, while Cinnamomum camphora, Magnolia grandiflora and Ginkgo biloba have relatively low emission potential. There were no significant differences in the emission patterns between native and nonnative tree species in Taizhou area. Species with low BVOC emission potential should therefore be chosen for future in urban greening initiatives. The annual BVOC emissions of Taizhou are much higher than those of Beijing (1.6×1010 g C/a) and Hong Kong (1.6×1010 g C/a) because these areas vary in climate and vegetation characteristics. This study also provides a scientific basis for the selection of tree species and improvement of air quality in cities in similar climate zones as Taizhou.