Keywords

1 Introduction

Nowadays, people live more in urban areas (54% in 2014) than in rural areas, and this trend is still growing with a prevision of 66% in 2050. The most urbanized regions include Northern America and Europe (respectively, with 82 and 73% of people living in urban areas), and all regions are expected to urbanize further over the coming decades (United Nations 2014).

Then, the control of the microclimate of urban environment became a very important matter as it can contribute to a better livability of the cities with an improved use of the outdoor spaces. In big cities, this phenomenon is strongly linked to the urban heat island (UHI) , and these aspects are treated in the first part of this work as state-of-art and literature review.

The second part of the work is dedicated to the evaluation of various factors influencing the air temperature and its variations in the urban context of Turin. Turin is one of the most populous cities in the northern part of Italy with an interesting various territory with parks, rivers, a historical centre, suburban areas and industrial zones.

The aim of this work is to elaborate a model for the prediction of air temperature as a function of urban variables, solar exposition and the characteristics of outdoor surfaces. Particular attention will be paid to factors influencing or closely related to the urban heat island, as already underlined in the previous works.

Different tools have been used to evaluate the microclimate at urban scale . ESRI ArcGIS 10.3 has been utilized to calculate the urban characteristics with a detailed representation of the built environment, starting from the Technical Map of the Metropolitan City of Turin. For the evaluation of the characteristics of outdoor urban surfaces, several images coming from the Aster satellite have been processed. Finally, five different zones in Turin with a weather station inside have been selected to elaborate a model to predict the air temperature as a function of the urban form, the solar exposure and the albedo coefficient of outdoor surfaces. The resulting model can be useful in the design phase in order to plan a more sustainable urban development or to control air temperature variation in the urban space. The model can also help in understanding how temperature changes depend on urban planning variables, as the presence of rivers and green surfaces, giving indications on the interventions that can mitigate the overheating effect.

Some of the results show the urban heat island effect with lower temperatures near the parks and the riversides and higher temperature in the high-density centre and industrial zones. These results will be further investigated on a larger zone around Turin considering more weather stations, other satellite images as the Landsat ones and additional indicators, for example, the normalized vegetation index (NDVI).

The analysis conducted in this work starts from previous researches on the influence of different materials used for outdoor urban spaces on the microclimate of urban environments (Mazzotta and Mutani 2015), on the evaluation of the urban form of buildings’ heritage and its influence on urban microclimate and heating energy consumptions of residential buildings (Delmastro et al. 2015a, b) and on the urban heat island effect in Turin (Mutani 2016; Mutani et al. 2016).

2 State of the Art and Literature Review

The urban overheating problem is a matter of fundamental importance that all the developed and the developing countries are studying to mitigate its effects. For big cities, this issue is strongly linked to the thermal phenomenon called urban heat island (UHI) .

Starting from the 1970s, thanks especially to the availability of data, tools and techniques were improved, and new algorithms and computational models have been formulated to understand and analyse more deeply the heat island phenomenon. The first studies on this phenomenon were based on the acquisition of satellite images (i.e. Landsat) to calculate the land surface temperatures (LSTs) and to observe how this parameter changes according to the different types of urban surfaces.

Several studies (Aniello et al. 1995; Iino and Hoyano 1996; Jusuf et al. 2007) analyse the relationship between land use and land cover (LULC) to understand how human activities can influence the intensity of the heat islands. The first results show that the heat island is mostly concentrated in urban areas, while the phenomenon is not observed in the surrounding rural areas (Li et al. 2009; Unger et al. 2010; Onischi et al. 2010): this is why the temperature measured in urban context is higher than in the peripheral zones.

Later studies begin to take into account different variables and factors that may affect the intensity of UHI as the relationship between land use and type of coverage (LULC), the normalized vegetation index (NDVI) and the land temperature surface (LST). All these factors were calculated by elaborating satellite images, such as Landsat and ASTER, through the use of specialized software capable in the localizing information, such as geographic information system (GIS) (Weng et al. 2004; Onischi et al. 2010; Li et al. 2011; Zhang et al. 2012, 2013; Xu et al. 2013; Effat and Hassan 2014; Kong et al. 2014; Shahidan et al. 2015; Fernàndez et al. 2015).

Together with these studies, a very important aspect to be taken into account is the variations of microclimate and UHI with the urban morphology . Depending on the different types of buildings and urban forms, the urban heat island has different intensities. Some researchers have investigated the changes of UHI with the urban characteristics, i.e. the urban density, the urban form and height of buildings (Aniello et al. 1995; Iino and Hoyano 1996; Weng et al. 2004; Unger et al. 2010; Li et al. 2011; Houet and Pigeon 2011; Zhang et al. 2012; Schwartz et al. 2012; Merbitz et al. 2012; Li et al. 2012; Chun B and Guldmann 2014; Ivajnsic et al. 2014; Allegrini et al. 2014; Kong et al. 2014; Sailor 2014; Dabaieh et al. 2015). These parameters are also used to describe the urban canyon phenomenon, characterized by high buildings facing on the same street. In this condition, the solar radiation is almost completely absorbed by urban surfaces, and the air stagnates, with an increase of air temperatures and, consequently, the urban overheating. As well as with these parameters, also the characteristics of materials, as the albedo coefficient, should be considered. This parameter is an intrinsic characteristic of the materials and represents the property of a material to reflect incident solar irradiation. In fact, a material such as asphalt, with very low of albedo value, tends to store and absorb all the incident solar irradiation, reaching higher surface temperatures than clay materials that absorb less heat and have less high surface temperatures.

Many studies evaluate algorithms to calculate albedo coefficients using satellite images on different electromagnetic bands (Li et al. 2009; Stathopoulou and Cartalis 2009; Mallick et al. 2013; Kolokotroni et al. 2014; Santamouris 2014a; Salata et al. 2015; Thophilou and Serghiedes 2015; Qin 2015; Coseo and Larsen 2015, Pisello 2015).

Nowadays, to investigate the UHI effect, the microclimate and the outside air temperature variations, the urban morphology , the properties of the open space materials (with the albedo coefficient) and the surfaces’ temperature have been analysed (Effat and Hassan 2014; Balogun et al. 2014; Zhang et al. 2014; Pisello and Cotana 2014, 2015; Santamouris 2014b; Fernàndez et al. 2015; Yang et al. 2015; Touchaei and Wang 2015; Debbage and Shepherd 2015; Kaloustian and Diab 2015; Fernàndez et al. 2015; Berger et al. 2015. Santamouris et al. 2015; Quan et al. 2015; Coseo and Larsen 2015; Mirzaei 2015; Raghavan et al. 2015; Peron et al. 2015; Yang et al. 2015).

While almost all of the analysed documents are proposing to investigate the heat islands and their effect on urban temperatures, other studies are focusing, instead, on the influence of the overheating on thermal comfort and on energy consumptions (Balogun et al. 2014; Zhang et al. 2014; Santamouris 2014a, b; Pisello 2015; Gracik et al. 2015; Santamouris et al. 2015; Misni 2015; Liu et al. 2015). These studies focus more on the heating than the cooling energy uses; obviously, the outside temperatures strongly influence the energy consumptions for space heating and cooling, but also the electrical consumptions are influenced by the outside microclimate. These researches evaluate solutions to reduce the heat island effect, air pollution, energy consumptions and costs. The main ideas concern the use of mitigation techniques to reduce the air temperature in urban areas. One method is to replace the traditional roofs with “cold roofs” made from tiles or slates in clay. In fact, these surfaces guarantee higher albedo coefficients able to reflect the solar irradiation with lower temperatures (Pisello and Cotana 2014, 2015; Dabaieh et al. 2015).

Other studies are focused, instead, on the use of the green on buildings’ roofs or facades or on parking structures; through evapotranspiration, external temperature is lower (Zhang et al. 2014; Kong et al. 2014; Fernàndez et al. 2015; Qin 2015; Raghavan et al. 2015).

It was also analysed the so-called cold asphalt, the last generation of asphalts with very-high albedo values, able to reflect solar irradiation and keep lower temperatures than traditional bituminous asphalt (Carnielo and Zinzi 2014; Qin 2015).

In almost the cited studies, the relationships between the outside air temperatures and some of the parameters mentioned above are investigated. Some of them provided also a model or an algorithm, to be used in the design phase, in order to create a more sustainable urban planning (Iino and Hoyano 1996; Unger et al. 2010; Yeo et al. 2013; Chun and Guldmann 2014; Feng et al. 2014; Ivajnsic et al. 2014; Allegrini et al. 2014; Zhang et al. 2014; Kolokotroni et al. 2014; Fernàndez et al. 2015; Quan et al. 2015; Morris et al. 2015).

In this work all the parameters characterizing the urban morphology , the solar exposure and the albedo coefficients will be analysed in relation to air temperature variation to define an algorithm for the evaluation and prediction of the microclimate in Turin.

3 Case Study: Turin

Turin is the fourth most populous Italian town in the North-West part of Italy with a particular territory characterized by surrounding mountains and hills, by many rivers and parks and an elegant built context.

Turin is called “the city of four rivers” because it is localized in a plain surrounded by Stura di Lanzo, Sangone and Po rivers with also the Dora Riparia flowing close to its historical centre. Its territory is also surrounded by the Alpine mountains, also connecting the city with France. Moreover, Turin is the Italian city with more public green with about 21.1 m2 of green per capita and about 160,000 trees along the streets and in the parks.

The climate is continental temperate with cold-dry winters and hot-humid summers, and, as the majority of urbanized territories, Turin is characterized by the UHI phenomenon with significantly higher temperatures than in the rural and hilly areas around the city. Considering the weather stations inside the town, there are differences on the monthly average air temperature principally due to the buildings’ density, the presence of industrial zones and the proximity of parks and rivers.

From the 1990s, the summers have registered a significant warming, especially from the year 2000; the maximum air temperature was recorded in the hot summer 2003 with a peak of 39.7 °C in August 11.

In Fig. 11.1, the monthly average air temperatures registered by the ARPA weather stations in Turin from June 2005 to June 2015 are represented (ARPA is the Regional Agency for the Protection of the Environment of Piedmont Region).

Fig. 11.1
figure 1

Average monthly air temperatures registered in five ARPA weather stations in Turin

Full size image

As it is possible to note, the temperature differences vary from 2 to 4 °C during the cold months and from 0 to 2 °C during the hot months. Much more differences are registered considering the minimum air temperature values with differences up to 9 °C from the different weather stations.

In Turin, in the last 10 years, the average value of heating degree days (HDD), for the heating seasons, is of about 2350 HDD with a standard deviation of about 200 HDD. The coldest area is Vallere (with parks and the Po river) and the warmest is via della Consolata (with a high buildings’ density).

In Figs. 11.2 and 11.3, the position of eight weather stations in the city of Turin is represented. Particularly, some weather stations are localized in the high buildings’ density zones, others in suburban zones and two of them in the parks; one of these, Vallere in the census section n. 2295, is near the Po river. The four rivers Po, Dora Riparia, Stura di Lanzo and Sangone can be recognized, respectively, in the east side near the hill, through Turin centre, in the northern part and in proximity to Turin’s southern boundary .

Fig. 11.2
figure 2

The eight weather stations in the city of Turin with the indication of the relative census sections and the ten districts (Source: author’s elaboration)

Full size image
Fig. 11.3
figure 3

The eight weather stations in the city of Turin with the representation of the urban context in which they are localized and the relative district (Source: author’s elaboration)

Full size image

In this work the causes of air temperature differences in the city of Turin are analysed considering also the different urban contexts in the census parcels in which the weather stations are localized.

4 Instruments and Methods

The aim of this work is to analyse how the built environment can influence the microclimate of different urban contexts. This analysis is focused on the census parcels, in which the weather stations of the city of Turin are localized (Fig. 11.2).

The variables analysed to characterize the built environment can be summarized in an urban morphology factor (U), a “solar exposure” (P) factor and the albedo coefficient (A) of outdoor materials.

The urban morphology characteristics (U) represent the contribution of the form of the built environment affecting the outdoor air temperature variation. The urban morphology is influenced mainly by the following variables: the buildings’ density (BD) and the average buildings’ distance (W) or the buildings’ coverage ratio (BCR) and the aspect ratio (height to buildings’ distance, H/W).

The urban morphology factor (U) can be expressed by

$$ U= BD\ /\ W= BC{R}^{.} H/ W\left[-\right]. $$
(11.1)

The solar exposure (P) is the function of the ratio between the height of the building and the height of the surroundings (H/H avg), the main buildings’ orientation (MBO) and the main orientation of the streets (MOS)

$$ P= H\ /\ {H_{avg}}^{.} MB{O}^{.} MOS\left[-\right]. $$
(11.2)

The optimal orientation for buildings and streets for solar exposure is the east-west direction; then different values have been assigned to the MBO and MOS as shown in Fig. 11.4.

Fig. 11.4
figure 4

Values for the main orientation of streets and buildings (MOS, MBO) (Source: author’s elaboration)

Full size image

For the calculation of the albedo coefficient A, the correlations founded in literature have been used (Liang 2000 and Liang et al. 2002) using ASTER images. The sensor Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) is a satellite remote sensor providing high-resolution images of the earth in 14 different wavelengths on the electromagnetic spectrum.

ASTER images, in the Tagged Image File Format (TIFF), contain information relating to the radiance measured according to the wavelength of the sensor and expressed in the form of digital numbers (DN). To convert the DN in spectral radiance and top of atmosphere (TOA) radiance, specific correlations and coefficients can be used (Ghulam 2009; Banerjee et al. 2014).

From the spectral radiance and TOA radiance, it is possible to apply the equations of Liang, in order to calculate the albedo coefficient for all the wavelength bands of ASTER sensor. In particular, in this work, the data have been proceeded according to the Liang correlations for near-infrared ASTER albedo coefficient, using the satellite images relating to VNIR 1–2–3 bands (visible and near infrared) and 4–5–6–7–8–9 SWIR (shortwave infrared). The images of the ASTER sensor were provided by CSI (Consortium for Information System of Piedmont Region) and are related to the acquisition of July 22nd, 2004.

Finally, each single urban variable is analysed with its average value considering its relative census section. For each census section, a global value G is then defined:

$$ G\left( U,\ P,\ A\right)\left[-\right] $$
(11.3)

considering the variables of “urban morphology factor ” (U), “solar exposure” (P) factor and albedo coefficient (A).

In this work, the urban variables of five different census sections are compared considering the average monthly air temperatures registered by the weather stations in Turin.

5 Results and Discussion

The city of Turin is localized in the Po Valley, and it is surrounded by beautiful mountains and hills. Turin is an important nineteenth-century city with historical buildings, beautiful squares and outdoor spaces but also industrial zones.

The area of the census sections in the city of Turin is not uniform due to districts with very large census sections as the hilly area (districts 7 and 8) and the industrial areas (districts 6 and 10) as shown in Figs. 11.2 and 11.3. The average size of the sections of the Turin census is about 35,000 m2, with 17% of roads and 19% of built area (which becomes 23% if only the blocks of buildings are considered). Among this analysis, the census sections that are closest to these average values are the weather stations of Reiss Romoli (census section n.1886 in the northern area of Turin) and Buon Pastore (census section n.152 that is now no longer active). The characteristics of the census sections analysed with the weather stations are reported in Table 11.1.

Table 11.1 Characteristics of the census sections with the analysed weather stations
Full size table

In Figs. 11.5 and 11.6, the values of the “urban morphology factor” (U) and “solar exposure” (P) factor are represented for each census section of the city of Turin. From Fig. 11.5, the high buildings’ density zone in the centre, the hilly part of the city in the south-east side and the industrial zones in the northern and southern parts can be recognized.

Fig. 11.5
figure 5

The urban morphology factor U for the census sections of the city of Turin

Full size image
Fig. 11.6
figure 6

The solar exposure factor P for the census sections of the city of Turin

Full size image

From Fig. 11.6, the solar exposure factor (P) is represented. The differences depend also by the orientation and the dimension of the census sections.

In Figs. 11.7 and 11.8, the albedo coefficient (A) is represented, respectively, with punctual values and average values for the census parcels. From Fig. 11.7, natural elements as the rivers and the hilly zone can be noticed; also the industrial roofs in the northern and southern part of the city can be noticed.

Fig. 11.7
figure 7

The albedo coefficient A in Turin (obtained from ASTER images of July 22nd, 2004)

Full size image
Fig. 11.8
figure 8

The average value of albedo coefficient A for the census sections of Turin

Full size image

Finally, in Fig. 11.8, the average value of the albedo coefficient was calculated for each census parcel. Higher values can be noticed in the central zone of the city and along the rivers; instead, lower values of albedo coefficient characterize the industrial zones with lighter colour roofs.

In Table 11.2, the main characteristics about the urban form (U), solar exposure (P) and albedo coefficient (A) of the census parcels analysed are reported (with the ARPA weather stations). In particular, the high buildings’ density (BD) values can be noticed for the areas with the weather stations of via della Consolata, Reiss Romoli, Alenia, Buon Pastore and Politecnico di Torino; lower values of BD can be observed in Giardini Reali and Vallere which are localized in parks and Vallere also near the Po river; finally, the Italgas weather station is in an ex-industrial zone.

Table 11.2 Characteristics of the census sections with the analysed weather stations
Full size table

The aspect ratio (H/W) is higher also for the high-density central zone, and, about the orientation of streets (MOS) and buildings (MBO), the main orientation is south, south-west and south-east.

Starting from these urban data, the aim of this work is to derive a model for the outdoor air temperature considering the average monthly data. This model could allow understanding, at the first design-planning phase, the effect of the materials, the geometric buildings’ forms, the characteristics of outdoor spaces and the solar exposition, on the performance of air temperatures; the model could be used as a tool to improve environmental sustainability and urban liveability in urban planning.

For a good design tool, the model should evaluate the average monthly air temperature for a prediction in all seasons, especially during wintertime and summertime.

A multiple linear regression analysis correlates the urban variables G(U, P, A) to the average air temperatures considering also a typical monthly air temperature behaviour for Turin (ΔT m )

$$ \begin{array}{l}{T}_{air}= f(GUPA)= f(\varDelta {T}_m,\ BCR,\ H/ W,\ MOS,\ H/{H}_{\mathrm{aver}},\ MBO,\ A)\\ {}=\left({\alpha_1}^{.}\varDelta {T}_m\right)+({\alpha_2}^{.} BCR)+({\alpha_3}^{.} H/ W)+({\alpha_4}^{.} MOS)+({\alpha_5}^{.} H/{H}_{\mathrm{aver}})+({\alpha_6}^{.} MBO)\\ {}\kern1.25em +\left({\alpha_7}^{.} A\right)[{}^{\circ}\mathrm{C}].\end{array} $$
(11.4)

where ΔT m is the monthly variation of the air temperatures considering weather data from 2006 to 2014 averaged on the actual functioning five ARPA weather stations in Turin: via della Consolata, Giardini Reali, Reiss Romoli, Vallere and Alenia.

In Table 11.3, the values of the monthly air temperature registered from the weather stations are reported with the monthly air temperature gradient. This value goes from 0 to 1, and it has been calculated from the relation:

$$ \varDelta {T}_m=\left({T}_m - {T}_{\min}\right)/\left({T}_{\max } - {T}_{\min}\right)\left[-\right] $$
(11.5)

and the ∆T m will be considered uniformly distributed in the all city of Turin.

Table 11.3 Monthly air temperatures (2005–2015) and the relative value of monthly gradients ΔT m
Full size table

In Table 11.4 the average characteristics of the census section with the considered five weather stations are reported.

Table 11.4 Statistical data of the census parcel with the five weather stations
Full size table

A regression analysis of the monthly air temperatures was performed considering the most influential urban variables in Eq. 11.4 to reach the measured values for each weather station in Table 11.3.

The resulting model is the following:

$$ \begin{array}{l}{T}_{air}= f(GUPA)=({22.75}^{.}\varDelta {T}_m)+({3.31}^{.} BCR)+{0.10}^{.} H/ W)+({2.38}^{.} MOS)\hfill \\ {} + \left({0.55}^{.} H/{H}_{\mathrm{aver}}\right)+(-{0.44}^{.} A)[{}^{\circ}\mathrm{C}]\hfill \end{array} $$
(11.6)

Using the model of Eq. 11.6, the air temperature is proportional with the building coverage ratio (BCR) (and then the building density (BD)), the aspect ratio (H/W), the main street orientation (MOS) and the ratio between building height and the height of the surrounding H/H aver, while the air temperature is inversely proportional to the albedo coefficient A, as expected.

Considering the effect of the water near the weather station of Vallere, another variable can be introduced (with water, H2O = 1; without water, H2O = 0). Then, the resulting model is the following:

$$ \begin{array}{l}{T}_{air}= f(GUPA)=({22.76}^{.}\varDelta {T}_m)+({2.19}^{.} BCR)+({0.99}^{.} H/ W)+({2.09}^{.} MOS)\hfill \\ {} + \left({0.74}^{.} H/{H}_{\mathrm{aver}}\right)+(-{0.56}^{.} A)+(-{0.53}^{.}{\mathrm{H}}_20)[{}^{\circ}\mathrm{C}].\hfill \end{array} $$
(11.7)

with the air temperature inversely proportional also to the presence of water, as expected.

Also the presence of vegetation was investigated, with a similar indicator as the one for the presence of water, but without better results.

As an example in Figs. 11.9 and 11.10, the results of the models (Eqs. 11.6 and 11.7) and the measured values of air temperatures are reported for the weather stations of via della Consolata and Vallere.

Fig. 11.9
figure 9

Comparison between measured air temperatures with the weather station of via della Consolata and the calculated air temperatures as a function of G(UPA) with the model (Eq. 11.6)

Full size image
Fig. 11.10
figure 10

Comparison between the measured air temperatures with the weather station of Vallere (with Po river) and the calculated air temperatures as a function of G(UPA) with the model (Eq. 11.7)

Full size image

In Table 11.5, the relative errors of the models used are reported for all the weather stations considered. As it is possible to notice, the relative errors are under 10% with both the models considering (with Eq. 11.7) or not (with Eq. 11.6) the presence of water.

Table 11.5 The relative errors between the measured air temperatures and the calculated values with the models (Eqs. 11.6 and 11.7) for the census sections with the ARPA weather stations
Full size table

In Figs. 11.11, 11.12, 11.13, 11.14, 11.6, 11.17 and 11.18, the evaluations of the outdoor air temperatures in the city of Turin are represented with the model of Eqs. 11.6 and 11.7 as a function of the urban morphology , the solar factor, the albedo coefficient and the presence of water for different months during the heating and cooling seasons.

Fig. 11.11
figure 11

Evaluation of outdoor air temperatures in January in the census sections of the city of Turin as a function of G(UPA) with Eq. 11.6

Full size image
Fig. 11.12
figure 12

Evaluation of outdoor air temperatures in January in the census sections of the city of Turin as a function of G(UPA) with Eq. 11.7 (considering the presence of water)

Full size image
Fig. 11.13
figure 13

Evaluation of outdoor air temperatures in April in the census sections of the city of Turin as a function of G(UPA) with Eq. 11.6

Full size image
Fig. 11.14
figure 14

Evaluation of outdoor air temperatures in April in the census sections of the city of Turin as a function of G(UPA) with Eq. 11.7 (considering the presence of water)

Full size image
Fig. 11.15
figure 15

Evaluation of outdoor air temperatures in July in the census sections of the city of Turin as a function of G(UPA) with Eq. 11.6

Full size image
Fig. 11.16
figure 16

Evaluation of outdoor air temperatures in July in the census sections of the city of as a function of G(UPA) Turin with the Eq. 11.7 (considering the presence of water)

Full size image
Fig. 11.17
figure 17

Evaluation of outdoor air temperatures in October in the census sections of the city of Turin as a function of G(UPA) with Eq. 11.6

Full size image
Fig. 11.18
figure 18

Evaluation of outdoor air temperatures in October in the census sections of the city of Turin as a function of G(UPA) with Eq. 11.7 (considering the presence of water)

Full size image

As it is possible to note, the areas near the rivers are the most cold zones with some of the areas in the hill (south-east) or near the parks. The areas near the centre of the city with high buildings’ density are the more warm together with some of the industrial zones in the southern and in the northern parts of the city. These results are confirmed in all months.

Comparing the results of the two models considering (Eq. 11.7) or not (Eq. 11.6) the presence of water in the areas, some differences can be noticed. The different influence of the aspect ratio H/W and of the presence of water must be further investigated in future works with more weather stations inside the city of Turin and comparing the results of the model with other data as the Landsat satellite images; also the presence of vegetation should be analysed.

6 Conclusion

In this work, the analysis of urban microclimate, the effect of urban variables and the presence of water and green surfaces have been conducted on the city of Turin, using information from the following sources: ARPA Piemonte weather stations, ASTER satellite data and urban variables from the Technical Map of the Metropolitan City of Turin.

The data acquired by weather stations are used to estimate the average monthly air temperatures measured in the areas with the weather stations in Turin; from the ASTER data, the albedo coefficient values were obtained for each point of the city and for each census sections of Turin. The Technical Map of the Metropolitan City of Turin has been used to evaluate all geometrical characteristics of the built heritage of the city of Turin to calculate the urban morphology factor U and the solar exposure P.

All the above information have been used to formulate two simplified models for the calculation of the outdoor air temperature: the first model calculates the air temperature taking into account the urban planning variables and the second model considering also the presence or the absence of the water.

The application of the model consents to map the air temperatures in the city of Turin for each month, according to the urban variables and the albedo coefficient of the outdoor spaces to understand the air temperature variations.

The model can be also applied in different design phases of the urban planning; it can be applied in the predesign phase to see how the project can influence the outside air temperature and then the liveability of the outdoor spaces in all the seasons. Moreover, the model can also be used to check how the different urban variables can influence the outdoor air temperatures and to evaluate how to improve the liveability of the urban environment with interventions that can mitigate the microclimate at blocks of building scale .

Future works will improve the models considering more areas around Turin with weather stations and comparing also the results of the models with other satellite images as the Landsat ones.