THIS – Tool for Heat Island Simulation: A GIS extension model to calculate urban heat island intensity based on urban geometry

https://doi.org/10.1016/j.compenvurbsys.2017.09.007Get rights and content

Highlights

  • THIS allows to verify the potential of urban geometry on influencing the development of urban heat islands.

  • The increasing of the roughness attenuates the growth of the UHI with the increasing H/W ratio.

  • THIS tool is an extension of a GIS to calculate the UHI intensity based on urban geometry.

  • The simplified way of inputting data is an advantage in the THIS tool.

  • THIS algorithm enables the user to insert new correction equations for new validations, when this is necessary.

Abstract

This paper presents the development of a simulation model, which was incorporated into a Geographic Information System (GIS) in order to calculate the maximum intensity of urban heat islands (UHImax) based on urban geometry data (using a H/W parameter). This tool is called THIS – Tool for Heat Island Simulation. The urban heat island phenomenon is defined by the temperature rise in dense city centers compared with the surrounding countryside. The methodology of this study is based on a theoretical-numerical basis (Oke model), followed by the development of a calculation algorithm incorporated into the GIS platform, which is then adjusted and applied as exemplification. This adjustment was made by calibrating the Oke model for a case study based on two Brazilian cities and different various trends for different roughness length ranges were found. As a consequence, this work has resulted in the automation of an algorithm to obtain maximum intensity values of heat islands based on a simplified model. After finishing the subroutine, the application of the THIS in a simulation of different urban scenarios showed different trends in the UHImax value for the H/W ratio and the roughness length. The UHImax increases when the H/W ratio increases, but the urban canyons with greater roughness (larger areas of facades and more heterogeneous heights, Z0  2.0) result in UHImax values of approximately two times smaller than canyons with less roughness (homogeneous with highest average areas occupied by buildings, Z0 < 2.0) for the same value as the H/W ratio. Overall, the developed tool has one aim: to simulate the effect of the isolated variable of urban geometry on the maximum intensity of nocturnal heat islands, considering different urban scenarios.

Introduction

The worldwide known phenomenon called ‘urban heat island’ (UHI) is still a concern for the quality of life. Due to the properties and arrangement of their elements, urban centers tend to store more heat and develop higher air temperatures than those found in the outskirts of the city (or surrounding rural areas). For Hamdi (2010) and Mendonça and Monteiro (2003), the growth of UHI due to the increase in urbanization is particularly important for the influence on the estimation of global warming.

Among the factors that influence the intensity of the heat island, urban geometry can be highlighted. Urban geometry interacts with the exchanging radiation between the Earth and Sky by the phenomena of reflection, absorption and thermal storage. The geometric combination of horizontal and vertical intra-urban surfaces is often referred to as ‘urban canyon’ and generally measured by the height and width (H/W) aspect ratio, the relationship between the average height of the building in an urban canyon and the street width.

Among the analytical studies correlating urban geometry and the formation of heat islands, it is worth highlighting the study conducted by Oke (1981). This author developed an empirically based model for predicting the intensity of nocturnal heat islands based on urban geometry. According to this author, the increase in the H/W ratio corresponds to the decrease in the cooling rate of the urban environment in relation to the rural area. Due to the potentiality of the Oke model, adapting it to a contemporary approach is one of the aims of this paper. Therefore, exploring the model as a computational tool may open up possibilities of urban analysis and make it easier for researchers to use.

Among the computational tools available, Geographic Information Systems (GIS) stand out due to the number of spatial and numerical interactions of geographic objects. In addition to the storage capacity of GIS, they are able to treat and represent tabular data and make it possible to incorporate new techniques and methods into territorial planning.

Associating the issue of urban heat island to the management possibilities offered by a GIS, the purpose of this article is to check the influence of urban geometry on the maximum intensity of urban heat islands. Thus, a computational algorithm was developed as an extension of a GIS, by exploiting its potential to analyze and manage geographical data. In order to do this, the following steps were taken: analyze the theoretical-numerical basis (Oke model); develop a calculation algorithm and implement it in the GIS platform; and adjust the tool by simulation testing of urban scenarios.

Section snippets

Theoretical basis

This section presents some approaches concerning the relationship between urban geometry and heat islands, as well as some examples of using GIS in urban studies.

Methodology

In order to develop this research, a study of the theoretical-numerical basis (Oke's model) shown in the previous sections was followed using a proposal of a calculation subroutine. This was based on the parameter of H/W ratio, as in Eq. (1). The subroutine was then created to identify the potential of urban geometries in developing urban heat islands. Thus, comparing simulated data to an actual condition of a tropical city, adjustments were made to the algorithm, so that the subroutine could

Results

The adapted model was developed to provide higher correlation results with the measured data than those calculated by the Oke model. During validation, it was found that for all the points, the simulated data by Oke's model resulted in an R2 = 0.63 with a standard deviation of 2.20 °C (Fig. 7a). The simulated data by the adapted model resulted in an R2 = 0.92 with a standard deviation of 1.01 °C (Fig. 7b).

The results of the proposed simulation (Fig. 8 and Table 3) demonstrate that all scenarios of Z0 <

Discussion

The results of the simulations of the hypothetical scenarios showed that the increase in roughness does not stimulate the development of greater heat island. By specifically assessing the parameter of ‘urban geometry’ and setting the roughness to a value > 2.0, an increasing trend, less marked than the scenarios of UHImax with lower roughness, was observed.

Some effects of increased building height on the mitigation of UHImax were also found by Theeuwes et al. (2014). These authors obtained a

Conclusion

This paper proposed to verify the influence of urban geometry on the maximum intensity of the nocturnal heat island, applying a computational tool developed as an extension of a GIS. The simplified Oke model (1981), incorporated into the calculation subroutine, was adjusted to provide more approximate results to the reality of two Brazilian cities, which served as the basis for calibrating the model. The comparison between real and simulated data using the Oke model showed a difference in the

Acknowledgments

The authors would like to express their gratitude to the São Paulo Research Foundation (2012/00594-5) (FAPESP), the Coordination for the Improvement of Higher Education Personnel (8802/13-0) (CAPES) and the National Council of Technological and Scientific Development (CNPq) for their financial support.

References (39)

  • CIIAGRO. Centro Integrado de Informações Agrometeorológicas
  • K. Fortuniak

    An application of the urban energy balance scheme for a statistical modeling of the UHI intensity

  • K. Giannopoulou et al.

    The impact of canyon geometry on intra urban and urban: suburban night temperature differences under warm weather conditions

    Pure and Applied Geophysics

    (2010)
  • J.C.S. Gonçalves et al.

    Adensamento urbano e desempenho ambiental no centro da cidade de São Paulo

  • R. Hamdi

    Estimating Urban Heat Island Effects on the Temperature Series of Uccle (Brussels, Belgium) Using Remote Sensing Data and a Land Surface Scheme

    Remote Sensing

    (2010)
  • R. Hamdi et al.

    Sensitive study of the urban heat island intensity to urban characteristics

    International Journal of Climatology

    (2008)
  • IPMET. Instituto de Pesquisas Meteorológicas
  • G.T. Johnson et al.

    Simulation of surface urban heat island under ‘Ideal’ conditions at night. Part 1, theory and tests against field data

    Boundary-Layer Meteorology

    (1991)
  • S.K. Jusuf et al.

    Development of empirical models for an estate level air temperature prediction in Singapore

  • Cited by (51)

    View all citing articles on Scopus
    View full text