Spatial definition of the cooling effect of urban green spaces using remote sensing: Case studies in the Barcelona Metropolitan Area

Abstract

Urban green spaces play a fundamental role in the climate change adaptation of the cities. Commonly, the high concentration of vegetation within cities is accompanied by an increase in humidity in the air and a greater projection of shadows on the surfaces. Which breaks the continuity of the artificialized ground cover distinctive of the cities and the high amount of solar radiation absorbed by it, which causes, in part, the urban heat island effect (UHI).

In this sense, the green spaces register a reduction in temperature in relation to their urban context, which commonly extends over the closest surroundings. This effect is known as the urban green spaces cool island (GCI) and is commonly addressed through two indicators of magnitude: the extent and intensity of cooling. The cooling extent (Lmax) refers to a spatial indicator that describes the distance between the perimeter of the green space and the furthest point of the spread of its microclimatic effect on its surroundings. While the cooling intensity (ΔT) describes the temperature difference between the urban context and the green space. Given this, the literature has addressed the cooling quantification through three types of analysis of the climatic behavior of urban spaces: 1) field measurements, 2) numerical modeling and 3) remote sensing. In general, there is a broad consensus on the quantification of the cooling effect of green spaces by calculating the ΔT in the three types of approximations. However, the spatial definition of the cooling extent has been evolving in recent decades and presents an open panorama for methodological proposals that are appropriate to different contexts. Particularly, the recovery of the Land Surface Temperature (LST) from satellite images has allowed the inclusion of larger-scale microclimatic studies to discuss the spatial definition of the thermal influence of urban spaces through statistical approaches.

Given this, the present work proposes a multiple-stage approach to spatial analysis for quantifying the cooling effect of green spaces in the metropolitan area of Barcelona from the LST of the Landsat-8 OLI/TIRs satellite. We select the summer period as a case study because of the increased vulnerability posed by climate change in cities during extremes heat waves episodes, which is accentuated by the UHI. We quantify the Lmax and ΔT of the cooling effect of seven green spaces in the conurbation of Viladecans, Gavà and Castelldefels through three analytical methods based on multiple stages of spatial subdivision of urban surroundings by concentric rings. The first results show a ΔT of 1.25ºC and 1.50ºC in relation to the concentric rings of 0-100m and 100-300m respectively. The Lmax calculated with the 10m-width concentric rings registered an average of 91.67m with a maximum ΔT (ΔTmax) of 1.22 ° C. Finally, with 10m-width cross sections in addition to the concentric rings of an arborized street, we identify an average ΔTmax of 2.21ºC in industrial areas, 1.05ºC in residential areas and 1.76ºC in spaces adjacent to another park. As well as an average Lmax of 109.00m to the northeast and 129.67m to the southwest, with a maximum of 170.00m in the industrial areas and 310.00m in the area adjacent to the other park. The ΔTmax records a correlation of 0.81R² (p<0.01) with the average LST of the closest surroundings to the perimeter of the park, while resulting in a non-significant correlation with the LST of the parks. In the conclusions, we discuss the differences between the methods applied and the considerations for their reproduction in larger-scale studies. The present study is part of the project “Urban-CLIMPLAN. The urban heat island: effects on climate change and modeling for territorial and urban planning strategies. Application to the Metropolitan Region of Barcelona”; financed by the Ministry of Economy and Competitiveness of Spain (MINECO) and the European Fund for Regional Development (FEDR).