Assessment of energy poverty in Andalusian municipalities: application of a combined indicator to detect priorities

Energy Reports 8 (2022) 5100–5116 Contents lists available at ScienceDirect Energy Reports journal homepage: www.elsevier.com/locate/egyr Research paper Assessment of energy poverty in Andalusian municipalities. Application of a combined indicator to detect priorities David Bienvenido-Huertasa, Ana Sanz Fernándezb, Carmen Sánchez-Guevara Sánchezb,∗, Carlos Rubio-Bellidoc aDepartment of Building Construction, University of Granada bSchool of Architecture (ETSAM), Universidad Politécnica de Madrid (UPM) cDepartment of Building Construction II, Higher Technical School of Building Engineering, University of Seville, Avd. Reina Mercedes 4A, 41012, Seville, Spain article info Article history: Received 24 September 2021 Received in revised form 17 January 2022 Accepted 8 March 2022 Available online xxxx Keywords: Energy poverty Income Gender gap Territorialisation Dwelling stock Energy requirements abstract This study aims to assess the energy poverty risk in the municipalities in Andalusia (Spain) to know their existing differences to prioritise the aids in the municipalities with greater needs. For this purpose, a composed indicator is developed. This new indicator is based on the so-called High Energy Requirements (HER) index, where proxy indicators related to the building energy efficiency are joint and where climate severity indicators, together with the income conditions and the gender gap of the population, are included. This new indicator assesses energy poverty globally, joining the indicators related to the characteristics of the dwelling stock with other factors related to energy poverty, such as monetary poverty or gender gap. The results are analysed at a municipal level, so the cases with greater problems both in general terms and those particularly from partial aspects (such as the conservation state or the type of climate) are identified. This characterisation of the phenomenon allows the most appropriate policies to be identified according to the indicators presenting the most unfavourable values, thus constituting a useful tool to prioritise and to implement measures adapted to each municipality’s needs. ©2022 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 1. Introduction The progress of today’s society is strongly related to energy consumption (Sorensen,2013), which is among the main aspects on which the policies of each country should be focused (Gatto and Drago,2021). Countries are more and more considering in their policies through energy poverty (EP) the relationship between energy and people’s welfare (Day et al.,2016). EP, also known as energy vulnerability (Castaño-Rosa et al.,2019), takes place when households are not capable of keeping acceptable thermal comfort conditions or having energy services at an acceptable price, and also when it is not possible to make essential activities because of the difficulty to access to energy (Bouzarovski and Petrova,2015). The European Union has included the importance of EP in the main challenges that the region should address in the next years (European Commission,2011). The creation of the European ∗Correspondence to: School of Architecture (ETSAM), Universidad Politécnica de Madrid. Avda. Juan de Herrera 4, 28040, Madrid, Spain. E-mail addresses: [email protected] (D. Bienvenido-Huertas), [email protected] (A. Sanz Fernández), [email protected] (C. Sánchez-Guevara Sánchez), [email protected] (C. Rubio-Bellido). Energy Poverty Observatory (EPOV) is useful for the member countries of the European Union to control and to reduce EP. In addition, European countries are increasingly developing policies and plans to reduce EP (Kyprianou et al.,2019). In Spain, the National Strategy against Energy Poverty 2019–2024 (The Government of Spain,2019) has established goals to reduce EP between 25 and 50% by 2025 in comparison with the current values. Thus, in the case of indicator 2M (disproportionate energy expenditure), the current value (17.3% of the population) should be reduced by between 12.9 and 8.6%. This strategy considers EP as the cases in which the family units are not capable of meeting the energy needs of their household due to their low income and the situations worsened by the deficient energy performance of dwellings, as well. Two aspects are therefore influencing EP: the energy efficiency of dwellings and the income of the family units. An essential aspect to develop and to apply policies to reduce EP is the way of quantifying it due to its interdisciplinary character (Thomson et al.,2017) that could vary among territories (Scarpellini et al.,2015). One of the main problems to develop these policies is to determine the regions of the country that present a greater EP risk, and therefore greater financing and more specific measures are required. This aspect can be done at different territorial scales. By way of example, neighbourhoods in https://doi.org/10.1016/j.egyr.2022.03.045 2352-4847/©2022 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 the same city may have different realities, with different levels of family income and the energy performance of buildings being common. Likewise, it is to be assumed that cities in the same region may have different characteristics. For these reasons, it is necessary to have methods for characterising energy poverty at the regional level. Some studies, which can be grouped at district and state levels, have addresses the possibility of analysing risk areas. In relation to the former, the studies by Martín-Consuegra et al. (2020,2019) developed a multidimensional index to assess EP in the underprivileged quarters in Madrid. A similar study was conducted by März (2018), who developed a combined methodology of multicriteria decision analysis with a geographic information system to assess the EP risk; that methodology was based on three dimensions (heating demand, building vulnerability, and socioeconomic aspects) in the quarters in the city of Oberhausen (Germany). However, its application in other regions could be something of a challenge as complex approaches with specific data of a region are used. Therefore, Reames (2016) assessed with public data the possibility of predicting the residential heating energy use intensity in census units in Kansas. Likewise, the aspects related to EP should be redefined. Sánchez-Guevara Sánchez et al. (2019) showed the need for considering variables related to high temperatures in summer to detect the EP risk at a district scale, particularly in cities located in warm climates. In another study, Sánchez-Guevara Sánchez et al. (2020a) stressed the importance of considering the gender perspective when assessing EP. Other regional studies have also assessed EP using various indicators. However, some works were not aimed to address EP in particular, such as Piai Paiva et al. (2019), who analysed the regions of Brazil with low-income population and the difficulties to access to energy. An approach more focused on EP are the studies by Walker et al. (2013,2012), who performed a regional analysis of EP in Northern Ireland. The results detected areas with policies slightly adapted to their level of energy requirement. In addition, Simoes et al. (2016) and Gouveia et al. (2019) assessed the EP in the cities of Portugal using a multidimensional indicator. Recently, Pérez-Fargallo et al. (2020) developed an approach to assess the EP in Chile; this approach was based on the energy demand of climate considering an adaptive approach of users and the population in situation of monetary poverty. These aspects were considered due to the similarities of the building stock in the various Chilean municipalities, but the authors doubted about the possibility of extrapolating the methodology to developed countries. Most research studies assessing EP are methodologies based on regional analyses or are focused on the individual analysis of municipalities. However, the way in which the most vulnerable municipalities stand out in comparison with the average values of the country is hardly included in the analysis. In addition, municipalities are not compared among themselves to identify those in a more unfavourable situation. If there is a methodology to assess the predominance of EP of one or several municipalities in a region, then the most vulnerable cases are known, and policies and financing programmes for these cases could be established (Morrison and Shortt,2008). A recent study by Sánchez-Guevara Sánchez et al. (2020b) developed the High Energy Requirements (HER) index to assess the districts in Madrid where the energy requirements were greater than the average values of the city. This index contributed to identify more quickly the districts with greater energy performance in their building stock. The main advantage of the indicator is that it allows the use of aggregated data, thus allowing adequate analysis between different areas. This aspect can allow a first decision-making when evaluating the areas most affected by energy poverty, although the subsequent analysis of individual cases (through existing indicators (Siksnelyte-Butkiene et al.,2021) allows obtaining a joint assessment of energy poverty. Although this approach was developed to be applied in the districts of a city, the possibilities of extrapolating the methodology to a regional context were not studied. This study analyses the possibility of application the HER index to a regional scale to assess the EP risk in the municipalities. Moreover, the HER index is focused on the analysis of the energy characteristics of the building stock, but without considering socioeconomic indicators. For this reason, the indicator is improved by including two socioeconomic indicators: population in monetary poverty and the existing gender gap. The latter was included because of the needs for including this dimension in the assessment of the EP at a regional level (Sánchez-Guevara Sánchez et al.,2020a). Although other social aspects can also influence the assessment of energy poverty (e.g., elderly population with low incomes (Chard and Walker,2016) or immigrants with low incomes (Fuller and McCauley,2016), the lack of quality data for these aspects makes their use difficult. In addition, this study also aims to broaden and to go more deeply into the regional analysis of the existing EP in one of the regions of Spain with greater predominance of EP: Andalusia (the Spanish region located to the south of the Iberian Peninsula). Although the values of state indicators, such as the number of inhabitants, unemployed people, and people in monetary poverty, are high in the region, the EP risk of this region has not been analysed in detail. Therefore, the importance of selecting Andalusia as a case study lies on two aspects: (i) it is the autonomous region with the greatest population of the country, divided into 778 municipalities; (ii) EP is estimated to be large in the region since it is one of the regions with the lowest income in Spain and with a deficient housing stock (Castaño-Rosa et al.,2020). Although Andalusia is located in a latitude with mild winters, the deficient building energy performance makes difficult to keep households adequately warm in winter, and is in turn combined with the severity of the summer conditions (Castaño-Rosa et al.,2020). Likewise, selecting Andalusia as a case study is of interest due to the situation taken place by the Covid-19 pandemic because its main economic activities are related to hotels, catering and tourism, and these activities have been therefore affected, thus reducing the number of employees and the income of the family units (The Government of Spain,2020); and (iii) the EP in the region has not been analysed. Although some studies have analysed certain buildings of the regions, such as Bienvenido-Huertas et al. (2020a,b), Castaño-Rosa et al. (2020), Sánchez-Guevara Sánchez et al. (2017), and Vilches et al. (2017), the complex character of its building stock and the most unfavourable regions have not been studied at a regional level in a similar way as the many studies focused on the city of Madrid (Martín-Consuegra et al., 2020,2019;Sánchez-Guevara Sánchez et al.,2020b,a). 2. Methodology 2.1. Study zone: Andalusia Andalusia is the autonomous region in the south of the Iberian Peninsula (Fig. 1), is divided into 8 provinces and has 778 municipalities, from which 29 exceed 50,000 inhabitants and 12 exceed 100,000 inhabitants. Most municipalities with more inhabitants are each province capital (e.g., Seville) and other important municipalities, such as Algeciras or San Fernando. The demographic distribution of the autonomous region is characterised by a high concentration of population in coastal zones and valleys, which is in turn characterised by the agri-cities with populations of up to 50,000 inhabitants. Likewise, there are regions with less inhabitants, such as the areas of Grazalema, Sierra Morena or 5101 D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 Fig. 1. Location of the autonomous region of Andalusia. Las Alpujarras. In terms of total population data, Andalusia is the autonomous region with the largest number of inhabitants in Spain (Fig. 2). The population data from the Spanish Institute of Statistics (2020a) shows that the number of inhabitants of the autonomous region is 9.62% greater than the second most populated autonomous region in the country (Catalonia). In demographic terms, selecting Andalusia as a case study is therefore significant because of its importance in the total number of inhabitants in Spain. However, this aspect becomes important when the data related to the economic situation of the region are analysed in comparison with the rest of the autonomous regions. Data of employment rate (Spanish Institute of Statistics,2020b) and the average annual income of the households (Spanish Institute of Statistics,2020c) of each autonomous region show that Andalusia is among the autonomous regions with the lowest values (Figs. 3 and 4). In the last 10 years, the employment rate has evolved from the year with the lowest value (2013) to the year with the greatest value (2019); however, Andalusia is among the autonomous regions with the worst values. Likewise, the rate of employees in Andalusia between 2010–2019 was 6.28% lower than the national average. The income of the family units is undoubtedly affected, thus implying that the values of the average national rate in Andalusia are lower than both the national average and most autonomous regions. Consequently, the EP risk in the region is high (Castaño-Rosa et al.,2020). The Spanish Institute of Statistics showed that the Andalusia has the highest population ratios in monetary poverty risk (13.1% greater than the national value in 2016 Spanish Institute of Statistics,2020d). Moreover, the report written by the Spanish Environmental Science Association indicates that the Andalusian population with income levels lower than the limit considered in 2014 was between 3 and 10% greater than the national average, thus constituting one of the autonomous regions with the most unfavourable values (Tirado Herrero et al.,2016). In demographic and economic terms, Andalusia is strongly affected by EP, so tools to assess it should be available. 2.2. Index developed to assess EP in municipalities One of the main challenges to establish policies to reduce EP is the economic distribution of the funds. The reason is that autonomous and provincial governments have funds for municipal governments to establish measures, such as the energy rehabilitation of buildings or the payment of energy bills. However, criteria for distributing funds among the municipalities are usually simple, and the reality of municipalities is not assessed. An example is the distribution of the funds of the Provincial Strategic Plan against Energy Poverty developed by the Cadiz Provincial Council in 2020 (Cadiz Provincial Council,2020), based on the number of inhabitants of each municipality (without considering the number of inhabitants in situation of monetary poverty or the state of the building stock). 2.2.1. Adoption of the pre-existing methodology: the HER index The lack of indicators to analyse EP at a regional or municipal level could contribute to the use of poor criteria to assess and to classify the distribution of funds and measures against the EP in the municipalities of a region. Pérez-Fargallo et al. (2020) indicated that many indicators to assess EP (focused on the individual analysis of the family units) make the use of indicators at a regional level something of a challenge. One of the main difficulties related to the analysis of the municipalities of a region is identifying their situation in relation to the economic poverty both in absolute terms (the influence of the phenomenon on the households located in the region) and in relative terms (in comparison with the municipalities with similar characteristics and belonging to the same region). From a simple perspective of the problem, EP is due to both the combination of high energy consumption (mainly due to a poor building energy performance) and the low income of the family units, so it is possible to establish indicators that analyse these aspects at a municipal level. Sánchez-Guevara Sánchez et al. (2020b) developed an indicator to assess the EP of the districts in Madrid: the HER index. This index was designed to assess the high risk in terms of energy requirement presented by the building stock of each district in Madrid in comparison with the average values of the city. The quarters with a greater risk were therefore detected. Five indirect indicators (construction year, maintenance conditions, availability of heating and cooling systems (differentiating those using electric heating systems because its use is more expensive than that of other heating systems), surface of the dwelling by household, and average temperature of the summer nights) of the building stock of each district were compared with the average values of Madrid, and a value was assigned to each indirect indicator according to the severity degree. Therefore, it could be quantitatively assessed the district that stands out in comparison with the city. Eqs. (1) and (2) show how the HER index used by Sánchez-Guevara Sánchez et al. (2020b) was determined. The rules established in Table 1 were applied to the five indirect indicators (Va,Vb,Vc,Vdand Ve) to determine the value assigned to each. A value of 10 in HER corresponded to the most unfavourable assumption, and a value of 0 corresponded to the less unfavourable assumption in comparison with the average data of the city. Regarding the importance of each indicator, Sánchez-Guevara Sánchez et al. (2020b) justified the importance of each in comparison with the characteristics of the dwelling stock of the households in EP in Madrid: (i) Va considered the buildings built before 1980 as these buildings were built without any requirement on energy efficiency (the first technical standard was applied from 1980 (The Government of Spain,1979)); (ii) Vbconsidered buildings with bad maintenance conditions as their energy performance is usually worse (e.g., the presence of damages in the façade or interstitial condensations contribute to a greater heat transfer (Litti et al.,2015;Rotilio et al.,2018); (iii) Vcwas determined by combining the other three subindicators: percentage of buildings without heating (V1), percentage of buildings without cooling (V2), and percentage of buildings with electric heating (V3). Considering buildings without heating or cooling systems could imply that family units use inefficient equipment (e.g., portable radiators) and face more thermal discomfort hours, as well. Buildings with electric heating were considered because the electrical energy is more expensive than gas, thus increasing the energy bills of the family units living in buildings with these heating systems; (iv) Vdconsidered the dwellings with a greater surface by member of the household as 5102 D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 Fig. 2. Number of inhabitants in Spain by autonomous regions. Average data of the period between 2010 and 2019. Source: Data obtained from the Spanish Institute of Statistics (2020a). Fig. 3. Evolution of the employment rate in Spain by autonomous regions in the last 10 years. Source: Data obtained from the Spanish Institute of Statistics (2020b). Fig. 4. Evolution of the annual net income of the Spanish households in each autonomous region in the last 10 years. Source: Data obtained from the Spanish Institute of Statistics (2020c). it could imply a greater energy expenditure for the family units due to the need of acclimatising a greater volume of air; and (v) Veconsidered the influence of the average temperature of the summer nights to assess the thermal recovery of the family units during the night, and it is also a possible indicator of the energy expenditure in summer. The first idea of Vewas the analysis of the impact of the heat island that reduces the passive cooling capacity of dwellings in summer, although the analysis was based on the average temperature of the summer nights. As a result, the indicator could be applied in other regions where there are no specific heat island data. HER =Va+Vb+Vc+Vd+Ve(1) Vc=V1+V2+V3 3(2) where Vais the indicator of the percentage of buildings built before 1980; Vbis the indicator of the percentage of buildings with a deficient conservation state; Vcis the indicator of the percentage of buildings with HVAC systems and determined by Eq. (2);Vdis the indicator of the average surface of the dwelling by household member; Veis the indicator of the average temperature of summer nights; and V1,V2and V3are the indicators used to assess the percentage of buildings with HVAC systems and their values are also determined with Table 1. Although the HER index was used to assess which districts of Madrid have a building stock with greater energy requirement, it could be used at a municipal level to carry out a regional comparative analysis as it is a methodology that aims to perform an analysis with greater disintegration and to compare territories among them. For this purpose, the approach by Sánchez-Guevara Sánchez et al. (2020b) was modified. Details are given below. 5103 D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 Table 1 Degree of severity and value assigned to each indicator in the classification of districts by Sánchez-Guevara Sánchez et al. (2020b). VeRest of indicators (Va,Vb,Vd,V1,V2and V3) Relation Degree of severity Assigned value Relation Degree of severity Assigned value DV −CA ≤0 Very Low 0.0 DV ≤CA Very Low 0.0 0<DV −CA ≤2 Low 1.0 CA <DV ≤1.5·CA Low 1.0 2<DV −CA ≤4 Medium 1.5 1.5·CA <DV ≤2·CA Medium 1.5 DV −CA >4 High 2.0 DV >2·CA High 2.0 DV: District Values; CA: City Average. 2.2.2. Modification of the HER index The HER was modified by including two factors related to the climate characteristics and socioeconomic factors. 2.2.2.1. Inclusion of the climate variability: HDH and CDH. As mentioned above, five indirect indicators related to the characteristics of the building stock were used, and only the indicator of the average temperature of the summer nights considered the influence of the climate characteristics of the area analysed. The reason was that the climate zone of Madrid is the same in all districts, but in the case of a region or autonomous region there could be various climate zones. For this reason, indirect indicators should be included to assess the greatest heating or cooling demand that a municipality could present in comparison with the average of the region. For this purpose, two indirect indicators were designed, based on the approach of heating degree hours (HDH) and cooling degree hours (CDH) developed by the Poverty-Adaptive Degree Hourly Index (Pérez-Fargallo et al.,2020). The advantage of using HDH and CDH instead of the degree days is that the actual building behaviour could be better known according to climate (Pérez-Fargallo et al.,2020). These degrees hours are determined by considering an adaptive-based temperature according to an adaptive thermal comfort standard (Bienvenido-Huertas et al., 2021,2020a). Adaptive thermal comfort models consider the possibility of people’s adaptation when the external temperature varies and are related to lower energy consumption as the use of HVAC systems is reduced. The use of this approach is appropriate for residential buildings. The surveys by Bienvenido-Huertas et al. (2020b) showed that most family units limit the use of HVAC systems in the hours of the greatest severity. For this reason, the use of an adaptive-based approach to determine the degree hours is appropriate to assess the energy demand presented by residential buildings. The EN 16798-1:2019 standard (European Committee for Standardization,2019) was used in this study. This standard establishes upper and lower limits among which the operative temperature should oscillate. To determine these limits, the running mean outdoor temperature (trm) (Eq. (3)) that determines the outdoor mean temperature of the days before that analysed is previously calculated. Upper and lower limits are determined with trm and using the linear correlations established in EN 16798-1:2019. To determine these limits, EN 16798-1:2019 considers several categories that vary the thermal differential between the upper and lower limit. The use of each category is established in EN 16798-1:2019, with Category III corresponding to the existing buildings. Therefore, the upper and lower limits from Category III were used (Eqs. (4) and (5)). With the values of these limits, HDH and CDH are determined by Eqs. (6) and (7). trm =(1−α)· n ∑ d=1 (α(i−1)·Text,d)(3) Lower limit =0.33 ·trm +13.8 (10≤trm≤30) (4) Upper limit =0.33 ·trm +22.8 (10≤trm≤30) (5) HDH = 8760 ∑ i=1 (Text,i−TAH,i)·XHA XHA =1if Text,i<TAH,i XHA =0if Text,i≥TAH,i (6) CDH = 8760 ∑ i=1 (TAC,i−Text,i)·XCA XCA =1if Text,i>TAC,i XCA =0if Text,i≤TAC,i (7) where Text,dis the daily outdoor mean temperature [◦C]; TAH,iis the daily adaptive-based temperature of the lower limit determined by Eq. (4) [◦C]; TAC,iis the daily adaptive-based temperature of the upper limit determined by Eq. (5) [◦C]; and XHA,XCA are logic values whose value is 1 if the condition given in the equations is met, and 0 if not. If HDH and CDH are determined, the new HER developed in this study could be established to assess the energy requirement of the municipalities in comparison with the average values of the autonomous region. Eq. (8) shows the indicator modified to be used in municipalities. The values assigned to each indicator are obtained from Table 2, which is adapted from the values assigned by Sánchez-Guevara Sánchez et al. (2020b) at a district level. In this case, the values of the municipalities are compared with the average values of the autonomous region. As the new HER could obtain a maximum value of 14 in the most unfavourable assumption, the indicator was normalised to use a result scale between 0 and 1. For this purpose, a min/max normalisation was used (Eq. (9)), thus developing the Normalised High Energy Requirements (NHER). HER =Va+Vb+Vc+Vd+Ve+Vf+Vg(8) NHER =HER −min(HER) max (HER)−min(HER)(9) where Vfis the value assigned to the indicator of the heating degree hours considering an adaptive-based temperature obtained by Category III from EN 16798-1:2019; and Vgis the value assigned to the indicator of the cooling degree hours considering an adaptive-based temperature obtained by Category III from EN 16798-1:2019. 2.2.2.2. Inclusion of socioeconomic aspects: income and gender. Although NHER is an appropriate indicator to assess the energy requirements of buildings, socioeconomic aspects are not considered. Sánchez-Guevara Sánchez et al. (2020b) analysed HER and the population in EP independently. However, if these two aspects are combined in an indicator, the energy requirements of the building stock and the socioeconomic requirements are known. In view of this circumstance, the normalised high energy requirements with population in monetary poverty and gender gap (NHERP,GG) index was designed. This indicator is obtained by correcting NHER according to both an indirect indicator of population in monetary poverty (P) and an indirect indicator of 5104 D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 Table 2 Degree of severity and value assigned to each indicator to be assessed at a municipal level. VeRest of indicators (Va,Vb,Vd,Vf,Vg,V1,V2and V3) Relation Degree of severity Assigned value Relation Degree of severity Assigned value MV −RA ≤0 Very Low 0.0 MV ≤RA Very Low 0.0 0<MV −RA ≤2 Low 1.0 RA <MV ≤1.5·RA Low 1.0 2<MV −RA ≤4 Medium 1.5 1.5·RA <MV ≤2·RA Medium 1.5 MV −RA >4 High 2.0 MV >2·RA High 2.0 MV: Municipality Values; RA: Region Average. the proportion of the female population in a situation of monetary poverty in comparison with the male population (GG). Eq. (10) shows the formulation to determine NHERP,GG. To consider the population in monetary poverty, the criterion established by Eurostat is used: this criterion is based on the fact that monetary poverty takes place when the net income of the family unit is lower than 60% of the median of income at a national level. In addition, the gender gap was considered because of the results obtained by Sánchez-Guevara Sánchez et al. (2020a) in their study on the feminisation of EP in Madrid. The results showed the significant gender gap in the households in EP in Madrid and suggested the need for establishing policies considering the gender perspective in EP. The two indirect indicators to weight the social perspective of EP modify the value obtained with NHER. For this purpose, the weighting values to be used should be established similarly to the indicators of the NHER. Data of the Andalusian population in monetary poverty (Fig. 5) and data of gender gap (Fig. 6) were collected. The data distribution shows that the interquartile range of the distribution is low, thus meaning a high concentration of data in the same values of population in monetary poverty or values of gender gap. Regarding the data of population in monetary poverty, however, interesting values are found in the municipalities located above the third quartile of the distribution. In these cases, the number of inhabitants in monetary poverty greatly oscillated, from 2786 to 171,669 inhabitants. For this reason, weightings were applied according to the location of the municipality with respect to the percentiles of the data distribution (Table 3). As the number of inhabitants in monetary poverty located below the first quartile corresponded to very low values (lower than 482 inhabitants), another correction was applied to these cases to reduce the severity of the EP of the municipality. In the gender gap, there was a distribution like that of the population in monetary poverty. Therefore, the distribution presented a low interquartile range, with low values of gender gap. From the 95th percentile, which corresponds to a gender gap of 361 inhabitants, significant values were found in the Andalusian municipalities. In this interval greater than the 95th percentile, the most significant cases of gender gap in the Andalusian municipality were concentrated. In view of this circumstance, a weighting was applied according to the rules established in Table 4. To compare the effect of the socioeconomic indicators of NHERP,GG in comparison with NHER, a same measurement scale should be established. As mentioned above, NHER is normalised between 0 and 1. Therefore, NHERP,GG should have the same value scale. For this purpose, NHERP,GG was limited to obtain a maximum value of 1 if the multiplication of NHER with the socioeconomic indicators obtains a greater value (Eq. (11)). NHERP,GG =NHER ·P·GG (10) NHERP,GG =1if NHER ·P·GG >1 (11) 2.3. Data collection To determine both NHER and NHERP,GG, data related to each indirect indicator were collected (e.g., percentage of buildings built before 1980). Different data sources were used according to Table 3 Degree of severity and value assigned to P. Relation Value MV >99th percentile 4.0 95th percentile <MV ≤99th percentile 2.0 75th percentile <MV ≤95th percentile 1.5 25th percentile ≤MV ≤75th percentile 1.0 MV <25th percentile 0.5 Table 4 Degree of severity and value assigned to GG. Relation Value MV >95th percentile 1.5 MV ≤95th percentile 1.0 the type of indirect indicator. Regarding the indicators related to the building stock and social aspects, data published by the Spanish Institute of Statistics were collected. Table 5 shows the data sources used for each indicator. As for the socioeconomic data, the statistical data sources provide data in both absolute and percentage value. Therefore, this study considered two approaches to analyse these indirect socioeconomic indicators: approach 1 for the number of inhabitants, and approach 2 for the percentages of inhabitants. Regarding the climate indicators, data were obtained through METEONORM, a software constituted by a database of climate files that is composed by 8325 weather stations distributed throughout the planet. This software has been widely used because it easily obtains climate data (Bellia et al.,2015;Hatwaambo et al.,2009;Kameni et al.,2019;Osman and Sevinc, 2019). From the data of the weather stations, the hourly values of external temperature in any location of the planet are obtained by interpolations (METEONORM,2019). The hourly data of the external temperature were obtained with METEONORM in the 778 Andalusian municipalities. Data generated by METEONORM were treated to obtain the average temperature values of the summer nights, HDH and CDH. The procedure to generate the dataset of the 778 Andalusian municipalities could be extrapolated to other regions in Spain, and even used in other countries. Obtaining data of the building stock through the statistical institutes of each country guarantees an easier workflow to obtain data than that based on an individual analysis performed by technicians and professionals of the existing buildings in each municipality. In this sense, several studies have highlighted the advantages of using this type of data (Camboni et al.,2021;Pérez-Fargallo et al.,2020). Likewise, climate data usually constitute one of the main limitations to determine the hourly degrees of a region because of the difficulties to obtain hourly data of the various climates of each municipality. For this reason, the use of METEONORM to generate hourly data implies that hourly climate data could be obtained in any place of the planet, and therefore the extrapolation of the methodology to other countries is guaranteed. 5105 D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 Fig. 5. Histogram of the population by municipalities in monetary poverty in 2017. Source: Data obtained from the Spanish Institute of Statistics (2019). Fig. 6. Histogram of the difference between the female and male population in monetary poverty by municipalities in 2017. Source: Data obtained from the Spanish Institute of Statistics (2019). Table 5 Data source by indirect indicator analysed. Indicator Variable Unit Reference VaPercentage of buildings built before 1980 % Population and housing census in Spain (Spanish Institute of Statistics,2011) VbPercentage of buildings with dilapidated, bad or deficient maintenance conditions % Population and housing census in Spain (Spanish Institute of Statistics,2011) V1Percentage of dwellings without heating systems % Population and housing census in Spain (Spanish Institute of Statistics,2011) V2Percentage of dwellings without cooling systems % Population and housing census in Spain (Spanish Institute of Statistics,2001) V3Percentage of dwellings with electric heating systems % Population and housing census in Spain (Spanish Institute of Statistics,2001) VdDwelling surface by household member m2Population and housing census in Spain (Spanish Institute of Statistics,2011) PPopulation in monetary poverty No. of people, % Distribution atlas of the household income (Spanish Institute of Statistics,2019) GG Gender gap between the female and the male population in monetary poverty No. of people, % Distribution atlas of the household income (Spanish Institute of Statistics,2019) 5106 D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 Fig. 7. Histograms of the data obtained in Andalusia in each variable considered by NHER. 3. Results and discussion 3.1. Analysis of the municipalities with greater EP risk due to building stock and climate indicators Firstly, the results obtained with NHER were analysed. As mentioned above, NHER is designed to analyse the EP risk presented by municipalities according to the characteristics of their building stock. To determine the EP of municipalities, several indirect indicators were analysed. The assessment of NHER implies to know the requirements of different aspects in a combined way, the individual analysis of each indirect indicator could be interesting to know more in detail the building stock of the Andalusian municipalities. Although the data used to generate the NHER are partially obsolete (the census, which should be updated, is dated from 2011), they were the best source available at a state level (thus leading to a replicability) and at a desegregation level lower than the regional level. As this is a comparative study among municipalities, this last characteristic of the data is also relevant as this source is useful for establishing magnitude orders that allow that comparison to be carried out. Fig. 7 shows the histograms of the data obtained for each indicator in the 778 municipalities. In the buildings built before 1980, most municipalities had high percentages in their building stock: the first quartile (Q1) was placed in 45.5% and the third quartile (Q3) in 67.8%, recording a maximum value of 97.9%. These values were consistent with the broad development of the building stock in the period posterior to the Spanish Civil War (Domínguez-Amarillo et al.,2016;Kurtz et al.,2015). Thus, a high percentage of the buildings of the Andalusian municipalities were built before 1980, and only some municipalities had percentage values lower than 45%. However, this was not correlated to the buildings with bad maintenance conditions, as the percentage of buildings in that state were not very high. The reason is that, although it is assumed that most of the building stock previous to 1980 has deficiencies in its thermal behaviour (due to the lack of insulation as the first standard regulating it, NBE-CT-79, was not in force yet), most buildings should not necessarily have a conservation state with deficiencies. Most data were concentrated between 4.5 and 7.7%, although percentages greater than 25% were found in some municipalities. Regarding the availability of HVAC systems in dwellings, the results were diverse. In general terms, there was a greater percentage of dwellings with heating systems rather than with cooling systems. Likewise, dwellings with electric heating systems presented many percentage distributions in the building stock of the municipalities, detecting municipalities with values close to 100%, whereas other municipalities obtained values of 0.4%. The values related to the mean surface area of the dwelling by household member showed that most dwellings had average values between 36.8 and 46.17 m2, mainly due to the fact that 5107 D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 Fig. 8. Spatial distribution of the values obtained in each variable of NHER:Va(buildings built before 1980); Vb(buildings with dilapidated, bad or deficient maintenance conditions); Vc(buildings with insufficient or inefficient heating and cooling systems); Vd(dwelling surface by household member); Ve(mean summer night temperature); Vf(hourly heating degrees), and Vg(hourly cooling degrees). the average surface of dwellings and the size of the family units are similar among the Andalusian municipalities, although sometimes the values of the surface were greater than the average, obtaining values of 69.86 m2. Finally, regarding the climate severity in Andalusia, the distribution of the degree hours presented various peaks, mainly due to the greater concentration of municipalities in geographic points with very similar characteristics of winter climate severity (e.g., the municipalities located in the Baetic Systems), whereas in summer the temperatures of cooling degree hours presented a greater concentration in the low values of the distribution because of the coastal municipalities. Likewise, the mean summer night temperature was high. Therefore, the Andalusian municipalities generally have a building stock with characteristics that could lead to a high energy consumption in dwellings. Financing policies and measures should be established according to a criterion of classification and priorities, so the Andalusian municipalities with negative values in comparison with the average values of the autonomous region should be analysed. For this reason, the distribution of the indirect indicators in the autonomous region was studied before analysing NHER (Fig. 8). The spatial distribution showed various aspects related to the building stock. When Va(buildings built before 1980) or Vd(dwelling surface by household member) had values in a high percentage of the Andalusian municipalities, the distribution in other indicators were different and geographic patterns were detected. As for Vb, there was a greater concentration of municipalities in the eastern part of the autonomous region with values greater than the Andalusian average. This could reflect the need for establishing specific plans to maintain and to repair the building stock of this region. As for the climate indicators (Ve,Vf, and Vg), the relations between the values obtained and the Baetic Systems were shown. Therefore, the municipalities at the greatest altitude of the autonomous region, such as those located in Sierra Morena or in the Penibaetic or Subbaetic System, obtained the greatest values of Vf(hourly heating degrees), and regarding Vg(hourly cooling degrees), the greatest values were obtained in the Baetic Depression (also known as the Guadalquivir Depression). Finally, as for Ve(mean summer night temperature), values equal to or greater than 1 were obtained, except in the municipalities at the greatest altitude, mainly located in the Penibaetic or Subbaetic System and some located in the Sierra de Cadiz. This spatial distribution of the indirect indicators showed that the characteristics of the Andalusian building stock are different, so establishing political intervention criteria could be something of a challenge. Although some aspects were clear, such as the need for reducing the heating loads in the municipalities at the greatest altitude or the cooling loads in the area of the Guadalquivir Depression, the many aspects interrelated to the energy performance of the building stock of the municipalities made difficult to establish a classification criterion. As a result, NHER allowed all these indicators to be combined. Fig. 9 shows the spatial distribution of NHER in the Andalusian municipalities. As mentioned in Section 2,NHER had values between 0 and 1. A total of 10 classification groups could be established by intervals of 0.1. The Andalusian municipalities obtained values of NHER between 0 and 0.6. For the methodology designed for NHER, no municipality therefore met the most unfavourable combination in all the indirect indicators. A total of 21 municipalities obtained the most unfavourable values of NHER, with the set of municipalities with values of NHER between 0.3 and 0.4 being the greatest (Fig. 10). After analysing the distribution of the indirect indicators in the groups of NHER (Fig. 11), the indicators most influencing the most unfavourable groups (G05 and G06) were detected. In these cases, the combinations of high values of buildings built before 1980, deficient buildings, average surface and climate variables were strongly related to the most unfavourable groups. Regarding the demographic characteristics of the municipalities with the greatest risk, the classifications of NHER did not correspond to the average economic level of the households by municipality. Table 6 shows the distributions of population in monetary poverty and gender gap in the groups of NHER. The groups did not consider socioeconomic factors greatly related to EP. The most outstanding aspect was the population in monetary 5108 D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 Bienvenido-Huertas, D., Pulido-Arcas, J.A., Rubio-Bellido, C., Pérez-Fargallo, A., 2021. Feasibility of adaptive thermal comfort for energy savings in cooling and heating: A study on europe and the mediterranean basin. Urban Clim. 36, http://dx.doi.org/10.1016/j.uclim.2021.100807. Bienvenido-Huertas, D., Rubio-Bellido, C., Pérez-Fargallo, A., Pulido-Arcas, J.A., 2020a. Energy saving potential in current and future world built environments based on the adaptive comfort approach. J. Clean. Prod. 249, http: //dx.doi.org/10.1016/j.jclepro.2019.119306. Bienvenido-Huertas, D., Sánchez-García, D., Rubio-Bellido, C., 2020b. Analysing natural ventilation to reduce the cooling energy consumption and the fuel poverty of social dwellings in coastal zones. Appl. Energy 279, http://dx.doi. org/10.1016/j.apenergy.2020.115845. Bouzarovski, S., Petrova, S., 2015. A global perspective on domestic energy deprivation: Overcoming the energy poverty–fuel poverty binary. Energy Res. Soc. Sci. 10, 31–40. http://dx.doi.org/10.1016/j.erss.2015.06.007. Cadiz Provincial Council, 2020. Provincial strategic plan against energy poverty. Camboni, R., Corsini, A., Miniaci, R., Valbonesi, P., 2021. Mapping fuel poverty risk at the municipal level. a small-scale analysis of Italian energy performance certificate, census and survey data. Energy Policy 155, 112324. http://dx.doi. org/10.1016/j.enpol.2021.112324. Castaño-Rosa, R., Solís-Guzmán, J., Marrero, M., 2020. Energy poverty goes south? Understanding the costs of energy poverty with the index of vulnerable homes in Spain. Energy Res. Soc. Sci 60, 101325. http://dx.doi.org/10.1016/j. erss.2019.101325. Castaño-Rosa, R., Solís-Guzmán, J., Rubio-Bellido, C., Marrero, M., 2019. Towards a multiple-indicator approach to energy poverty in the European union: A review. Energy Build. 193, 36–48. http://dx.doi.org/10.1016/j.enbuild.2019.03. 039. Chard, R., Walker, G., 2016. Living with fuel poverty in older age: Coping strategies and their problematic implications. Energy Res. Soc. Sci. 18, 62–70. http://dx.doi.org/10.1016/j.erss.2016.03.004. Day, R., Walker, G., Simcock, N., 2016. Conceptualising energy use and energy poverty using a capabilities framework. Energy Policy 93, 255–264. http: //dx.doi.org/10.1016/j.enpol.2016.03.019. Domínguez-Amarillo, S., Sendra, J.J., Oteiza, I., 2016. La envolvente térmica de la vivienda social. El caso de sevilla, 1939 a 1979. Editorial CSIC: Madrid. Dubois, G., Sovacool, B., Aall, C., Nilsson, M., Barbier, C., Herrmann, A., Bruyère, S., Andersson, C., Skold, B., Nadaud, F., Dorner, F., Moberg, K.R., Ceron, J.P., Fischer, H., Amelung, D., Baltruszewicz, M., Fischer, J., Benevise, F., Louis, V.R., Sauerborn, R., 2019. It starts at home? Climate policies targeting household consumption and behavioral decisions are key to low-carbon futures. Energy Res. Soc. Sci. 52, 144–158. http://dx.doi.org/10.1016/j.erss.2019.02.001. European Commission, 2011. A roadmap for moving to a competitive low carbon economy in 2050. Brussels, Belgium. European Committee for Standardization, 2019. EN 16798-1:2019 energy performance of buildings - ventilation for buildings - part 1: Indoor environmental input parameters for design and assessment of energy performance of buildings addressing indoor air quality, thermal environment, lighting and acous. Fuller, S., McCauley, D., 2016. Framing energy justice: Perspectives from activism and advocacy. Energy Res. Soc. Sci. 11, 1–8. http://dx.doi.org/10.1016/j.erss. 2015.08.004. Gatto, A., Drago, C., 2021. When renewable energy, empowerment, and entrepreneurship connect: Measuring energy policy effectiveness in 230 countries. Energy Res. Soc. Sci. 78, 101977. http://dx.doi.org/10.1016/j.erss. 2021.101977. Gouveia, J.P., Palma, P., Simoes, S.G., 2019. Energy poverty vulnerability index: A multidimensional tool to identify hotspots for local action. Energy Rep. 5, 187–201. http://dx.doi.org/10.1016/J.EGYR.2018.12.004. Hatwaambo, S., Jain, P.C., Perers, B., Karlsson, B., 2009. Projected beam irradiation at low latitudes using meteonorm database. Renew. Energy 34, 1394–1398. http://dx.doi.org/10.1016/j.renene.2008.09.011. Kameni, M., Yvon, A., Kalameu, O., Asadi, S., Choudhary, R., Reiter, S., 2019. Impact of climate change on demands for heating and cooling energy in hospitals, : An in-depth case study of six islands located in the Indian ocean region. Sustain. Cities Soc. 44, 629–645. http://dx.doi.org/10.1016/j.scs.2018. 10.031. Kurtz, F., Monzón, M., López-Mesa, B., 2015. Energy and acoustics related obsolescence of social housing of Spain’s post-war in less favoured urban areas. The case of zaragoza. Inf. la construcción 67, m021. http://dx.doi.org/ 10.3989/ic.14.062. Kyprianou, I., Serghides, D.K., Varo, A., Gouveia, J.P., Kopeva, D., Murauskaite, L., 2019. Energy poverty policies and measures in 5 EU countries: A comparative study. Energy Build. 196, 46–60. http://dx.doi.org/10.1016/j.enbuild.2019.05. 003. Litti, G., Khoshdel, S., Audenaert, A., Braet, J., 2015. Hygrothermal performance evaluation of traditional brick masonry in historic buildings. Energy Build. 105, 393–411. http://dx.doi.org/10.1016/j.enbuild.2015.07.049. Martín-Consuegra, F., Gómez Giménez, J.M., Alonso, C., Hernández, R.Córdoba., Aja, A.Hernández., Oteiza, I., 2020. Multidimensional index of fuel poverty in deprived neighbourhoods. Case study of madrid. Energy Build. 224, http: //dx.doi.org/10.1016/j.enbuild.2020.110205. Martín-Consuegra, F., Hernández-Aja, A., Oteiza, I., Alonso, C., 2019. Distribución de la pobreza energética en la ciudad de madrid (españa). Eure 45, 133–152. http://dx.doi.org/10.4067/S0250-71612019000200133. März, S., 2018. Assessing the fuel poverty vulnerability of urban neighbourhoods using a spatial multi-criteria decision analysis for the german city of oberhausen. Renew. Sustain. Energy Rev. 82, 1701–1711. http://dx.doi.org/ 10.1016/J.RSER.2017.07.006. Mastropietro, P., 2019. Who should pay to support renewable electricity? Exploring regressive impacts, energy poverty and tariff equity. Energy Res. Soc. Sci. 56, 101222. http://dx.doi.org/10.1016/j.erss.2019.101222. METEONORM, 2019. Handbook Part II: Theory (Version 7.3.1). Bern, Switzerland. Morrison, C., Shortt, N., 2008. Fuel poverty in Scotland: Refining spatial resolution in the scottish fuel poverty indicator using a GIS-based multiple risk index. Health Place 14, 702–717. http://dx.doi.org/10.1016/j.healthplace. 2007.11.003. Osman, M.M., Sevinc, H., 2019. Adaptation of climate-responsive building design strategies and resilience to climate change in the hot/arid region of Khartoum. Sudan. Sustain. Cities Soc. 47, 101429. http://dx.doi.org/10.1016/j.scs. 2019.101429. Pérez-Fargallo, A., Bienvenido-Huertas, D., Rubio-Bellido, C., Trebilcock, M., 2020. Energy poverty risk mapping methodology considering the user’s thermal adaptability: The case of Chile. Energy Sustain. Dev. 58, 63–77. http://dx.doi. org/10.1016/j.esd.2020.07.009. Piai Paiva, J.C., Jannuzzi, G.D.M., de Melo, C.A., 2019. Mapping electricity affordability in Brazil. Util. Policy 59, 100926. http://dx.doi.org/10.1016/J.JUP.2019. 100926. Reames, T.G., 2016. Targeting energy justice: exploring spatial, racial/ethnic and socioeconomic disparities in urban residential heating energy efficiency. Energy Policy 97, 549–558. http://dx.doi.org/10.1016/j.enpol.2016.07.048. Rotilio, M., Cucchiella, F., Berardinis, P.De., Stornelli, V., 2018. Thermal transmittance measurements of the historical masonries: Some case studies. Energies 11 (2987), http://dx.doi.org/10.3390/en11112987. Sánchez-Guevara Sánchez, C., Mavrogianni, A., Neila González, F.J., 2017. On the minimal thermal habitability conditions in low income dwellings in Spain for a new definition of fuel poverty. Build. Environ. 114, 344–356. http://dx.doi.org/10.1016/j.buildenv.2016.12.029. Sánchez-Guevara Sánchez, C., Núñez Peiró, M., Taylor, J., Mavrogianni, A., Neila González, J., 2019. Assessing population vulnerability towards summer energy poverty: Case studies of madrid and London. Energy Build. 190, 132–143. http://dx.doi.org/10.1016/j.enbuild.2019.02.024. Sánchez-Guevara Sánchez, C., Sanz Fernández, A., Núñez Peiró, M., Gómez Muñoz, G., 2020a. Feminisation of energy poverty in the city of madrid. Energy Build. 223, http://dx.doi.org/10.1016/j.enbuild.2020.110157. Sánchez-Guevara Sánchez, C., Sanz Fernández, A., Núñez Peiró, M., Gómez Muñoz, G., 2020b. Energy poverty in madrid: Data exploitation at the city and district level. Energy Policy 144, http://dx.doi.org/10.1016/j.enpol.2020. 111653. Scarpellini, S., Rivera-Torres, P., Suárez-Perales, I., Aranda-Usón, A., 2015. Analysis of energy poverty intensity from the perspective of the regional administration: Empirical evidence from households in southern Europe. Energy Policy 86, 729–738. http://dx.doi.org/10.1016/j.enpol.2015.08.009. Siksnelyte-Butkiene, I., Streimikiene, D., Lekavicius, V., Balezentis, T., 2021. Energy poverty indicators: A systematic literature review and comprehensive analysis of integrity. Sustain. Cities Soc. 67, http://dx.doi.org/10.1016/j.scs. 2021.102756. Simoes, S.G., Gregório, V., Seixas, J., 2016. Mapping fuel poverty in Portugal. Energy Procedia 106, 155–165. http://dx.doi.org/10.1016/J.EGYPRO.2016.12. 112. Sorensen, B., 2013. A history of energy: Northern Europe from the stone age to the present day. Routledge. Sorrell, S., 2007. Improving the evidence base for energy policy: The role of systematic reviews. Energy Policy 35, 1858–1871. http://dx.doi.org/10.1016/ j.enpol.2006.06.008. Sorrell, S., 2015. Reducing energy demand: A review of issues, challenges and approaches. Renew. Sustain. Energy Rev. 47, 74–82. http://dx.doi.org/10. 1016/j.rser.2015.03.002. Spanish Institute of Statistics, 2001. Census of population and housing (2001). Spanish Institute of Statistics, 2011. Census of population and housing (2011) [www document]. URL https://www.ine.es/censos2011_datos/cen11_datos_ resultados.htm. Spanish Institute of Statistics, 2019. Atlas of distribution of household income [www document]. URL https://www.ine.es/experimental/experimental.htm. 5115 D. Bienvenido-Huertas, A. Sanz Fernández, C. Sánchez-Guevara Sánchez et al. Energy Reports 8 (2022) 5100–5116 Spanish Institute of Statistics, 2020a. Official population figures from the revision of the municipal register. [www document]. URL https://www.ine.es/jaxiT3/Tabla.htm?t=2915. Spanish Institute of Statistics, 2020b. Employment rates by autonomous community [www document]. URL https://www.ine.es/jaxiT3/Tabla.htm?t=4942 & L=0. Spanish Institute of Statistics, 2020c. Income per household by autonomous communities [www document]. URL https://www.ine.es/jaxiT3/Tabla.htm?t=9949. Spanish Institute of Statistics, 2020d. Poverty risk rate by autonomous communities [www document]. URL https://www.ine.es/jaxiT3/Tabla.htm?t=9963. The Government of Spain, 1979. Royal decree 2429/79. Approving the basic building norm NBE-CT-79, about the thermal conditions in buildings. Normas básicas de la edificación. The Government of Spain, 2019. National strategy against energy poverty 2019–2024 (Spain). The Government of Spain, 2020. Employment regulation statistics advance. 2020–2020 [www document]. URL http://www.mites.gob.es/estadisticas/Reg/ reg20ene_ago/reg_08_2020.pdf. Thomson, H., Bouzarovski, S., Snell, C., 2017. Rethinking the measurement of energy poverty in europe: A critical analysis of indicators and data. Indoor Built. Environ. 26, 879–901. http://dx.doi.org/10.1177/1420326X17699260. Tirado Herrero, S., Jiménez Meneses, L., López Fernández, J.L., Perero Van Hove, E., Irigoyen Hidalgo, V.M., Savary, P., 2016. Pobreza, vulnerabilidad y desigualdad energética. Nuevos enfoques de análisis. Asociación de ciencias ambientales. Madrid. Ürge-Vorsatz, D., Tirado Herrero, S., 2012. Building synergies between climate change mitigation and energy poverty alleviation. Energy Policy 49, 83–90. http://dx.doi.org/10.1016/j.enpol.2011.11.093. Vilches, A., Barrios Padura, Á., Molina Huelva, M., 2017. Retrofitting of homes for people in fuel poverty: Approach based on household thermal comfort. Energy Policy 100, 283–291. http://dx.doi.org/10.1016/j.enpol.2016.10.016. Walker, R., Liddell, C., McKenzie, P., Morris, C., 2013. Evaluating fuel poverty policy in Northern Ireland using a geographic approach. Energy Policy 63, 765–774. http://dx.doi.org/10.1016/j.enpol.2013.08.047. Walker, R., McKenzie, P., Liddell, C., Morris, C., 2012. Area-based targeting of fuel poverty in Northern Ireland: An evidenced-based approach. Appl. Geogr 2012 (34), 639–649. http://dx.doi.org/10.1016/j.apgeog.2012.04.002. 5116