Transpiration rates of urban trees, Aesculus chinensis.

Transpiration patterns of Aesculus chinensis in relation to explanatory variables in the microclimatic, air quality, and biological phenomena categories were measured in Beijing, China using the thermal dissipation method. The highest transpiration rate measured as the sap flux density of the trees took place from 10:00 am to 13:00 pm in the summer and the lowest was found during nighttime in the winter. To sort out co-linearity, principal component analysis and variation and hierarchical partitioning methods were employed in data analyses. The evaporative demand index (EDI) consisting of air temperature, soil temperature, total radiation, vapor pressure deficit, and atmospheric ozone (O3), explained 68% and 80% of the hourly and daily variations of the tree transpiration, respectively. The independent and joint effects of EDI variables together with a three-variable joint effect exerted the greatest influences on the variance of transpiration rates. The independent effects of leaf area index and atmospheric O3 and their combined effect exhibited minor yet significant influences on tree transpiration rates.