Evaluating the species-specific microclimatic performance of urban street trees in the city of Dresden, Germany

Abstract

This study evaluated the microclimatic performance of three urban street tree species: Sophora japonica, Koelreuteria paniculata, and Tilia cordata, in Dresden, Germany, focusing on their cooling and humidifying effects under current and projected climate scenarios. Using on-site microclimatic measurements and ENVI-met simulation modeling, the research examined how species identity, age class, and canopy presence influence air temperature and relative humidity. Results show that both species and age class significantly affect microclimatic regulation. Young Koelreuteria paniculata and Tilia cordata provided the greatest midday cooling, reducing air temperatures by 0.6164°C (p < 0.0001, d = 1.015) and 0.7965°C (p < 0.0001, d = 1.20), respectively, while mature stands had a lesser effect (less than 0.4°C). At night, all species and age classes contributed to cooling, with reductions from 0.3276°C to 0.7828°C (p < 0.05), and young Tilia cordata remaining the most effective. A positive correlation between leaf area index (LAI) and cooling in young stands (R² = 0.461) suggests that denser canopies enhance temperature reduction. Humidifying effects peaked at midday. Young Sophora japonica showed the highest humidity increase at 1.606% (p < 0.0001, d = -0.4871), though with a minor effect size. In contrast, both young and mature Koelreuteria paniculata (young: 1.347%, d = -2.47; mature: 0.923%, d = -1.0981) and Tilia cordata (young: 1.347%, d = -2.4681; mature: 1.478%, d = -0.8361) exhibited larger, more significant humidifying effects. Nighttime humidification persisted across all species and age groups, ranging from 0.475% to 1.24%. A strong negative correlation was found between air temperature and relative humidity in both open areas (r = -0.777) and under canopies (r = -0.778). Three-way ANOVA indicated that canopy presence significantly reduced air temperature (F = 3.976, p = 0.0462), but species (F = 18.495, p < 0.0001) and age class (F = 40.671, p < 0.0001) had even stronger effects, with a highly significant interaction (F = 265.452, p < 0.0001). For relative humidity, species (F = 717.297, p < 0.0001), age class (F = 563.681, p < 0.0001), and their interaction (F = 369.212, p < 0.0001) were all highly significant, while canopy presence showed a marginal effect (F = 3.465, p = 0.0627). ENVI-met models were validated for simulating urban microclimates in Dresden, showing high agreement with observed temperatures (R² = 0.889–0.975; RMSE = 1.05–1.84°C; IoA = 0.953–0.991), though some discrepancies appeared under future scenarios. The study recommends integrated, adaptive, and multidisciplinary approaches to urban tree planning, with ongoing research, standardized monitoring, and strong collaboration among stakeholders to ensure the long-term resilience of urban green infrastructure.