The study explored the effects of urban riparian buffers (RBs) with different structures on air temperature (T) and relative humidity (RH) in the North Mort of Beijing. From May to October in 2011, four RB types were selected on both the north and south RBs based on structural differences (such as width, slope, plant species, vegetation coverage, and distance and height to road). By both sides of the roads vertical to these RBs, the corresponding control areas (CAs) were set, adjacent to surrounding different urban land cover types. From 8:00 to 18:00, 10-hour measurements for T and RH were taken in different RB types and their CAs on sunny or cloudy days of late each month. Statistical analysis was used to test significance of differences in T, RH, and Thermohygrometric Index (THI) between RBs and CAs, between north and south RB types, and among different RB types and cross sectional zones. The results demonstrated that RBs had significant adjusting effects on T and RH compared with CAs, and could effectively improve human comfort. Firstly, from May to September, T and RH for RBs were extremely significantly lower and higher than those for CAs, respectively. Secondly, most daily variations in T for CAs showed a bimodal trend, but they had a single peak for RBs; the highest T for RBs occurred between 14:00 and 15:00, which was one hour later than that for CAs; RBs could significantly reduce the highest daily T, although it changed the lowest daily T little; meanwhile it could also effectively decrease high-temperature days. Thirdly, daily variations in T and RH for RBs were less than those for CAs. These differences depended on season, that is, the adjusting effects of RBs on microclimate range were especially strong in spring and autumn, but not in summer. Finally, THI for different RB types were significantly different from those for their CAs: all the CAs made human feel uncomfortable in the summer; the south RB types could effectively enhance human comfort level in June and August, and the north RB types could also do that in August. The results also showed that differences in adjusting effects on T and RH also existed in different RB types. Firstly, south RBs had stronger adjusting effects than north RBs, that is, T and RH for the north RB types were significantly higher and lower than those for the south RB types, respectively. It was mainly due to tree density, canopy structure, and riparian width. The wider south RB with higher tree density and denser upper canopy could partly block direct solar radiation. Secondly, the differences in T and RH among different RB types changed with seasons: there were significant differences in May and October, but not from June to September, related to vegetation coverage, riparian width, distance to the road, and surrounding land cover types. When vegetation coverage was higher than a certain threshold, its increase would not obviously influence T and RH for RB. Wider RB with open surroundings and larger distance to road would lead to stronger effects on microclimate. Finally, the differences in THI among different RB types were almost the same as those in T. In addition, there were no significant differences among different cross sectional zones within each RB type because of the limited RB width. The study may play an essential role on ecosystem service assessment of urban riparian buffer, and further provide theoretical basis for the planning, restoration and management of urban riparian buffers.