Elsevier

Urban Climate

Volume 33, September 2020, 100648
Urban Climate

A novel method to improve temperature forecast in data-scarce urban environments with application to the Urban Heat Island in Beirut

https://doi.org/10.1016/j.uclim.2020.100648Get rights and content

Highlights

  • We investigate key parameters that contribute to the Urban Heat Island in Beirut.

  • The parameters are the anthropogenic heat, roof and wall albedos, and roof thermal conductivity.

  • The parameters are tuned to match the forecasted temperature and wind speeds to the measured ones.

  • The forecasted temperature and wind speeds are obtained using the single-layer UCM with WRF.

  • We found that the anthropogenic heat is the major parameter contributing to the UHI.

Abstract

The urban heat island (UHI) effect has been the subject of much research due to its adverse effects on health, energy, and the environment. The objective of this work is to investigate the key parameters contributing to the urban heat island effect in Beirut city and to improve the temperature forecast in the city by taking these parameters into account. This is accomplished by coupling the single – layer urban canopy model (UCM) with the Weather Research and Forecasting model (WRF), where the urban parameters in UCM are calculated or fine – tuned to minimize the difference between the measured and the forecasted temperature. Urban parameters that were tuned included the roof albedo, wall albedo, roof thermal conductivity, in addition to the anthropogenic heat. To get optimum values for these parameters, simulations with different values for each parameter were performed during the four seasons and compared with observations from stations that are distributed across the city using a novel method that has not been used before to the best of our knowledge. Comparison is based on the mean, standard deviation, mean bias, root mean square error, and correlation coefficient. We also present an assessment of the UHI in Beirut based on some metrics of urban complexity.

Introduction

Beirut, the capital of Lebanon, is a coastal city that lies on the eastern shore of the Mediterranean Sea. Beirut city (administrative area) spans an area of about 18 km2 and sits atop of two hills, Al-Achrafieh (East Beirut) and Al-Msaitbeh (West Beirut) (Beirut [Online], n.d.).

The city has witnessed rapid urbanization during the past thirty years. One of the main causes of this phenomenon was the civil war which started in 1975 and led to a catastrophic destruction of many buildings in Beirut. Consequently, a massive construction movement started after the war ended in 1990. Another reason for the rapid urban expansion in Beirut is the fact that economic levels in Beirut in terms of availability of jobs and salaries are generally better than those in other regions in Lebanon. Therefore, people from rural areas tend to move to Beirut for the sake of having a better living. As a matter of fact, according to the United Nations Human Settlements Programme (UN-Habitat) (UN-Habitat (2014, 2017)), 88% of Lebanon's population lived in urban areas in 2014. With the need for proper accommodation, the previous figure clearly reflects the general urban status in Lebanon, particularly in Beirut. In addition, urban populations in Lebanon increased further due to the influx of refugees from conflict zones in neighboring countries. Based on the most recent data from the United Nations High Commissioner for Refugees (UNHCR) (UNHCR. (2/2/2019), n.d.) and ReliefWeb (ReliefWeb. (2/9/2019), n.d.), the influx of Syrian refugees resulted in 22% and 5% increase in the population of Lebanon and administrative Beirut, respectively. The increase in energy demand associated with this rapidly increasing urban population and the inability of the Lebanese government to meet this increase has led to the sprouting of diesel generators to fill the supply-demand gap (AEMS, 2017).

Notwithstanding its contribution to the urban economy (Liang and Yang, 2019), urbanization has a negative impact on the human health and the environment (Alberti, 2005; Lu, 2019). A key phenomenon is the urban heat island (UHI) effect which characterizes urban areas that are significantly warmer than their rural surroundings (Oke, 1967). The energy consumption to meet the demands of transportation, space heating, cooling, and domestic hot water use, is a major contributor to the UHI due to the anthropogenic heat release associated with the combustion of fossil fuel. Other aspects influencing the UHI include man-made structures and surfaces that affect the thermal dynamics and solar radiation interaction between the urban environment and the atmosphere. Human exposure to emissions of various pollutants (such as carbon monoxide (CO), hydrocarbons (HC), particulate matter (PM), nitrogen oxides (NOx), sulfur dioxide (SO2) and volatile organic compounds (VOCs)) from traffic, power plants, and diesel generators increases the risk of respiratory and cardiovascular diseases, neurological disorders, cancer, and fertility problems (Rumana et al., 2014; Calderon-Garciduenas et al., 2008; Nakano and Otsuki, 2013; Deng et al., 2016).

In Lebanon, and especially in Beirut, electricity is generated, during outage hours (i.e. when there is no electricity supplied by state-owned electricity suppliers, such as Electricité du Liban (EDL)), through privately owned diesel generators. Recent surveys (Baayoun et al., 2019; Shihadeh et al., 2012) showed that there is on average one diesel generator per two buildings in Beirut. These generators are usually located on street level, in basements, or on rooftops. In addition to releasing anthropogenic heat, emissions from these generators in Hamra, a dense neighborhood in Beirut, were shown to lead to additional daily exposure to airborne carcinogens such as Particle-bound Polycyclic Aromatic Hydrocarbons (PPAH) by about 65% for an outage period of 3 h (Helou, 2012). Furthermore, heavy traffic in the city is another source of anthropogenic heat and pollution. Given the lax governmental regulations that control pollutant emissions from the car fleet, traffic is a significant contributor to both the UHI and air pollution (Halabi et al., 2018).

In this paper, we propose a novel methodology to find optimum values of key urban parameters relevant to the UHI effect in data-scarce urban environments and to improve temperature forecasts in such regions using the obtained values. The objectives of this work are to (1) assess the UHI effect in Beirut in terms of the associated temperature rise, (2) investigate and quantify the key parameters contributing to this effect, and (3) improve the temperature forecast in Beirut city by taking these parameters into account. This is accomplished by comparing predictions, using the single-layer urban canopy model (UCM) coupled with the Weather Research and Forecasting model (WRF), with measurements. To provide a more accurate representation of Beirut in UCM, the urban morphology needs to be accounted for and the urban parameters need to be calibrated to reflect the urban characteristics of Beirut and to minimize the difference between the measured and the forecasted temperature. This process is particularly important for urban environments where data is scarce, as is the case with Beirut city. As a matter of fact, one study (Waked et al., 2013) conducted in Beirut to model pollution dispersion in the city during the summer, using WRF to get meteorological data, relied on default urban parameters due to the lack of data corresponding to such parameters. Another study (Abdallah et al., 2018) that aimed at assessing air quality modeling over Lebanon, also using WRF as a meteorological driver, did not use UCM at all due to the scarcity of geometrical and thermal data that UCM depends on. This proves the need for obtaining such urban parameters for Beirut city.

This paper is organized as follows: Section 2 discusses the methodology used in this work in terms of the modeling tool, domain setup, initial and boundary conditions, tuning process, observational data, and model evaluation. Section 3 highlights optimum values that resulted from the tuning process using statistical analysis, compares temperature predictions – using the tuned parameters – with observations, provides a validation of the obtained values based on physical arguments and concludes with some metrics of urban complexity that contribute to the UHI effect in Beirut. Finally, section 4 provides a summary and sheds the light on some future work.

Section snippets

The modeling tool

In this study, the Weather Research and Forecasting model (Skamarock et al., 2008) (WRF, version 3.9) was used. WRF is a nonhydrostatic, compressible model that uses a mass coordinate system with a terrain following coordinates. It is the next generation open-source mesoscale numerical weather prediction tool designed for both atmospheric research and operational forecasting applications and can produce simulations based on actual atmospheric conditions or idealized conditions with high

Results and discussion

During July and October, the tuning process showed that the optimum values of the roof and wall albedos, which corresponded to the minimum weighted average of the NRMSE, were below 0.1. Unlike the albedo parameters, obtained values for the roof thermal conductivity were of the same order of magnitude as the default values. Regarding the anthropogenic heat, fine-tuned values ranged between 210 W/m2 and 610 W/m2 during the months of July, October, January, and April. Optimum values for the

Conclusion

This paper presented a novel approach to find optimal values of key urban parameters relevant to the UHI effect in urban environments that are characterized by data scarcity. Consequently, obtained numbers were used to improve temperature forecasts. Optimum values of different physical parameters characterizing the urban environment in Beirut were determined by performing numerous simulations using the numerical weather forecasting tool WRF – coupled with the urban canopy model UCM – and

Funding

This work was funded by the Collaborative Research Stimulus (CRS award #103321) from the American University of Beirut (AUB). The sponsor did not have any involvement at any stage of writing this article.

Acknowledgements

The authors would like to acknowledge Fatima Hussein for providing measurements from the Chemistry Building at AUB.

Declaration of Competing Interests

None.

References (66)

  • W.M. Angevine et al.

    Uncertainty in Lagrangian pollutant transport simulations due to meteorological uncertainty from a mesoscale WRF ensemble

  • A. Baayoun

    Emission inventory of key sources of air pollution in Lebanon

  • Beirut [Online]
  • H.A. Bekhet et al.

    Assessing the Elasticities of electricity consumption for rural and urban areas in Malaysia: a non-linear approach

    Int. J. Econ. Financ.

    (2011)
  • Berkeley Lab Heat Island Group
  • F. Chen

    The integrated WRF/urban modelling system: development, evaluation, and applications to urban environmental problems

    Int. J. Climatol.

    (2011)
  • Z. Chun-sheng et al.

    Effects of household energy consumption on environment and its influence factors in rural and urban areas

  • J. Crétat et al.

    Uncertainties in simulating regional climate of southern Africa: sensitivity to physical parameterizations using WRF

    Clim. Dyn.

    (2012)
  • J. Dudhia

    Numerical study of convection observed during the winter monsoon experiment using a Mesoscale two-dimensional model

    J. Atmos. Sci.

    (1989)
  • E. Erell et al.

    Urban Microclimate: Designing the Spaces between Buildings

    (2012)
  • L.E. Halabi et al.

    Developing Cleaner & Efficient Vehicle Policies in Lebanon

    (2018)
  • K. Hammerberg et al.

    Implications of employing detailed urban canopy parameters for mesoscale climate modelling: a comparison between WUDAPT and GIS databases over Vienna, Austria

    Int. J. Climatol.

    (2018)
  • M.A. Helou

    Impact of distributed urban generators on household exposure to carcinogenic airborne particles during rolling blackout episodes

  • S.-Y. Hong et al.

    A revised approach to ice microphysical processes for the bulk parameterization of clouds and precipitation

    Mon. Weather Rev.

    (2004)
  • S.-Y. Hong et al.

    A new vertical diffusion package with an explicit treatment of entrainment processes

    Mon. Weather Rev.

    (2006)
  • P.A. Jiménez et al.

    A revised scheme for the WRF surface layer formulation

    Mon. Weather Rev.

    (2012)
  • J.S. Kain

    The Kain–Fritsch convective parameterization: an update

    J. Appl. Meteorol.

    (2004)
  • N. Kaloustian et al.

    Local climatic zoning and urban Heat Island in Beirut

  • N. Kaloustian et al.

    Effects of urbanization on the urban heat island in Beirut

  • H. Kusaka et al.

    Coupling a single-layer urban canopy model with a simple atmospheric model: impact on urban Heat Island simulation for an idealized case

    Journal of the Meteorological Society of Japan. Ser. II

    (2004)
  • H. Kusaka et al.

    Thermal effects of urban canyon structure on the nocturnal Heat Island: Numerical experiment using a Mesoscale model coupled with an urban canopy model

    J. Appl. Meteorol.

    (2004)
  • H. Kusaka et al.

    A simple single-layer urban canopy model for atmospheric models: comparison with multi-layer and slab models

    Boundary-Layer Meteorology, journal article

    (December 01 2001)
  • P. Lin et al.

    The impact of urban design descriptors on outdoor thermal environment: a literature review

    Energies

    (2017)
  • Cited by (11)

    View all citing articles on Scopus
    View full text