Energy Retrofit of Social Housing with Cultural Value in Spain: Analysis of Strategies Conserving the Original Image vs. Coordinating Its Modification

Abstract: Two renovation strategies were considered and compared in this paper for the energy renovation of social housing in condo buildings built in 1945–1969, situated in residential estates that have been declared urban areas of interest. One of the strategies, named here the image conservation approach, consists of renovating the building with very efficient solutions, except for the brick-facing walls, which will only have insulation in the existing cavities. The other strategy, the image coordination approach, consists of renovating the whole building with efficient solutions, including external insulation for existing brick-facing walls, and drawing up coordination plans for each of the residential estates in order to guarantee a coordinated new image of the residential estates. By means of three case studies in the city of Zaragoza and four scenarios of improvement, the two strategies are compared regarding the regulation compliance, the energy use, the emissions reduction, the thermal comfort, and the cost-effectiveness of the measures. The results show that the image coordination approach is the best strategy in regard to the studied aspects, achieving a 69%–79% reduction of CO2 emissions and a 75%–97% reduction of heating use, as well as being the only one capable of transforming these buildings into nearly zero energy buildings (nZEBs).


Introduction
The European Union (EU) has set targets for reducing its greenhouse gas emissions progressively up to 2050 by means of the 2020 climate and energy package, the 2030 climate and energy framework, and the 2050 long-term strategy [1]. Sustainable construction is expected to play a big part in meeting the European Union's climate and energy goals since the built environment is estimated to account for approximately 40% of energy consumption and 36% of CO 2 emissions in the EU [2]. Sustainable, climate-proofed buildings are needed on a massive scale to meet the objectives for 2020 and 2030, and to achieve a climate-neutral Europe by 2050 [2]. Now, the EU is urging countries to bring the building renovation wave into postpandemic recovery plans.
The majority of the building floor area in Spain is composed of residential buildings (82.67% in 2013) [3]. Since the existing residential buildings at present will represent the major part of the housing stocks of Spanish cities in 2050 [4,5], energy retrofit has become one of the most important strategies for the decarbonization of the residential sector. The European Directive 2010/31/EU on Energy Efficiency in Buildings introduced the need to transform buildings into nearly zero energy (nZEB) by 2050, and the European Directive 2018/844/EU amended the existing Directive 2010/31/EU in order to accelerate the decarbonization of the parks in the EU.
The energy retrofit of multifamily residential buildings built in 1945-1969 is of great importance in Spain because it is one of the countries in Europe with the highest share of multifamily dwellings (70.78%     The tensions and conflicts between the sustainability goals and the preservation of the modern built heritage have broadly been addressed in the literature [14][15][16][17][18][19][20][21]. Some authors took a stand against the demolition of old housing buildings, arguing that important synergies can be found between sustainability and conservation, e.g., [4,5,18,21]. Other authors focused on the design challenge of finding an architectural balance between existing and newly added qualities when designing the renovation of housing blocks with specific architectural heritage values, e.g., [14,15,19,20]. Oorschot et al. [20] considered the careful analysis of the architectural and cultural value of the original building by recording which characteristics must be preserved and which could be modified to be of great importance. In Amsterdam, buildings with historic and visual values were observed to offer fewer possibilities for interventions on the outside, especially the façades and the roofs, whereas inside insulation creates more inconvenience, especially if tenants stay in their homes during the intervention [14]. Dowson et al. [16] estimated that in UK there are restrictions on external wall insulation for a further 1.2 million dwellings in conservation areas and 300,000 listed dwellings [22] because of the change of appearance. Ben and Steemers [17] argued that some measures are generally not considered appropriate for listed buildings in UK, such as external wall insulation and double (or triple) glazing, and proposed several viable improvement strategies, including cavity wall insulation, roof insulation, secondary glazing, and draughtproofing, for the Brunswick Centre in London.
In the case of the UAI of Zaragoza, the main cons of inside insulation strategies discussed by the technicians of the municipality and the researchers were the following: • It reduces the already small usable floor areas. This disadvantage is very difficult to avoid by merging different dwellings in one, as in the case of Amsterdam [14] because of the condominium regime of the buildings, or by expanding the area of the dwellings because the floor additions would affect the image as much as, or even more than, the insulation.

•
It obliges occupants to temporarily leave their homes during the intervention. This is an inconvenience that could be partially mitigated by carefully planning the stages with more impact on the occupants' daily life and reducing their duration as much as possible.
For these reasons, the internal insulation strategies were discarded in Zaragoza and this position was accepted in this paper as a starting point. The disadvantage of external insulation is, as already mentioned, the potential loss of the residential estates' character of unity, and two possible options were considered in this study and have been discussed with the ZMHC to overcome the problem:

•
One of the options is to use image conservation-oriented energy retrofit strategies that do not compromise the façades image, as in the case of the Brunswick Centre in London [17], but without using inside insulation so that the usable floor area remains equal after the intervention. This implies energy improvement of roofs, windows, plastered walls, and existing wall cavities. Plastered walls are externally insulated, and brick-facing walls are improved by pumping insulation into the existing wall cavities, which are 3 cm thick [13].

•
The other option is to assume the preponderance of environmental matters and mitigate their negative impact on cultural aspects as much as possible by drawing up coordination plans for each of these areas in order that the individual interventions in each of the buildings follow some common rules that ensure a final image of unity. This implies that the whole envelope is deeply renovated. All façades, including brick-facing walls, are externally insulated. This paper focused on comparing different energy retrofit strategies for the case of the UAI in Zaragoza that do not reduce the usable floor area, with the aim to know if the image conservation approach and the image coordination one can ensure comparable levels of performance regarding the regulation compliance, the energy use, the emissions reduction, the thermal comfort, and the cost-effectiveness of the measures, in order to support with data the decisions of the local administration. The scientific contribution of this paper is in the field of design research [23,24], and focuses on the improvement of the observed phenomenon in architectural design of dealing with the tensions between the sustainability goals and the preservation of the modern built heritage.

Materials and Methods
In order to compare the different energy retrofit strategies, three case studies were used of real social housing buildings located in three of the 19 multifamily UAI of Zaragoza. The research methodology is composed of the next steps, which we develop in the following sections:
Definition of comparison criteria and simulation methods 4.
Results analysis and discussion 5. Conclusions.

Selection of Cases Studies
In previous research works, we had cataloged the 19 multifamily residential estates of Zaragoza declared as UAI [25], identified the construction characteristics of the buildings [13,26], done an energy and acoustic characterization of their envelopes [27,28], and simulated the buildings energy use through a simplified methodology [7,26,27]. Important conclusions from these previous studies have been useful to define the criteria to select our case studies.
On the one hand, we tried to select cases studies representative of the most common energy efficiency performances found among the buildings of the 19 residential estates. The energy performance rating for the indicator of nonrenewable primary energy was found to be an E for 88.5% of the buildings, F for 5.2%, and G for 3.1%, with the remaining 3.2% being renovated buildings in 2016 [29]. Regarding the energy performance rating for the indicator CO 2 emissions, the share was 80.9% of E and 15.9% of D [29]. As can be seen, by far the majority of buildings had a rating of E for nonrenewable primary energy and a rating of E for CO 2 emissions. Our three case studies had an E estimation for nonrenewable primary energy rating and an E estimation for CO 2 emissions rating, except for one of them (the Alférez Rojas case), which had an F estimation for nonrenewable primary energy rating (Table 1). On the other hand, we selected cases studies with different construction solutions observed in the 19 residential estates in order to check whether these different solutions can make an important difference or not regarding the energy performance of the renovated building:

•
Most of the buildings presented façades with thick brick-bearing walls without thermal insulation, and only some of the buildings, built at the end of the studied period, presented structures of pillars with the corresponding thinning of the façade without thermal insulation [13]. Two of our case studies (the Alférez Rojas and Girón cases) presented thick brick-bearing walls, and one of them presented thinner walls (the Balsas de Ebro Viejo case), as can be seen in Table 1.

•
With regard to the grounds, most of the buildings (65% of them) presented concrete slab-on-grounds, some of them (26%) had suspended reinforced concrete one-way floors, and the remaining part (9%) had reinforced concrete one-way slabs in contact with uninhabitable rooms, such as storage rooms [13]. Each of our case studies presented one of these three solutions (Table 1).
The general image and urban layout of each of these case studies can be observed in Figure 3. The specific U-value of the different construction solutions for the envelope can be found in Appendix A. Since the thermal mass is important in the energy retrofit of buildings from this period [30], the internal areal heat capacity and periodic thermal transmittance for the different façades and grounds are provided in Appendix A to evaluate their thermal inertia, as suggested by [31,32]. The calculations were made according to EN ISO 13786:2017 [33]. the 19 residential estates in order to check whether these different solutions can make an important difference or not regarding the energy performance of the renovated building:


Most of the buildings presented façades with thick brick-bearing walls without thermal insulation, and only some of the buildings, built at the end of the studied period, presented structures of pillars with the corresponding thinning of the façade without thermal insulation [13]. Two of our case studies (the Alfé rez Rojas and Giró n cases) presented thick brick-bearing walls, and one of them presented thinner walls (the Balsas de Ebro Viejo case), as can be seen in Table 1.  With regard to the grounds, most of the buildings (65% of them) presented concrete slab-ongrounds, some of them (26%) had suspended reinforced concrete one-way floors, and the remaining part (9%) had reinforced concrete one-way slabs in contact with uninhabitable rooms, such as storage rooms [13]. Each of our case studies presented one of these three solutions ( Table  1).
The general image and urban layout of each of these case studies can be observed in Figure 3. The specific U-value of the different construction solutions for the envelope can be found in Appendix A. Since the thermal mass is important in the energy retrofit of buildings from this period [30], the internal areal heat capacity and periodic thermal transmittance for the different façades and grounds are provided in Appendix A to evaluate their thermal inertia, as suggested by [31,32]. The calculations were made according to EN ISO 13786:2017 [33].

Scenarios Definition
Five scenarios were considered for each of the three case studies ( Figure 4):

•
Scenario 0: original state. The envelope is the original one, described in Table 1 and thermally characterized in Appendix A. With regard to the heating and domestic hot water (DHW) systems, all of them are individual, and so are the cooling systems when they exist. For the sake of comparison, we considered, for the heating and DHW systems, an individual regular gas boiler in each of the dwellings, since it is the most extensive system in Spain [34]. For the cooling system, an electrical single zone system with a seasonal energy efficiency of 2.52 was considered, as recommended by the user guide for the official tool for energy rating in Spain when the system is not known or nonexistent [35].
• Scenario 1: the envelope is improved, avoiding the external insulation of the façades made out of facing bricks. Efficient solutions are considered for the rest of elements of the envelope, even when installing very thick insulations is technically difficult, such as in the case of the external insulation of plastered façades and grounds, in order to compensate the thin insulation of brick facing walls. This scenario follows the image conservation approach. Specifically, the energy improvement of the envelope is made by means of: • Scenario 2: the envelope is improved, insulating externally the whole façade. Efficient solutions are used for the elements of the envelope that they are technically easy to install (roofs and windows), whereas for more problematic elements of the envelope (grounds and external insulation of walls) more technically feasible solutions are considered. This scenario is thought to follow the image coordination approach. Specifically, the improvement is made by means of: • Same solutions for roofs and windows as in scenario 1.
• Grounds are improved for the case of suspended floors with 10 cm thick external PUR insulation, and for the case of slabs in contact with uninhabitable rooms with 10 cm thick external PUR insulation. The U-value of these two types of renovated grounds is 0.28 W/m 2 ·K (Appendix A). For the slab-on-ground, perimeter insulation is used. • External insulation of 10 cm thick for façades. The U-values of our renovated façades in this scenario range from 0.21 to 0.23 W/m 2 ·K (Appendix A).
• Scenario 3: the envelope improvement is the same as in scenario 1, and additionally: • The heating and DHW system is substituted by gas condensing boilers.
• Solar panels are installed in order to guarantee that at least 60% of the energy for DHW is solar, as prescribed in the Spanish Technical Code for this location if the system is changed. • Scenario 4: the envelope improvement is the same as in scenario 2, and as in scenario 3 the heating system is substituted by gas condensing boilers and 60% of the energy for DHW is solar.
The values for the parameters of the envelope and the heating, cooling, and DHW systems that define their contribution to the buildings energy efficiency can be found in Appendix A.
scenario range from 0.21 to 0.23 W/m 2 ·K (Appendix A).


Scenario 3: the envelope improvement is the same as in scenario 1, and additionally: o The heating and DHW system is substituted by gas condensing boilers. o Solar panels are installed in order to guarantee that at least 60% of the energy for DHW is solar, as prescribed in the Spanish Technical Code for this location if the system is changed.
 Scenario 4: the envelope improvement is the same as in scenario 2, and as in scenario 3 the heating system is substituted by gas condensing boilers and 60% of the energy for DHW is solar.
The values for the parameters of the envelope and the heating, cooling, and DHW systems that define their contribution to the buildings energy efficiency can be found in Appendix A.

Definition of Comparison Criteria and Simulation Methods
The five scenarios of the three case studies were compared in terms of: energy regulation compliance; energy use, CO2 emissions, and thermal comfort performance; and cost-effectiveness of the measures.

Methodology to Check the Energy Regulation Compliance
In December 2019, a royal decree modifying the Basic Document on Energy Saving (DB-HE) of the Buildings Technical Code was approved in Spain. This recently updated regulation sets, on the one hand, more restrictive requirements for major renovations-which are compulsory when more than 25% of the envelope is improved, as is our case-and, on the other hand, defines for the first time specific requirements for the transformation of existing buildings into nZEBs-which are not mandatory, but must be met if an nZEB renovation is aimed for.
In this paper, we checked the compliance or not of the five scenarios with the two set of requirements: the compulsory major renovation requirements and the optional nZEB renovation requirements.

Definition of Comparison Criteria and Simulation Methods
The five scenarios of the three case studies were compared in terms of: energy regulation compliance; energy use, CO 2 emissions, and thermal comfort performance; and cost-effectiveness of the measures.

Methodology to Check the Energy Regulation Compliance
In December 2019, a royal decree modifying the Basic Document on Energy Saving (DB-HE) of the Buildings Technical Code was approved in Spain. This recently updated regulation sets, on the one hand, more restrictive requirements for major renovations-which are compulsory when more than 25% of the envelope is improved, as is our case-and, on the other hand, defines for the first time specific requirements for the transformation of existing buildings into nZEBs-which are not mandatory, but must be met if an nZEB renovation is aimed for.
In this paper, we checked the compliance or not of the five scenarios with the two set of requirements: the compulsory major renovation requirements and the optional nZEB renovation requirements.
For both levels of renovation (major renovation and nZEB), the main parameters that are regulated are: • The envelope global heat transfer coefficient K-value (W/m 2 ·K) must be less than a limit value Klimit. The K-value depends on the U-values of the envelope elements and thermal bridges and it is the heat transfer coefficient of ISO 14683:2017, divided by the exchange area [39]. The maximum allowable Klimit depends on the climate zone and the compactness of the building and is the same for major renovation and nZEB renovation.

•
The value of the solar control of windows will be lower than 2 kW·h/m 2 ·month for the month of July for households, for both major renovation and nZEB renovation. The solar control is the Sustainability 2020, 12, 5579 9 of 24 ratio between the solar gains in July of all windows of the envelope with their solar protections activated and the useful floor area of the building. • Nonrenewable primary energy (kW·h/m 2 ·year) and total primary energy (kW·h/m 2 ·year) use are limited by a fixed value depending on the climate zone. The values for nZEB renovation are more restrictive than for major renovation. Specifically, for our climate zone D, the limit values for nonrenewable primary energy are 70 kW·h/m 2 ·year for major renovation and 38 kW·h/m 2 ·year for nZEB renovation, and the limit values for total primary energy are 105 kW·h/m 2 ·year for major renovation and 76 kW·h/m 2 ·year for nZEB renovation.
Additionally, the regulation sets other requirements, such as: • A maximum U-value is set for each element of the envelope that is added, replaced, or modified ( Table 2). Our four scenarios of improvement fulfill these values except for the brick facing walls in scenarios 1 and 3. • In our climate zone, at least 60% of the energy demand for DHW has to be obtained from renewable sources, with local origin or nearby the building. This is applicable to those renovations that change the building installations. We made scenarios 3 and 4 fulfil this requirement.

•
The accepted permeability of the windows depends on the climate zone. In our location, class 3 or higher, as defined in EN 12207:2017, is demanded. We chose class 4 with lower permeability and therefore better energy performance.
We made our four scenarios of improvement fulfil the additional requirements (except for the U-value of the brick-facing walls in scenarios 1 and 3), and afterwards we checked whether the main requirements were met or not for major renovation and nZEB renovation. To check the compliance of these main requirements, the official tool for energy certification of buildings in Spain was used, named HULC version 2.0.1960.1156 [39].

Methodology to Estimate Energy Use, CO 2 Emissions, and Thermal Comfort
To estimate energy use and CO 2 emissions, we used the HULC version 2.0.1960.1156 software [37], and to characterize the thermal comfort the Design Builder version 6 software [40]. This implies a total of 30 simulations.
The same building models were introduced in the two software tools regarding the shapes and orientations of the buildings, their construction solutions (Table 1 and  The result, 33 L/s, was the same for the five scenarios (scenarios 0-4), since it depends on the household layout, which was kept the same in the improved scenarios. It includes the infiltrations plus the ventilation through the use of the windows. Additionally, we considered the following values for the infiltration air flow rate per internal volume through the envelope (n50), both in HULC and Design Builder: 5.72 h −1 for scenario 0 (original building), as reported by Fernández-Agüera et al. [41] for multifamily social housing in Mediterranean Europe, and 4.91 h −1 for the improved scenarios (1 to 4), since according to Almeida et al. [42], the average window contribution for the room permeability is 15% and we changed windows with air permeability of 50 m 3 /h·m 2 to 3 m 3 /h·m 2 (a reduction of 94%). Night ventilation of four renovations per hour during summer was considered in the two software tools, since all the studied buildings allow for crossed ventilation.
With regard to the solar control of windows, we defined two parameters: • The total transmittance of solar energy of windows when all the solar devices are activated: according to the support document of the Spanish Technical Code for calculation of envelope parameters [43], it is of 0.15 for scenario 0 (characterized by a dark exterior shutter with single glass) and 0.05 for the improved scenarios 1 to 4 (characterized by a light shutter and a window with triple glass).

•
The activation of the solar protection: the activation values by default in HULC, for 30% for the summer period, were considered in both software tools. The way this parameter is introduced in Design Builder can be seen in Appendix C.
To estimate the thermal comfort, we used the adaptative comfort approach according to the ASHRAE55-2017 standard included in the Design Builder software for naturally ventilated buildings. This method has been proven to be useful for mixed-mode buildings such as ours (i.e., a combination of operable windows and mechanical conditioning) [44]. An annual simulation was run at intervals of one hour. Once the simulation was done, the data (air temperature, mean radiant temperature, operative temperature, exterior air temperature, renovations per hour by means of mechanical ventilation, natural ventilation, and infiltrations) in csv format were exported to the software CONFADAPT-ASH55 version 1.0 [45], which allowed us to obtain the hours inside and outside the 80% and 90% acceptability limits, indicating the percentage of occupants expected to be comfortable at the indicated indoor. We used the 90% acceptability limit to do the estimations with a higher standard of thermal of comfort.

Methodology to Estimate the Cost-Effectiveness of the Scenarios
We first estimated the total investment renovation costs of the measures for each of the four improvement scenarios of the three case studies by means of the CYPE software and database [46], and afterwards we divided it by the total surface to obtain the renovation investment per m 2 for each case. To estimate the cost-effectiveness of each scenario of improvement, the costs-to-effectiveness ratio [47] that we established divided the total investment costs by the saved kW·h of nonrenewable primary energy for the five scenarios of the three case studies.

Energy Regulation Compliance
Only one of the scenarios complied with the nZEB renovation requirements, scenario 4, and two of the scenarios, scenarios 2 and 4, complied with the major renovation requirements (Table 3). For the case study of Balsas de Ebro Viejo, it was necessary to increase the percentage of solar energy for domestic hot water (DHW) up to 65% in order to make it comply with the nZEB renovation requirements (we name this scenario 4* in Tables 3 and 4 and Tables A1-A3). This building presented a worse energy performance in all the scenarios (with 27%-59% more nonrenewable primary energy and total primary energy use), which could be explained by the fact that it is the only one with thinner façades and a corresponding lower thermal mass, as shown by the values of the periodic thermal transmittance in Appendix A.
This implies that if social housing in Zaragoza is renovated following the image conservation approach (scenarios 1 and 3), the energy regulation for major renovation, which is compulsory, could not be fulfilled, unless additional measures were taken on the top of those that we have studied here. Only the image coordination approach provides solutions that would comply with current regulations (scenarios 2 and 4) and could even transform existing social housing buildings into nZEBs when improving both the envelope and the systems (scenario 4). This was the case for the two case studies with massive load-bearing façades (Girón and Alférez Rojas). For the case of the building with thinner façades and a structure of pillars (Balsas de Ebro Viejo), the only scenario fulfilling the major renovation requirements was scenario 4. In order for the image conservation approach (scenarios 1 and 3) to fulfil the major renovation requirements, it would be necessary to reduce the K-value (Table 3). This would require, for example, adding internal insulation, but as already discussed this is not considered a feasible solution due to the small size of households.

Energy Use, CO 2 Emissions, and Thermal Comfort
When looking at the CO 2 emissions of the three case studies for the five scenarios considered (Table 4), we observed that:

•
The energy retrofit of the envelope following the image conservation approach (scenario 1) could be responsible for the reduction of 23%-39% of CO 2 emissions, depending on the case study, whereas the improvement of the envelope following the image coordination approach (scenario 2) could entail a reduction of 39%-58% of CO 2 emissions, according to simulations. • If, additionally, the systems are improved, the total reduction of CO 2 emissions following the image conservation approach (scenario 3) could be up to 53%-64%, whereas following the image conservation approach (scenario 4) it could be up to 69%-79%, according to simulations.
Therefore, the image coordination approach could bring about an additional 13%-24% reduction of CO 2 emissions, as compared to the image conservation approach, which is quite a significant figure considering the objectives of the EU to reduce emissions by 80%-95% for 2050.
When the CO 2 emissions are broken down into heating, cooling, and DHW emissions for the three case studies and five scenarios ( Figures 5-7), we observed that:

•
The envelope energy retrofit would significantly contribute to the reduction of emissions caused by the use of heating systems: a 45%-64% reduction in heating CO 2 emissions with the image conservation approach, and 75%-96% with the image coordination approach. On the other hand, it would increase the emissions caused by the use of cooling systems: a 9%-27% increase in cooling CO 2 emissions with the image conservation approach and 13%-39% with the image coordination approach. However, these increases are not so significant because CO 2 emissions due to cooling only represent a 6%-8% of total emissions in the original buildings (scenario 0) for this city and these case studies. The reason for this low need of cooling systems may be due to the combination of high internal thermal inertia, night ventilation, and solar protection, which together would significantly reduce the cooling demand in the five scenarios [48].

•
The proposed systems improvement would reduce 5%-20% of CO 2 emissions of heating systems and 69% of DWH systems.
Therefore, the most significant CO 2 emissions reductions would be caused by the deep energy retrofit of the envelope in the image coordination approach and the improvement of the DHW production systems.
When analyzing the energy use (broken down into heating, cooling, and DHW, as well as into renewable and nonrenewable) and thermal comfort for the three case studies and five scenarios (Figures 8-10), we observed that:

•
The reason why the emissions get so significantly reduced by the deep renovation of the envelope is the significant saving in heating energy use: a 45%-64% saving with the image conservation approach (scenario 1) and 75%-97% for the image coordination one (scenario 2), according to simulations. This reduction in heating energy use could be achieved without compromising the comfort. In fact, it would actually increase the percentage of hours of 90% acceptability, with an additional 48%-54% of hours for the image conservation approach (scenario 1) and 45%-55% for the image coordination approach (scenario 2). The comfort achieved would be quite similar for both approaches.

•
The reason why the emissions get so significantly reduced by the improvement of the DHW production systems is mainly the installation of renewable sources. As can be seen in Figures 8-10, the energy use for DHW would only be reduced by 21% in all the cases, and the rest of the emissions reduction (48%) would be achieved by the use of sun as energy source. 10, the energy use for DHW would only be reduced by 21% in all the cases, and the rest of the emissions reduction (48%) would be achieved by the use of sun as energy source.     10, the energy use for DHW would only be reduced by 21% in all the cases, and the rest of the emissions reduction (48%) would be achieved by the use of sun as energy source.

Cost-Effectiveness of the Scenarios
To complete the analysis, we performed a cost-effectiveness analysis of the scenarios for the three case studies. As can be seen in Table 5, the most cost-effective scenario is number 2, corresponding to the deep renovation of the envelope following the image coordination approach, with a cost of 0.69€-2.02€ per nonrenewable kW·h saved, depending on the case study. The next most cost-effective scenario is the one improving the envelope following the image conservation approach (scenario 1), followed by the scenarios that also improve systems (scenarios 4 and 3). Again, the image coordination approach was proven to be more advantageous from the sustainability point of view than the image conservation one.

Cost-Effectiveness of the Scenarios
To complete the analysis, we performed a cost-effectiveness analysis of the scenarios for the three case studies. As can be seen in Table 5, the most cost-effective scenario is number 2, corresponding to the deep renovation of the envelope following the image coordination approach, with a cost of 0.69€-2.02€ per nonrenewable kW·h saved, depending on the case study. The next most cost-effective scenario is the one improving the envelope following the image conservation approach (scenario 1), followed by the scenarios that also improve systems (scenarios 4 and 3). Again, the image coordination approach was proven to be more advantageous from the sustainability point of view than the image conservation one.

Conclusions
How to conduct the energy renovation of social housing of the period 1945-1969 in Spain, without compromising the character of unity of the residential estates to which they belong, is a challenge that some city councils are facing.
In Spain, the energy retrofit of social housing with cultural value presents different challenges from those identified in other countries, such as England or the Netherlands [14][15][16][17][18][19][20][21][22], since inside insulation is not considered a feasible option because of the small size of the households and the difficulty of merging dwellings in condo buildings.
Two possible approaches were considered in this paper to deal with the tensions between the sustainability goals and the preservation of the modern built heritage. One of them, named here the image conservation approach, consists of renovating the building with very efficient solutions, except for the brick-facing walls, which will only have insulation in their existing 3 cm thick cavities. The other strategy, the image coordination approach, consists of renovating the whole building with efficient solutions, including external insulation for existing brick-facing walls, and drawing up coordination plans for each of the residential estates in order to guarantee a coordinated new image of the residential estates.
By means of three case studies in the city of Zaragoza and four scenarios of improvement, the two strategies were compared regarding the regulation compliance, the energy use, the emissions reduction, the thermal comfort, and the cost-effectiveness of the measures.
The results show that from the points of view of environment (energy use and emissions) and cost-effectiveness, the image coordination approach is the best strategy because: • It would bring about an additional 13%-24% reduction of CO 2 emissions as compared to the image conservation approach. In total, it could contribute to up to 69%-79% reduction of CO 2 emissions in comparison with the original buildings, which is quite close to the objectives of the EU for 2050 to reduce emissions in the building sector by 80%-95%.

•
It implies an additional reduction of heating energy use of 29%-51% as compared to the image conservation approach. The heating energy use can be reduced in total up to 75%-97% in comparison with the original buildings with this approach, according to simulations.

•
Even if it implies an additional increase of the cooling energy use of 4%-13% as compared to the image conservation approach and a total increase of 13%-39% with regard to the original buildings (scenario 0), this is not so significant when looking at the whole picture because cooling only represents a 6%-8% of total emissions in scenario 0 thanks to the high thermal inertia, night ventilation, and solar protection of these types of buildings, which are maintained or improved in the subsequent scenarios.
Regarding thermal comfort, both strategies provide high levels of comfort, reaching 96%-100% of hours within the 90% acceptability.
The analysis also shows that the image coordination strategy is the only one that can ensure the fulfilment of the compulsory major renovation requirements recently modified in Spain, and can even transform existing social housing buildings into nZEBs by improving both the envelope and the systems.
For these reasons, the image coordination approach is the only strategy recommended in this paper, unless the boundary conditions were changed, for example by defining innovative solutions to facilitate merging households in condo buildings and using inside insulation solutions.
The buildings from this period with thick brick-bearing walls show a better energy performance before and after renovation than buildings with structures of pillars and the corresponding thinning of the façade, due to the higher thermal inertia of their façades making it easier and cheaper (up to 14%-62%) to achieve the major renovation and nZEB standards. Funding: This research was funded by the Ministry of Economy and Competitiveness of Spain, project "BIA2013-44001-R Integrated design protocol for social housing retrofit and urban regeneration"; and by FEDER 2014-2020, project "Construyendo Europa desde Aragón", research group T37_17R "Grupo de Investigación en Arquitectura".  Note: * Additional 5% of DHW (in total 65%).