Assessing Urban Greening Strategies to Mitigate Heatwave Impacts in Greater Athens Metropolitan Area, Greece ^†

Abstract

As cities grow, natural surfaces are replaced by heat-retaining materials, raising urban temperatures and intensifying heatwave impacts. The present study investigates the effectiveness of urban greening strategies, including green roofs, street vegetation and metropolitan parks, in enhancing climate resilience in Athens, a coastal Mediterranean city characterized by complex heatwave dynamics. The strategies were evaluated through simulations using the WRF model coupled with the BEP-BEM urban canopy model and a detailed land-cover map that is uses the 11 urban Local Climate Zones (LCZ) categories (CLIMPACT) tailored for Athens. Simulations focused on a significant heatwave event that affected the region in 2021 assessed the thermal impacts of the different greening scenarios. Results show that expanding green areas reduces peak temperatures and modifies local thermal circulations, highlighting the potential of greening in mitigating urban heat island effects.

1. Introduction

Urban heating is one of the major concerns for Mediterranean cities. Compact urban areas with limited vegetation in conjunction with the increasingly frequent heatwaves driven by climate change, intensify thermal stress on populations and infrastructure. In coastal urban regions, such as the Greater Athens Area (GAA), these challenges are further intensified by complex interactions between synoptic-scale winds (e.g., Etesians), mesoscale sea-breeze systems and urban-induced heat island circulations [1,2]. These dynamic interactions can amplify or modify the urban thermal environment in non-linear ways, especially during prolonged heatwave events. Thus, it becomes important to assess urban adaptation strategies based on site-specific microclimatic conditions for the effective mitigation of urban heating.

This study aims at investigating the cooling potential of different urban greening strategies designed to reduce heat stress during extreme heat events. The scenarios selected include the integration of green roofs and the enhancement of vegetated areas across the city. These interventions are assessed through mesoscale atmospheric modeling with urban canopy parameterizations to capture both local and regional-scale thermal dynamics. The analysis focuses on the GAA as a representative case where complex synergies between natural ventilation processes (via sea breezes and Etesians) and anthropogenic heat accumulation coexist. By examining the thermal impacts of each greening strategy, this research aims to support evidence-based urban planning for climate-resilient Mediterranean cities.

2. Materials and Methods

2.1. Study Area

The GAA, the capital of Greece, covers 361 km² and comprises 40 municipalities divided into five regional units: North, South, East, West Athens, and Piraeus. It is a densely urbanized coastal city located in the eastern Mediterranean, bordered by mountains to the W-NW and E-NE (Figure 1). Heatwaves are frequent in July and August, often intensified by a strong Urban Heat Island (UHI) effect during hot, dry conditions [3]. Local summer circulations are influenced by sea breezes (SB) and UHI, or by synoptic northerly winds (Etesians). Strong Etesians can suppress SB development, while weaker flows allow local thermally induced circulations to form [4]. To evaluate the impact of different greening scenarios on urban heating, a prolonged heatwave (HW) event was selected, spanning from 28 July to 5 August 2021 and peaked on 3 August. This was one of the most intense HWs in Greece over the past decade, marked by extreme temperatures. During this period, strong and persistent anticyclonic conditions dominated the Eastern Mediterranean, driven by a ridge extending from southwestern Libya toward northeastern regions, affecting both Greece and Turkey. This ten-day pattern transported very hot and dry air masses into the area, with 850 hPa temperatures exceeding ~27 °C on 3 August 2021 [2].

2.2. Modeling System

This study employs the WRF model version 4.3.3, integrated with the BEP-BEM urban canopy scheme to provide a detailed representation of urban dynamics, thermodynamics, and radiative processes [5]. The BEP-BEM scheme combines the multilayer Building Effect Parameterization (BEP) with the Building Energy Model (BEM). BEP simulates the influence of urban structures, including horizontal surfaces such as roofs and canyon floors, and vertical elements like building walls, on momentum, heat fluxes, and turbulent kinetic energy [6,7,8]. BEM complements this by incorporating anthropogenic heat emissions, accounting for processes such as air conditioning and thermal exchanges between building interiors and the surrounding atmosphere. These energy exchanges play a significant role in shaping urban climate conditions. Finally, the BEP-BEM, incorporates a land-surface scheme for Green Roofs (GRs) [9,10]. It is a 1D model that calculates the energy and moisture budget by taking into account the incoming net radiation, the precipitation or irrigation, the evapotranspiration heat exchange and diffusion of heat and moisture from soil [10].

To account for the heterogeneity inherent in complex urban environments and fully leverage the capabilities of an advanced urban canopy model, accurate input data on land use and urban morphology is essential. In this study, an adapted Local Climate Zone (LCZ) map was used, developed within the CLIMPACT project by modifying the standard Local Climate Zone classification to better represent the urban morphology of Athens [11,12,13]. The map includes 11 urban categories with enhanced spatial detail (Figure 1a). It was generated using remote sensing data, digital elevation models from the Hellenic Cadastre, and GIS layers to derive key morphological parameters such as building and vegetation height, density, and canyon aspect ratio. A decision tree classification was applied to assign each pixel to an LCZ, linking urban form to surface parameters essential for urban climate modeling [11].

2.3. Sensitivity Analysis

To simulate the selected event, the WRF-BEP-BEM was employed in a three-level, two-way nested configuration. The outermost domain, operated at a 9 km resolution, encompassing the broader region of Europe and North Africa, ensuring the accurate representation of large-scale atmospheric patterns. Nested within this, a 3 km grid was used to resolve meteorological processes over Greece. The innermost domain, with a horizontal resolution of 1 km, was centered over the Attica region, enabling a detailed representation of urban-scale dynamics (Figure 2). Each domain was configured with 45 vertical levels to capture the boundary layer and lower tropospheric processes effectively. Boundary and initial atmospheric conditions were provided by the ERA5 reanalysis product (ECMWF), available at 0.25° spatial resolution and updated every 6 h. While the sea surface temperature (SST) fields were retrieved from CMEMS (Copernicus Marine Environment Monitoring Service) data. All model diagnostics and evaluations were based on the hourly outputs of the 1 km resolution domain. The modeling set-up has been evaluated in [2].

Three urban greening scenarios were evaluated: (i) extensive green roof implementation (vegetation type: grass-herbaceous lawn) with 80% roof coverage (grass80), (ii) morphological transformation from LCZ2 to LCZ5 to reflect increased street-level vegetation (change) and (iii) the addition of the large-scale metropolitan park “The Ellinikon” (greening) (Figure 1,

Table 1. Characteristic parameters of the modified LCZ categories.

Parameter	LCZ2	LCZ5	LCZ9
Building Height	3–9 stories	3–9 stories	1–2 stories
Impervious Surface Fraction	99%	70%	30%
Street Width	10 m	20 m	10 m
Vegetation Cover	Minimal	Substantial	Dominant

). Among these, grass80 primarily tends to enhance building-level cooling, LCZ2–5 targets to extend the urban ventilation and surface greening, while the Metropolitan park scenario represented by LCZ9 aims to offer localised cooling effects.

3. Results

The evaluation of the three urban greening strategies revealed distinct impacts on near-surface air temperature (T2m) during the heatwave period, as shown in Figure 3 and Figure 4. The green roof scenario (grass80), assuming 80% vegetation coverage on rooftops, resulted in daytime temperature reductions of approximately 0.5–1.5 °C in the urban hotspots of Athens and Kallithea (Figure 3a,b). These reductions were primarily driven by enhanced evapotranspiration, which is the main physical process that increases latent heat flux and reduces sensible heat flux. The implementation of GRs altered the spatial distribution of temperature over the GAA (Figure 5a,c). The effectiveness of GRs is modulated by site-specific climatic conditions, including ambient temperature and wind patterns, as well as geographic factors such as proximity to the sea. During nighttime the effect of GRs on the ambient temperature was negligible compared to daylight hours (Figure 6a,c). This is because under conditions of limited solar radiation, the cooling contribution of vegetation is substantially reduced, as evapotranspiration is predominantly driven by incoming shortwave radiation.

In contrast, the change scenario, where LCZ2 (Compact Midrise) was replaced by LCZ5 (Open Midrise), exhibited stronger and more spatially extensive cooling during nighttime (Figure 3, Figure 5 and Figure 6). The transformation introduced wider streets and increased vegetation cover (Table 1), enabling improved surface ventilation and reduced heat storage. As illustrated in Figure 3, this scenario led to nighttime T2m reductions of up to 2.0 °C in the urban hotspots of Athens and Kallithea.

The Metropolitan Park scenario (greening), represented by the planned Ellinikon Park and modeled using LCZ9 characteristics (Table 1), demonstrated intense localized cooling effects only for one day, as shown in Figure 4. Areas located within approximately 2 km of the park’s boundary experienced significant daytime T2m reductions, reaching 1.5 °C, particularly on 4 August. However, the spatial and temporal extent of the cooling seems that is strongly affected by the complex interaction between the mesoscale and local circulation patterns in the region. The park is situated in a coastal corridor bounded to the north by orographic terrain, which restricts the inland penetration of the sea breeze and limits the advection of park-induced cool air. This indicates the sensitivity of large-scale urban greening to local wind conditions, topographic influences, and the synergy between park-generated cooling and background atmospheric circulations.

4. Conclusions

In this study the effectiveness of different greening strategies to mitigate the urban heating over the GAA was assessed, during a HW event utilizing the WRF mesoscale model coupled with the BEP-BEM urban canopy model. For this analysis a refined and tailored to GAA land use map provided by the CLIMPACT project based on LCZ classification was utilized. The study examined three distinct urban greening strategies: the introduction of green roofs with 80% grass-herbaceous coverage aimed at enhancing cooling at the building scale; a morphological transformation of the urban structure by replacing LCZ2 (Compact Midrise) with LCZ5 (Open Midrise), intended to increase vegetation cover and promote street-level ventilation; and the incorporation of the planned Ellinikon Metropolitan Park, represented as LCZ9, designed to provide strong but localized cooling effects within the urban landscape.

The results of this analysis indicated that the urban greening strategies proved effective in reducing near-surface air temperature (T2m) during heatwave conditions in Athens, though the performance of urban greening interventions is strongly controlled by local climatic and topographic conditions, including sea breezes, Etesians, and urban heat island circulation.

Among the tested scenarios, green roofs (grass80) produced daytime cooling effects—up to 1.5 °C in dense urban areas like Athens and Kallithea—primarily through increased evapotranspiration. However, their influence was limited to daylight hours and had minimal nighttime impact because their cooling mechanism relies on solar-driven energy exchange processes.

The morphological transformation scenario (LCZ2 to LCZ5) presented the most substantial and spatially extensive cooling benefits, with reductions in T2m reaching 2.0 °C, particularly during the night. This effectiveness was attributed to the introduction of wider streets, increased vegetation, and improved ventilation capacity, which also contributed to lower urban heat storage.

Finally, the Metropolitan Park scenario (greening), represented by the addition of the Ellinikon Park (LCZ9), showed localized daytime cooling of approximately 1.5 °C, but only on a single day (4 August). However, it remains unclear to what extent the surrounding urban areas consistently benefit from the park’s presence, as the cooling effect appears to be highly sensitive to the complex interactions between local wind regimes, sea breeze penetration, and thermally induced circulations. These dynamic and site-specific atmospheric processes complicate the attribution of stable or widespread thermal benefits to the park under varying meteorological conditions.