In urban environments, pedestrian health is affected by both temperature and pollutants during commuting, and the shape of urban structures and afforestation can mitigate these health risks. Computational fluid dynamics (CFD) was used to establish a method considering different aspect ratios (building height/street width, AR = H/W =1, 1.25; W = 20 m) and leaf area density (LAD) were used to obtain the distribution of pollutant concentration and temperature in the street valley. Meanwhile, Environmental Quality index (EQI) was introduced to better evaluate the health risk to pedestrians under the coupling of the two. The results show that high LAD can improve the thermal environment of urban street valley and can reduce the pedestrian activity area by up to 2 ℃, but it also leads to the accumulation of pollutants. Trees can reduce the air flow in the street canyon, and the vortex effect leads to the accumulation of pollutants mainly on the leeward side, and the concentration of leeward side is significantly higher than that on windward side. In the deep street valley, vegetation has great influence on the area above the tree canopy but has limited influence on the area below the tree canopy. In the deep street valley, EQI is significantly lower than that in the shallow street valley. At the pedestrian level, EQI almost does not change with the change of trees. Finally, we found that the deep street valley is more suitable for taller trees and the shallow street valley is more suitable for pedestrians.