|
Global Warming is causing increasing Heat Waves (HW) that affect human health. In this context, urban heat islands (UHI) increase the effects of heatwaves, representing a serious inconvenience to human health and comfort. For these reasons the study of UHI is of great importance in the context of climate change and global warming. The literature on urban climate has highlighted the singular importance of the nighttime UHI. High nighttime temperatures have a negative effect on human comfort and health. Situation that is aggravated in extreme heat events, such as Heat Waves. For this reason, this work seeks to study the spatial distribution of temperature in a situation of maximum nocturnal thermal stress, in order to know the real impact of UHI in heat wave episodes. Given the practical identity between air temperature (that is, that experienced by humans) and land surface temperature (obtained by remote sensing), the study of the nocturnal LST is of great importance. This paper aims to develop a nighttime land surface temperature (LST) model using Landsat imagery in order to study UHI contribution in HW episodes. Once nighttime LST (supplied by the Landsat 8 TIR sensor) has been obtained, the analysis of the mean temperature of land uses (obtained by Urban Atlas) allows a first approach to the UHI at night. First of all, a "geographic" OLS model is developed, with explanatory variables such as longitude, latitude, altitude, slope, orientation and distance to the sea. Model that allows knowing the impact of the "physical" variables in the spatial distribution of the nocturnal LST, without considering the urban aspects. Next, the real surface temperature of each type of urban and rural landscape is compared with that obtained in the "geographic" model. Comparison that allows knowing not only which landscapes are hotter (or colder), but also their heat balance in relation to their physical-geographical characteristics. Finally, an OLS model is developed integrating, together with the "geographical" variables, "urbanterritorial" variables (such as NDVI, NDBI, albedo, imperviousness, …). "Hybrid" model that allows to know in detail the spatial distribution of the UHI, as well as the contribution of each type of urban landscape to the night UHI. The case study is the Metropolitan Area of Barcelona (636 km2, 3.3 million inhabitants).
|
