- Email: [email protected]
Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies
G Model
ARTICLE IN PRESS
SCS-535; No. of Pages 8
Sustainable Cities and Society xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Sustainable Cities and Society journal homepage: www.elsevier.com/locate/scs
Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies Stefania Bonafoni, Giorgio Baldinelli ∗ , Paolo Verducci Department of Engineering, University of Perugia, via G. Duranti 93, 06125 Perugia, Italy
a r t i c l e
i n f o
Article history: Received 23 September 2016 Received in revised form 7 November 2016 Accepted 8 November 2016 Available online xxx Keywords: Albedo Cool roofs Surface urban heat island Satellite observations
a b s t r a c t In recent years the use of satellite remote sensing techniques has proven to be a useful tool for monitoring urban surface parameters: data provided on the reflective and thermal state of the urban texture, both at local and global scale, give fundamental information on the surface urban heat island (SUHI) control of the urban planning. In this work, the retrieval of the urban albedo and land surface temperature (LST) from Landsat 7 satellite data is performed over a selected area of a town in Central Italy (Terni), exhibiting a significant urban change during the last 10 years. Comparing two satellite images on 2005 and on 2015, the spatial pattern of albedo and LST shows an average albedo decrease of 0.03 during this period and a daytime SUHI increase of 2.3 ◦ C. As highlighted by a focused local scale analysis, built-up area modifications moved towards both a reduction and an increase of the surface albedo, comparing the previous situation of the area and the reflective properties of materials chosen for the new settlements or refurbishments. The proposed analysis with remote sensing data may be considered an effective indicator able to point out if urban changes like interventions and new constructions move towards an urban sustainable development in terms of SUHI mitigation. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Sustainability is one of those fundamental issues whereby modernity is challenged and affects different aspects: land consumption and environment (overloading), energy (gradually depleting), culture (on mass lines), landscape (scarred and overcrowded), urban settings and infrastructures (congested), local resources of a cognitive, aesthetic and motivational nature (devalued). Cities constitute the main driving force of economic development, but they also represent a problem from an environmental point of view: in fact, they are responsible for 60–80% of global energy consumption and around the same share of the global carbon emission (http://ec.europa.eu/clima/policies/international/ paris protocol/cities/index en.htmhttp://ec.europa.eu/clima/ policies/international/paris protocol/cities/index en.htm). Studying new strategies for reducing the environmental impact represents an ethical commitment well before a scientific one. The
∗ Corresponding author. E-mail address: [email protected] (G. Baldinelli).
new paradigm “smart city” refers to an urban model sustainable, intelligent, competitive, inclusive, creative, hyper-connected, technological, efficient, e-governed, and open. In this paradigm shift, moving from the city to the smart city, the purpose is to favour the re-use of the existing avoiding the consumption of land, protecting and enhancing the urban green, promoting energy efficiency and reducing polluting emissions: in short, to improve the life of people who live in cities (Verducci & Desideri, 2012). In the urban sustainable development issue, the mitigation of the urban heat island (UHI) is a key point. UHI is a phenomenon caused by the increase of urbanization process, together with an increase of air pollution and anthropogenic heat sources. The city growth has changed the nature of surfaces reducing the presence of vegetation, with building structures and materials trapping solar radiation during the day, determining a significant temperature differences between urban and rural areas (Anniballe, Bonafoni, & Pichierri, 2014; Oke, 1982; Rizwan, Dennis, & Chunho, 2008; Stathopoulou et al., 2009). The UHI effect exacerbates during the summer, increasing energy consumption and producing dangerous effects for human health.
http://dx.doi.org/10.1016/j.scs.2016.11.005 2210-6707/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
2
Currently, this phenomenon affects not only large metropolitan areas, but also smaller cities. There are several strategies to mitigate UHI phenomenon, based on wisely designing the urban residential environment to obtain significant long-term energy savings and health benefits (Gaitani et al., 2014; Mackey, Lee, & Smith, 2012; Smith & Levermore, 2008; Synnefa, Santamouris, & Livada, 2006; Takebayashi & Moriyama, 2007). Different urban cooling strategies have been proposed and developed and, among the most effective ones, the increasing of urban surface reflectivity (Akbari, Damon Matthews, & Seto, 2012; Bretz, Akbari, & Rosenfeld, 1998; Dimoudi et al., 2014; Suehrcke, Peterson, & Selby, 2008; Taha, Akbari, Rosenfeld, & Huang, 1988; Wang & Akbari, 2016) and the intensification of urban vegetation (green roofs, street trees, and green spaces) (Dimoudi et al., 2014; Wang & Akbari, 2016;; Givoni, 1991) play a fundamental role. Generally, a built-up area exhibits a variable thermal pattern with high and low Land Surface Temperature (LST) zones, corresponding to low and high reflectivity impervious surfaces, respectively. A high reflective surface is typically light in colour and absorbs less solar radiation than a conventional dark-coloured one: less absorbed radiation brings to a lower LST, directly reducing building heat gain and air-conditioning demand. The albedo is a parameter that quantitatively describes the reflective behaviour of a surface: it represents the total hemispherical reflection of a surface integrated over the solar spectrum. It can be assumed that the average albedo of existing roofs does not exceed 0.30 but it can be increased to about 0.55 ÷ 0.60 with appropriate refurbishing and interventions (Akbari, Menon, & Rosenfeld, 2009). In particular, it is possible to apply coverings such as cool roof paintings to increase the albedo of residential and industrial building roofs and consequently mitigate the UHI. Remote sensing techniques and data processing methodologies allow retrieving surface and air parameters, with the aim of monitoring Earth surface both at local and global spatial scale, and during different temporal intervals (Kikon, Singh, Singh, & Vyas, 2016; Raghavan, Mandla, & Franco, 2015). In this work, the retrieval from Landsat 7 satellite data of the urban albedo and LST is carried out over a small area of a town in Central Italy (Terni), characterised by a clear urban change during the last 10 years. An analysis of the albedo variations and its impact on LST using satellite images on 29th of July 2005 and on 25th of July 2015 is provided, both at local and global scale. Spaceborne remote sensing techniques and methods allow monitoring the spatial pattern of surface reflective and thermal parameters and their temporal evolution that can be related to the urban texture change. Therefore, this analysis may constitute an effective indicator pointing out if urban changes like refurbishments, interventions and new constructions move towards an urban sustainable development.
2. Study area The study area is the Maratta zone of the city of Terni, Central Italy (Fig. 1), covering a rectangular surface of 3.0 km x 1.5 km. The UTM WGS84 coordinates (33T zone) of the selected area (Fig. 1 b) and c)) are: longitude range [m E]: 302300–305300 and latitude range [m N]: 4714350–4715850. The role of Terni, in the landscape of average Italian cities, is connected to its industrial history that saw, in the second half of the 19th century, a transformation of an average size historic town centre, still structured along the lines of the urban layout of the ancient Roman town, into the dynamic city of the 1930s, at the top of the national table in the manufacturing of iron and steel and production of electricity. The area of Maratta, the subject of study and analysis, developed as a settlement, is characterised by the presence of two major subsystems: the first one consists of an urban
layout made up of large industrial structures (a warehouse with open area) placed among small local handicraft activities and parts of a residential fabric, and small service areas. The second one is represented by its particular environmental situation, rich in stretches of water and areas of agricultural vegetation. These two factors should tie in more closely, especially during urban changes like refurbishments, interventions and new constructions. In this case, an accurate analysis, both at local and global scale, on reflective behaviour of building coverings (or, more generally, of impervious surfaces), and on its impact on LST changes was carried out on July 2005 and ten years later, using satellite images. The assessment of urban changes and their effects over a 10 years span is the first step to set up an urban sustainable design strategy in terms of SUHI mitigation for the future.
3. Data description and processing 3.1. Satellite data: Landsat 7 ETM+ The Landsat 7 satellite, operational since 1999, carries the Enhanced Thematic Mapper Plus (ETM + ) sensor. ETM+ is composed by six reflective bands in the visible (VIS), near infrared (NIR) and short-wavelength infrared (SWIR), a band in the thermal infrared (TIR) region, and includes a panchromatic band (Table 1). It has a spatial resolution of 30 m for the six reflective bands, 60 m for the thermal band, and 15 m for the panchromatic one. TIR band data are also delivered at 30 m, after resampling with a cubic convolution, by the USGS (http://earthexplorer.usgs.gov). In this work, the Landsat 7 passages over Terni reported in Table 2 were considered. The two images, downloaded from USGS at 30 m resolution, were processed according to the calibration technique proposed by (Chander, Markham, & Helder, 2009), in order to convert digital number values to at-sensor spectral radiance values. The computation of the surface reflectance of the ETM+ reflective bands (Chander et al., 2009; Chavez, 1996) was carried out applying the atmospheric correction described in (Chavez, 1996), that is characterised by the advantage of requiring only data from the same satellite scenes, without needing contemporary in-situ measurements.
Table 1 Landsat 7 ETM + bands. Band description (30 m native resolution unless otherwise denoted)
Spectral range (m)
B1–VIS (blue) B2 − VIS (green) B3 − VIS (red) B4 − NIR B5 − SWIR B6 − TIR (60 m) B7 − SWIR B8 − panchromatic (15 m)
0.45–0.51 0.52–0.60 0.63–0.69 0.77–0.90 1.55–1.75 10.31–12.36 2.06–2.35 0.52–0.90
Table 2 Data description of the two selected Landsat 7 scenes. Date
Scene Center Time (CEST: Central European Summer Time)
29 July 2005 11:42 CEST 25 July 2015 11:52 CEST
Landsat scene ID
Path Row
LE71910302005210EDC00 191 LE71910302015206NSG00 191
30 30
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model
ARTICLE IN PRESS
SCS-535; No. of Pages 8
S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
3
Fig. 1. (a) Terni location (city centre coordinate 42.56◦ N, 12.64◦ E) and Martta zone (yellow box, 3 × 1.5 km). (b)Regione Umbria (RU) image of Maratta zone on 2005 (Regione Umbria, 2005); red marks indicate measurement points described in sub- Section 3.4; yellow marks indicate changed surfaces analysed in Section 4. (c) Google Earth (GE) image of Maratta on 2015 (Google Earth, 2015); red marks indicate measusrement points described in sub-Section 3.4; yellow marks indicate changed surfaces analysed in Section 4.
3.2. Albedo retrieval from satellite data The albedo is the bihemispherical reflectance, i.e. the ratio of the radiant flux reflected from a unit surface area into the whole hemisphere to the incident radiant flux of hemispherical angular extent. The Landsat surface spectral reflectances correspond to the hemispherical (incoming radiation) − conical (reflected radiation) configuration (Schaepman-Strub, Schaepman, Painter, Dangel, & Martonchik, 2006). With the assumption of Lambertian surface, the Landsat spectral reflectances can be considered as narrowband albedos from which broadband albedo is estimated. The total broadband albedo for the study area is retrieved using the Liang (Liang, 2001) relation, as a function of the following coefficients for the Landsat ETM+ bands:
radiance. LST is retrieved from (2) by inversion of the Planck’s law (Jimenez-Munoz & Sobrino, 2003), once the surface emissivity ↓ ↑ is known. The atmospheric parameters , L and L were computed using a web-based tool (http://atmcorr.gsfc.nasa.gov) that takes the National Centers for Environmental Prediction (NCEP) atmospheric profiles as input to the MODTRAN radiative transfer code (Barsi, Barker, & Schott, 2003). Land surface emissivity was estimated by the normalized difference vegetation index (NDVI) threshold method (Bonafoni, Anniballe, Gioli, & Toscano, 2016; Sobrino, Jimenez-Munoz, & Paolini, 2004), with NDVI computed as: (B4-B3)/(B4 + B3)
(3)
3.4. In situ albedo measurements ␣ = 0.356 xB1 + 0.130 xB3 + 0.373 xB4 + 0.085 xB5 + 0.072 xB7–0.0018(1) From the previous equation, an albedo map is produced for each Landsat 7 scene. 3.3. LST retrieval from satellite data The LST from B6 band data was obtained by inverting the following radiative transfer equation:
↓
↑
Lsens, = ε B (Ts ) + (1 − ε ) L + L
(2)
where Lsens, is the at-sensor radiance at the top-of-atmosphere, εl is the surface emissivity, B (Ts ) is the Planck’s law where Ts is the ↓ LST, L is the downwelling atmospheric radiance, is the total ↑
atmospheric transmissivity and L is the upwelling atmospheric
An experimental campaign was carried out to assess the reliability of albedo estimation from satellite observations using Eq. (1). In situ measurements of albedo, red points of Fig. 1 b) and c), were realized by an albedometer Delta Ohm Pyra 05, made up of two Class 1 pyranometers (International Organization for Standardization, 1990) and operating within 0.3 m ÷ 3.0 m spectral range. The ground campaign was carried out on the 26th of July 2016, at around h.11:00 local time, in correspondence to a Landsat 7 passage above the zone. The results of the in situ measurements are reported in Table 3, where the albedo values retrieved from the Landsat 7 ETM+ observations are also indicated. The points chosen for the analysis felt within the zone under investigation, trying to measure low and high albedo areas with a
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model
ARTICLE IN PRESS
SCS-535; No. of Pages 8
S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
4
Table 3 Results of albedo ground measurements and comparison with Landsat ETM+ albedo retrieval. Point
Description
Measured albedo
Landsat albedo
Difference
Difference (%)
V1 V2
White surface parking lot Rural field
0.383 0.170
0.370 0.150
0.013 0.020
+3% +11%
Fig. 2. View of the two points (V1 and V2) chosen for the validation with ground albedo measurements.
It is evident how the ground measurements validated the albedo estimation from the satellite data as already highlighted in a previous work of the authors (Baldinelli et al., 2015), with a maximum deviation of 0.02 in terms of absolute values and 11% as a percentage between the two methodologies. The reasons for the light shift registered between the Landsat albedo and the ground measurements can be ascribed to the satellite footprint averaging inside a 30 m pixel, as well as in the intrinsic hypothesis of Lambertian surface necessary to compare the two methodologies, which is not perfectly respected for the zones investigated. 4. Results
Fig. 3. Picture of the ground albedo measurement on the point V2.
surface wider than the satellite pixel size for the reflective bands (30 × 30 m). Fig. 2 shows the two points analysed with the relative coordinates, while Fig. 3 reports a picture of the measurement setup on point V2.
4.1. Albedo and LST temporal variation: detailed analysis of significant sub-areas The main territorial changes of the area under investigation are often accompanied by surface albedo modifications. In Fig. 4, five points of interest are reported, as they appear in the visible images of years 2005 (Regione Umbria, 2005) and 2015 (Google
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
5
Fig. 4. Detail of visible image, albedo and lst variation in five points of interest between the years 2005 and 2015.
Earth, 2015), jointly with the albedo and LST variation between the same period. The albedo variation (␣) is intended as the difference between the values of 2005 and 2015; the LST variation (LST) is reported, on the contrary, between 2015 and 2005, since an albedo decrease corresponds generally to a LST enhancement. The pictures in cyan-red scale show clearly the direct connection between the surface albedo and the LST values of the buildings analysed. In P1, for instance, an albedo increase in 2015 of 0.08 linked to a surface modification produced a LST diminution of around 0.8 ◦ C, in P4, on the contrary, the roof albedo diminished strongly (0.13), causing a temperature enhancement of 7.7 ◦ C; finally, the new construction in P2, although producing an albedo enhancement (0.04),
an increase of LST (5.3 ◦ C) is registered. This circumstance could be easily explained if it is considered that the building footprint was previously covered by vegetation, the latter producing a cooling effect despite a low albedo value. Thus, it emerges that the fine scale modifications could move towards both a reduction and an enhancement of the surface albedo and LST, depending on the previous situation of the area and on the materials chosen for the new settlements. It is worthy of particular interest the photovoltaic roof, as many industrial constructions hosted this solution all over Europe: in point P3, the photovoltaic roof caused an albedo diminution of 0.08 and a consequent LST increase of 6.1 ◦ C. If from one side the roof
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8 6
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
Fig. 5. Distribution of albedo for the selected Terni area (Maratta Alta) for 2005 (red) and 2015 (blue). Main statistics are reported. Vegetation pixels are discarded. Fig. 6. Distribution of albedo difference ∝ (2015–2005) for the selected area.
became a renewable energy source, from the other side diminished its albedo, giving a negative contribution to the greenhouse effect reduction: a deeper analysis is needed to understand the relative weight of each component. In the work of Akbari (Akbari et al., 2009) it is stated that a 0.25 surface albedo diminishing of 1 square meter brings to an equivalent emission of 64 kg CO2 . The area analysed lowered its albedo of about 0.07 (the roof was white before the photovoltaic plant set up), therefore, an increase of around 18 kg CO2 is caused by the installation of the PV panels. The same square meter may host in Italy approximately 100 W of photovoltaic panels, which produce in the same site around 130 kWh/year of electric energy. Hypothesizing a life cycle of 20 years and considering that in Italy, the energetic mix brings to emit 0.350 kg/kWh of electric energy produced (ENEA, 2016), the panels avoid 910 kg CO2 (the CO2 produced in the construction and disposal phase is neglected). As a result, the CO2 emission avoided is orders of magnitude higher than its increase linked to the effect of the albedo reduction. 4.2. Albedo and LST temporal variation: full area analysis The albedo and LST variations, described in the previous section at local scale, are now analysed at a global scale. The average urban change for the whole Maratta area is examined considering all the albedo and LST pixels of the Landsat 7 images on 29th of July 29, 2005, and on 25th of July 2015. For this analysis, vegetation pixels were removed, since they are characterised by low albedo values with different impact with respect to the low albedo of impervious surfaces. In fact, despite the low reflective properties, the energy stored by plants and foliage is mainly used for their life processes and evapotranspiration, producing a cooling effect of the vegetation cover. The vegetation filtering was performed selecting pixels with NDVI < 0.35. Fig. 5 shows the distribution of albedo values for the satellite scene of 2005 (red) and 2015 (blue). Mean and standard deviation (std) are also reported. Fig. 6 reports the distribution of the albedo difference (2015–2005) for the selected area. It is evident the change of the albedo during 2015, with an average reduction of 0.03. This means that the fine scale modifications, showing both a reduction and a growth of the surface albedo with respect to the previous situation, move globally towards a worsening of the albedo pattern. Even though the albedo drop of built-up areas leads
Fig. 7. Distribution of SUHI for the selected Terni area (Maratta Alta) for 2005 (red) and 2015 (blue).Main statistics are reported. Vegetation pixels are discarded.
to an increase of surface and air temperature, albedo reduction due to photovoltaic panel presence has not a negative impact on the greenhouse effect, as previously pointed out. The LST global pattern is analysed through the surface urban heat island (SUHI) intensity, defined as the difference in LST between the urban pixels and the surrounding rural background. Fig. 7 shows the distribution of SUHI values for 2005 (red) and 2015 (blue) satellite scenes, reporting mean and std. Fig. 8 shows the distribution of the SUHI difference (2015–2005). An increase of the SUHI intensity of 2.3 ◦ C is revealed, highlighting global heating of the area due to the urban texture changes after 10 years. Fig. 9 investigates the correspondence between LST and albedo for the two satellite scenes. Since each image has about 1700 pixels, a scatterplot LST-albedo would produce a data-cloud of hard interpretation: to better analyze the relationship, an average study was carried out evaluating the mean LST at each fixed increment of
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
7
Table 4 Mean and std of albedo from 2005 to 2016. Landsat 7 passages at around 11:50 CEST. Clear sky summer image of 2008 is not available.
Fig. 8. Distribution of SUHI difference (2015–2005) for the selected area.
Fig. 9. LST vs. albedo (2015 in blue, 2005 in red) scatterplot for the selected area and linear fitting.
the albedo. In particular, the incremental step is 0.004 in order to ensure about 100 evaluation points in the albedo range. The average scatterplot and the correspondent linear fitting reveal the expected “inverse” linear relation between LST and albedo. This relation changes for the two different Landsat 7 images, confirming the variability of these parameters in relation to the different urban texture and construction materials. Finally, the mean and std albedo for July–August Landsat passages from 2005 to 2016 are reported in Table 4. This temporal sequence of albedo average for the selected zone points out that the albedo reduction started from 2014, after the new urbanistic intervention. 5. Discussion The study showed that the retrieval from satellite data of the urban albedo and LST, or equivalently SUHI, provides information on their spatial pattern and on their temporal evolution to be linked to the urban texture change, and so to the urban sustainable development.
Landsat Image
Mean albedo
Std albedo
29 July 2005 1 August 2006 19 July 2007 16 July 2009 11 July 2010 23 August 2011 16 July 2012 3 July 2013 7 August 2014 25 July 2015 27 July 2016
0.26 0.27 0.26 0.26 0.26 0.26 0.27 0.26 0.22 0.23 0.22
0.042 0.042 0.039 0.039 0.043 0.038 0.047 0.038 0.034 0.030 0.035
Within an urban changing and development, if the albedo is not analysed with care, the risk of a SUHI increase arises. The refurbishing of existing construction should consider, among the other issues, the enhancement of the surface albedo, a low-cost intervention with positive outcomes on the greenhouse effect. When the city planning foresees the replacement of rural areas with anthropic installations (roofs, roads, pavements, squares, etc . . .), the negative environmental impact is unquestionable, firstly for the loss of the effect of CO2 capture connected to the photosynthesis. Furthermore, the foliage produce on the other hand the strong cooling effect of leaves, thanks to the evapotranspiration. Thus, a particular attention to the new impervious surfaces albedo is needed, with the aim of limiting their impact on the SUHI raise and on the greenhouse effect worsening. In addition to the results discussed in the previous section, an interesting question regards the relation between SUHI and heat island of the air, i.e how a measured LST increase can be translated into an air temperature increase. As Voogt and Oke (Voogt & Oke, 2003) summarized, the relation between surface and air temperature is empirical and no simple general relation exists, even if their correlation improved at night when microscale advection is reduced. Advection, i.e. the horizontal transport of heat by wind, depends on the geometry of the urban surface, surface moisture and roughness, thermal admittance and mean wind velocity. Therefore, air and LST relation can be solved if a detailed knowledge of the surface micrometeorology is available. Overall, the results of the Maratta area pointed out an average daytime SUHI increase of 2.3 ◦ C during the last 10 years, from which a lower increase of the daytime air temperature is expected (Stathopoulou et al., 2009). During night time, SUHI and heat island of the air should be similar (Stathopoulou et al., 2009). 6. Conclusions Satellite remote sensing techniques and methods proved their effectiveness in the spatial pattern analysis of albedo and LST: their temporal evolution allows to understand if the urban texture change moves towards a sustainable development in terms of SUHI reduction. The validation of spaceborne observations with ground measurements demonstrated that albedo retrieval from satellite data can be used with a satisfactory level of accuracy, with the enormous advantage of providing a wide spatial coverage, making possible to monitor the albedo and LST changes with time. The study pointed out that if the albedo is not carefully analysed within urban surface changes, the risk of a SUHI increase arises; furthermore, the technical and economical effort to make impervious surfaces highly reflective results almost negligible, especially within a new construction ore refurbishment framework.
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
8
Although albedo decrease of impervious surfaces leads to a LST increase, albedo reduction of photovoltaic panels installed on roofs has not a long-term negative impact on the greenhouse effect due to the CO2 emission reduction. Overall, the assessment of urban change effects can suggest in future how to give a contribution to set up a sustainable strategy in designing smart cities: foreseeing both mitigating urban heat islands and evaluating the renewable energy efficiency, the use of remote sensors may constitute a resource. Anyway, designing smart cities with solutions and technologies that implement the issues above is only a first step: design smart cities means also favouring the use of the existing and avoiding the consumption of land, protecting and enhancing the urban green areas, by promoting energy efficiency and reducing polluting emissions. Currently, great efforts are being made to retrieve air pollution from satellite data, confirming the potentiality of this technology in support of strategies for urban sustainable development. Acknowldegments The work has been supported by “Fondo di Ricerca di Base di Ateneo”, funded by the Department of Engineering of University of Perugia-Italy. Authors wish to thank Eng. Roberta Anniballe for the data processing support and Eng. Andrea Presciutti for the ground measurements. References Akbari, H., Menon, S., & Rosenfeld, A. (2009). Global cooling: Increasing world-wide urban albedos to offset CO2 . Climatic Change, 94, 275–286. Akbari, H., Damon Matthews, H., & Seto, D. (2012). The long-term effect of increasing the albedo of urban areas. Environmental Research Letters, 7, 1–10. Anniballe, R., Bonafoni, S., & Pichierri, M. (2014). Spatial and temporal trends of the surface and air heat island over Milan using Modis data. Remote Sensing of Environment, 150, 163–171. Baldinelli, G., Bonafoni, S., Anniballe, R., Presciutti, A., Gioli, B., & Magliulo, V. (2015). Spaceborne detection of roof and impervious surface albedo: Potentialities and comparison with airborne thermography measurements. Solar Energy, 113, 281–294. Barsi, J. A., Barker, J. L., & Schott, J. R. (2003). An atmospheric correction parameter calculator for a single thermal band earth-sensing instrument. In Proceedings of IEEE international geoscience and remote sensing symposium 21–25 july (pp. 3014–3016). Bonafoni, S., Anniballe, R., Gioli, B., & Toscano, P. (2016). Downscaling landsat land surface temperature over the urban area of florence. European Journal of Remote Sensing, 49, 553–569. Bretz, S., Akbari, H., & Rosenfeld, A. (1998). Practical issues for using solar-reflective materials to mitigate urban heat islands. Atmospheric Environment, 32(1), 95–101. Chander, G., Markham, B. L., & Helder, D. L. (2009). Summary of current radiometric calibration coefficients for landsat MSS, TM, ETM+, and EO-1 ALI sensors. Remote Sensing of Environment, 113, 893–903. Chavez, P. S. (1996). Image-based atmospheric corrections-revisited and improved. Photogrammetric Engineering and Remote Sensing, 62(9), 1025–1036. Dimoudi, A., Zoras, S., Kantzioura, A., Stogiannou, X., Kosmopoulos, P., & Pallas, C. (2014). Use of cool materials and other bioclimatic interventions in outdoor places in order to mitigate the urban heat island in a medium size city in Greece. Sustainable Cities and Society, 13, 89–96. ENEA −Italian National Agency for New Technologies, Energy and Sustainable Economic Development, Analisi trimestrale del Sistema Energetico italiano −1◦ trimestre 2016. Ed. ENEA, Rome 2016 in Italian.
http://ec.europa.eu/clima/policies/international/paris protocol/cities/index en. htm (retrieved on 7th of September 2016). Gaitani, N., Santamouris, M., Cartalis, C., Pappas, I., Xyrafi, F., Mastrapostoli, E., et al. (2014). Microclimatic analysis as a prerequisite for sustainable urbanisation: Application for an urban regeneration project for a medium size city in the greater urban agglomeration of Athens, Greece. Sustainable Cities and Society, 13, 230–236. Givoni, B. (1991). Impact of planted areas on urban environmental quality: A review. Atmospheric Environment, 25, 289–299. Google Earth, 7.1.5.1557. Terni 303,530 m E and 4,715,285 m N, 7th of August 2015. International Organization for Standardization. (1990). Standard ISO 9060 Solar energy – Specification and classification of instruments for measuring hemispherical solar and direct solar radiation. Jimenez-Munoz, J. C., & Sobrino, J. A. (2003). A generalized singlechannel method for retrieving land surface temperature from remote sensing data. Journal of Geophysical Research, 108, 1–9. Kikon, N., Singh, P., Singh, S. K., & Vyas, A. (2016). Assessment of urban heat islands (UHI) of Noida City, India using multi-temporal satellite data. Sustainable Cities and Society, 22, 19–28. Liang, S. (2001). Narrowband to broadband conversions of land surface albedo − I algorithms. Remote Sensing of Environment, 76, 213–238. Mackey, C. W., Lee, X., & Smith, R. B. (2012). Remotely sensing the cooling effects of city scale efforts to reduce urban heat island. Building and Environment, 49, 348–358. Oke, T. R. (1982). The energetic basis of the urban heat island. Quarterly Journal of the Royal Meteorological Society, 108, 1–24. Raghavan, K., Mandla, V., & Franco, S. (2015). Influence of urban areas on environment: Special reference to building materials and temperature anomalies using geospatial technology. Sustainable Cities and Society, 19, 349–358. Regione Umbria–Giunta Regionale – SIAT Sistema Informativo Regionale Ambientale e Territoriale 2005. Rizwan, A. M., Dennis, L. Y. C., & Chunho, L. I. U. (2008). A review on the generation, determination and mitigation of Urban Heat Island. Journal of Environmental Science, 20, 120–128. Schaepman-Strub, G., Schaepman, M. E., Painter, T. H., Dangel, S., & Martonchik, J. V. (2006). Reflectance quantities in optical remote sensing-definitions and case studies. Remote Sensing of Environment, 103, 27–42. Smith, C., & Levermore, G. (2008). Designing urban spaces and buildings to improve sustainability and quality of life in a warmer world. Energy Policy, 36, 4558–4562. Sobrino, J. A., Jimenez-Munoz, J. C., & Paolini, L. (2004). Land surface temperature retrieval from LANDSAT TM 5. Remote Sensing of Environment, 90, 434–440. Stathopoulou, M., Synnefa, A., Caralis, C., Santamouris, M., Karless, T., & Akbari, H. (2009). A surface heat island study of Athens using high-resolution satellite imagery and measurements of the optical and thermal properties of commonly used building and paving materials. International Journal of Sustainable Energy, 28(1), 59–76. Suehrcke, H., Peterson, E. L., & Selby, N. (2008). Effect of roof solar reflectance on the building heat gain in a hot climate. Energy and Buildings, 40, 2224–2235. Synnefa, A., Santamouris, M., & Livada, I. (2006). A study of the thermal performance of reflective coatings for the urban environment. Solar Energy, 80, 968–981. Taha, H., Akbari, H., Rosenfeld, A., & Huang, J. (1988). Residential cooling loads and the urban heat island: The effects of albedo. Building and Environment, 23, 271–283. Takebayashi, H., & Moriyama, M. (2007). Surface heat budget on green roof and high reflection roof for mitigation of urban heat Island. Building and Environment, 42, 2971–2979. Verducci, P., & Desideri, U. (2012). Architecture, urban design, energetic systems. Design of eco-sustainable industrial parks with very low environmental impact, reduced use of water and energy resources and optimized waste management. Perugia: Ed. Morlacchi Editore. Voogt, J. A., & Oke, T. R. (2003). Thermal remote sensing of urban climates. Remote Sensing of Environment, 86, 370–384. Wang, Y., & Akbari, H. (2016). Analysis of urban heat island phenomenon and mitigation solutions evaluation for Montreal. Sustainable Cities and Society, 26, 438–446.
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
ARTICLE IN PRESS
SCS-535; No. of Pages 8
Sustainable Cities and Society xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Sustainable Cities and Society journal homepage: www.elsevier.com/locate/scs
Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies Stefania Bonafoni, Giorgio Baldinelli ∗ , Paolo Verducci Department of Engineering, University of Perugia, via G. Duranti 93, 06125 Perugia, Italy
a r t i c l e
i n f o
Article history: Received 23 September 2016 Received in revised form 7 November 2016 Accepted 8 November 2016 Available online xxx Keywords: Albedo Cool roofs Surface urban heat island Satellite observations
a b s t r a c t In recent years the use of satellite remote sensing techniques has proven to be a useful tool for monitoring urban surface parameters: data provided on the reflective and thermal state of the urban texture, both at local and global scale, give fundamental information on the surface urban heat island (SUHI) control of the urban planning. In this work, the retrieval of the urban albedo and land surface temperature (LST) from Landsat 7 satellite data is performed over a selected area of a town in Central Italy (Terni), exhibiting a significant urban change during the last 10 years. Comparing two satellite images on 2005 and on 2015, the spatial pattern of albedo and LST shows an average albedo decrease of 0.03 during this period and a daytime SUHI increase of 2.3 ◦ C. As highlighted by a focused local scale analysis, built-up area modifications moved towards both a reduction and an increase of the surface albedo, comparing the previous situation of the area and the reflective properties of materials chosen for the new settlements or refurbishments. The proposed analysis with remote sensing data may be considered an effective indicator able to point out if urban changes like interventions and new constructions move towards an urban sustainable development in terms of SUHI mitigation. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Sustainability is one of those fundamental issues whereby modernity is challenged and affects different aspects: land consumption and environment (overloading), energy (gradually depleting), culture (on mass lines), landscape (scarred and overcrowded), urban settings and infrastructures (congested), local resources of a cognitive, aesthetic and motivational nature (devalued). Cities constitute the main driving force of economic development, but they also represent a problem from an environmental point of view: in fact, they are responsible for 60–80% of global energy consumption and around the same share of the global carbon emission (http://ec.europa.eu/clima/policies/international/ paris protocol/cities/index en.htmhttp://ec.europa.eu/clima/ policies/international/paris protocol/cities/index en.htm). Studying new strategies for reducing the environmental impact represents an ethical commitment well before a scientific one. The
∗ Corresponding author. E-mail address: [email protected] (G. Baldinelli).
new paradigm “smart city” refers to an urban model sustainable, intelligent, competitive, inclusive, creative, hyper-connected, technological, efficient, e-governed, and open. In this paradigm shift, moving from the city to the smart city, the purpose is to favour the re-use of the existing avoiding the consumption of land, protecting and enhancing the urban green, promoting energy efficiency and reducing polluting emissions: in short, to improve the life of people who live in cities (Verducci & Desideri, 2012). In the urban sustainable development issue, the mitigation of the urban heat island (UHI) is a key point. UHI is a phenomenon caused by the increase of urbanization process, together with an increase of air pollution and anthropogenic heat sources. The city growth has changed the nature of surfaces reducing the presence of vegetation, with building structures and materials trapping solar radiation during the day, determining a significant temperature differences between urban and rural areas (Anniballe, Bonafoni, & Pichierri, 2014; Oke, 1982; Rizwan, Dennis, & Chunho, 2008; Stathopoulou et al., 2009). The UHI effect exacerbates during the summer, increasing energy consumption and producing dangerous effects for human health.
http://dx.doi.org/10.1016/j.scs.2016.11.005 2210-6707/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
2
Currently, this phenomenon affects not only large metropolitan areas, but also smaller cities. There are several strategies to mitigate UHI phenomenon, based on wisely designing the urban residential environment to obtain significant long-term energy savings and health benefits (Gaitani et al., 2014; Mackey, Lee, & Smith, 2012; Smith & Levermore, 2008; Synnefa, Santamouris, & Livada, 2006; Takebayashi & Moriyama, 2007). Different urban cooling strategies have been proposed and developed and, among the most effective ones, the increasing of urban surface reflectivity (Akbari, Damon Matthews, & Seto, 2012; Bretz, Akbari, & Rosenfeld, 1998; Dimoudi et al., 2014; Suehrcke, Peterson, & Selby, 2008; Taha, Akbari, Rosenfeld, & Huang, 1988; Wang & Akbari, 2016) and the intensification of urban vegetation (green roofs, street trees, and green spaces) (Dimoudi et al., 2014; Wang & Akbari, 2016;; Givoni, 1991) play a fundamental role. Generally, a built-up area exhibits a variable thermal pattern with high and low Land Surface Temperature (LST) zones, corresponding to low and high reflectivity impervious surfaces, respectively. A high reflective surface is typically light in colour and absorbs less solar radiation than a conventional dark-coloured one: less absorbed radiation brings to a lower LST, directly reducing building heat gain and air-conditioning demand. The albedo is a parameter that quantitatively describes the reflective behaviour of a surface: it represents the total hemispherical reflection of a surface integrated over the solar spectrum. It can be assumed that the average albedo of existing roofs does not exceed 0.30 but it can be increased to about 0.55 ÷ 0.60 with appropriate refurbishing and interventions (Akbari, Menon, & Rosenfeld, 2009). In particular, it is possible to apply coverings such as cool roof paintings to increase the albedo of residential and industrial building roofs and consequently mitigate the UHI. Remote sensing techniques and data processing methodologies allow retrieving surface and air parameters, with the aim of monitoring Earth surface both at local and global spatial scale, and during different temporal intervals (Kikon, Singh, Singh, & Vyas, 2016; Raghavan, Mandla, & Franco, 2015). In this work, the retrieval from Landsat 7 satellite data of the urban albedo and LST is carried out over a small area of a town in Central Italy (Terni), characterised by a clear urban change during the last 10 years. An analysis of the albedo variations and its impact on LST using satellite images on 29th of July 2005 and on 25th of July 2015 is provided, both at local and global scale. Spaceborne remote sensing techniques and methods allow monitoring the spatial pattern of surface reflective and thermal parameters and their temporal evolution that can be related to the urban texture change. Therefore, this analysis may constitute an effective indicator pointing out if urban changes like refurbishments, interventions and new constructions move towards an urban sustainable development.
2. Study area The study area is the Maratta zone of the city of Terni, Central Italy (Fig. 1), covering a rectangular surface of 3.0 km x 1.5 km. The UTM WGS84 coordinates (33T zone) of the selected area (Fig. 1 b) and c)) are: longitude range [m E]: 302300–305300 and latitude range [m N]: 4714350–4715850. The role of Terni, in the landscape of average Italian cities, is connected to its industrial history that saw, in the second half of the 19th century, a transformation of an average size historic town centre, still structured along the lines of the urban layout of the ancient Roman town, into the dynamic city of the 1930s, at the top of the national table in the manufacturing of iron and steel and production of electricity. The area of Maratta, the subject of study and analysis, developed as a settlement, is characterised by the presence of two major subsystems: the first one consists of an urban
layout made up of large industrial structures (a warehouse with open area) placed among small local handicraft activities and parts of a residential fabric, and small service areas. The second one is represented by its particular environmental situation, rich in stretches of water and areas of agricultural vegetation. These two factors should tie in more closely, especially during urban changes like refurbishments, interventions and new constructions. In this case, an accurate analysis, both at local and global scale, on reflective behaviour of building coverings (or, more generally, of impervious surfaces), and on its impact on LST changes was carried out on July 2005 and ten years later, using satellite images. The assessment of urban changes and their effects over a 10 years span is the first step to set up an urban sustainable design strategy in terms of SUHI mitigation for the future.
3. Data description and processing 3.1. Satellite data: Landsat 7 ETM+ The Landsat 7 satellite, operational since 1999, carries the Enhanced Thematic Mapper Plus (ETM + ) sensor. ETM+ is composed by six reflective bands in the visible (VIS), near infrared (NIR) and short-wavelength infrared (SWIR), a band in the thermal infrared (TIR) region, and includes a panchromatic band (Table 1). It has a spatial resolution of 30 m for the six reflective bands, 60 m for the thermal band, and 15 m for the panchromatic one. TIR band data are also delivered at 30 m, after resampling with a cubic convolution, by the USGS (http://earthexplorer.usgs.gov). In this work, the Landsat 7 passages over Terni reported in Table 2 were considered. The two images, downloaded from USGS at 30 m resolution, were processed according to the calibration technique proposed by (Chander, Markham, & Helder, 2009), in order to convert digital number values to at-sensor spectral radiance values. The computation of the surface reflectance of the ETM+ reflective bands (Chander et al., 2009; Chavez, 1996) was carried out applying the atmospheric correction described in (Chavez, 1996), that is characterised by the advantage of requiring only data from the same satellite scenes, without needing contemporary in-situ measurements.
Table 1 Landsat 7 ETM + bands. Band description (30 m native resolution unless otherwise denoted)
Spectral range (m)
B1–VIS (blue) B2 − VIS (green) B3 − VIS (red) B4 − NIR B5 − SWIR B6 − TIR (60 m) B7 − SWIR B8 − panchromatic (15 m)
0.45–0.51 0.52–0.60 0.63–0.69 0.77–0.90 1.55–1.75 10.31–12.36 2.06–2.35 0.52–0.90
Table 2 Data description of the two selected Landsat 7 scenes. Date
Scene Center Time (CEST: Central European Summer Time)
29 July 2005 11:42 CEST 25 July 2015 11:52 CEST
Landsat scene ID
Path Row
LE71910302005210EDC00 191 LE71910302015206NSG00 191
30 30
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model
ARTICLE IN PRESS
SCS-535; No. of Pages 8
S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
3
Fig. 1. (a) Terni location (city centre coordinate 42.56◦ N, 12.64◦ E) and Martta zone (yellow box, 3 × 1.5 km). (b)Regione Umbria (RU) image of Maratta zone on 2005 (Regione Umbria, 2005); red marks indicate measurement points described in sub- Section 3.4; yellow marks indicate changed surfaces analysed in Section 4. (c) Google Earth (GE) image of Maratta on 2015 (Google Earth, 2015); red marks indicate measusrement points described in sub-Section 3.4; yellow marks indicate changed surfaces analysed in Section 4.
3.2. Albedo retrieval from satellite data The albedo is the bihemispherical reflectance, i.e. the ratio of the radiant flux reflected from a unit surface area into the whole hemisphere to the incident radiant flux of hemispherical angular extent. The Landsat surface spectral reflectances correspond to the hemispherical (incoming radiation) − conical (reflected radiation) configuration (Schaepman-Strub, Schaepman, Painter, Dangel, & Martonchik, 2006). With the assumption of Lambertian surface, the Landsat spectral reflectances can be considered as narrowband albedos from which broadband albedo is estimated. The total broadband albedo for the study area is retrieved using the Liang (Liang, 2001) relation, as a function of the following coefficients for the Landsat ETM+ bands:
radiance. LST is retrieved from (2) by inversion of the Planck’s law (Jimenez-Munoz & Sobrino, 2003), once the surface emissivity ↓ ↑ is known. The atmospheric parameters , L and L were computed using a web-based tool (http://atmcorr.gsfc.nasa.gov) that takes the National Centers for Environmental Prediction (NCEP) atmospheric profiles as input to the MODTRAN radiative transfer code (Barsi, Barker, & Schott, 2003). Land surface emissivity was estimated by the normalized difference vegetation index (NDVI) threshold method (Bonafoni, Anniballe, Gioli, & Toscano, 2016; Sobrino, Jimenez-Munoz, & Paolini, 2004), with NDVI computed as: (B4-B3)/(B4 + B3)
(3)
3.4. In situ albedo measurements ␣ = 0.356 xB1 + 0.130 xB3 + 0.373 xB4 + 0.085 xB5 + 0.072 xB7–0.0018(1) From the previous equation, an albedo map is produced for each Landsat 7 scene. 3.3. LST retrieval from satellite data The LST from B6 band data was obtained by inverting the following radiative transfer equation:
↓
↑
Lsens, = ε B (Ts ) + (1 − ε ) L + L
(2)
where Lsens, is the at-sensor radiance at the top-of-atmosphere, εl is the surface emissivity, B (Ts ) is the Planck’s law where Ts is the ↓ LST, L is the downwelling atmospheric radiance, is the total ↑
atmospheric transmissivity and L is the upwelling atmospheric
An experimental campaign was carried out to assess the reliability of albedo estimation from satellite observations using Eq. (1). In situ measurements of albedo, red points of Fig. 1 b) and c), were realized by an albedometer Delta Ohm Pyra 05, made up of two Class 1 pyranometers (International Organization for Standardization, 1990) and operating within 0.3 m ÷ 3.0 m spectral range. The ground campaign was carried out on the 26th of July 2016, at around h.11:00 local time, in correspondence to a Landsat 7 passage above the zone. The results of the in situ measurements are reported in Table 3, where the albedo values retrieved from the Landsat 7 ETM+ observations are also indicated. The points chosen for the analysis felt within the zone under investigation, trying to measure low and high albedo areas with a
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model
ARTICLE IN PRESS
SCS-535; No. of Pages 8
S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
4
Table 3 Results of albedo ground measurements and comparison with Landsat ETM+ albedo retrieval. Point
Description
Measured albedo
Landsat albedo
Difference
Difference (%)
V1 V2
White surface parking lot Rural field
0.383 0.170
0.370 0.150
0.013 0.020
+3% +11%
Fig. 2. View of the two points (V1 and V2) chosen for the validation with ground albedo measurements.
It is evident how the ground measurements validated the albedo estimation from the satellite data as already highlighted in a previous work of the authors (Baldinelli et al., 2015), with a maximum deviation of 0.02 in terms of absolute values and 11% as a percentage between the two methodologies. The reasons for the light shift registered between the Landsat albedo and the ground measurements can be ascribed to the satellite footprint averaging inside a 30 m pixel, as well as in the intrinsic hypothesis of Lambertian surface necessary to compare the two methodologies, which is not perfectly respected for the zones investigated. 4. Results
Fig. 3. Picture of the ground albedo measurement on the point V2.
surface wider than the satellite pixel size for the reflective bands (30 × 30 m). Fig. 2 shows the two points analysed with the relative coordinates, while Fig. 3 reports a picture of the measurement setup on point V2.
4.1. Albedo and LST temporal variation: detailed analysis of significant sub-areas The main territorial changes of the area under investigation are often accompanied by surface albedo modifications. In Fig. 4, five points of interest are reported, as they appear in the visible images of years 2005 (Regione Umbria, 2005) and 2015 (Google
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
5
Fig. 4. Detail of visible image, albedo and lst variation in five points of interest between the years 2005 and 2015.
Earth, 2015), jointly with the albedo and LST variation between the same period. The albedo variation (␣) is intended as the difference between the values of 2005 and 2015; the LST variation (LST) is reported, on the contrary, between 2015 and 2005, since an albedo decrease corresponds generally to a LST enhancement. The pictures in cyan-red scale show clearly the direct connection between the surface albedo and the LST values of the buildings analysed. In P1, for instance, an albedo increase in 2015 of 0.08 linked to a surface modification produced a LST diminution of around 0.8 ◦ C, in P4, on the contrary, the roof albedo diminished strongly (0.13), causing a temperature enhancement of 7.7 ◦ C; finally, the new construction in P2, although producing an albedo enhancement (0.04),
an increase of LST (5.3 ◦ C) is registered. This circumstance could be easily explained if it is considered that the building footprint was previously covered by vegetation, the latter producing a cooling effect despite a low albedo value. Thus, it emerges that the fine scale modifications could move towards both a reduction and an enhancement of the surface albedo and LST, depending on the previous situation of the area and on the materials chosen for the new settlements. It is worthy of particular interest the photovoltaic roof, as many industrial constructions hosted this solution all over Europe: in point P3, the photovoltaic roof caused an albedo diminution of 0.08 and a consequent LST increase of 6.1 ◦ C. If from one side the roof
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8 6
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
Fig. 5. Distribution of albedo for the selected Terni area (Maratta Alta) for 2005 (red) and 2015 (blue). Main statistics are reported. Vegetation pixels are discarded. Fig. 6. Distribution of albedo difference ∝ (2015–2005) for the selected area.
became a renewable energy source, from the other side diminished its albedo, giving a negative contribution to the greenhouse effect reduction: a deeper analysis is needed to understand the relative weight of each component. In the work of Akbari (Akbari et al., 2009) it is stated that a 0.25 surface albedo diminishing of 1 square meter brings to an equivalent emission of 64 kg CO2 . The area analysed lowered its albedo of about 0.07 (the roof was white before the photovoltaic plant set up), therefore, an increase of around 18 kg CO2 is caused by the installation of the PV panels. The same square meter may host in Italy approximately 100 W of photovoltaic panels, which produce in the same site around 130 kWh/year of electric energy. Hypothesizing a life cycle of 20 years and considering that in Italy, the energetic mix brings to emit 0.350 kg/kWh of electric energy produced (ENEA, 2016), the panels avoid 910 kg CO2 (the CO2 produced in the construction and disposal phase is neglected). As a result, the CO2 emission avoided is orders of magnitude higher than its increase linked to the effect of the albedo reduction. 4.2. Albedo and LST temporal variation: full area analysis The albedo and LST variations, described in the previous section at local scale, are now analysed at a global scale. The average urban change for the whole Maratta area is examined considering all the albedo and LST pixels of the Landsat 7 images on 29th of July 29, 2005, and on 25th of July 2015. For this analysis, vegetation pixels were removed, since they are characterised by low albedo values with different impact with respect to the low albedo of impervious surfaces. In fact, despite the low reflective properties, the energy stored by plants and foliage is mainly used for their life processes and evapotranspiration, producing a cooling effect of the vegetation cover. The vegetation filtering was performed selecting pixels with NDVI < 0.35. Fig. 5 shows the distribution of albedo values for the satellite scene of 2005 (red) and 2015 (blue). Mean and standard deviation (std) are also reported. Fig. 6 reports the distribution of the albedo difference (2015–2005) for the selected area. It is evident the change of the albedo during 2015, with an average reduction of 0.03. This means that the fine scale modifications, showing both a reduction and a growth of the surface albedo with respect to the previous situation, move globally towards a worsening of the albedo pattern. Even though the albedo drop of built-up areas leads
Fig. 7. Distribution of SUHI for the selected Terni area (Maratta Alta) for 2005 (red) and 2015 (blue).Main statistics are reported. Vegetation pixels are discarded.
to an increase of surface and air temperature, albedo reduction due to photovoltaic panel presence has not a negative impact on the greenhouse effect, as previously pointed out. The LST global pattern is analysed through the surface urban heat island (SUHI) intensity, defined as the difference in LST between the urban pixels and the surrounding rural background. Fig. 7 shows the distribution of SUHI values for 2005 (red) and 2015 (blue) satellite scenes, reporting mean and std. Fig. 8 shows the distribution of the SUHI difference (2015–2005). An increase of the SUHI intensity of 2.3 ◦ C is revealed, highlighting global heating of the area due to the urban texture changes after 10 years. Fig. 9 investigates the correspondence between LST and albedo for the two satellite scenes. Since each image has about 1700 pixels, a scatterplot LST-albedo would produce a data-cloud of hard interpretation: to better analyze the relationship, an average study was carried out evaluating the mean LST at each fixed increment of
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
7
Table 4 Mean and std of albedo from 2005 to 2016. Landsat 7 passages at around 11:50 CEST. Clear sky summer image of 2008 is not available.
Fig. 8. Distribution of SUHI difference (2015–2005) for the selected area.
Fig. 9. LST vs. albedo (2015 in blue, 2005 in red) scatterplot for the selected area and linear fitting.
the albedo. In particular, the incremental step is 0.004 in order to ensure about 100 evaluation points in the albedo range. The average scatterplot and the correspondent linear fitting reveal the expected “inverse” linear relation between LST and albedo. This relation changes for the two different Landsat 7 images, confirming the variability of these parameters in relation to the different urban texture and construction materials. Finally, the mean and std albedo for July–August Landsat passages from 2005 to 2016 are reported in Table 4. This temporal sequence of albedo average for the selected zone points out that the albedo reduction started from 2014, after the new urbanistic intervention. 5. Discussion The study showed that the retrieval from satellite data of the urban albedo and LST, or equivalently SUHI, provides information on their spatial pattern and on their temporal evolution to be linked to the urban texture change, and so to the urban sustainable development.
Landsat Image
Mean albedo
Std albedo
29 July 2005 1 August 2006 19 July 2007 16 July 2009 11 July 2010 23 August 2011 16 July 2012 3 July 2013 7 August 2014 25 July 2015 27 July 2016
0.26 0.27 0.26 0.26 0.26 0.26 0.27 0.26 0.22 0.23 0.22
0.042 0.042 0.039 0.039 0.043 0.038 0.047 0.038 0.034 0.030 0.035
Within an urban changing and development, if the albedo is not analysed with care, the risk of a SUHI increase arises. The refurbishing of existing construction should consider, among the other issues, the enhancement of the surface albedo, a low-cost intervention with positive outcomes on the greenhouse effect. When the city planning foresees the replacement of rural areas with anthropic installations (roofs, roads, pavements, squares, etc . . .), the negative environmental impact is unquestionable, firstly for the loss of the effect of CO2 capture connected to the photosynthesis. Furthermore, the foliage produce on the other hand the strong cooling effect of leaves, thanks to the evapotranspiration. Thus, a particular attention to the new impervious surfaces albedo is needed, with the aim of limiting their impact on the SUHI raise and on the greenhouse effect worsening. In addition to the results discussed in the previous section, an interesting question regards the relation between SUHI and heat island of the air, i.e how a measured LST increase can be translated into an air temperature increase. As Voogt and Oke (Voogt & Oke, 2003) summarized, the relation between surface and air temperature is empirical and no simple general relation exists, even if their correlation improved at night when microscale advection is reduced. Advection, i.e. the horizontal transport of heat by wind, depends on the geometry of the urban surface, surface moisture and roughness, thermal admittance and mean wind velocity. Therefore, air and LST relation can be solved if a detailed knowledge of the surface micrometeorology is available. Overall, the results of the Maratta area pointed out an average daytime SUHI increase of 2.3 ◦ C during the last 10 years, from which a lower increase of the daytime air temperature is expected (Stathopoulou et al., 2009). During night time, SUHI and heat island of the air should be similar (Stathopoulou et al., 2009). 6. Conclusions Satellite remote sensing techniques and methods proved their effectiveness in the spatial pattern analysis of albedo and LST: their temporal evolution allows to understand if the urban texture change moves towards a sustainable development in terms of SUHI reduction. The validation of spaceborne observations with ground measurements demonstrated that albedo retrieval from satellite data can be used with a satisfactory level of accuracy, with the enormous advantage of providing a wide spatial coverage, making possible to monitor the albedo and LST changes with time. The study pointed out that if the albedo is not carefully analysed within urban surface changes, the risk of a SUHI increase arises; furthermore, the technical and economical effort to make impervious surfaces highly reflective results almost negligible, especially within a new construction ore refurbishment framework.
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005
G Model SCS-535; No. of Pages 8
ARTICLE IN PRESS S. Bonafoni et al. / Sustainable Cities and Society xxx (2016) xxx–xxx
8
Although albedo decrease of impervious surfaces leads to a LST increase, albedo reduction of photovoltaic panels installed on roofs has not a long-term negative impact on the greenhouse effect due to the CO2 emission reduction. Overall, the assessment of urban change effects can suggest in future how to give a contribution to set up a sustainable strategy in designing smart cities: foreseeing both mitigating urban heat islands and evaluating the renewable energy efficiency, the use of remote sensors may constitute a resource. Anyway, designing smart cities with solutions and technologies that implement the issues above is only a first step: design smart cities means also favouring the use of the existing and avoiding the consumption of land, protecting and enhancing the urban green areas, by promoting energy efficiency and reducing polluting emissions. Currently, great efforts are being made to retrieve air pollution from satellite data, confirming the potentiality of this technology in support of strategies for urban sustainable development. Acknowldegments The work has been supported by “Fondo di Ricerca di Base di Ateneo”, funded by the Department of Engineering of University of Perugia-Italy. Authors wish to thank Eng. Roberta Anniballe for the data processing support and Eng. Andrea Presciutti for the ground measurements. References Akbari, H., Menon, S., & Rosenfeld, A. (2009). Global cooling: Increasing world-wide urban albedos to offset CO2 . Climatic Change, 94, 275–286. Akbari, H., Damon Matthews, H., & Seto, D. (2012). The long-term effect of increasing the albedo of urban areas. Environmental Research Letters, 7, 1–10. Anniballe, R., Bonafoni, S., & Pichierri, M. (2014). Spatial and temporal trends of the surface and air heat island over Milan using Modis data. Remote Sensing of Environment, 150, 163–171. Baldinelli, G., Bonafoni, S., Anniballe, R., Presciutti, A., Gioli, B., & Magliulo, V. (2015). Spaceborne detection of roof and impervious surface albedo: Potentialities and comparison with airborne thermography measurements. Solar Energy, 113, 281–294. Barsi, J. A., Barker, J. L., & Schott, J. R. (2003). An atmospheric correction parameter calculator for a single thermal band earth-sensing instrument. In Proceedings of IEEE international geoscience and remote sensing symposium 21–25 july (pp. 3014–3016). Bonafoni, S., Anniballe, R., Gioli, B., & Toscano, P. (2016). Downscaling landsat land surface temperature over the urban area of florence. European Journal of Remote Sensing, 49, 553–569. Bretz, S., Akbari, H., & Rosenfeld, A. (1998). Practical issues for using solar-reflective materials to mitigate urban heat islands. Atmospheric Environment, 32(1), 95–101. Chander, G., Markham, B. L., & Helder, D. L. (2009). Summary of current radiometric calibration coefficients for landsat MSS, TM, ETM+, and EO-1 ALI sensors. Remote Sensing of Environment, 113, 893–903. Chavez, P. S. (1996). Image-based atmospheric corrections-revisited and improved. Photogrammetric Engineering and Remote Sensing, 62(9), 1025–1036. Dimoudi, A., Zoras, S., Kantzioura, A., Stogiannou, X., Kosmopoulos, P., & Pallas, C. (2014). Use of cool materials and other bioclimatic interventions in outdoor places in order to mitigate the urban heat island in a medium size city in Greece. Sustainable Cities and Society, 13, 89–96. ENEA −Italian National Agency for New Technologies, Energy and Sustainable Economic Development, Analisi trimestrale del Sistema Energetico italiano −1◦ trimestre 2016. Ed. ENEA, Rome 2016 in Italian.
http://ec.europa.eu/clima/policies/international/paris protocol/cities/index en. htm (retrieved on 7th of September 2016). Gaitani, N., Santamouris, M., Cartalis, C., Pappas, I., Xyrafi, F., Mastrapostoli, E., et al. (2014). Microclimatic analysis as a prerequisite for sustainable urbanisation: Application for an urban regeneration project for a medium size city in the greater urban agglomeration of Athens, Greece. Sustainable Cities and Society, 13, 230–236. Givoni, B. (1991). Impact of planted areas on urban environmental quality: A review. Atmospheric Environment, 25, 289–299. Google Earth, 7.1.5.1557. Terni 303,530 m E and 4,715,285 m N, 7th of August 2015. International Organization for Standardization. (1990). Standard ISO 9060 Solar energy – Specification and classification of instruments for measuring hemispherical solar and direct solar radiation. Jimenez-Munoz, J. C., & Sobrino, J. A. (2003). A generalized singlechannel method for retrieving land surface temperature from remote sensing data. Journal of Geophysical Research, 108, 1–9. Kikon, N., Singh, P., Singh, S. K., & Vyas, A. (2016). Assessment of urban heat islands (UHI) of Noida City, India using multi-temporal satellite data. Sustainable Cities and Society, 22, 19–28. Liang, S. (2001). Narrowband to broadband conversions of land surface albedo − I algorithms. Remote Sensing of Environment, 76, 213–238. Mackey, C. W., Lee, X., & Smith, R. B. (2012). Remotely sensing the cooling effects of city scale efforts to reduce urban heat island. Building and Environment, 49, 348–358. Oke, T. R. (1982). The energetic basis of the urban heat island. Quarterly Journal of the Royal Meteorological Society, 108, 1–24. Raghavan, K., Mandla, V., & Franco, S. (2015). Influence of urban areas on environment: Special reference to building materials and temperature anomalies using geospatial technology. Sustainable Cities and Society, 19, 349–358. Regione Umbria–Giunta Regionale – SIAT Sistema Informativo Regionale Ambientale e Territoriale 2005. Rizwan, A. M., Dennis, L. Y. C., & Chunho, L. I. U. (2008). A review on the generation, determination and mitigation of Urban Heat Island. Journal of Environmental Science, 20, 120–128. Schaepman-Strub, G., Schaepman, M. E., Painter, T. H., Dangel, S., & Martonchik, J. V. (2006). Reflectance quantities in optical remote sensing-definitions and case studies. Remote Sensing of Environment, 103, 27–42. Smith, C., & Levermore, G. (2008). Designing urban spaces and buildings to improve sustainability and quality of life in a warmer world. Energy Policy, 36, 4558–4562. Sobrino, J. A., Jimenez-Munoz, J. C., & Paolini, L. (2004). Land surface temperature retrieval from LANDSAT TM 5. Remote Sensing of Environment, 90, 434–440. Stathopoulou, M., Synnefa, A., Caralis, C., Santamouris, M., Karless, T., & Akbari, H. (2009). A surface heat island study of Athens using high-resolution satellite imagery and measurements of the optical and thermal properties of commonly used building and paving materials. International Journal of Sustainable Energy, 28(1), 59–76. Suehrcke, H., Peterson, E. L., & Selby, N. (2008). Effect of roof solar reflectance on the building heat gain in a hot climate. Energy and Buildings, 40, 2224–2235. Synnefa, A., Santamouris, M., & Livada, I. (2006). A study of the thermal performance of reflective coatings for the urban environment. Solar Energy, 80, 968–981. Taha, H., Akbari, H., Rosenfeld, A., & Huang, J. (1988). Residential cooling loads and the urban heat island: The effects of albedo. Building and Environment, 23, 271–283. Takebayashi, H., & Moriyama, M. (2007). Surface heat budget on green roof and high reflection roof for mitigation of urban heat Island. Building and Environment, 42, 2971–2979. Verducci, P., & Desideri, U. (2012). Architecture, urban design, energetic systems. Design of eco-sustainable industrial parks with very low environmental impact, reduced use of water and energy resources and optimized waste management. Perugia: Ed. Morlacchi Editore. Voogt, J. A., & Oke, T. R. (2003). Thermal remote sensing of urban climates. Remote Sensing of Environment, 86, 370–384. Wang, Y., & Akbari, H. (2016). Analysis of urban heat island phenomenon and mitigation solutions evaluation for Montreal. Sustainable Cities and Society, 26, 438–446.
Please cite this article in press as: Bonafoni, S., et al. Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society (2016), http://dx.doi.org/10.1016/j.scs.2016.11.005