Abstract

Urban forests affect land surface temperature (LST) within a city due to the cooling effect of transpiration. The latter depends on tree health, but it can also be affected by the structure and composition of forest, as a monospecies environment may potentially worsen the health of urban forest. The following hypotheses are therefore proposed: a) greater tree diversity within urban forest results in lower LST at the city block scale; and b) the state of biotic disturbance of urban forest is negatively correlated with LST. The present research explores the relationship between urban forest tree diversity and health based on a survey of 38,950 individuals in the district of Providencia in the city of Santiago, Chile, and compares this information against LST data from the ASTER satellite instrument at the city block scale. The health of the urban forest was determined by expert knowledge means of a field survey that collected data concerning growth stage, phytosanitary state, and state of biotic disturbance. The first hypothesis could not be tested by the lack of urban tree diversity which showed strong domination of three species with more than 52 % of abundance (Robinia pseudoacacia, Platanus orientalis and Acer negundo). The second hypothesis was proved since the results revealed a positive and significant correlation between urban forest diversity and LST, with a Spearman?s correlation coefficient of between 0.56 and 0.7. A positive and significant correlation of 0.55 was found between mean biotic disturbance (BDSm) and median LST (Med), indicating a direct relationship between higher LST and poorer urban forest health. A possible explanation is that, among the trees surveyed within the urban forest, the effect of biotic disturbance is greater than that of species diversity. As such, it may be concluded that planting of trees on city streets as a means of temperature moderation is made less effective if specimens are maintained in a poor general condition of health.