Urban residents in tropical or subtropical climates may suffer from serious overheating issues due to global warming and urban heat island effects, causing high mortality, poor health, high energy load, and social disorder. Researchers have proposed various methods to mitigate overheating in outdoor environments, including altering the urban geometry, using cool surfaces, incorporating waterbodies, and using vegetation. Trees are highly effective at cooling. Previous studies demonstrate that the cooling effects of trees depend upon their physical characteristics and/or surrounding environments. This project conducted a comprehensive investigation into tree cooling effects at different levels, including the cooling effects of grouped trees in urban green spaces and individual trees in open spaces. They were studied in separation through field monitoring studies on the Gold Coast, a coastal city in Australia with subtropical climate features. The first part of the project studied grouped trees in urban green spaces. Eighteen urban parks were investigated. These parks are characterised by various tree coverages (PTC) and different distances from the sea (DFS). Monitoring results demonstrated that parks with a higher proportion of tree coverage or closer to a waterbody were cooler. After measuring each park for one full day and analysing the hourly data, the cooling intensity of parks (PCI), defined by the air temperature differences between the park and the Griffith University metrological station, demonstrated periodic variations influenced by PTC or DFS. Analyses of variance (ANOVA) found that PTC significantly correlated with PCI in the evening while, DFS had an important effect in the morning. For further exploration, multiple linear regression models were established. PTC was increasingly significant from morning to evening, its R2 generally increased from 0.01 (at 10:00) to 0.2 (at 16:00). In contrast, the effect of DFS weakened after peaking at noon. Trees showed poorer cooling effects during the daytime compared to the cooling effects of the sea. The second part of the project focused on the characteristics of a tree component (crown) that accounts for a large portion of the cooling effect. The tree crown characteristics were indexed by the crown diameter, trunk height, sky view factor (SVF), and leaf area index (LAI). Four new indicators that describe crown characteristics were proposed: the gross canopy index (GCI), leaf layer index (LLI), crown volume index (CVI), and total canopy index (TCI). The former two are basic indicators, and the latter two are sophisticated indicators. The results showed that the two sophisticated indicators were significantly associated with the mean radiant temperature beneath the tree crown. Further, increasing the value of each indicator by 10% could help reducing the mean radiant temperature by 0.67°C (CVI) and 0.83°C (TCI), respectively, when the other predictors were fixed. Similarly, increasing the value of each basic indicator by 10% could reduce the mean radiant temperature by 0.78°C (GCI) and 0.06°C (LLI), respectively. Still, their cooling effects were less significant than CVI and TCI. The thermal effect of the two sophisticated indicators was more significant than that of the two basic ones. Among all, TCI was the best predictor of the outdoor thermal environment underneath the tree crown. A similar result was found using air temperature as the index. However, these indicators can rarely predict air velocities and relative humidity. Part three explored temperature distributions under individual tree canopies. It was found that the mean radiant temperature (MRT) gradually increased when moving away from the trunk towards the shade. This increased intensity was affected by the crown size (TCD), tree height, sunlight angles, and distance from the trunk (DFT). The MRT followed a polynomial function with DFT. A sharp increase in the MRT was witnessed due to the disappearance of the shade. For a tree with a TCD of 16.6 m, MRT differences between a location near the trunk and a location outside the tree canopy were up to 18°C. This finding has confirmed that the cooling effect from individual trees was due to their canopied areas. This project has investigated the cooling effects of trees at various levels and the factors influencing these effects in a coastal city of Australia. The results echoed that some urban planning and landscape policies and strategies improve sustainability and habitability of the city. Findings provide a rational basis for future urban planning and landscaping practices.