Urban Tree CO₂ Compensation by Albedo

Abstract

Urban form and surface properties significantly influence city liveability. Material choices in urban infrastructure affect heat absorption and reflectivity, contributing to the urban heat island (UHI) effect and residents’ thermal comfort. Among UHI mitigation strategies, urban parks play a key role by modifying the microclimate through albedo and evapotranspiration. Their effectiveness depends on their composition, such as tree cover, herbaceous layers, and paved surfaces. The selection of tree species affects the radiation dynamics via foliage color, leaf persistence, and plant morphology. Despite their ecological potential, park designs often prioritize aesthetics and cost over environmental performance. This study proposes a novel approach using CO₂ compensation as a decision-making criterion for surface allocation. By applying the radiative forcing concept, surface albedo variations were converted into CO₂-equivalent emissions to allow for a cross-comparison with different ecosystem services. This method, applied to four parks in two Italian cities, employed reference data, drone surveys, and satellite imagery processed through the Greenpix software v1.0.6. The results showed that adjusting the surface albedo can significantly reduce CO₂ emissions. While dark-foliage trees may underperform compared to certain paved surfaces, light-foliage trees and lawns increase the reflectivity. Including evapotranspiration, the CO₂ compensation benefits rose by over fifty times, supporting the expansion of vegetated surfaces in urban parks for climate resilience.

1. Introduction

As a consequence of the rapid growth of urban areas [1,2], an increasing portion of the global population is exposed to extreme living conditions due to climate change [3,4]. The growing frequency and severity of heatwaves [5] is often exacerbated by the intrinsic characteristics of cities [3,6], leading to adverse health consequences for residents and increased pollution [7]. The changes in surface properties due to urban expansion reduce the cooling effects provided by vegetation and moist soil, making it essential to better understand the complex processes at the intersection of urbanization, the climate, and human health [8]. These changes can lead to negative feedback loops driven by heat-coping mechanisms, such as the use of air conditioning [9], as well as other anthropogenic activities [10,11,12,13]. Throughout the past century, the urban heat island (UHI) phenomenon, defined as the tendency of urban areas to experience higher temperatures than their rural surroundings due to human activities and modified land surfaces [14], has been progressively identified, conceptually refined, and systematically studied [15,16], with growing emphasis on its primary drivers, including urban form, surface material characteristics, and reduced albedo, thereby evolving into a complex and interdisciplinary field of research addressing its causes, impacts, and mitigation strategies [17]. Urban form, defined as the spatial configuration of human activities, influences land cover and use, building density and disposition [18], and urban surface properties, thereby affecting the intensity of the UHI effect [19]. Urban surfaces, defined as horizontal and vertical surfaces of building envelopes and the ground within cities, play a key role in this process [20]. The replacement of natural, permeable surfaces with impervious materials characterized by a low albedo and a high thermal inertia leads to greater heat absorption and reduced evaporative cooling [21,22]. These conditions contribute significantly to elevated urban temperatures and the persistence of the UHI phenomenon. In particular, the albedo, the fraction of incident solar radiation reflected by a surface, has been identified as one of the main contributing factors to UHI mitigation [14,23]. Many strategies have been developed in order to increase the albedo surface in urban environments; such strategies involve the use of high-albedo materials, cool pavements [24,25], and vegetation [3]. However, the urban green infrastructure effect on the UHI is not limited to passive cooling mechanisms [26]; it also comprises several ecosystem services (ESs), i.e., evapotranspiration, shading, and carbon storage [27,28,29], which can result in a significant temperature reduction in urban environments. Nevertheless, a holistic comparison of different infrastructure potentialities for increasing the albedo is still difficult due to the lack of common measurement parameters [17,20]. This study proposes a novel approach using CO₂ compensation as a decision-making criterion for surface allocation, applying the radiative forcing (RF) concept. Radiative forcing represents the net balance between the solar radiation absorbed by the Earth and the radiation re-emitted to the atmosphere [30], and it is deeply correlated with the terrestrial albedo and the atmospheric chemical composition [31].

The Intergovernmental Panel for Climate Change (IPCC) commonly deploys RF to quantify (and compare) anthropogenic and natural drivers of climate change by assessing the rise in the global mean temperature (in Watts/meters) as a consequence of greenhouse gas (GHG) emissions (i.e., CO₂) on Earth’s overall energy balance [32]. Accordingly, the variation in the temperature of the atmosphere due to the cooling of different surfaces can be translated in terms of CO_2eq emissions [33]. In this study, this principle was applied to four case studies in Italy, two parks in Perugia and two in Bologna, to quantify the CO_2eq compensation of different surface allocation compositions depending on their albedo. The expected result of this method is a value representing the cooling effect of an increased reflectivity of urban surfaces in terms of the CO_2eq compensation in order to ensure that the results are comparable to other cooling benefits derived by urban parks.

The present research aimed to (i) assess different compositions of surface typology allocation among four existing parks, including tree cover, lawns, and paved surfaces; (ii) quantify the cooling effect due to albedo in terms of the CO₂ compensated for by the different surfaces; and (iii) investigate the cooling effectiveness of the most common urban tree species from an albedo point of view.

2. Materials and Methods

2.1. Study Areas

The study areas were selected among the Italian pilot cities participating in the LIFE+ Clivut project (CLIVUT LIFE 18 GIC/IT/001217), a participatory European project aimed at identifying the climate values of urban trees that also involves public administrations and citizens in urban green census activities through a web app (http://lifeclivut.treedb.eu/index.php accessed on 10 July 2025) [34,35] to quantify the ecosystem services that they provide. The selected parks were previously investigated by deploying a similar approach to quantify the CO₂ compensation of urban trees through trees’ evapotranspiration (ETP) [33]. Therefore, the same park boundaries were considered in order to allow for results comparisons with this study.

In this study, two urban parks were selected from each Italian city included in the Clivut project web app dataset. The parks chosen include Chico Mendez (43.1048, 12.3627) and Pescaia (43.1044, 12.3810) Parks in Perugia; and 11 September 2001 (44.5004, 11.3365) and Villa Ghigi (44.4769, 11.3266) Parks in Bologna (Figure 1), which were individuated basing on the representativeness of the present tree species and of types of green infrastructure, typical of these Italian pilot cities.

Chico Mendez Park, located in Perugia, spans 16 hectares and was established in the 1980s on flat terrain, featuring pathways and extensive tree-lined avenues. Pescaia Park, also in Perugia, was built in the 1970s near the historic centre, known for its diverse Mediterranean flora spread across varying elevations due to the terrain slope.

In Bologna, 11 September 2001 Park is a centrally located equipped area that emerged from the redevelopment of a former industrial site in the 1970s. Villa Ghigi Park, notable for its historical park characteristics, has been accessible to the public since 1974. In this study, only the wooded area surrounding Villa Ghigi is considered, while the entire holding covers over 30 hectares of agricultural land surrounding Bologna.

2.2. Assessment of the Different Parks’ Surface Areas

As part of the spatial analysis, high-resolution satellite imagery of the studied urban parks was sourced from Google Earth™. The images were processed and analysed using the Greenpix software environment, to enable the categorization and measurement (in square meters) of the different surfaces characterizing the parks (i.e., tree canopy cover, lawn covered surface, streets and paths, and, finally, building by exclusion) [36]. The Greenpix 0.3 software, developed by the Institut de Recerca i Tecnologia Agroalimentàries—IRTA, Barcellona, Spain, is a digital analysis software according to Casadesús et al. [37]

Greenpix evaluates canopy cover from digital images [36], identifying the number of pixels in different hue ranges and converting them in a surface unit of measure [37]. As shown in Figure 2, in a portion of Chico Mendez Park in Perugia given as an example, by selecting a different range of color between 0 and 360 degrees of the HSI (hue, saturation, and intensity) color model, it is possible to isolate and measure (in m²) different typologies of surfaces, i.e., the dry (yellow) grass of the lawn, from the bright-green lawn [38], or the white paths and paved surfaces. The HSI model, currently implemented in the Greenpix tool, is a color model that represents colors in a way that is more aligned with human perception than the RGB model, and it has already been used as an objective indicator for turf quality evaluations [37].

Satellite images of the four parks, captured in June, were cropped into a 200 × 146 m grid and analyzed through the Greenpix software in order to assure a high level of detail and avoid reference length conversion errors. Areas of the photos exceeding the park borders, as represented in the Clivut web app, were excluded from the analysis. Buildings that encroach on the park’s surface were manually excluded from the Greenpix software analysis. The resulting net surface of the park was then calculated as the difference between the total park area and the building surface, as determined by the GIS transposition of the Clivut census areas. Paved surfaces and white paths were found to respond to the saturation S > 12.0 exclusion criteria of the software, while the yellowish dry lawn area responded to the hue range from 20 to 60 (H25–H55), and the green lawn responded better to the hue range from 60 to 180 (H65–H175), similarly to values found in the literature [36,38].

Considering the difficulty of the software to distinguish different vertical layers in the same hue range, the tree canopy area was instead calculated separately based on the Clivut morphological data of the studied trees as approximation of the maximum crown diameter of the trees noted during 2020–2021 LIFE+ Clivut campaign. Additionally, the project web app allows for the collection of several categories of data about trees, i.e., h = geolocation data, morphological parameters, and others, deriving the calculation of various ES provided by each tree species, i.e., particulate matter (PM10) absorption, tree shading, increase in biodiversity, potential heat reduction (Q) ability of the trees, and CO₂ storage. A detailed description of the applied data collection methodology can be found in the Clivut protocol https://www.lifeclivut.eu/public_download/download/8/file_download_en.pdf accessed on 20 November 2024. All web app databases and functions of the web app are open-source and accessible through any internet browser through https://lifeclivut.treedb.eu/, accessed on 20 November 2024 and they are also suitable for GIS applications.

2.3. Albedo Assessment

Albedo measurements of tree crowns were carried out using a drone equipped with an albedometer (Figure 3); specifically, a drone setup (Figure 3a) was used to measure the ability of each tree species present in the studied parks to reflect solar radiation in the visible spectrum.

An albedometer is an instrument consisting of two pyranometers: one facing up-ward, toward the sky, and one facing downward, toward the surface under investigation. In Figure 3b, the two white-colored sensors (pyranometers) are clearly visible, positioned on the upper and lower parts of the drone, respectively. In Figure 3c, the drone in flight mode can be seen, along with the two monitors of the control system that allow for the real-time visualization of the footage from both the drone’s front camera and its thermal camera. The data measured by the two pyranometers are stored in the drone’s internal memory; the albedo value is also calculated in real time as the ratio between the intensity of the radiation incident on the upper sensor and the reflected radiation captured by the lower sensor. Albedo is usually measured using either a fixed sensor or a mobile sensor mounted on a pole and carried over the surface of interest. Using a drone equipped with an albedometer made it possible to collect data over various types of vegetation without stepping on it and covering larger areas.

The albedo values collected this way are reported in

Table 1. Species-specific tree crown albedo values.

Species Name	Albedo (α)	Mean Value	Approximated Value	Reference
Acer campestre	0.21	x		Deng et al., 2021 [39]
Acer negundo 1	0.22		x	Deng et al., 2021 [39]
Acer opalus 1	0.22		x	Deng et al., 2021 [39]
Acer platanoides	0.22	x		Deng et al., 2021 [39]
Acer pseudoplatanus 1	0.22		x	Deng et al., 2021 [39]
Acer rubrum	0.15	x		Breuer et al., 2003 [40]
Acer saccharinum 1	0.22		x	Deng et al., 2021 [39]
Acer saccharum 1	0.22		x	Deng et al., 2021 [39]
Acer spp. 1	0.22		x	Deng et al., 2021 [39]
Aesculus hippocastanum	0.18	x		Drone survey
Aesculus x carnea	0.18	x		Drone survey
Albizia julibrissin	0.60	x		Palomo Amores et al., 2023 [41]
Carpinus betulus	0.19	x		Deng et al., 2021 [39]
Celtis australis	0.18	x		Palomo Amores et al., 2023 [41]
Cercis siliquastrum	0.60	x		Palomo Amores et al., 2023 [41]
Cupressus sempervirens	0.16	x		Vatani et al., 2019 [42]
Cupressus sempervirens pyramidalis 2	0.16		x	Vatani et al., 2019 [42]
Diospyros kaki 3	0.23		x	Breuer et al., 2003 [40]
Fagus sylvatica	0.17	x		Otto et al., 2014 [43]
Fraxinus angustifolia	0.18	x		Weis et al., 2022 [44]
Fraxinus excelsior	0.18	x		Weis et al., 2022 [44]
Fraxinus intermedia	0.18	x		Weis et al., 2022 [44]
Fraxinus ornus	0.18	x		Weis et al., 2022 [44]
Fraxinus spp.	0.18	x		Weis et al., 2022 [44]
Liquidambar styraciflua	0.30	x		Wicklein et al. 2012 [45]
Malus communis 4	0.15		x	Zanotelli et al., 2019 [46]
Malus domestica	0.15	x		Zanotelli et al., 2019 [46]
Malus sylvestris 4	0.15		x	Zanotelli et al., 2019 [46]
Olea europaea	0.19	x		Ramírez-Cuesta et al., 2019 [47]
Picea abies	0.14	x		Weis et al. 2022 [44]
Pinus halepensis	0.14	x		Drone survey
Pinus pinaster 5	0.17		x	Breuer et al., 2003 [40]
Pinus pinea	0.17	x		Drone survey
Pinus sylvestris	0.16	x		Otto et al., 2014 [43]
Platanus acerifolia	0.18	x		Palomo Amores et al., 2023 [41]
Populus alba	0.31	x		Breuer et al., 2003 [40]
Populus nigra	0.31	x		Breuer et al., 2003 [40]
Populus nigra ‘Italica’	0.31	x		Breuer et al., 2003 [40]
Prunus amygdalus	0.40	x		Silva et al., 2025 [48]
Prunus persica	0.28	x		Breuer et al., 2003 [40]
Pyrus calleryana	0.40	x		Silva et al., 2025 [48]
Pyrus communis 6	0.40		x	Silva et al., 2025 [48]
Pyrus pyraster 6	0.40		x	Silva et al., 2025 [48]
Quercus crenata	0.34	x		Breuer et al., 2003 [40]
Quercus ilex	0.15	x		Drone survey
Quercus petraea	0.34	x		Breuer et al., 2003 [40]
Quercus pubescens	0.34	x		Breuer et al., 2003 [40]
Quercus robur	0.34	x		Breuer et al., 2003 [40]
Quercus spp.	0.34	x		Breuer et al., 2003 [40]
Robinia pseudoacacia	0.12	x		Gao et al., 2022 [49]
Tilia cordata 7	0.20		x	Deng et al., 2021 [39]
Tilia intermedia 7	0.20		x	Deng et al., 2021 [39]
Tilia platyphyllos	0.20	x		Deng et al., 2021 [39]
Ulmus carpinifolia	0.28	x		Breuer et al., 2003 [40]
Ulmus laevis	0.28	x		Breuer et al., 2003 [40]
Ulmus minor	0.28		x	Breuer et al., 2003 [40]
Ulmus pumila	0.28	x		Breuer et al., 2003 [40]

¹ Based on A. platanoides; ² based on C. sempervirens; ³ based on Magnolia virginiana; ⁴ based on Malus domestica; ⁵ based on P. pinea; ⁶ based on Pyrus calleryana; ⁷ based on T. platyphillos.

. Considering that further albedo measurements were limited by COVID-19 restrictions, species-specific albedo values were taken from the literature when available in order to evaluate the albedo of the majority of the trees present in the studied parks (approximately over 70%).

The albedo values as listed in Table 1, derived both from the literature and drone survey, allowed for the assessment of the species-specific tree reflectance of 68% of the trees in Villa Ghigi Park, 69% in 11 September 2001 Park, 81% in Chico Mendez Park, and 83% in Pescaia Park.

A reference value was used to model the “Asphalt scenario” and the “Lawn scenario”, as described in Section 2.4. The albedo value for the Asphalt scenario was taken from the literature (α = 0.11) [50], as were the values for green and dry lawn, with α = 0.165 and α = 0.265, respectively [51]. However, the Lawn scenario albedo value was calculated proportionally to the percentage of green and dry lawn actually present in the analyzed park at the time, as assessed through the Greenpix software.

2.4. Albedo Influence on Radiative Forcing (RF)

As described in Fornaciari et al.’s (2024) study [33], the increase in radiative forcing (RF) can be translated into an equivalent compensation of CO₂ [52]. Similarly, in this case study, the model developed by the “Mauro Felli” Interuniversity Research Centre on Pollution and the Environment (CIRIAF) of the University of Perugia was used to determine the cooling effect due to albedo, expressed in CO₂-equivalent terms, by using a latitude-dependent coefficient.

The RF model simulation indicates that an average increase in surface albedo of 0.01 (or 1%) results in a reduction in radiative forcing equal to ΔRFα = −1.27 W/m² [53]. Conversely, the emission of 1 t of CO₂ into the atmosphere produces an average increase in RF equal to 1.63 kW, i.e., ΔRFCO₂ = 1.63 kW per ton of CO₂.

If interpreted in terms of radiative forcing reduction, it can be stated that the removal of 1 ton of CO₂ equivalent from the atmosphere results in a decrease in RF of 1.63 kW, or, equivalently, the removal of 1 kg of CO₂eq translates into a reduction of only 1.63 W in the overall radiative forcing.

Although this value may seem extremely small, if we assume that such a reduction is concentrated on a portion of solar radiation affecting just 1 m² of Earth’s surface, the effect would slightly exceed that produced by increasing the albedo of the same area by one percentage point. Specifically, from the equivalence of the two average ΔRF values, it can be deduced that increasing the albedo of a 1 m² surface by one percentage point is radiatively equivalent to removing approximately 0.87 kg of CO₂ equivalent from the atmosphere (ΔRFf).

Considering that, on average, half of the CO₂ emitted into the atmosphere is absorbed by terrestrial and oceanic carbon sinks [30], it is more appropriate to quantify the compensatory effect of albedo changes by doubling this value. Therefore, the compensation factor becomes ΔRFf = 1.74 kgCO_2eq/m²·Δα. This value should be interpreted solely as an average, since the effect of an albedo variation is strongly influenced by the geographic location (latitude) and the inclination of the surface. In the present case study, the location-specific ΔRFf value was estimated to be 1.57 kg/m² of CO₂ equivalent in Bologna and 1.60 kg/m² of CO₂ equivalent in Perugia. Consequently, the CO₂ equivalent impact of a surface albedo modification can be estimated by multiplying this RF coefficient by the change in albedo and/or the variation in the surface area involved (in m²). Accordingly, the albedo of tree canopies, measured via drone or obtained from the literature, was compared with that of asphalt (the Asphalt scenario) and that of grass (the Lawn scenario). Increases in albedo contribute positively to the RF balance only when green infrastructure enhances surface reflectance. Conversely, decreases in albedo determine an increase in atmospheric temperature, which can be expressed by the amount of CO_2eq emissions necessary to determine the same atmospheric temperature increase.

3. Results

The results of the park surface assessment carried out through the Greenpix software are presented in

Table 2. Study parks surface distribution.

City	Park Name	Total Park Area (m2)	Streets and Paths (m2)	Buildings (m2)	Tree Canopy (m2)	Lawn (m2)
PG	Chico Mendez Park	160,235.88	38,359.61	162.14	39,398.12	82,316.01
			(24%)	(0%)	(25%)	(51%)
PG	Verbanella Park	30,868.47	3276.18	465.16	24,534.95	2592.19
			(11%)	(2%)	(79%)	(8%)
BO	Villa Ghigi Park	25,474.33	1694.31	934.77	15,596.02	7249.22
			(7%)	(4%)	(61%)	(28%)
BO	11 September Park	15,965.77	4561.64	1555.41	7044.42	2804.29
			(29%)	(10%)	(44%)	(18%)

. The table highlights the different surface distributions within the parks of the features described in the Materials and Methods section: tree canopy, lawn, paved streets and paths, and buildings.

As shown by the Greenpix results, in the same seasonal period, the lawn surface (m²) of the studied parks was constituted by dry grass (yellow hues) at 51% in Chico Mendez Park, 52% in Pescaia park, 85% in 11 September 2001 Park, and only 6% in Villa Ghigi Park.

All the image outputs from the Greenpix assessment are reported in the Supplementary Material (Figures S1–S14).

The following results (

Table 3. CO₂ compensated by albedo change in Chico Mendez Park, Perugia.

Tree Species	Number of Trees	Crown Area (m2)	Compensated CO2eq (Ton) Trees vs. Asphalt Scenario	Compensated CO2eq (Ton) Trees vs. Lawn Scenario
Acer campestre	72	2153.17	0.33	−0.04
Acer opalus	2	22.78	0.00	0.00
Acer platanoides	4	80.90	0.01	0.00
Acer pseudoplatanus	7	208.13	0.04	0.00
Acer rubrum	1	9.62	0.00	0.00
Aesculus hippocastanum	189	4919.51	0.55	−0.28
Aesculus × carnea	10	444.93	0.05	−0.03
Carpinus betulus	16	506.58	0.06	−0.02
Cupressus sempervirens	170	538.19	0.04	−0.05
Malus communis	1	7.07	0.00	0.00
Malus domestica	1	12.57	0.00	0.00
Olea europea	54	616.14	0.08	−0.02
Pinus halepensis	5	392.70	0.02	−0.05
Pinus pinea	2	226.19	0.01	−0.03
Populus alba	19	1545.66	0.48	0.22
Populus nigra	124	1573.35	0.49	0.23
Populus nigra ‘Italica’	18	72.26	0.02	0.01
Prunus amygdalus	7	119.77	0.06	0.04
Prunus persica	6	18.85	0.01	0.00
Quercus crenata	6	91.50	0.03	0.02
Quercus ilex	141	9855.87	0.63	−1.03
Quercus petraea	3	356.57	0.13	0.07
Quercus pubescens	18	995.88	0.37	0.20
Quercus robur	6	204.99	0.08	0.04
Robinia pseudoacacia	39	691.54	0.01	−0.11
Tilia platyphyllos	5	249.76	0.03	−0.01
Ulmus carpinifolia	14	455.53	0.12	0.04
Ulmus laevis	15	743.77	0.20	0.07
Ulmus pumila	3	368.35	0.10	0.04
Total	958	27,482.13	3.95	−0.69

,

Table 4. CO₂ compensated by albedo change in Pescaia Park, Perugia.

Tree Species	Number of Trees	Crown Area (m2)	Compensated CO2eq (Ton) Trees vs. Asphalt Scenario	Compensated CO2eq (Ton) Trees vs. Lawn Scenario
Acer opalus	5	237.19	0.04	0.00
Acer pseudoplatanus	1	19.63	0.00	0.00
Acer saccharinum	5	106.81	0.02	0.00
Acer saccharum	5	296.10	0.05	0.00
Aesculus hippocastanum	6	230.91	0.03	−0.01
Cercis siliquastrum	41	1425.50	1.12	0.87
Cupressus sempervirens	42	296.88	0.02	−0.03
Cupressus sempervirens pyramidalis	21	103.08	0.01	−0.01
Fraxinus ornus	9	450.82	0.05	−0.03
Fraxinus sp.	3	76.18	0.01	0.00
Olea europaea	66	2568.25	0.33	−0.11
Pinus pinaster	24	4265.50	0.41	−0.32
Pinus pinea	38	4228.58	0.20	−0.52
Quercus ilex	47	2849.42	0.18	−0.31
Robinia pseudoacacia	6	117.81	0.00	−0.02
Ulmus carpinifolia	42	1935.61	0.51	0.18
Ulmus pumila	2	66.76	0.02	0.01
Total	363	19,275.05	3.01	−0.30

,

Table 5. CO₂ compensated by albedo change in 11 September 2001 Park, Bologna.

Tree Species	Number of Trees	Crown Area (m2)	Compensated CO2eq (Ton) Trees vs. Asphalt Scenario	Compensated CO2eq (Ton) Trees vs. Lawn Scenario
Acer campestre	3	75.40	0.01	−0.01
Acer negundo	4	453.96	0.08	−0.02
Acer platanoides	2	166.50	0.03	−0.01
Aesculus hippocastanum	4	201.06	0.02	−0.02
Albizia julibrissin	1	50.27	0.04	0.03
Celtis australis	17	1624.20	0.18	−0.18
Cercis siliquastrum	3	37.70	0.03	0.02
Fraxinus angustifolia	1	12.57	0.00	0.00
Fraxinus excelsior	4	87.96	0.01	−0.01
Fraxinus intermedia	3	37.70	0.00	0.00
Fraxinus ornus	1	12.57	0.00	0.00
Liquidambar styraciflua	2	25.13	0.01	0.00
Malus sylvestris	3	37.70	0.00	−0.01
Pinus sylvestris	3	195.56	0.02	−0.03
Platanus acerifolia	7	559.20	0.06	−0.06
Populus nigra	5	457.10	0.14	0.04
Populus nigra ‘Italica’	1	7.07	0.00	0.00
Pyrus calleryana	1	12.57	0.01	0.00
Quercus ilex	1	12.57	0.00	0.00
Quercus robur	2	226.98	0.08	0.03
Robinia pseudoacacia	2	183.00	0.00	−0.04
Tilia cordata	3	113.10	0.02	−0.01
Tilia intermedia	4	245.83	0.03	−0.02
Tilia platyphyllos	2	25.13	0.00	0.00
Ulmus carpinifolia	2	190.07	0.05	0.01
Total	81	5050.90	0.82	−0.28

and

Table 6. CO₂ compensated by albedo change in Villa Ghigi Park, Bologna.

Tree Species	Number of Trees	Crown Area (m2)	Compensated CO2eq (Ton) Trees vs. Asphalt Scenario	Compensated CO2eq (Ton) Trees vs. Lawn Scenario
Acer campestre	7	227.77	0.03	0.01
Acer opalus	15	714.71	0.12	0.05
Acer sp.	1	78.54	0.01	0.01
Carpinus betulus	2	355.00	0.04	0.01
Celtis australis	21	2203.04	0.24	0.03
Cercis siliquastrum	1	19.63	0.02	0.01
Cupressus sempervirens	1	50.27	0.00	0.00
Diospyros kaki	20	565.49	0.10	0.05
Fagus sylvatica	2	57.33	0.01	0.00
Fraxinus angustifolia	1	153.94	0.02	0.00
Picea abies	1	50.27	0.00	0.00
Pinus sylvestris	1	38.48	0.00	0.00
Pyrus communis	1	113.10	0.05	0.04
Pyrus pyraster	3	139.80	0.06	0.05
Quercus pubescens	13	2787.38	1.01	0.74
Quercus robur	2	267.04	0.10	0.07
Quercus spp.	1	78.54	0.03	0.02
Tilia cordata	13	1361.88	0.18	0.05
Tilia platyphyllos	5	547.42	0.07	0.02
Ulmus carpinifolia	1	28.27	0.01	0.00
Ulmus laevis	1	28.27	0.01	0.00
Ulmus minor	1	78.54	0.02	0.01
Total	114	9944.71	2.14	1.19

) show the CO_2eq compensated by the albedo change as a consequence of the presence of tree canopies with respect to two different surfaces: (i) asphalt (Asphalt scenario) and (ii) lawn (Lawn scenario), characterized by different values of albedo.

Table 3 shows the results regarding the amounts of CO₂ compensated by the albedo change in Chico Mendez Park, comparing the albedo of the asphalt to the tree canopy albedo (Asphalt scenario) and comparing that of the tree canopy albedo to the grass albedo (Lawn scenario). The overall value of the Asphalt scenario, considering all trees (958 individuals), was 3.95 CO_2eq (ton). A particularly high contribution was registered due to the presence of two wood species (Quercus ilex, Aesculus hippocastanum). On the other hand, the large presence of Cupressus sempervirens trees (170 individuals) contributed to reaching only 0.04 CO_2eq. (ton). The overall value of the Lawn scenario was −0.69 CO_2eq. (ton) highlighting a small cooling contribution due to the presence of grass albedo in the park compared to the tree crown effect. Consequently, in this park, the presence of grass determined an increase in the albedo coefficient, leading to a favorable environmental condition.

Table 4 presents the conditions in Pescaia Park, where the highest CO_2eq. compensation was due to the Cercis siliquastrum species (1.12 CO_2eq. ton). The overall value of the Asphalt scenario considering all trees (363 individuals) was about 3 CO_2eq (ton). In this park, the Lawn scenario evidenced an increase in the cooling effect related to the presence of grass −0.3 CO_2eq. (ton).

As shown in Table 5, in 11 September 2001 Park, the lowest CO_2eq compensated value was recorded when comparing the albedo of asphalt to the tree canopy albedo (0.82 CO_2eq. ton), while the “Lawn scenario” value was in line with the previous parks considered, with a similar increase in the cooling effect due to the presence of grass (−0.28 CO_2eq. ton).

Finally, as shown in Table 6, the Villa Ghigi Park showed an interesting phenomenon concerning the comparison between the tree canopy albedo and the grass albedo. The CO_2eq compensated by the tree crown was positive, with a cooling effect of wood species, while the presence of grass evidenced a positive value (1.19 CO_2eq. ton) derived by the presence of herbaceous species with the lowest albedo coefficient (with darker colors) in comparison to the tree species crown.

4. Discussion

Surface albedo plays a pivotal role in modulating the Earth’s radiative energy balance [54] and, consequently, in influencing the trajectory of global climate change [4,31]. The negative feedback loop associated with decreased surface reflectivity as a consequence of urban expansion [55] determines the reduction in net radiative forcing, as is well established in the literature [12,56]. Changes in the albedo of surfaces have been extensively analyzed across a wide range of natural biomes [45,57,58]. In particular, numerous studies have examined the albedo dynamics of forested ecosystems [29,59,60], including mixed deciduous and coniferous stands [30], where the cumulative effect of multiple species with varying leaf optical properties is accounted for.

Within the urban context, albedo has been taken into consideration as part of engineered strategies to mitigate the UHI effect [14,21,23,24]. This parameter has predominantly been explored in relation to the deployment of highly reflective materials, such as cool roofs and pavements [61]. The contribution of urban vegetation, and particularly arboreal components, to surface albedo has received relatively limited attention, despite trees being widely acknowledged as central to urban climate resilience [62]. While a growing body of the literature has assessed the thermal performance of different vegetation types relative to reflective surfaces [1,20], limitations persist regarding the creation of a consistent comparative framework. The difficulty is mainly due to the complexity of quantifying the multifaceted ecosystem services delivered by urban trees [29,55,63]. This study aimed to address this gap by estimating the CO₂-equivalent (CO_2eq) compensation due to albedo variation in four urban park settings, comparing vegetated scenarios with alternative surface uses, including asphalt and grass [64,65]. Negative values are associated with lower reflectivity of the tree canopies compared to lawns. The latter are characterized by lighter shades of green, which may also reach significantly higher albedo values during dry seasons, exhibiting substantially higher reflectivity than most tree canopies as a consequence of seasonal yellowing. The conversion of grassy areas into tree-covered zones may thus lead to a reduction in albedo, contributing to local surface warming, similarly to what is reported in recent studies [66].

A particularly counterintuitive outcome emerges when modelling hypothetical scenarios in which vegetated park areas are entirely replaced with reflective artificial materials. Under such circumstances, the theoretical albedo-induced CO_2eq compensation would increase substantially, suggesting that, in terms of radiative forcing, highly reflective surfaces could outperform vegetative cover. However, such scenarios neglect critical ecosystem functions and fail to account for the broader suite of climate-regulating mechanisms associated with urban vegetation.

While a number of studies have explored the thermal implications of tree species selection versus reflective surfaces, a direct comparison is inherently complex. Trees offer multifaceted ecosystem services, such as shading, evapotranspiration, carbon sequestration, and biodiversity support, which cannot be directly replicated by reflective materials. Trees, unlike inert reflective materials, provide direct microclimatic cooling via evapotranspiration [33,49], sequester atmospheric carbon, and offer shading benefits that contribute significantly to human thermal comfort and the overall habitability of urban environments [13,41,67,68,69]. When the albedo-derived CO_2eq compensation from tree cover is evaluated alongside carbon storage and ETP cooling, the combined climate mitigation potential grows dramatically.

An increase of up to 50 times has been calculated based on the CO₂ compensation results from Fornaciari et al. (2024) for the same urban parks. For example, the tree spe-cies from Chico Mendez Park were globally able to stock 431.22 tonnes of CO₂ and compensate for 403.39 tonnes of CO₂ equivalents through their evapotranspiration cooling effect [33]. Similarly, Pescaia Park trees in Perugia were responsible for the storage of 366.35 tonnes of CO₂ and the compensation of 233.40 tonnes of CO₂ equivalent [33]. The 11 September Park and Villa Ghigi Park in Bologna allowed for compensation of 104.80 and 206.28 tonnes of CO₂ by stocking and 74.37 and 151.28 tonnes of CO₂ by evapotranspiration, respectively [33]. This highlights the necessity of adopting a multi-criteria assessment framework when evaluating urban greening strategies. In addition to the impact of trees on albedo compared to the surface, their contributions through CO₂ capture and canopy evapotranspiration ecosystem services must also be considered. A view limited to the first factor may lead to privileging inert solutions that are energetically efficient from the point of view of albedo but ecologically and socially suboptimal. An additional implication of this study pertains to species selection in urban forestry. Leaf spectral characteristics, particularly those associated with high visible and near-infrared reflectance, may confer a measurable advantage in optimizing albedo performance without compromising vegetative cooling and sequestration functions. Nevertheless, species-specific albedo values remain under-characterized, especially across phenological stages and varying canopy structures. Further empirical research is warranted to construct a robust, seasonally resolved database of urban tree albedo. Such a tool would support evidence-based planning and enhance the integration of biophysical performance into species selection protocols, informing adaptive land-use strategies under accelerating urbanization and climate stress [70].

5. Conclusions

-

Albedo Contribution of Urban Parks:

The study evaluated how variations in surface albedo contribute to CO₂-equivalent (CO₂eq) compensation across four urban parks.

Surface cover scenarios were compared, highlighting differences between high-albedo surfaces (e.g., lawns) and low-albedo ones (e.g., tree canopies).

-

Limited Albedo-Based Mitigation from Trees:

While trees offer multiple ecosystem services, their standalone contribution to climate mitigation via increased surface albedo is quantitatively limited.

In some cases, increasing tree cover at the expense of highly reflective vegetation (e.g., lawns) may lead to a net negative radiative effect.

-

Importance of Synergistic Ecosystem Services:

Despite their limited albedo impact, trees remain essential for climate-resilient urban design due to their synergistic ecosystem services.

These include evapotranspiration, carbon sequestration, urban cooling, air quality improvement, and biodiversity support.

-

Call for Integrative Assessment Approaches:

Future methodologies should integrate radiative, ecological, and social dimensions to more accurately evaluate the role of urban vegetation.

A multi-criteria approach is necessary to capture the full climate regulating potential of trees.

-

Species Selection and Albedo Optimization:

Selecting tree species with lighter foliage may help improve surface albedo, partially mitigating radiative drawbacks.

However, detailed knowledge of species-specific albedo characteristics especially across seasons is currently lacking.

-

Need for Further Research and Data:

More context-specific and phenology-aware albedo datasets are needed for urban vegetation.

Such research would enhance the capacity of planners and policymakers to optimize urban greening not only for biodiversity and thermal comfort but also for its contribution to long-term climate goals.