Next Article in Journal
Blockchain-Powered Incentive System for JIT Arrival Operations and Decarbonization in Maritime Shipping
Previous Article in Journal
Simulation Study of a Novel Solar Air-Source Heat Pump Heating System Based on Phase-Change Heat Storage
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 22
10.3390/su152215685
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Systematic Review

The Impact of Urban Form and Density on Residential Energy Use: A Systematic Review

by
Sina Narimani Abar
*,
Martin Schulwitz
and
Martin Faulstich
Resource and Energy Systems Group, Spatial Planning Department, Technical University of Dortmund, 44227 Dortmund, Germany
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(22), 15685; https://doi.org/10.3390/su152215685
Submission received: 5 September 2023 / Revised: 24 October 2023 / Accepted: 1 November 2023 / Published: 7 November 2023
Browse Figures
Review Reports Versions Notes

Abstract

:
The first step for reducing residential energy use is identifying the effective variables in this sector. This review paper extracts 10 urban form variables and discusses their correlations, interpretations, and frequencies alongside methodologies used to quantify their impacts. This review also identifies the parameters that cause mixed effects of density on residential energy use in different studies. Multinomial logistic regression is used to quantify the odds of obtaining a positive or non-significant association rather than a negative correlation. The model can predict the impact of density on residential energy consumption in almost 80% of the cases based on the identified parameters, namely the number of indicators considered in the model as the proxy of density, type of energy, unit of measurement, methodology, data reliability, published year, relevancy, geographical location of case studies and their climate classifications. The result shows that while density correlates negatively with residential energy use in cold climates, its impact could be considered positive in temperate regions.

1. Introduction

Since climate change was recognized as a critical global challenge [1], environmental preservation and energy conservation as the two complementary goals have resulted in a wide range of policies and regulations that involve various levels [2] and different subjects, from industrial processes to mobility and the building sector. The Paris Agreement and Glasgow Climate Pact are well-known examples of collaborative international efforts in 2015 and 2021, respectively. At the city level, the urban form would be an essential element that should be considered in urban planning in order to improve energy efficiency in the building sector [2].
Although only 0.65% of the world’s surface is covered by built-up areas [3], 75% of total energy is consumed, and 80% of global GHG (Greenhouse Gas) emissions are produced in urbanized areas [4]. In particular, energy consumption mainly fell into the building, transport, and industry sectors. The buildings sector is responsible for nearly one quarter of total energy consumption, and its share of CO2 emissions would reach almost 30% if its indirect emissions were also included [5]. Notably, the residential sector is a sub-division of the building sector, which is responsible for three quarters of its total energy consumption [6] and consists of cooling, heating, lighting, and appliances. In 2011, 23% of TFC (Total Final Consumption) was consumed in the residential sector [7]. In this regard, urban planners have recently paid close attention to the role of urban forms in declining REC (Residential Energy Consumption), which is still controversial among scholars [8]. Although the association between urban form and energy used in the transportation sector has been vigorously studied, the association between urban form and REC is still ambivalent [9]. In the past decade, many publications have been released in this field, but their variables, methodologies, and case studies are varied. With this in mind, extracting a specific fact supported by these publications is quite difficult. Hence, this paper endeavors to bridge this void by addressing commonalities, as well as providing substantiation for the disparities observed among scholars. With this intention, this paper conducts a systematic review to find the prevailing debates and challenges through the most cited papers written in this field. Additionally, it explores the methodologies employed in recent studies. Based on the systematic approach, it begins with the clear articulation of research questions. Subsequently, the need for this review is justified by identifying gaps in previous research. A detailed account of the search strategy and the specific inclusive and exclusive criteria used is then provided. Moving forward, this review thoroughly discusses various parameters contributing to mixed effects, employing numerous illustrative examples. Furthermore, each urban form variable is evaluated based on its frequency, correlation, and interpretation. Finally, the paper employs multinomial logistic regression to quantify the effects of the previously discussed parameters, followed by a comprehensive interpretation of the obtained results, shedding light on this topic.

1.1. Previous Reviews and the Gaps

Although conducting a review in this field is not novel, some contrasting findings and interpretations among reviews should be justified to make a better understanding and clearer picture of this field. Noticeably, only 25% of the reviews are published in urban-planning-related journals. Some scholars focus only on certain urban form variables [10,11,12,13,14,15,16] or a particular country [10,17,18], while some others consider the transportation sector [10,13,16,19,20,21,22,23,24], solar access [11,19,23] or embodied energy [10,11,13] alongside residential energy use. In this regard, their conclusions are affected by inordinate considerations which are not in the scope of this review. Moreover, some of these reviews were conducted more than ten years ago, and therefore, the number of papers taken into account is considerably low. Notably, in the recent decade, richer databases and new methodologies have allowed scholars to conduct more research on this topic. Given that, a newer review paper is needed to evaluate recent findings in this field.
Rickwood et al., for instance, exclusively discussed the association between housing types and carbon dioxide emissions. They statistically could not reach a certain conclusion because of the uncontrolled socio-economic characteristics of households in their studies [10]. R. V. Jones et al. conducted a comprehensive investigation encompassing 62 factors that fell into three categories: socio-economic, dwelling characteristics, and household appliances. However, spatial variables were not taken into account, and they did not focus on the probable reasons why there is a discrepancy among scholars about the role of each variable (especially housing type and housing age, which were considered in this review). On the other hand, they tried to elicit specific facts by counting the number of papers confirming similar findings, while the studies show mixed effects [12]. In contrast to Jones et al., Gassar & Cha considered geographic and climatic factors alongside those three categories, but they broke down the geographic factors into geographical location and urbanization level, which are usually used in urban scale studies [14]. H. Zhang et al. classified the influential factors into macro, mid, and micro levels. The macro level consists of three factors: the urban pattern, the urban structure, and the population density. The urban pattern and the urban structure have been used in urban scale studies, and its conclusion about the population density is too general to be informative because it neglects the case studies, methodologies, and incompatible role of density on cooling and heating energy use [16]. Quan & Li defined nine different classifications for 54 reviewed articles and compared the number of papers placed in each group with each other. Notably, they highlighted the limited depth of discussion regarding discrepant findings and underscored the imperative need for additional research to enhance our comprehension of this subject matter [23].
In conclusion, from the urban planning side, there should be a review paper that focuses on urban form variables and methodologies used to model the complexity of urban form characteristics and describe their contrasting effects in different contexts. This paper aims to fill this gap by reviewing relevant papers systematically and meta-analyzing their findings to answer the first research question, which focuses on the role of density in residential energy consumption.

1.2. Research Questions

Eliciting repetitive findings and justifying the contradictory conclusions among different research could facilitate further studies in this field. Moreover, clarifying the methodologies used in modeling the complexity of urban form and the variables taken into account would be another contribution of this study. In particular, this review paper intends to find answers to these questions:
  • Is there any consensus among scholars over their findings about the impact of density on residential energy use? Which parameters lead them to contradictory conclusions?
  • Which urban form variables are frequently used to represent the urban form’s complexity, and what are their correlations and interpretations?
  • Which methodologies are used in the literature to elicit the association between urban form and residential energy use?

2. Materials and Methods

Improving energy efficiency is one of the main concerns among scholars from different educational backgrounds. Urban planners have recently joined 25 other academic disciplines that had already been active in this realm [25]. Given that, drawing a comprehensive conclusion entails scrutinizing publications from various journals. In other words, scholars consider every significant variable based on their units of analysis and observations regardless of the exact research objectives. Otherwise, the probability of bias occurring and over-interpreting the effect of a single variable would be increased. Hence, it is reasonable to find urban form variables in modeling energy consumption, even in journals, which are not the core focus of urban planning. As shown in Table 1, some crucial journals in this research field cannot be disregarded. These journals focus on this topic and consist of top-cited papers that could be considered to be the primary debates.
The systematic review adheres to the PRISMA (The Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, which delineate the procedures for identifying, selecting, evaluating, and synthesizing studies [26]. Figure 1 shows the step-by-step process of the selection of studies using specific inclusion and exclusion criteria to generate the final number of publications for analysis [27].
This review carried out a four-step procedure to search and select literature. The initial step was searching Google Scholar by this expression (“urban” OR “urban form”) AND (“energy” OR “energy consumption” OR “energy use” OR “residential energy”). In the second step, the publications were preliminarily evaluated using two criteria: relevant to the topic and written in English. In the third step, studies using aggregated data (the unit of analysis is not households) or focusing on appliances and socio-economic characteristics of households regardless of spatial features were excluded because these studies usually come from higher scales and their methodologies would be varied regarding the data they use. Moreover, publications that focused on the transportation sector without considering REC, as well as conference papers and non-peer-reviewed papers, were also excluded. Finally, 70 of 207 publications were found eligible and were scrutinized to distinguish the repetitive findings or contradictory conclusions. Although there is no time period limitation applied in this review paper, relevant papers can barely be found before 2008 [23]. Likewise, the selected articles were mainly published after 2008.
Figure 2 shows the co-citation network of the 70 chosen articles. Each line represents a citation of one article in another. As can be seen, the more citations each article has, the bigger the bubble becomes. The horizontal axis is the time of publication, which means earlier articles are on the left and the newer ones are on the right. Interestingly, they have a hierarchical relationship. This association among publications facilitates the monitoring of central discourses and challenges within the subject, as illuminated in recent literature. It is reasonable to expect that well-conducted inquiries have drawn upon the most significant prior studies.

3. Results

3.1. Urban Form and Density on Residential Energy Use

Urban form is one of the main interests of urban planners to address different city-level problems. Improving energy efficiency to reduce energy consumption and GHG emissions is an international goal. Given that, divulging whether urban form could significantly affect REC would be a primary step for urban planners. In this regard, scholars consider one or more different variables of urban form, based on their data accessibility, in modeling residential energy consumption.
As shown in Table 2, there exists a disparity in the assessment of how urban form and population density influence residential energy consumption among researchers, a phenomenon stemming from various underlying causes.

3.1.1. Limitation of the Data Sources

It confines the scholars to detect the exact effects of urban form and density on REC [87], which has different aspects. First, privacy and security issues make the required databases partly inaccessible. Second, there is no single data source that includes REC, spatial features, socio-economic characteristics of households, and other relevant factors altogether [8]. Therefore, connecting the existing databases would be another challenge for researchers. One example is Reams’ study in Kansas. He used the RECS (U.S. Residential Energy Consumption Survey) database to extract household-level energy consumption data, alongside the U.S. Census Bureau and American Community Survey (ACS), to gather spatial and socio-economic characteristics of households [71]. As can be seen in Figure 3, the case studies investigated in various publications are not evenly distributed due to the availability of accessible data sources that have been gathered and archived in specific countries. For example, in the United States, there is a dataset called RECS, which includes the residential energy consumption of more than 4000 households by their states. In conclusion, although scholars frequently study certain countries, other countries and even continents are partly or completely disregarded (e.g., Africa and South America).

3.1.2. The Units of Measurement

It could also significantly affect the eventual outcome and its interpretation [70,86]. For instance, Kaza compares the energy consumption in distinct housing types from 1978 to 2002 by different units, namely per household, per square meter, and per capita in the United States. It is evident that “Single-Family Detached Housing” consumes more annual energy per household than “Single-Family Attached” and “Apartments in Buildings with five or More Units”. In contrast, this relationship becomes quite the opposite in annual energy per square meter [53]. Noticeably, the higher the percentage of “Apartments in Buildings with 5 or More Units” in housing stock, the denser urban form can be assumed. Ko also calculated the REC by housing type in different units of measurement. It implies that the unit of measurement could change the interpretation of the role of density in reducing REC [19]. With this in mind, discerning the role of urban form and discussing whether density has a positive or negative correlation is controversial.

3.1.3. Climate Zone

Investigated cities and regions are in various locations with different climatic characteristics (Figure 3). The climate could directly affect the impact of density on residential energy use. In severe cold regions, for example, where heating energy consumption dominates the other types of energy usage, density is prone to have a negative association with annual REC [47]. According to the Koppen-Geiger climate classification, the world climate can be divided into five main parts, and this classification is in regular use by researchers across various fields [88]. Hence, it serves the purpose of categorizing the case studies examined in this review paper. To achieve this objective, we overlay the updated Koppen-Geiger climate classification map onto the layer representing the cases in ArcGIS Pro. As shown in Figure 4, 45% of the cases are located in temperate regions, while tropical and arid areas are less studied. Notably, some studies take a country or a set of cities in distinct climate zones determined by “Varied”.

3.1.4. Energy Type

The data related to residential energy consumption can be classified into three distinct categories, namely cooling energy consumption, heating energy consumption, and annual energy use. Regarding the type of energy taken as a dependent variable, the role of density could be varied. Figure 4 summarizes the reasons that led the scholars to contradictory conclusions in 5 different columns with their classifications. As depicted in Figure 4, most studies (71%) utilize annual energy use as their dependent variable. This choice is often necessitated by limitations in their data sources or the complexities involved in disaggregating annual energy use into cooling and heating energy consumption. Ko & Radke, for example, estimated cooling energy consumption by subtracting the average monthly electricity use in the months of March, April, September, and October (it is assumed that households do not have air-conditioning energy usage in these months) from the total electricity consumption in the months of June, July, and August [60]. This approach was employed to address the absence of detailed data.

3.1.5. Data Reliability

Some countries do not provide any reliable data source regarding household energy consumption. Hence, scholars should find some ways to cope with the shortage of needed information. For instance, some publications estimate the REC by considering household energy expenditures [34,41,54,70,73] or taking total energy use in summer months as the proxy of the cooling energy consumption [59,60], while others use existing datasets.

3.1.6. Relevancy

If an article is published in urban planning-related journals, it shows that the main focus of that article is on urban planning concepts. With this in mind, it can be assumed that urban form variables are selected and interpreted more accurately. Therefore, this parameter is also considered to describe the variation of eventual outcomes.

3.1.7. Variables and Methodologies

Converting the complexity of urban form to some measures and methods to clear out the relationship between density and REC is explained in the following steps.
  • Urban Form variables
Converting urban form’s complexity to some numeric or categorical variables is a great challenge for researchers. Data accessibility, preferences, and study objectives could justify the differences between publications. As seen in Figure 5, plenty of variables are taken into account as the proxy of urban form in various studies.
Considerably, certain variables could not be included simultaneously in modeling the REC due to issues related to multicollinearity. Notably, multicollinearity occurs when two or more explanatory variables correlate with each other and violate the assumption of independence. Ecological and spatial features are prone to multicollinearity, enhancing the necessity of actions to make them less biased [89]. In the following step, each urban form variable is discussed based on its definition, impact on REC, and interpretation.
  • Housing Type
Although housing type is a building characteristic, it could also be considered to be an urban form variable. Dwelling type as a categorical variable could be directly used in modeling REC or indirectly in defining different homogenous urban forms. Wang (2021), for example, used 13 parameters plus housing type to create six homogenous urban forms ranging from class A (green spaces) to class F (high-density compact buildings). Although the study shows that class F emits more carbons per square meter than others, he did not calculate the correlation between housing type and REC [82], p. 67.
Housing type is a well-known variable that is the subject of extensive research. The majority of studies have concluded that the degree of detachment of the dwellings positively affects the amount of REC [8,9,30,33,35,41,42,43,44,50,52,54,56,76,78]. In other words, families residing in detached houses consume more energy than semi-detached houses and apartments [12], while a few studies reached a contradictory conclusion [34,90].
2.
Housing Age
Housing Age is one of the most famous indicators among publications. It could also be assumed as a proxy of construction technology [8]. Holden & Norland argue that the different building types constructed after 1980 in the case study of Norway have fewer distinctions in energy consumption. Therefore, the technology used in recent buildings approximately compensates for the malfunctions in energy efficiency in various building types [30]. On the other hand, Lee & Lee debate that, although developing technologies and new building materials help improve energy efficiency, this amount of saved energy could not be sufficient to tackle climate change [8].
Some researchers conclude that newer buildings use more energy than older ones [31,37,44,54,57,75,78]. They ascribe the increase in energy demand to the penetration of air-conditioning and high-consumption appliances [12]. In contrast, other studies argue that there is a positive correlation between the age of the building and REC [8,9,30,35,46,48,51,53,56,60,61,71,73,74] which means energy demand would go up by increasing the dwelling age. They interpret this observed result by emphasizing improving insulation [91] and energy efficiency [35] in recent buildings, thus lowering the total energy demand.
3.
Housing Size
On the one hand, larger buildings consume more energy than smaller ones [9,30,34,35,37,40,44,49,50,53,54,56,57,59,60,76]. Based on Reid Ewing’s findings, a household residing in an 186 m2 house consumes 16% more for space heating and 13% more for space cooling than a household living in a 93 m2 one [9]. On the other hand, when considering energy use per square meter as the reference unit, it mitigates the influence of housing size, necessitating a distinct interpretation. With this in mind, the negative correlation between REC/m2 and dwelling size implies that energy is used more efficiently in larger buildings than in smaller ones [51,75].
4.
Urbanity Index
The urbanity index is the most straightforward indicator, which delineates whether the sample is in rural, suburban, or urban areas. Wiedenhofer et al. consider this variable and reckon that suburban and rural lives consume 10% more energy than urban life. Interestingly, they ascribe it to the use of more energy-intensive appliances by the average person in a rural household and argue that necessities are more energy-intensive than luxuries [36]. Furthermore, this phenomenon can be attributed to sprawl development, as well as substantial disparities in housing types and population density between urban areas and suburban or rural regions [28,31,37,46,49,50,52,76,77]. In contrast, Tso & Guan divulge that if a house were in an urban area rather than a rural area, the expected average household energy consumption would be increased by 61.722 kWh/year [35]. Additionally, Huang found rural areas less energy-intensive than urban areas in Taiwan. Interestingly, he also ascribes the results to other factors rather than urban form variables, namely the differences in income and lifestyle between urban and rural residents [34].
According to Heinonen & Junnila, each housing type appears less energy-intensive in rural areas. This conclusion comes from comparing the energy purchases per household for each housing type in rural and urban areas. This study conducts a descriptive analysis in Finland and uses energy expenditure to estimate the actual energy consumption [70]. Counter-intuitively, as shown in Table 3, the average family size in an urban area in Finland is bigger than in rural areas. Therefore, energy purchases per person contradicts the first conclusion and upholds the argument that the unit of measurement could affect the results and their interpretations.
5.
Floor Area Ratio
The floor area ratio (FAR) is defined as the ratio of the total combined floor areas of all buildings to the area of the sampled district [47]. It describes the vertical density of the urban form. Some scholars disclose the negative correlation between FAR and REC [30,33,47,48], which reinforces the urban growth policies in order to reduce energy consumption, while others found the impact of FAR on REC insignificant [45,51,57,60,61,63]. On the other hand, Kontokosta & Tull argue that taller buildings with more storeys, controlling for floor area, consume more energy per square meter [75]. Mostafavi et al., for example, work on six different cities in the United States and integrate three databases to evaluate the role of urban form and density on REC in these cities separately. Although they reached a negative correlation between FAR and REC in Washington DC, the result in other cities was quite the opposite. Hence, they have concluded that the association between FAR and REC varied across cities [45].
6.
Division Group
Most investigated studies reveal that urban form has a significant effect on REC. Hence, spatial elements play an essential role in residential energy consumption, which should be considered in modeling REC to describe the variation of the dependent variable. Some of these elements could not be taken into account because of data accessibility and methods’ assumptions. One of the approaches that researchers use to cope with this problem is dividing their case studies into some homogenous regions and examining whether there is a meaningful difference between these regions [33,82].
Another approach is adding a categorical variable containing the name of regions, which usually are based on administrative [46,54,55,74,75,92] or geographic divisions [34,35,52]. As shown in Figure 6, more scholars prefer to use administrative division because it has a determined boundary, and the eventual result could be found relevant by state or local authorities. Moreover, the model would cover some unmeasured political or managerial variables by considering administrative regions.
7.
Population Density
Population Density is the number of humans in each hectare or square kilometer [93]. Because of its simplicity, it could be used interchangeably with other urban form elaborate variables [36]. Put differently, population density, when accounting for housing size and household size, emerges as the most comprehensive indicator of urban density, encompassing other density variables such as FAR or BCR. Many scholars conclude that increasing population density could result in improving energy efficiency per household [8,31,36,44,57,76,77]. Counter-intuitively, Ko & Radke [60] considered only cooling energy use as the dependent variable; they found a negative correlation between population density and REC in Sacramento. This result contradicts the assumptions of the Urban Heat Island effect that leads to lower energy consumption for heating in cold seasons and higher energy use in summer for cooling [94].
8.
Green Space Ratio
Green Space Ratio is the percentage of an area covered by green spaces. This variable usually comes from land-use attributes or imagery approaches. Ko & Radke reveal that there is a negative correlation between Green Space Ratio and summer cooling electricity use in Sacramento, California [48,60], while Leng et al. find a positive relationship between Green Space Ratio and REC in Harbin, a severe cold region city [47]. The discrepancy among these studies might be justified by the different climate situations and could be solved by considering cooling and heating separately.
9.
Community Layout
Building Orientation or Community Layout is vital to maximizing solar gain, which is essential during winter [95]. In particular, the result illustrates that there is not a significant relationship between CL and summer-time energy consumption [48]. On the other hand, Quang Minh Nguyen argues that orientation can immensely affect household energy consumption. He reveals that row houses and apartments facing the Southeast consume 47.7% and 26.8% less energy than the same size row houses and apartments facing the Southwest in Vietnam, respectively, because of the use of warm morning sunlight in winter and cool wind in summer months [96], Ch. 3.
10.
Building Coverage Ratio
BCR is the ratio of built ground to that of the sample district area [47], and it is also defined as the horizontal density of urban form. Regardless of how many floors a building has, it shows the percentage of built area in the region. Some scholars argue that BCR has a negative association with REC [33,47,51], while others reach the opposite affiliation [45]. Notably, all research supporting the negative correlation between BCR and REC has used per square meter as the unit of REC. Given that, the more building-covered areas we have in a certain region, the less energy we need for each square meter.
11.
Methodologies in the Literature
The methodology constitutes a pivotal component in every study, comprising a series of intricate steps. From a meta-perspective, methodologies could be divided into two main groups: simulation and empirical studies. Simulation enables scholars to manage the intricacies of urban forms, albeit with the caveat of dealing with a substantial degree of uncertainty due to limited access to detailed data [79]. Moreover, as simulation provides finer estimates of energy consumption at the individual building scale, this method is hugely conducted by architects, building scientists, and mechanical engineers [11,19].
In empirical studies, scholars would use real-world data and, therefore, integrate various datasets to describe the variation of the dependent variable. In contrast to Quan & Li [23], the majority of investigated articles in this review paper are empirical studies. The most commonly used inferential statistic is multiple linear regression. Although linear regression is a simple and elegant tool, it depends on assumptions and details requiring close attention [97], p. 223. One of these assumptions is independent sample distribution. In other words, there must be no relationship among the individual observations in the sample [98], p. 23. Noticeably, revealing the impact of urban form on residential energy use entails a dataset that not only encompasses the built environment features at broader scales, such as neighborhoods, 100 m airline buffers, and urban districts, but also includes socio-economic characteristics at the individual household level. With this in mind, in these cases, samples are prone to have nested structures. Nesting occurs when individual households in each neighborhood are more similar to each other than households from different neighborhoods [99], p. 155. In conclusion, multilevel modeling is conducive to reaching more reliable outcomes. As shown in Figure 7, although almost half of the studies use multiple linear regression, which is the most common method utilized in the literature, 11% of them implement multilevel modeling. It indicates that nested structure is recognized and controlled in more than 10% of studies.

3.2. Classification Model

Density is one of the most substantial factors studied in various research. According to Lee, although density significantly affects the amount of energy consumption, the effects of other spatial variables are estimated to be small [8]. As seen in Table 2, the majority of scholars reveal a negative correlation between density and REC. In contrast, some others argue that density positively impacts REC, and the third group elicits a non-significant relationship between them. Based on the investigated publications, 67% of studies reveal that increasing density leads to less energy consumption, while 10% reach the opposite conclusion about this relationship. As the dependent variable is categorical, logistic regression would be the preferred method [100], p. 308, and multinomial logistic regression is selected because the response variable has three different levels, namely negative, positive, and non-significant relationship between urban density and residential energy consumption. Conducting this method is not novel in urban planning. For instance, Mustafa et al. divided Wallonia into eight classes based on various density ranges and employed multinomial logistic regression to investigate the effects of 16 drivers on the probability of urban development along different densities [101]. Additionally, Geraghty & Mahony implemented multinomial logistic regression for examining the relative importance of the spatial and temporal variables of locations in predicting urban noise levels [102].
This review paper employs multinomial logistic regression to assess the likelihood of observing a positive or non-significant relationship, as opposed to a negative one, between density and Residential Energy Consumption (REC) by considering some study characteristics. The recognized parameters which are taken into account are:
  • Number of indicators considered in the model as the proxy of density,
  • Type of energy, which is taken as a dependent variable,
  • Unit of measurement,
  • Methodology,
  • Data reliability,
  • Geographical location and its climate classification,
  • Published year,
  • Relevancy (a binary variable that determines whether the article is published in urban planning-related journals).
Notably, the information from each article is meticulously archived within a Google Sheet organized in a tabular format (Table 4). Subsequently, inferential statistics are rigorously conducted in RStudio through the utilization of R programming. It is noteworthy that certain scholars compute the correlation between urban density and REC for different cities independently. In such instances, the study characteristics, together with their corresponding results, are treated as individual observations within the model. Conversely, when samples are drawn from diverse regions and combined into a single model, it constitutes an observation categorized under the “Varied” climate classification.
According to the model (Table 5), for each additional indicator related to the urban density included in the analysis, the odds of revealing a negative impact of the density on REC becomes 3.5 times higher than failing to disclose the significant relationship. Moreover, for an article published in urban planning-related journals, the odds of obtaining a non-significant relationship are almost one third of the odds that the study reveals a negative association. In other words, in urban planning-related journals, scholars usually aim to divulge the impact of density rather than take it as a control variable. With this in mind, they would probably take more suitable indicators and proxies. Furthermore, in studies estimating residential energy use instead of using existing datasets, the odds of obtaining a non-significant impact are 4.18 times higher than the odds that the density correlates negatively with REC. In addition, if a study is published one year later, the odds of obtaining a negative association are 11% higher than the odds of obtaining a non-significant relationship. It could be ascribed to more accurate analysis in newer publications.
The odds of revealing a positive correlation between density and REC in the cases in temperate climates is more than 10 times higher than of revealing a negative one. On the other hand, this formulation is quite the opposite in cold climates. It is upheld by the hypothesis that increasing density intensifies the urban heat island effects and moderates the heating loads.
Interestingly, using any inferential analysis instead of a descriptive analysis decreases the odds of obtaining a positive impact. In this regard, it could be assumed that more accurate computational methods led to a negative association. This argument is upheld by the odds ratio of the estimated dependent variable. It implies that if a study uses estimated energy consumption, the odds of obtaining positive affiliation would be multiplied by 3.57 times.

4. Discussion

Urbanized areas consume 70% and emit 80% of GHG emissions. With this in mind, improving energy efficiency in these areas deserves close attention. From the urban planning view, it is essential to reveal the impact of urban form and density on residential energy use. With this intention, this review paper systematically reviews articles with a determined search strategy to find an answer to its research questions.
Is there any consensus among scholars over their findings about the impact of density on residential energy use? Which parameters lead them to contradictory conclusions?
Density, from a spatial view, would be defined simply as the number of units in a given area. However, there is a wide variety of definitions depending on the kind of density being sought. Boyko has documented Twenty-three different definitions associated with density, of which three are frequently used in this field: population density, floor area ratio, and building coverage ratio [103]. Additionally, other indicators have a meaningful relationship with density. For instance, an urban area is clustered with at least 1500 inhabitants per sq. km, while a rural area would have 300 inhabitants per sq. km [104]. Therefore, identifying whether the sample is in an urban area would be a proxy of higher density. Moreover, a higher percentage of single detached houses in a neighborhood results in higher residential densities [105].
During the twentieth century, planning theories and practices were strongly opposed to density in the urban context [106]. This trend started reversing with Jane Jacobs advocating for higher densities in thriving neighborhoods [107], p. 217. In America, the term “Smart Growth” first appeared in Maryland in 1997 to limit the sprawling patterns of low-density residential development [108].
Density has a significant impact on various sectors. Enhancing accessibility, making transit more efficient, making better use of resources and existing infrastructures, and improving housing choices are its main advantages. On the other hand, exacerbating traffic congestion, increasing parking problems, and worsening various types of pollution are its focal drawbacks. In particular, density is neither inherently positive nor negative [106]. It should be evaluated based on the context of planning. The findings of this review paper can also support this argument.
Although scholars have reached contradictory conclusions about the role of density in reducing residential energy consumption, some elements can justify this discrepancy. The number of indicators considered in the model as the proxy of density, type of energy taken as a dependent variable, unit of measurement, methodology, data reliability, geographical location, and climate classification, alongside published year and relevancy, are the recognized parameters. Subsequently, the multinomial logistic regression is conducted to quantify the odds of obtaining a positive or non-significant impact rather than a negative effect of density on REC.
In this review paper, 77% of the studies examined establish either a positive or negative association between density and residential energy use. Noticeably, the multinomial logistic regression model reveals that using estimated data instead of reliable datasets, considering few urban form indicators and lack of urban planning knowledge in some studies, resulted in a reduction in this figure. According to the model, whose accuracy is almost 80%, it can be concluded that in regions characterized by cold climates, where the predominant energy consumption stems from residential heating demands, there is a negative correlation between density and REC, while this relationship is quite the opposite in temperate regions.
As mentioned, climate classification is derived from Koppen-Geiger climate classification, which divides the world’s climate into five distinct regions, namely A (tropical), B (arid), C (temperate), D (continental), and E (polar). By overlapping the location of cases studied by scholars and polygons of different climate zones, it reveals that the cases are not fairly distributed in these regions. As shown in Figure 4, the percentage of cases located in tropical and arid regions is fewer than cases in temperate and continental zones.
Which urban form variables are frequently used to represent the urban form’s complexity, and what are their correlations and interpretations?
This review paper investigated each urban form’s variable used in the literature. Likewise, the disagreement on their correlations, interpretations, and frequencies among various studies is well discussed. Housing type, housing age, and housing size are the most frequently used variables in the literature because these characteristics are also in the scope of other majors, and their information is gathered and accessible through various datasets. Although articles published in urban planning-related journals tried to consider more specific indicators (e.g., population density), researchers in not-related journals used more superficial indicators to keep the variation described by the model high (e.g., urbanity index).
Only 35% of the publications have been published in urban planning-related journals. Consequently, the interpretations of urban form variables in the remaining 65% of publications may differ from the perspective of urban planning. It could also be concluded that there is a misunderstanding of some concepts. For example, Wiesmann et al. use the average number of rooms as the proxy for total floor area [54], and Kaza presented an argument regarding the influence of neighborhood density, while this concept is operationalized based on whether the sample is situated in urban, rural, suburban, or town areas [53].
This review paper identifies 10 urban form variables that are frequently used in publications, namely, Housing type, Housing age, Housing size, Urbanity Index, Floor Area Ratio, Division group, Population Density, Green Space Ratio, Community Layout, and Building Coverage Ratio. Additionally, some other variables are taken by fewer studies, such as the Composite Index, Inside Lot, and Entropy Index.
Composite index is the incorporation of different variables into one index. Ewing & Rong, for example, introduce a composite index named sprawl index as a measure of urban form, which incorporates six different variables [9]. The sprawl index incorporates population density, urbanity index, and inside lots in metropolitan areas. Although it could describe the high percentage of variability of the dependent variable, the effect of each sub-indicator could not be detected. Therefore, considering composite indexes is only suggested when the study is trying to answer general questions regarding the association of urban form or density with residential energy use.
The inside-lot variable, collected at the zip-code level, denotes whether the building is situated in the middle of the block or at a corner. A negative correlation between inside lots and REC could lead urban planners to encourage building large blocks to improve energy efficiency. Building large blocks increases the percentage of buildings located in the middle of the blocks.
The entropy index measures the balance between different land uses. The index ranges from 0, where all land is in a single use, to 1, where land is equally divided among all the uses [99], p. 15. According to Holden & Norland, land use is more effective in the transportation sector than in the residential sector [30]. Given that, if scholars focus on REC, they barely consider this indicator in their model, while Zhang et al. reveal the positive correlation between EI and residential blocks’ carbon emissions in Changxing City, China [40].
Which methodologies are used in the literature to elicit the association between urban form and REC?
This review paper also reveals the methodological heterogeneity among different studies and clarifies the odds of achieving a positive or non-significant impact rather than a negative association by implementing multinomial logistic regression. The result shows that nearly half of scholars conducted multiple linear regression, which is very common in various research. Although scholars frequently use the multiple regression model, it could not provide an opportunity to consider spatial variables calculated in higher scales. Considering the attributes of individual households nested in their neighborhoods with their specific features entails conducting more accurate methods. In this regard, multilevel modeling is strongly recommended.

5. Conclusions

In conclusion, it would be difficult for urban planners to go through the details of making an urban planning guideline as long as there is not a solid effect of each urban form variable on residential energy use in all cities with various climatic, geographical, and cultural features. This review paper aims to extract the role of each unique trait in improving energy efficiency. For instance, the odds of having a positive association between density and residential energy use in temperate regions is more than ten times higher than being a negative relationship between these two. Hence, considering these facts, which are elicited by this review paper, would give insight to urban planners to compile urban plans aligned with mitigating climate change and the 11th sustainable development goal.
This review paper puts forward the argument that, counter-intuitively, there is an acceptable consensus among scholars if the aforementioned parameters are controlled and considered. The model which is conducted in this paper can predict 80% of the outcomes related to the impact of urban density on REC correctly. Gathering reliable data and accessing disaggregated information are still some of the challenges in this field. Consequently, some countries or even continents are disregarded, especially ones that are in tropical and arid regions.
This review paper also recognizes some urban form variables which are used frequently in various studies. Interestingly, their impacts and interpretations are varied in different research, as extensively discussed in this review paper. Some of these different interpretations can be ascribed to the contexts’ characteristics, which have not been controlled yet, and some others rely on distinct points of view from different disciplines. Moreover, methods used frequently in this field are extracted, and a more reliable model based on the data structure is suggested for further studies.

6. Further Research in this Field

There are too few studies on cases in tropical and arid regions. Therefore, the impact of urban form and density on residential energy consumption in these regions is still controversial and needs more investigations. One of the primary constraints encountered during the endeavor to ascertain similarities among scholars lies in the restricted access to the raw data from each study. Therefore, it would be great to analyze the various datasets used by scholars and compare the results.

Funding

This research received no external funding.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Hamin, E.M.; Gurran, N. Urban form and climate change: Balancing adaptation and mitigation in the U.S. and Australia. Habitat Int. 2009, 33, 238–245. [Google Scholar] [CrossRef]
  2. Anderson, W.P.; Kanaroglou, P.S.; Miller, E.J. Urban Form, Energy and the Environment: A Review of Issues, Evidence and Policy. Urban Stud. 1996, 33, 7–35. [Google Scholar] [CrossRef]
  3. Liu, Z.; He, C.; Zhou, Y.; Wu, J. How much of the world’s land has been urbanized, really? A hierarchical framework for avoiding confusion. Landsc. Ecol. 2014, 29, 763–771. [Google Scholar] [CrossRef]
  4. Fang, C.; Wang, S.; Li, G. Changing urban forms and carbon dioxide emissions in China: A case study of 30 provincial capital cities. Appl. Energy 2015, 158, 519–531. [Google Scholar] [CrossRef]
  5. IEA. World Energy Outlook 2021. 2021. Available online: www.iea.org/weo (accessed on 4 September 2023).
  6. Nejat, P.; Jomehzadeh, F.; Taheri, M.M.; Gohari, M.; Muhd, M.Z. A global review of energy consumption, CO2 emissions and policy in the residential sector (with an overview of the top ten CO2 emitting countries). Renew. Sustain. Energy Rev. 2015, 43, 843–862. [Google Scholar] [CrossRef]
  7. IEA. Energy Efficiency Indicators: Fundamentals on Statistics. 2014. Available online: https://www.iea.org/termsandconditionsuseandcopyright/ (accessed on 4 September 2023).
  8. Lee, S.; Lee, B. The influence of urban form on GHG emissions in the U.S. household sector. Energy Policy 2014, 68, 534–549. [Google Scholar] [CrossRef]
  9. Ewing, R.; Rong, F. The impact of urban form on U.S. residential energy use. Hous. Policy Debate 2008, 19, 1–30. [Google Scholar] [CrossRef]
  10. Rickwood, P.; Glazebrook, G.; Searle, G. Urban structure and energy-a review. Urban Policy Res. 2008, 26, 57–81. [Google Scholar] [CrossRef]
  11. Sanaieian, H.; Tenpierik, M.; Van Den Linden, K.; Seraj, F.M.; Mofidi Shemrani, S.M. Review of the impact of urban block form on thermal performance, solar access and ventilation. Renew. Sustain. Energy Rev. 2014, 38, 551–560. [Google Scholar] [CrossRef]
  12. Jones, R.V.; Fuertes, A.; Lomas, K.J. The socio-economic, dwelling and appliance related factors affecting electricity consumption in domestic buildings. Renew. Sustain. Energy Rev. 2015, 43, 901–917. [Google Scholar] [CrossRef]
  13. Doherty, M.; Nakanishi, H.; Bai, X.; Meyers, J. Relationship Between from, Morphology, Density and Energy in Urban Environments; CSIRO Sustainable Ecosystems: Canberra, Australia, 2009; pp. 1–28. [Google Scholar]
  14. Gassar, A.A.A.; Cha, S.H. Energy prediction techniques for large-scale buildings towards a sustainable built environment: A review. Energy Build. 1102, 2020, 22438. [Google Scholar] [CrossRef]
  15. Khan, I. Household factors and electrical peak demand: A review for further assessment. Adv. Build. Energy Res. 2021, 15, 409–441. [Google Scholar] [CrossRef]
  16. Zhang, H.; Peng, J.; Wang, R.; Zhang, J.; Yu, D. Spatial planning factors that influence CO2 emissions: A systematic literature review. Urban Clim. 2021, 36, 100809. [Google Scholar] [CrossRef]
  17. Cai, M.; Shi, Y.; Ren, C.; Yoshida, T.; Yamagata, Y.; Ding, C.; Zhou, N. The need for urban form data in spatial modeling of urban carbon emissions in China: A critical review. J. Clean. Prod. 2021, 319, 128792. [Google Scholar] [CrossRef]
  18. Alsaadani, S. A statistical review of a decade of residential energy research in Egypt. Energy Rep. 2022, 8, 95–102. [Google Scholar] [CrossRef]
  19. Ko, Y. Urban Form and Residential Energy Use: A Review of Design Principles and Research Findings. J. Plan. Lit. 2013, 28, 327–351. [Google Scholar] [CrossRef]
  20. Anderson, J.E.; Wulfhorst, G.; Lang, W. Energy analysis of the built environment—A review and outlook. Renew. Sustain. Energy Rev. 2015, 44, 149–158. [Google Scholar] [CrossRef]
  21. Gargiulo, C.; Russo, L. Cities and Energy Consumption: A Critical Review. TeMA J. Land Use Mobil. Environ. 2017, 10, 259–278. [Google Scholar]
  22. Silva, M.; Oliveira, V.; Leal, V. Urban Form and Energy Demand: A Review of Energy-relevant Urban Attributes. J. Plan. Lit. 2017, 32, 346–365. [Google Scholar] [CrossRef]
  23. Quan, S.J.; Li, C. Urban form and building energy use: A systematic review of measures, mechanisms, and methodologies. Renew. Sustain. Energy Rev. 2021, 139, 110662. [Google Scholar] [CrossRef]
  24. Sun, C.; Zhang, Y.; Ma, W.; Wu, R.; Wang, S. The Impacts of Urban Form on Carbon Emissions: A Comprehensive Review. Land 2022, 11, 1430. [Google Scholar] [CrossRef]
  25. Geng, Y.; Chen, W.; Liu, Z.; Chiu, A.S.F.; Han, W.; Liu, Z.; Zhong, S.; Qian, Y.; You, W.; Cui, X. A bibliometric review: Energy consumption and greenhouse gas emissions in the residential sector. J. Clean. Prod. 2017, 159, 301–316. [Google Scholar] [CrossRef]
  26. Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.; Brennan, S.E.; et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Syst. Rev. 2021, 10, 1–11. [Google Scholar] [CrossRef] [PubMed]
  27. Harris, J.D.; Quatman, C.E.; Manring, M.M.; Siston, R.A.; Flanigan, D.C. How to write a systematic review. Am. J. Sports Med. 2014, 42, 2761–2768. [Google Scholar] [CrossRef] [PubMed]
  28. Glaeser, E.L.; Kahn, M.E. The greenness of cities: Carbon dioxide emissions and urban development. J. Urban Econ. 2010, 67, 404–418. [Google Scholar] [CrossRef]
  29. Ratti, C.; Baker, N.; Steemers, K. Energy consumption and urban texture. Energy Build. 2005, 37, 762–776. [Google Scholar] [CrossRef]
  30. Holden, E.; Norland, I.T. Three challenges for the compact city as a sustainable urban form: Household consumption of energy and transport in eight residential areas in the Greater Oslo Region. Urban Stud. 2005, 42, 2145–2166. [Google Scholar] [CrossRef]
  31. Jones, C.; Kammen, D.M. Spatial distribution of U.S. household carbon footprints reveals suburbanization undermines greenhouse gas benefits of urban population density. Environ. Sci. Technol. 2014, 48, 895–902. [Google Scholar] [CrossRef]
  32. Güneralp, B.; Zhou, Y.; Ürge-Vorsatz, D.; Gupta, M.; Yu, S.; Patel, P.L.; Fragkias, M.; Li, X.; Seto, K.C. Global scenarios of urban density and its impacts on building energy use through 2050. Proc. Natl. Acad. Sci. USA 2017, 114, 8945–8950. [Google Scholar] [CrossRef]
  33. Rode, P.; Keim, C.; Robazza, G.; Viejo, P.; Schofield, J. Cities and energy: Urban morphology and residential heat-energy demand. Environ. Plan. B Plan. Des. 2014, 41, 138–162. [Google Scholar] [CrossRef]
  34. Huang, W.H. The determinants of household electricity consumption in Taiwan: Evidence from quantile regression. Energy 2015, 87, 120–133. [Google Scholar] [CrossRef]
  35. Tso, G.K.F.; Guan, J. A multilevel regression approach to understand effects of environment indicators and household features on residential energy consumption. Energy 2014, 66, 722–731. [Google Scholar] [CrossRef]
  36. Wiedenhofer, D.; Lenzen, M.; Steinberger, J.K. Energy requirements of consumption: Urban form, climatic and socio-economic factors, rebounds and their policy implications. Energy Policy 2013, 63, 696–707. [Google Scholar] [CrossRef]
  37. Min, J.; Hausfather, Z.; Lin, Q.F. A high-resolution statistical model of residential energy end use characteristics for the United States. J. Ind. Ecol. 2010, 14, 791–807. [Google Scholar] [CrossRef]
  38. Sajadian, A.; Niazmardi, S.; Argany, M.; Arababadi, R.; Pignatta, G. Investigating the Relative Importance of Spatial and Environmental Factors on Energy Consumption in the Residential Sectors Using GIS (Case Study: Kerman). Environ. Sci. Proc. 2021, 12, 6. [Google Scholar]
  39. Ahmadian, E.; Sodagar, B.; Bingham, C.; Elnokaly, A.; Mills, G. Effect of urban built form and density on building energy performance in temperate climates. Energy Build. 2021, 236, 110762. [Google Scholar] [CrossRef]
  40. Zhang, X.; Yan, F.; Liu, H.; Qiao, Z. Towards low carbon cities: A machine learning method for predicting urban blocks carbon emissions (UBCE) based on built environment factors (BEF) in Changxing City, China. Sustain. Cities Soc. 2021, 69, 102875. [Google Scholar] [CrossRef]
  41. Belaid, F.; Rault, C. Energy Expenditure in Egypt: Empirical Evidence Based on a Quantile Regression Approach. Environ. Model. Assess. 2021, 26, 511–528. [Google Scholar] [CrossRef]
  42. Berrill, P.; Gillingham, K.T.; Hertwich, E.G. Linking housing policy, housing typology, and residential energy demand in the United States. Environ. Sci. Technol. 2021, 55, 2224–2233. [Google Scholar] [CrossRef]
  43. Rokseth, L.S.; Manum, B. Patterns of dwelling types, location, and spaciousness of living in norway. Critical remarks on the practice of measuring energy performance per floor area only. Buildings 2021, 11, 394. [Google Scholar] [CrossRef]
  44. Voskamp, I.M.; Visscher, M.N.; Vreugdenhil, C.; Van Lammeren, R.J.A.; Sutton, N.B. Spatial, infrastructural and consumer characteristics underlying spatial variability in residential energy and water consumption in Amsterdam. Sustain. Cities Soc. 2021, 72, 102977. [Google Scholar] [CrossRef]
  45. Mostafavi, N.; Heris, M.P.; Gándara, F.; Hoque, S. The relationship between urban density and building energy consumption. Buildings 2021, 11, 455. [Google Scholar] [CrossRef]
  46. Tilov, I.; Farsi, M.; Volland, B. From frugal Jane to wasteful John: A quantile regression analysis of Swiss households’ electricity demand. Energy Policy 2020, 138, 111246. [Google Scholar] [CrossRef]
  47. Leng, H.; Chen, X.; Ma, Y.; Wong, N.H.; Ming, T. Urban morphology and building heating energy consumption: Evidence from Harbin, a severe cold region city. Energy Build. 2020, 224, 110143. [Google Scholar] [CrossRef]
  48. Douthat, T.; Morgan, J.D.; Alibašić, H.; Grant, A. Passive building characteristics, and summertime residential energy use: A spatial analysis of energy efficiency in Gainesville, FL. Build. Environ. 2020, 169, 106542. [Google Scholar] [CrossRef]
  49. Rong, P.; Zhang, Y.; Qin, Y.; Liu, G.; Liu, R. Spatial patterns and driving factors of urban residential embedded carbon emissions: An empirical study in Kaifeng, China. J. Environ. Manag. 2020, 271, 110895. [Google Scholar] [CrossRef] [PubMed]
  50. Belaid, F.; Youssef, A.B.; Omrani, N. Investigating the factors shaping residential energy consumption patterns in France: Evidence form quantile regression. Eur. J. Comp. Econ. 2020, 17, 127–151. [Google Scholar] [CrossRef]
  51. Ahn, Y.J.; Sohn, D.W. The effect of neighbourhood-level urban form on residential building energy use: A GIS-based model using building energy benchmarking data in Seattle. Energy Build. 2019, 196, 124–133. [Google Scholar] [CrossRef]
  52. Shi, X.; Wang, K.; Cheong, T.S.; Zhang, H. Prioritizing driving factors of household carbon emissions: An application of the LASSO model with survey data. Energy Econ. 2020, 92, 104942. [Google Scholar] [CrossRef]
  53. Kaza, N. Understanding the spectrum of residential energy consumption: A quantile regression approach. Energy Policy 2010, 38, 6574–6585. [Google Scholar] [CrossRef]
  54. Wiesmann, D.; Azevedo, I.L.; Ferrão, P.; Fernández, J.E. Residential electricity consumption in Portugal: Findings from top-down and bottom-up models. Energy Policy 2011, 39, 2772–2779. [Google Scholar] [CrossRef]
  55. Kavousian, A.; Rajagopal, R.; Fischer, M. Determinants of residential electricity consumption: Using smart meter data to examine the effect of climate, building characteristics, appliance stock, and occupants’ behavior. Energy 2013, 55, 184–194. [Google Scholar] [CrossRef]
  56. Brounen, D.; Kok, N.; Quigley, J.M. Residential energy use and conservation: Economics and demographics. Eur. Econ. Rev. 2012, 56, 931–945. [Google Scholar] [CrossRef]
  57. Chen, Y.J.; Matsuoka, R.H.; Liang, T.M. Urban form, building characteristics, and residential electricity consumption: A case study in Tainan City. Environ. Plan. B Urban Anal. City Sci. 2018, 45, 933–952. [Google Scholar] [CrossRef]
  58. Quan, S.J. Energy efficient neighborhood design under residential zoning regulations in Shanghai. Energy Procedia 2017, 143, 865–872. [Google Scholar] [CrossRef]
  59. Li, C.; Song, Y.; Kaza, N. Urban form and household electricity consumption: A multilevel study. Energy Build. 2018, 158, 181–193. [Google Scholar] [CrossRef]
  60. Ko, Y.; Radke, J.D. The effect of urban form and residential cooling energy use in Sacramento, California. Environ. Plan. B Plan. Des. 2014, 41, 573–593. [Google Scholar] [CrossRef]
  61. You, Y.; Kim, S. Revealing the mechanism of urban morphology affecting residential energy efficiency in Seoul, Korea. Sustain. Cities Soc. 2018, 43, 176–190. [Google Scholar] [CrossRef]
  62. Vartholomaios, A. A parametric sensitivity analysis of the influence of urban form on domestic energy consumption for heating and cooling in a Mediterranean city. Sustain. Cities Soc. 2017, 28, 135–145. [Google Scholar] [CrossRef]
  63. Quan, S.J.; Wu, J.; Wang, Y.; Shi, Z.; Yang, T.; Yang, P.P.J. Urban form and building energy performance in Shanghai neighborhoods. Energy Procedia 2016, 88, 126–132. [Google Scholar] [CrossRef]
  64. Wilson, B. Urban form and residential electricity consumption: Evidence from Illinois, USA. Landsc. Urban Plan. 2013, 115, 62–71. [Google Scholar] [CrossRef]
  65. Rickwood, P. Residential operational energy use. Urban Policy Res. 2009, 27, 137–155. [Google Scholar] [CrossRef]
  66. Javanroodi, K.; Mahdavinejad, M.; Nik, V.M. Impacts of urban morphology on reducing cooling load and increasing ventilation potential in hot-arid climate. Appl. Energy 2018, 231, 714–746. [Google Scholar] [CrossRef]
  67. Asfour, O.S.; Alshawaf, E.S. Effect of housing density on energy efficiency of buildings located in hot climates. Energy Build. 2015, 91, 131–138. [Google Scholar] [CrossRef]
  68. Zhang, L.; Li, R. Impacts of green certification programs on energy consumption and GHG emissions in buildings: A spatial regression approach. Energy Build. 2022, 256, 111677. [Google Scholar] [CrossRef]
  69. Cao, J.; Zhou, W.; Wang, W.; Pan, X.; Jing, C.; Qian, Y. Spatially heterogeneous effect of temperature on electricity consumption in Shenzhen, China. Build. Environ. 2023, 241, 110468. [Google Scholar] [CrossRef]
  70. Heinonen, J.; Junnila, S. Residential energy consumption patterns and the overall housing energy requirements of urban and rural households in Finland. Energy Build. 2014, 76, 295–303. [Google Scholar] [CrossRef]
  71. Reames, T.G. Targeting energy justice: Exploring spatial, racial/ethnic and socioeconomic disparities in urban residential heating energy efficiency. Energy Policy 2016, 97, 549–558. [Google Scholar] [CrossRef]
  72. Makido, Y.; Dhakal, S.; Yamagata, Y. Relationship between urban form and CO 2 emissions: Evidence from fifty Japanese cities. Urban Clim. 2012, 2, 55–67. [Google Scholar] [CrossRef]
  73. Tiwari, P. Architectural, Demographic, and Economic Causes of Electricity Consumption in Bombay. J. Policy Model. 2000, 22, 81–98. [Google Scholar] [CrossRef]
  74. Bartusch, C.; Odlare, M.; Wallin, F.; Wester, L. Exploring variance in residential electricity consumption: Household features and building properties. Appl. Energy 2012, 92, 637–643. [Google Scholar] [CrossRef]
  75. Kontokosta, C.E.; Tull, C. A data-driven predictive model of city-scale energy use in buildings. Appl. Energy 2017, 197, 303–317. [Google Scholar] [CrossRef]
  76. Goldstein, B.; Gounaridis, D.; Newell, J.P. The carbon footprint of household energy use in the United States. Proc. Natl. Acad. Sci. USA 2020, 117, 19122–19130. [Google Scholar] [CrossRef] [PubMed]
  77. Boukarta, S.; Berezowska, E. Exploring the Energy Implication of Urban Density in Residential Buildings. J. Appl. Eng. Sci. 2017, 7, 7–14. [Google Scholar] [CrossRef]
  78. Li, C.; Song, Y.; Kaza, N.; Burghardt, R. Explaining Spatial Variations in Residential Energy Usage Intensity in Chicago: The Role of Urban Form and Geomorphometry. J. Plan. Educ. Res. 2019, 43, 317–331. [Google Scholar] [CrossRef]
  79. Kavgic, M.; Mavrogianni, A.; Mumovic, D.; Summerfield, A.; Stevanovic, Z.; Djurovic-Petrovic, M. A review of bottom-up building stock models for energy consumption in the residential sector. Build. Environ. 2010, 45, 1683–1697. [Google Scholar] [CrossRef]
  80. Tian, W.; Heo, Y.; De Wilde, P.; Li, Z.; Yan, D.; Park, C.S.; Feng, X.; Augenbroe, G. A review of uncertainty analysis in building energy assessment. Renew. Sustain. Energy Rev. 2018, 93, 285–301. [Google Scholar] [CrossRef]
  81. Li, Q.; Hu, J.; Yu, B. Spatiotemporal patterns and influencing mechanism of urban residential energy consumption in China. Energies 2021, 14, 3864. [Google Scholar] [CrossRef]
  82. Wang, G. Low-Carbon City Development based on Land Use Planning. Ph.D. Thesis, Technische Universiteit Eindhoven, Eindhoven, The Netherlands, 2021. [Google Scholar]
  83. Liu, X.; Wang, M.; Qiang, W.; Wu, K.; Wang, X. Urban form, shrinking cities, and residential carbon emissions: Evidence from Chinese city-regions. Appl. Energy 2020, 261, 114409. [Google Scholar] [CrossRef]
  84. Muñiz, I.; Dominguez, A. The impact of urban form and spatial structure on per capita carbon footprint in U.S. larger metropolitan areas. Sustainbility 2020, 12, 389. [Google Scholar] [CrossRef]
  85. Quan, S.J.; Economou, A.; Grasl, T.; Yang, P.P.J. An exploration of the relationship between density and building energy performance. Urban Des. Int. 2020, 25, 92–112. [Google Scholar] [CrossRef]
  86. Norman, J.; MacLean, H.L.; Kennedy, C.A. Comparing High and Low Residential Density: Life-Cycle Analysis of Energy Use and Greenhouse Gas Emissions. J. Urban Plan. Dev. 2006, 132, 10–21. [Google Scholar] [CrossRef]
  87. Sun, B.; Han, S.; Li, W. Effects of the polycentric spatial structures of Chinese city regions on CO2 concentrations. Transp. Res. Part D 2020, 82, 102333. [Google Scholar] [CrossRef]
  88. Peel, M.C.; Finlayson, B.L.; Mcmahon, T.A. Updated world map of the Koppen-Geiger climate classification. Hydrol. Earth Syst. Sci. 2007, 11, 1633–1644. [Google Scholar] [CrossRef]
  89. Graham, M.H. Confronting Multicollinearity in Ecological Multiple Regression. Ecology 2003, 84, 2809–2815. [Google Scholar] [CrossRef]
  90. Qu, Y.; Liu, Y.; Wang, W.; Cang, Y. Sustainability assessment of urban residential consumption in China megacity. Environ. Dev. Sustain. 2021, 23, 7509–7523. [Google Scholar] [CrossRef]
  91. Baker, N.; Steemers, K. Energy and Environment in Architecture: A Technical Design Guide. 2016. Available online: https://books.google.com.hk/books?hl=en&lr=&id=FJp5AgAAQBAJ&oi=fnd&pg=PP1&dq=Energy+and+Environment+in+Architecture:+A+Technical+Design+Guide&ots=6jEKrCxW6L&sig=mh58EtR4e0GTB1ZlKo11_qLOzl4&redir_esc=y#v=onepage&q=Energy%20and%20Environment%20in%20Architecture%3A%20A%20Technical%20Design%20Guide&f=false (accessed on 4 September 2023).
  92. Han, J.; Meng, X.; Zhou, X.; Yi, B.; Liu, M.; Xiang, W.N. A long-term analysis of urbanization process, landscape change, and carbon sources and sinks: A case study in China’s Yangtze River Delta region. J. Clean. Prod. 2017, 141, 1040–1050. [Google Scholar] [CrossRef]
  93. Xu, L.; Du, H.; Zhang, X. Driving forces of carbon dioxide emissions in China’s cities: An empirical analysis based on the geodetector method. J. Clean. Prod. 2021, 287, 125169. [Google Scholar] [CrossRef]
  94. Crawford, T.W. Urban form as a technological driver of carbon dioxide emission: A structural human ecology analysis of onroad and residential sectors in the conterminous U.S. Sustainbility 2020, 12, 7801. [Google Scholar] [CrossRef]
  95. Abanda, F.H.; Byers, L. An investigation of the impact of building orientation on energy consumption in a domestic building using emerging BIM (Building Information Modelling). Energy 2016, 97, 517–527. [Google Scholar] [CrossRef]
  96. Motoasca, E.; Agarwal, A.K.; Breesch, H. Energy Sustainability in Built and Urban Environments; Springer: Berlin/Heidelberg, Germany, 2019; p. 329. Available online: http://link.springer.com/10.1007/978-981-13-3284-5 (accessed on 4 September 2023).
  97. Ewing, R.; Park, K. Basic Quantitative Research Methods for Urban Planners; Routledge: London, UK, 2020. [Google Scholar] [CrossRef]
  98. Finch, W.H.; Bolin, J.E.; Kelley, K. Multilevel Modeling Using R, 2nd ed.; Routledge: London, UK, 2019. [Google Scholar]
  99. Ewing, R.; Park, K. Advanced Quantitative Research Methods for Urban Planners; Routledge: London, UK, 2020. [Google Scholar]
  100. Boslaugh, S.; Watters, P.A. Statistics in a Nutshell. 2008. Available online: https://theswissbay.ch/pdf/Gentoomen%20Library/Maths/Statistics/OReilly.Statistics.in.a.Nutshell.A.Desktop.Quick.Reference.Aug.2008.pdf (accessed on 4 September 2023).
  101. Mustafa, A.M.; Cools, M.; Saadi, I.; Teller, J. Urban Development as a Continuum: A Multinomial Logistic Regression Approach. In Proceedings of the Computational Science and Its Applications, Banff, AB, Canada, 22–25 June 2015; pp. 729–744. [Google Scholar] [CrossRef]
  102. Geraghty, D.; Mahony, M.O. Urban Noise Analysis Using Multinomial Logistic Regression. J. Transp. Eng. 2016, 142, 04016020. [Google Scholar] [CrossRef]
  103. Boyko, C.T.; Cooper, R. Clarifying and re-conceptualising density. Prog. Plann. 2011, 76, 1–61. [Google Scholar] [CrossRef]
  104. United Nations Human Settlements Programme. World Cities Report 2022: Envisaging the Future of Cities; United Nations Human Settlements Programme (UN-Habitat): Nairobi, Kenya, 2022. [Google Scholar]
  105. Aurand, A. Density, housing types and mixed land use: Smart tools for affordable housing? Urban Stud. 2010, 47, 1015–1036. [Google Scholar] [CrossRef]
  106. Moroni, S. Urban density after Jane Jacobs: The crucial role of diversity and emergence. City Territ. Archit. 2016, 3, 13. [Google Scholar] [CrossRef]
  107. Jacobs, J. The Death and Life of Great American Cities. 1961. Available online: http://www.petkovstudio.com/bg/wp-content/uploads/2017/03/The-Death-and-Life-of-Great-American-Cities_Jane-Jacobs-Complete-book.pdf (accessed on 4 September 2023).
  108. Daniels, T. Smart growth: A new American approach to regional planning. Plan. Pract. Res. 2001, 16, 271–279. [Google Scholar] [CrossRef]
Sustainability 15 15685 g001
Figure 1. Study selection.
Figure 1. Study selection.
Sustainability 15 15685 g001
Sustainability 15 15685 g002
Figure 2. The relationship between 70 eligible articles considered in this paper [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80]. Note: Credited by https://www.litmaps.com/ (accessed on 4 September 2023), the small figure shows the articles that cite Ewing & Rong [9]. Considerably, these highlighted papers might be interconnected with each other.
Figure 2. The relationship between 70 eligible articles considered in this paper [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80]. Note: Credited by https://www.litmaps.com/ (accessed on 4 September 2023), the small figure shows the articles that cite Ewing & Rong [9]. Considerably, these highlighted papers might be interconnected with each other.
Sustainability 15 15685 g002
Sustainability 15 15685 g003
Figure 3. Different case studies were taken by scholars.
Figure 3. Different case studies were taken by scholars.
Sustainability 15 15685 g003
Sustainability 15 15685 g004
Figure 4. The number of negative, positive, and non-significant impacts of density on REC by different study characteristics. Note: The Climate Zone consists of five various groups, namely Tropical (A), Arid (B), Temperate (C), Cold (D), and Polar (E). Obviously, there is not any case study in the Polar (E) zone.
Figure 4. The number of negative, positive, and non-significant impacts of density on REC by different study characteristics. Note: The Climate Zone consists of five various groups, namely Tropical (A), Arid (B), Temperate (C), Cold (D), and Polar (E). Obviously, there is not any case study in the Polar (E) zone.
Sustainability 15 15685 g004
Sustainability 15 15685 g005
Figure 5. How many times an urban form variable is used in various publications (n = 70).
Figure 5. How many times an urban form variable is used in various publications (n = 70).
Sustainability 15 15685 g005
Sustainability 15 15685 g006
Figure 6. The percentage of each dividing type.
Figure 6. The percentage of each dividing type.
Sustainability 15 15685 g006
Sustainability 15 15685 g007
Figure 7. The percentage of each method.
Figure 7. The percentage of each method.
Sustainability 15 15685 g007
Table 1. The names of leading journals in this field.
Table 1. The names of leading journals in this field.
Advances in Building Energy ResearchJournal of Environmental Management
Applied EnergyJournal of Industrial Ecology
Building and EnvironmentJournal of Planning Education and Research
EnergyJournal of Planning Literature
Energy and BuildingsJournal of the American Planning Association
Energy EconomicsJournal of Urban Economics
Energy PolicyJournal of Urban Planning and Development
Energy ProcediaLandscape and Urban Planning
Environment and Planning BRemote Sensing Applications: Society and Environment
Environmental Science and TechnologyRenewable and Sustainable Energy Reviews
Housing Policy DebateSustainable Cities and Society
Intelligent Buildings InternationalUrban Climate
Journal of Applied Engineering ScienceUrban Policy and Research
Journal of Cleaner ProductionUrban Studies
Note: The names of journals sorted alphabetically. These journals publish the most relevant and cited papers.
Table 2. The discrepancy among scholars about the impact of Urban Form and density on REC.
Table 2. The discrepancy among scholars about the impact of Urban Form and density on REC.
The Role of Urban Form
on REC
The role of urban formpublications
It affects the amount of REC[8,9,11,14,15,16,17,19,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,81,82,83,84,85,86]
It does not affect the amount of REC[10,23,71,72,73,74]
The role of density
on REC
The role of densitypublications
Positive correlation[34,35,40,54,67,75]
Negative correlation[8,9,28,30,31,32,33,36,37,38,39,41,42,43,46,47,49,50,51,52,56,57,60,64,66,68,76,77,78]
Not disclosed
(Not. Sig)		[29,45,48,53,55,59,61,62,65,69,70,71,72,73,74,82]
Note. The names of scholars sorted alphabetically.
Table 3. Comparing Urban and Rural housing types in energy purchases.
Table 3. Comparing Urban and Rural housing types in energy purchases.
UIHTEnergy Purchases (EUR/Household/Year)Average Family SizeLiving Space (m2) per HouseholdEnergy Purchases (EUR/Person/Year)Energy Purchases (EUR/m2/Year)
UrbanDetached house17332.75128.0630.213.5
Row-/terraced house12452.1980.2568.515.5
Apartment building8661.6558.3524.814.9
RuralDetached house16882.62124.8644.313.5
Row-/terraced house10551.7662.8599.416.8
Apartment building7761.5755.0494.314.1
Note. Table 2 and Table 4 in Heinonen, J., & Junnila, S. (2014) [70].
Table 4. The study characteristics of the 25 out of the 70 articles analyzed in this review.
Table 4. The study characteristics of the 25 out of the 70 articles analyzed in this review.
ResearchersNumber of Density IndicatorsEnergy TypeUnits of MeasurementCase StudyClimate ZoneMethodologyRelevancyReliabilityy_Publish
Ewing [9]2Cooling, Heating, and Annual energy usePer householdUSAvariedMultilevel Modelingyesdataset2008
Glaeser [28]1Annual energy usePer householdUSAvariedMultiple regressionyesdataset2010
Holden [30]2Heating energy usePer CapitaGreater Oslo RegionDfbMultiple regressionyesdataset2005
Jones [31]2Annual energy usePer householdUSAvariedMultiple regressionnodataset2014
Lee [8]2Cooling and HeatingPer householdUSAvariedMultilevel Modelingnoestimated2014
Kaza [53]2Cooling and HeatingPer householdUSAvariedQuantile regressionnodataset2010
Reames [71]1Heating energy usePer square meterKansasCfaMultiple regressionnoestimated2016
Sajadian [38]1Annual energy usePer householdKermanBSkDescriptive Analysisnodataset2022
Wiedenhofer [36]1Annual energy usePer householdAustraliaBWhMultiple regressionnodataset2013
Tso [35]2Annual energy usePer householdUSAvariedMultilevel Modelingnodataset2014
Huang [34]2Annual energy usePer householdTaiwanCwbQuantile regressionnoestimated2015
Rode [33]3Heating energy usePer square meterParis, London, Berlin, and IstanbulCPearson correlationyesestimated2014
Min [37]1Cooling and HeatingPer householdUSAvariedMultiple regressionnodataset2010
Heinonen [70]2Annual energy usePer household/Per square meter/Per capitaFinlandDfcDescriptive Analysisnoestimated2014
Kontokosta [75]3Annual energy usePer square meterNew YorkCfaMultiple regressionnodataset2017
Makido [72]1Annual energy usePer CapitaJapanDfbMultiple regressionyesdataset2012
Belaid [41]2Annual energy usePer householdEgyptBWhQuantile regressionnoestimated2021
Voskamp [44]2Annual energy usePer householdAmsterdamCfbMultiple regressionyesdataset2021
Berrill [42]1Cooling and HeatingPer householdUSAvariedMultiple regressionnodataset2021
Rokseth [43]1Annual energy usePer square meter/Per capitaOslo, TrondheimDfb, DfcDescriptive Analysisnodataset2021
Wang [82]3Annual energy usePer square meterEindhovenCfbMultiple regressionnodataset2021
Zhang [40]2Annual energy usePer householdChangxingCfaSpatial autocorrelationyesdataset2021
Ahmadian [39]1Heating energy usePer square meterLondonCfbSimulationnoestimated2021
Wiesmann [54]3Annual energy usePer CapitaPortugueseCsaMultiple regressionnoestimated2011
Table 5. Multinomial logistic regression outcome.
Table 5. Multinomial logistic regression outcome.
KERRYPNX[Positive Correlation][Not. Sig]
VariableOdd Ratiop-ValueOdd Ratiop-Value
Num_indicators1.840.2270.28 *0.017
Type [annual]
  • Cooling
0.11 ***<0.0010.36 *0.031
  • Heating
0.09 ***<0.0011.550.426
Units [m2]
  • Per capita
1.540.1800.570.223
  • Per household
0.34 **0.0021.360.454
Climate Zone [Arid (B)]
  • Temperate (C)
>10 ***<0.0017.90 ***<0.001
  • Cold (D)
0.00 ***<0.0017.31 ***<0.001
  • Varied
>10 ***<0.0010.760.357
Methodology [Descriptive Analysis]
  • Multilevel Modeling
0.01 ***<0.0010.750.451
  • Multiple Regression
0.00 ***<0.0010.850.799
  • Pearson Correlation
0.00 ***<0.0010.540.075
  • Quantile Regression
0.00 ***<0.0011.040.901
  • Spatial Autocorrelation
0.00 ***<0.0010.510.226
Relevancy [No]
  • Yes
1.460.5060.36 *0.046
Reliability [Dataset]
  • Estimated
3.57 **0.0054.18 **0.002
Y-publish1.03 ***<0.0010.89 ***<0.001
Lat0.980.6440.970.135
Long1.010.0920.990.076
Observations120
log-Likelihood −72.524
AIC221.047
Note. The dependent variable is the impact of density on REC, and it has three different levels. The reference level is the “Negative correlation”. The significant correlations are bold. * Significance p < 0.05, ** Significance p < 0.01, *** Significance p < 0.001.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Narimani Abar, S.; Schulwitz, M.; Faulstich, M. The Impact of Urban Form and Density on Residential Energy Use: A Systematic Review. Sustainability 2023, 15, 15685. https://doi.org/10.3390/su152215685

AMA Style

Narimani Abar S, Schulwitz M, Faulstich M. The Impact of Urban Form and Density on Residential Energy Use: A Systematic Review. Sustainability. 2023; 15(22):15685. https://doi.org/10.3390/su152215685

Chicago/Turabian Style

Narimani Abar, Sina, Martin Schulwitz, and Martin Faulstich. 2023. "The Impact of Urban Form and Density on Residential Energy Use: A Systematic Review" Sustainability 15, no. 22: 15685. https://doi.org/10.3390/su152215685

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop