Scientometric Analysis on Climate Resilient Retrofit of Residential Buildings

Abstract

This study aims to understand the impacts of climate change and extreme climate events on residential buildings and explore how existing buildings can be adapted to resist these negative impacts. A bibliometric and scientometric analysis was conducted on resilient residential retrofits to highlight the prevalent themes, critical directions, and gaps in the literature, which can inform future research directions. The resilient residential retrofit publications from 2012 to 2023 were retrieved and analyzed using text-mining software. In all, 4011 publications and 2623 patents were identified. The analysis revealed an average annual publication growth rate of 11%, indicating increasing interest in resilient residential retrofits. Four central topics were explored specifically throughout the study, as they are known to be the most prevalent climate risks for residential buildings: Overheating, Flooding, Wind, and Wildfires. The research trends analysis reveals that emerging interests in resilient residential retrofit encompass nature-based solutions, energy efficiency, thermal comfort, microclimates, durability, post-disaster recovery, and extreme events. Nearly half of the publications reference urban context and over one-third mention costs. The building envelope is the most frequently discussed housing component. Although energy retrofit was not the primary focus of this study and was not specifically searched for, energy concerns were still prevalent in the dataset, highlighting the critical importance of energy efficiency and management in resilient residential retrofits. The analysis of R&D momentum revealed several research gaps. Despite high growth rates, there are low publication rates on key topics such as durability, holistic approaches, microclimates, nature-based solutions, and traditional homes, to name a few. These areas could benefit from further research in the context of climate-resilient residential retrofits. Additionally, the analysis indicates a lack of publications on cross-themed research specific to rural and suburban settings. There are also few studies addressing combinations of themes, such as overheating in high-rise buildings, wildfires in Nordic climates, and flooding risk in smart homes within the scope of resilient residential retrofits. The United States leads in publication output, followed by China and the UK, with China dominating the patent landscape. This scientometric analysis provides a comprehensive overview of the research landscape in resilient residential retrofit, systematically maps and analyzes the vast amount of research output, and identifies the key trends and gaps, enabling us to see a type of quantitative snapshot of the research in a field at a certain point in time and thus providing a unique point of view. This study helps stakeholders prioritize efforts and resources effectively for guiding future research, funding decisions, informing policy decisions, and ultimately enhancing the resilience of residential buildings to climate-related challenges.

1. Introduction

The changing climate calls for innovative building design. Determining evidence-based building methods to resist the increasing occurrence of natural hazards is a complex challenge with regional imperatives. Nonetheless, existing buildings are expected to constitute the majority of building stock for years to come, highlighting the relevance of residential retrofits. Resilient residential retrofitting is the concept of upgrading existing buildings to decrease the buildings’ vulnerabilities to physical climate risks. Resilient retrofits include structural hardening (e.g., wind-resistant roofs and windows), resource conservation (e.g., enhanced insulation), and energy supply (e.g., backup generation). The climate risks can be acute (e.g., hurricanes) or chronic (e.g., extreme heat). There are different resilient retrofit strategies for every major hazard, and ideally, occupants should plan to make their homes resilient to multiple hazards according to their regional situation. Best practices call for comprehensive resilient retrofit planning with incremental implementation.

In order to better understand the impacts of climate change and extreme climate events on residential buildings as a whole and how the building industry can enable the existing buildings to resist the negative impacts of climate change, a scientometric study was conducted.

Scientometrics is the quantitative study of scientific output, such as publications, patents, and citations. It can be defined as metaresearch or research on research. It is a validated and widely recognized method for analyzing research trends, evaluating scholarly impact, and assessing the dynamics of knowledge production. It employs statistical analysis, including bibliometric techniques, to provide evidence-based metrics, which are pivotal for understanding scientific progress, informing policy and funding decisions, or establishing R&D priorities. Scientometric methods are especially insightful for analyzing large bodies of evidence that could not be analyzed otherwise while providing a bird-eye view of a research field. As science becomes increasingly interdisciplinary and complex, scientometrics is a valuable tool to ensure its ongoing credibility and societal relevance. Its validity as a scientific method is supported by the rigorous frameworks outlined in foundational works like Porter, Zhang, and Newman (2024) [1], Lewis, Templeton, and Luo (2007) [2], Chen (2006) [3], and Serenko and Bontis (2004) [4].

Scientometric studies systematically map and analyze research output, providing an evidence-based macro-view of a field. Known limitations of scientometric studies include the quality of each of the publications included in the dataset since the analysis uses the metadata and is not based on the full text of the references, disparities in citation patterns between disciplines, an overemphasis on quantitative metrics, and a bias toward English references. By analyzing thousands of references and looking at different indicators at once, the potentially poorer data should have a negligible weight in the study. Moreover, even a poorly designed study would nonetheless help to show the trends of research interests in the field, even if it had a minimal contribution to the field. Disparities in citation patterns among research disciplines are accounted for by using field-weighted citation impact (FWCI) instead of the number of citations. Although scientometric studies rely on quantitative data, the examples of findings mentioned throughout the study help to present a story outside of numbers only. As for the bias towards English publications, this study is not limited to English references, although the metadata needed to be available in English to be included. The English bias in science is a necessary limitation in the context of this scientometric study, and further research in specific geographical areas should make sure to include data in the relevant languages and publication paths outside of scientific journals.

This study identifies and classifies international research domains in the field of resilient residential retrofit as they pertain to resistance to weather events, including the impacts of climate change and natural disasters. The study analyzes the research trends and related advances in the understanding of resilient retrofit for residential purposes during the past decade (2012–2023). It is not a critical evaluation of the methods or theories being reviewed but rather reports on trends in the literature based on an analysis of the metadata associated with the bibliographic references. There are a few recent scientometric studies already published in the field of residential building retrofit, but none explore the overall dimensions of the research field. They focus on one dimension of the research field, such as retrofitting methods, environmental performance of buildings, or a specific climate change impact (earthquake, flooding, etc.) [5,6,7].

1.1. Methodology

Literature searches were conducted primarily in Scopus and Questel-Orbit’s FamPat worldwide patents database in February 2023. Additionally, internet searches were conducted to gather information that is not usually published through standard distribution channels or indexed in commercial databases.

To identify relevant resilient residential retrofit publications and build an initial dataset, a search combining the “resilience” concept with the “climate” concepts was conducted and combined with a search on “residential housing” and “retrofit”. Different keywords that imply a similar meaning were also used in the search to maximize the capture of relevant information. For example, “modernize”, “upgrade”, “renovation”, “alteration”, “restoration”, etc., were used in parallel to the word retrofit. Similarly, “housing”, “townhouse”, “dwelling”, “apartment”, and several similar words were used in parallel to “residential housing”. Techniques to reduce or mitigate the environmental impacts of global warming, such as emission reductions, net zero materials, energy-saving measures, etc., were out of scope and excluded from the results. The search strategy used in the Scopus database is available is available upon request.

Literature search terms were vetted by experts in the field for accuracy and were searched in the title, abstract, and index/author keywords fields. The data collection included publications and patents from 2012 to early 2023.

Literature and patent references were imported into the VantagePoint software (version 15) for cleaning and analysis. VantagePoint enables the creation of various groupings, statistical analyses, matrices, graphs, and cross-correlations to analyze the data and profile the activities of the major players. Additionally, other analytical tools such as Tableau (version 2022.4) and Gephi (version 0.10) were utilized to generate graphs based on statistical operations performed in VantagePoint.

References were manually reviewed to assess their relevancy. As such, irrelevant publications using, for example, construction terms without being related to the field (e.g., “the construction of the dataset”, “building on the findings”) were excluded from the study. In order to be part of the final dataset, the publications had to mention the three core concepts of the study: resiliency to climate events, residential buildings, and retrofitting in the title, abstract, or index or author keywords.

The references were categorized by topics using natural language processing (NLP) to the titles, author keywords, abstracts, and regexes (regular expressions or sequences of characters used to find matching patterns in a text) to capture all the occurrences of a concept (including different spelling, pluralization, and synonyms). Manual verification was then conducted to remove irrelevant terms caught by the regexes, and the relevancy of the topic groups was assessed by an expert in the field. The complete list of topics is available upon request.

Although topic groups were carefully crafted from readings, discussion with experts, and the analysis of the NLP list of terms, some themes could be underrepresented in the analysis. In this sense, solutions to retrofit non-residential buildings could very well be applicable to residential buildings as well. Such cases were not caught by the data collection. The software VantagePoint also enables the cleaning of the data by identifying duplicated records and harmonizing the countries and affiliations that are not systematically indexed the same way in the Scopus database. It is also used for analysis by creating matrices to put one field against another (e.g., how many publications per year for each topic, how many times two topics are discussed in the same publications, etc.) and by calculating the average field-weighted citation impact (FWCI), for example, for each topic.

The primary climate hazards that residential buildings encounter are Overheating, Flooding, Wind, and Wildfires. These four central topics were used throughout the scientometric study to illustrate the connections between them and other research topics; each also has its own section featuring an in-depth review of the literature.

1.2. R&D Momentum Indicator

The R&D momentum indicator is designed to identify rapidly emerging topics with relatively few publications or patents [8]. The challenge in identifying such topics lies in the volume, as their rapid growth and evolution are overshadowed by the high number of established subjects. Specifically, an “emerging” topic is characterized not only by a sharply rising trend line but also by a small footprint in the domain of interest. This small footprint often causes emerging subjects to be overlooked until their disruptive impacts become evident. In the momentum indicator, the two parameters are (1) growth rate, which is the slope of a subject’s trend line (horizontal axis), and (2) volume, which is the cumulated total number of publications (vertical axis). The growth rate is determined by the number of publications per year, and year-to-year variations influence a topic’s position in the R&D momentum.

After separating growth rate and volume, a two-dimensional coordinate system can be used to plot a group of subjects. To achieve this, the two parameters were normalized using z-scores. The normalization process converts values in different units into a common measure based on standard deviation, thereby standardizing the variations for each parameter. The resulting four-quadrant visualization offers a structured view of the relative positions of these subjects within the group.

1.3. Datasets

1.3.1. Publications Dataset

To address the key questions of this study, a comprehensive literature search for peer-reviewed publications on resilient residential retrofit was conducted in Scopus in February 2023, covering the years 2012 to 2023, including the references already published in early 2023. It should be noted that 2022 was not fully indexed when the dataset was created. A year is usually considered completely indexed by Scopus in June of the subsequent year to ensure that all relevant publications are included; this study analyzed publications on the retrofitting or renovation of any type of residence to create or improve building resilience to climate change and its related climatic events. Although data from 2022 are incomplete, the observable trends from the bulk of publications captured in 2022 have high odds of remaining the same after including the publications that were not indexed at the moment of data collection. The term “building” was not searched on its own because it generated significant false results, including publications where “building” was used as a verb instead of a noun to describe a place where people live. In all, 4011 publications were retrieved to create the following dataset:

• Articles: 2514 (67%);
• Conference papers: 1014 (25%);
• Book chapters: 251 (6%);
• Reviews: 172 (4%);
• Books: 40 (1%);
• Six editorials, five short surveys, four data papers, four notes, and one letter (less than 1% each).

The publication metadata (including title, author, affiliation, keywords, abstract, etc.) were imported into text-mining software for cleaning, topical grouping, and analysis.

1.3.2. Topic Analysis of Patent Families

A dataset of patent families on resilient residential retrofits was created in the FamPat worldwide patents database from Questel-Orbit in February 2023, covering the years 2012 to 2022. A patent family is defined as a group of substantively equivalent depositions made by the same applicant within a specified time period. Each patent family essentially describes one invention and includes a priority patent (the first deposition) along with any other applications made in other countries. After reviewing for relevancy, 2623 patent families were kept for the 2012 to 2022 time period. There is an 18-month delay between a patent application and its publication; therefore, 2021 and 2022 are incomplete. It is an irremediable blind spot when analyzing recent patent activity trends. Even if the interpretation of 2021 and 2022 data should be reviewed with caution, data from the whole analyzed time frame highlight the trends in patent activities.

1.3.3. Temporal Analysis

Figure 1 illustrates the annual distribution of the resilient residential retrofit scholarly peer-reviewed publications and patent families retrieved in February 2023. Since 2012, the publication count has had a positive growth trend and has increased, on average, by 11% per year (compound annual growth rate, CAGR), mainly due to the high growth seen in 2017 and 2018 (+19% each). Although the reasons leading to this increase in publications are probably multifold, one of them could be the result of initiatives that stemmed from the publication of the 2015 Paris Agreement by the United Nations, Article 7 in particular. Effectively, Article 7 is about planning adaptation and resilience to the effects of climate change [9]. Even if 2022 has a higher number of publications than 2021, this number is very likely underestimated. Effectively, the search in the database Scopus was conducted in February 2023, and a year is usually considered completely indexed in Scopus in June of the following year. Since data for 2023 are considerably incomplete, 2023 is not included in Figure 1. From 2012 to 2021, the patent count has increased by 15% per year. Given that there is generally an 18-month delay between publishing patent applications and their earliest filing date, the years 2021 and 2022 are considered incomplete in the patent dataset and can at least partially explain the drop visible in 2022. This delay in patent application could also at least partially explain why the sharper increase in patents in 2019 (28%) is visible later than in the publications (19% increase in 2017 and 2018).

2. Results and Discussion

2.1. Topic Categories

To gain insight into the dataset’s content (4011 publications), words and phrases were extracted from the article titles, abstracts, and keywords. These terms were compiled into a separate field containing a total of 170,711 keywords. Using text-mining techniques, the terms were then classified into 156 topic groups. These groups cover 99% of the dataset and represent the major subjects of interest in the research field of resilient residential retrofit. The topics were then manually classified into the five categories listed and defined in

Table 1. Topic categories.

Category	# of Topics	Share of Dataset	Definition
Problems and Issues	45	99%	Challenges associated with climate events to retrofit buildings, e.g., disasters, corrosion, and heat islands.
Housing	47	93%	Types of buildings, location of the building, and building components, e.g., building envelopes, urban settings, and high rises.
Solutions	36	65%	Technological and natural solutions to tackle challenges and issues, e.g., AI, nature-based solutions, and sensors.
Tools	20	71%	Non-technological tools to support decision-making and planning, e.g., building regulations, incentives, and risk assessments.
Resilience qualities	8	53%	Built environment resiliency qualities. Inspired by the classification presented in Al-Humaiqani, M. M., and Al-Ghamdi, S. G. [10], e.g., flexibility, redundancy, and inclusion.

. The complete list of topics is available upon request.

Grouping subjects into broader categories allows for the comparison of similar themes, facilitating topic analyses and the application of statistical techniques to identify new and emerging research areas. Each topic can only be assigned to one category even if there are topics that could fit into more than one (e.g., Air conditioning is assigned to Housing but could also be considered a Solution).

Most resilient residential retrofit publications mention at least one Problems and Solutions topic (99%) and one Housing topic (93%). The Resilience qualities were taken from a 2022 paper by M.M. Al-Humaiqani and S.G. Al-Ghamdi [10], and they are mentioned in slightly more than half of the publications (53%).

2.2. Topic Analysis of Patent Families

The following section reviews recent patents in the field of resilient residential retrofit. A search strategy focusing on resilient residential retrofit was applied in the FamPat worldwide patents database from Questel-Orbit. The search resulted in 2623 patent families from 2012 to 2022 (earliest application year). China is the priority country for 50% of the patent families, followed by the United States, with about 21% of the patent families. There are 13 patents that have Canada as a priority country. The patent families dataset was manually reviewed to identify patent families in the scope of the study. Figure 2 shows the top topics in the patent families dataset. The top topics are mainly from the Housing and Problems and Issues categories. There are 1472 patent families discussing Water, which represent 56% of the dataset. Roofing is the second most frequent topic, with 51%, followed by Urban settings (50%) and Houses and Interior walls (49% each).

2.3. Topic Analysis of Publications

Peer-reviewed publications are classified by broad subject areas using the All Science Journal Classification system (ASJC) available in Scopus. The ASJC system categorizes journals according to their main topic of interest. The ASJC comprises 27 subject categories, which are, in turn, divided into 364 subcategories. This analysis is limited to publications indexed in Scopus. The main subject area found in the publication dataset is Building and Construction (25%), followed by Civil and Structural Engineering (22%), Geography, Planning, and Development (15%), and Renewable Energy, Sustainability, and the Environment (14%).

Elsevier and allied researchers have developed the SciVal Topic Prominence model, which assigns a SciVal Topic and a SciVal Topic Cluster to each publication indexed in Scopus based on their direct citation network and a clustering algorithm. SciVal is a research performance assessment tool that allows analysis of the data from Scopus. These SciVal Topics are defined as a collection of documents with a common intellectual interest or research problem. Thus far, over 94,000 topics have been included in the model. The SciVal Topics are multidisciplinary and can be large or small, new or old, growing or declining. Related SciVal Topics have been merged together to create the SciVal Topic Clusters. There are 1500 SciVal Topic Clusters, and each topic is assigned only one cluster.

Table 2. Top SciVal Topic Clusters.

SciVal Topic Cluster	# Resilient Residential Retrofit Pubs	Share of Resilient Residential Retrofit Pubs
Building; Air Conditioning; Ventilation	916	23%
Disasters; Floods; Risks	718	18%
Roofs; Heat Island; Buildings	242	6%
Electricity; Energy; Economics	146	4%
Particulate Matter; Air Pollution; Air Pollutants	117	3%
Wind; Wind Stress; Wind Effects	91	2%
Stormwater; Storm Sewers; Rainwater	75	2%

presents the most frequent SciVal Topic Clusters found in the resilient residential retrofit dataset. Ventilation, disasters, roofs, heat islands, energy, air pollution, wind, and stormwater are the research themes covered by these Topic Clusters and are the most common research themes found in the publication dataset according to this model. Resilient residential retrofit publications account for a very small share of the overall worldwide publications in these Topic Clusters, from 1.50% to 0.12%. These are, therefore, much larger areas of research in which resilient residential retrofit research accounts for a very small portion.

2.4. Top Topics in Publications

Figure 3 displays the top 15 topics found in the dataset and their corresponding number of publications, along with the percent of the dataset they cover (share). Urban settings is the most frequently occurring topic (46% of the dataset), followed by Costs (36%) and Risks (32%). Costs, Risks, and Effectiveness are the most prominent topics in the Problems and Issues category, while Urban settings, Houses, and Building envelopes are the most prominent topics in the Housing category. Reflectivity is the only topic in the Resilience qualities category that appears in the top 15.

It is important to note that this study can only analyze the topics’ patterns of occurrence and cannot comment on the context in which the topics are used. For example, while there is a clear interest in the resilient residential retrofit research on Risks, without reading the Risks publications, it is impossible to know in what way the term is being used. The publications may discuss the risk of flooding in a certain area, the risks involved in adding a certain retrofit to a building, or the building occupant’s willingness to take risks.

Figure 4 lists the topics with the highest field-weighted citation impact (FWCI) average (note that the FWCI analysis was limited to topics with at least 20 publications). The FWCI is a normalized citation metric that indicates how the number of citations received by a publication compares with the average number of citations received by all other similar publications in the Scopus database. An FWCI of 1.00 means that the publication performs just as expected for the global average; more than 1.00 means that the publication is cited more often than expected according to the global average. The FWCI is an indicator that can be used as a proxy to assess research impact or research excellence. Overall, the publications in the resilient residential retrofit field have an FWCI score of 1.16, which means that they are cited 14% more than similar publications (i.e., from the same area of research, document type, and publication period).

A third of the highly cited topics are from the Housing category, including Bricks, Reinforced concrete, Metals, Structural components, and Compressive strength. The same goes for the Solutions category, which represents one-third of the highly cited topics with Nature-based solutions, Energy storage, Living roofs, Emission control, and AI. The top-cited topic in the dataset is Bricks (FWCI of 3.61). Even though there is a small number of topics related to specific building materials (Wood, Reinforced concrete, Metals, Glass, Bricks, and Stones), it is interesting to note that three out of the six are among the topics with the highest FWCIs.

2.5. Topic Co-Occurrence Analysis in Publications

Figure 5 displays a cluster map of topics based on a topic autocorrelation matrix using cosine similarity. This cluster map was generated with Gephi software (version 0.10), utilizing an autocorrelation matrix that shows a bi-directionally normalized correlation percentage (0% to 100%) to compare the relationship and overlap between two topics. The topics are represented by nodes (bubbles), which are color-coded by cluster (clusters were assigned statistically through a modularity analysis using the Leiden algorithm (best for text-based terms)) and sized according to the number of underlying publications. The thickness of the lines between nodes indicates the degree of correlation between subjects (i.e., thicker lines signify stronger correlations). Nodes are clustered and positioned based on their correlation with other subjects (attraction/repulsion) and gravity (central, highly connected nodes move to the center, while less connected nodes are pushed to the periphery). Nodes within clusters tend to be close together, while nodes in separate clusters are farther apart. To increase visual clarity and to highlight only the most important relationships, the coefficient of correlation has been filtered to a minimum of 25% and includes 108 topics. In all, Urban settings, Costs, Risks, Houses, and Effectiveness are the topics with the most connections to the other topics in the graph (in fact, these topics are connected to all other topics with one or two exceptions, notably Effectiveness).

Using a social network analysis clustering algorithm, the topics found in the resilient residential retrofit literature were divided into four distinct research themes (i.e., clusters) with strong co-occurrence relationships:

• The Housing and occupant comfort cluster (green):

Includes: Thermal comfort, Insulation, Cooling, Apartments, Architecture, Building envelope, Energy efficiency, Health, Investments, Ventilation, Passive houses, Winter, and Living roofs.

• The Bricks and mortar cluster (orange):

Includes: Compressive strength, Freeze–thaw, Durability, Bricks, Materials, Metals, Corrosion, Reinforced building, Concrete buildings, and Beams.

• The Disaster resilience cluster (purple):

Includes: Urban settings, Integrated solutions, Water management, Landscaping, Droughts, Erosion, Flooding, Vulnerabilities, Damages, Disasters, Decision-making, and Resourcefulness.

• The Sustainability and technologies cluster (blue):

Includes: Sustainability, Sustainable development, and Technologies.

2.6. R&D Momentum in Publications

While the volume and impact of publications in topic groups and cluster maps offer some insight into research activity levels, these methods alone do not fully capture the momentum of a topic compared to others in the same dataset. In other words, the number of records does not necessarily indicate which topics are “hot”, fading, or emerging. To determine momentum in the literature, we look at the relative velocity of each group. It involves plotting the standard deviation of standardized publication counts and velocity (rate of publication increase) on two axes. Nodes to the left of the Y-axis intersection have below-average velocity, and those below the X-axis have relatively smaller publication counts. A third dimension is added by sizing nodes based on the total number of underlying publications. Even a small node in the lower-right quadrant can be significant, as emerging topics often start small but gain research attention and increase in velocity.

The four quadrants are defined as follows:

• Established Subjects (top left): topics with a high number of publications but experiencing negative growth.
• Hot Subjects (top right): topics with a high number of publications and high acceleration.
• Emerging Subjects (bottom right): topics with a low number of publications but high acceleration
• Disappearing/Brand New Subjects (bottom left): topics with a low number of publications and negative growth.

The R&D momentum analysis incorporates a growth dimension alongside the previously discussed volume dimension to identify topics that are emerging or declining within the studied domain. It is important to note that nodes in the Emerging or Brand New quadrants may not necessarily represent original or novel subjects. In fact, these topics might be quite mature but are gaining increased research interest due to factors like solving previously intractable problems or breakthroughs in supporting technologies or processes. The results of the R&D momentum analysis should be interpreted relative to the domain under study, not in absolute terms. The position of topics in the R&D momentum analysis is relative to all other topics. Therefore, publications on the same topic in a slightly different field or analyzed within a different timeframe could be positioned differently.

The Established and Hot quadrants reflect prevalent themes in the literature; they have many publications and are well researched. The Hot quadrant shows topics with a high acceleration, showing the actual critical directions in the analyzed time frame. The Emerging quadrant can hint at the upcoming critical directions or gap-filling in the field as the topics have a high acceleration without a significant number of publications yet. The Disappearing/Brand New quadrant shows topics with potential early signals of future directions or areas where gaps could be filled, with the possibility of them fading out of interest.

The R&D momentum model was applied to resilient residential retrofit topics, Encompassing a total of 156 topics. Figure 6 illustrates the distribution of all topics within the dataset across the indicator’s four quadrants. Overall, 18 topics (12% of the topics) are plotted in the Hot quadrant, and 41 topics (26%) are plotted in the Emerging quadrant. Due to the large number of topics, not all node labels are clearly visible on the complete graph, especially those in the Established, Brand New, and Emerging quadrants.

The Emerging topics with the highest growth rate (i.e., above 1.5) are: Microclimates, Nature-based solutions, Thermal retrofits, Windows and doors, Energy storage, Net-zero buildings, Holistic approaches, AI, Durability, Drainage, Traditional houses, Glass, Post-disasters, and Extreme events. The Hot topics with the highest growth rates are Historic buildings, Cooling, and Energy retrofits. The Hot and Emerging topics with the highest growth rates, when viewed together, offer a window into the top interests in the past ten years and show a willingness to tackle residential resiliency for different types of buildings in sustainable and energy-efficient ways.

In summary, the categories are well spread across the four quadrants, with none having most of their topics in a single category. Problems and Issues have an equal share between Established and Emerging topics that are almost as much as the topics in Disappearing/Brand new. Most Solutions topics are almost equally distributed between Emerging and Disappearing/Brand New, which could indicate recent innovations in retrofit solutions.

2.7. Current and Emerging Topics

The following sections analyze the resilient residential retrofit research topics by category.

2.7.1. Housing

Figure 7 shows the top 15 Housing topics identified in this study and their share of the dataset. Close to half of the dataset mentions Urban settings (46%), the dataset’s top topic. Other settings, including Rural settings (5%) and Suburban settings (1%), do not occur frequently in the dataset and might represent a gap in the resilient residential retrofit literature. In both Rural settings and Suburban settings, the strongest co-occurrence is with Urban settings. The topics Costs, Risks, Disasters, Reflectivity, and Effectiveness rank high in all settings, suggesting concurrent interests in all types of settings. Next, two topics relate to building type (Houses, 31%, and Historic buildings, 9%), and six relate to materials and building parts (Building envelopes, 21%; Materials, 17%; Infrastructures, 11%; Roofing, 11%; Insulation, 10%; and Interior walls, 10%). As one-third of the top Housing topics pertain to occupant comfort, this seems to be an area of focus in resilient residential retrofit research (Comfort, 15%; Heating, 15%; Ventilation, 13%; Cooling, 9%; and Thermal comfort, 8%).

Figure 8 presents the R&D momentum filtered to show only those topics in the Housing category.

Building envelopes, Comfort, Heating, Ventilation, Insulation, Interior walls, Historic buildings, Cooling, and Thermal comfort fall in the Hot quadrant. The topic of Building envelopes is mentioned in 836 resilient residential retrofit publications and is positioned the highest on the vertical axis in the Hot quadrant. It co-occurs frequently with Energy efficiency (46%), Energy usage (36%), Effectiveness (34%), and Air quality (33%). In terms of topics of interest, Building envelopes is discussed along with Overheating in 13% of the dataset, Wind in 11%, and Flooding in 6% of the dataset; additionally, two publications mention Wildfires. The second publication with the highest FWCI of the dataset (9.79) is the Building envelopes topic, which discusses retrofitting with external green walls to mitigate urban heat islands, with a daytime temperature difference of up to 8.4 °C noticed during observation [11]. Research from the University of South Australia also shows the potential of vegetated roofs to not only provide a significant cooling effect in summer time but also to function as an insulation layer to keep buildings warmer in the winter and reduce energy demand [12].

Historic buildings are mentioned in 375 resilient residential retrofit publications, and the topic has the highest growth rate in the Hot quadrant. It co-occurs the most with Costs (32%), Air quality (30%), Risks (30%), and Reflectivity (28%). In terms of topics of interest, Flooding has a 14% co-occurrence with Historic buildings, and Wind has 5%. Wildfires appear in six Historic buildings publications and Overheating is in four Historic buildings publications. A publication with a very high FWCI (9.10) from the University of Rome La Sapienza (Italy) discusses retrofitting historical buildings with solar technologies without compromising the original typological and functional characteristics [13]. Other retrofitting of historic buildings includes insulation of brick walls with vacuum insulation panels, which have been studied in combination with resistance to driving rain [14].

There are 138 resilient residential retrofit publications in the dataset discussing Windows and doors, the topic with the highest growth rate in the Emerging quadrant. Half of the topic’s publications also mention Energy efficiency. Other top co-occurrences with Windows and doors are Effectiveness (38%), Air quality, and Costs and Energy usage (36% each). For the topics of interest, 19% discuss Overheating, 7% discuss Flooding, 5% discuss Wind, and no publications discuss both Windows and doors and Wildfires. Publications in Windows and doors include how window opening behavior affects indoor air quality in renovated wooden houses in Norway in order to motivate homeowners to renovate their ventilation systems [15]. Also, in Norway, the Norwegian University of Science and Technology found a higher risk of overheating when buildings were retrofitted to higher energy standards due to airtightness, which was too good [16].

Glass has 61 publications in the dataset and co-occurs the most with Investments (34%), Costs (34%), and Air quality, as well as with Energy usage and Energy efficiency (33% each). Also, from the Emerging quadrant, Nordic climate has 67 publications, which are primarily concerned with efficiency and energy, co-occurring with Energy efficiency (46%), Energy usage (43%), Air quality (42%), and Effectiveness (37%). Nordic climate publications discuss Overheating in 21% of the publications, Wind and Wildfires in 6% each, and Flooding in one publication. The University of Perugia, Italy, published a study about retro-reflective materials, including base ceramic tiles coated with glass beads, with the potential to reflect the striking energy backward and reduce the thermal energy kept inside the urban canopy [17]. Still in the retro-reflective area, the Tokyo University of Science, Japan, discussed preventing heat islands with a solar heat-shielding film, as transparent as glass, that can be applied to existing buildings [18].

2.7.2. Problems and Issues

The 45 Problems and Issues topics identified in the dataset cover 99% of the resilient residential retrofit publications. Figure 9 shows the top 15 topics from the category. The top three are Costs, with 36% of the dataset, Risks (32%), and Effectiveness (31%). Effectiveness excludes Energy efficiency, which is categorized under Solutions. Close to one-third of the top 15 Problems and Issues topics are climatic conditions or events, such as Disasters, Air quality, Water, Flooding, Stormwater, and Earthquakes. Sustainability also ranks high at sixth and is mentioned in 22% of the dataset. It is also interesting to note that, ranked 15th, Health is mentioned in 11% of the dataset, potentially implying interest in research related to occupant health in resilient residential retrofit research.

Figure 10 presents the R&D momentum filtered to show only those topics in the Problems and Issues category.

Effectiveness, Sustainability, and Flooding all plot in the Hot quadrant, with Effectiveness very close to the Established quadrant. Microclimates shows the highest growth rate by far and has 85 publications in the dataset, 75% of which also mention Urban settings and often co-occur with Heat islands (56% overlap). The publications on both topics (Microclimates and Urban settings) mostly discuss cooling strategies such as vertical greenery systems and green roofs, as well as nature-based solutions such as trees, reflective walls, and cooling materials. In terms of topics of interest, the co-occurrence of Microclimates with Overheating and Wind is 12%, Flooding is 4%, and none with Wildfires.

There are thirteen topics in the Emerging quadrant, with Microclimates, Durability, Post-disasters, and Extreme events all showing a high growth rate in the dataset (indicating they have growth rate values above 1.5). Durability is mostly discussed along with Materials (61%), Building envelopes (41%), and Costs (34%). In terms of topics of interest, 12% of the Durability publications mention Flooding, 8% mention Wind, and none mention Overheating or Wildfires. A publication with a very high FWCI (9.89) is among the Durability group. It is a review paper about using fiber-reinforced polymer (FRP) to strengthen reinforced concrete structures (Reinforced concrete is in the Housing category and plots in the Disappearing/Brand New quadrant) [19]. Materials durability is a recurring theme in the Historic building topic, encompassing a large range of materials, including basalt fiber-reinforced concrete, textile-reinforced cementitious composites, hempcrete and regular concrete, lime-based mortar, drywalls, clay, aerogel-based coating mortars, nano-based insulation, and other advanced materials [20,21,22,23,24,25,26,27]. Research has been evaluating these materials and their resistance to climatic events such as acid rain, freezing–thawing, earthquakes, and flooding, as well as their efficiency in mitigating heat islands or improving air quality [24,28,29].

There are 231 publications that discuss Post-disasters, with a 46% overlap with Rapidity. The main type of disaster discussed in Post-disasters is Earthquakes (42%), followed by Hurricanes (19%), Flooding, and Tsunamis (14% each). This area of research includes mostly case study research discussing approaches to reconstruction, sometimes specifically with the goal of making buildings resilient to climatic events [30,31,32,33,34]. The Post-disasters topic is particularly related to human involvement and planning through policies and community engagement, for example [35,36,37,38]

Challenges and barriers associated with post-disaster reconstruction are usually multifold and typically consist of financial and economic, social, infrastructure and housing, environment, or coordination and resources nature [39]. Combining the deployment of resilient reconstruction and population relocation while keeping cultural identity integrity during an urgent crisis requires multiple levels of organization and engagement [40].

All the topics of interest are from the Problems and Issues category and appear in different quadrants. First, Overheating (298 publications) appears in the Established quadrant and is most likely to discuss Heating (64%), Comfort (59%), Simulations (48%), and Energy efficiency (46%). Highly cited papers on this topic discuss measures against overheating in the UK, including a 2012 paper from Loughborough University (FWCI 28.8) about the efficiency of actions against overheating from best (external wall insulation and solar reflective coatings) to worst (internal wall insulation). Another paper from Oxford Brookes University (FWCI: 14.29) focuses on combining fabric improvements with internal heat gain reduction [41,42].

Second, Flooding (669 publications) plots in the Hot quadrant and most frequently discusses Risks (64%), Urban settings (55%), and Disasters (50%). The two publications with the highest FWCI in Flooding (21.70 and 12.95) are both about China and discuss sponge cities. Sponge cities are “urban areas with abundant natural areas such as trees, lakes and parks or other good design intended to absorb rain and prevent flooding” [43]. One paper published in 2018, in collaboration with the IHE Delft Institute for Water Education and Southeast University, China, presents an overview of the challenges and opportunities of sponge cities, including the long-term planning required to transform existing building stock [44]. A collaboration between researchers from the University of Exeter, the University of Oxford, and China’s Dalian University of Technology (FWCI 12.95) argues that bottom-up community-based measures are essential to surpass the limits of sponge cities, enabling them to become flood-resilient cities [45].

Next, Wildfires (56 publications) plots in the Emerging quadrant. Papers in this category are likely to discuss Risks (71%), Urban settings (64%), Disasters (43%), and Reflectivity (34%). A cost estimation framework to optimize the retrofit of residential buildings is presented in a 2018 research paper from the University of New Mexico. It integrates multi-attribute vulnerability rating systems, on-site wildfire vulnerability assessment, property characteristics, and homeowners’ preferences [46]. Another publication from 2022 by a collaboration of Canadian researchers from the University of Alberta and Natural Resources Canada details the results of a survey about existing incentives and barriers to uptake of FireSmart Canada recommended wildfire mitigation activities in Fort McMurray, Alberta. Since most respondents feel they have a low to moderate risk of experiencing a catastrophic fire, they believe that they are already doing enough to reduce the immediate risk [47].

Finally, Wind (231 publications) plots in the Brand New/Disappearing quadrant. Wind publications are most likely to discuss Urban settings (42%), Hurricanes (42%), Risks (40%), Stormwater (39%), and Building envelopes (39%). Wind is often discussed as a factor to which a certain material or solution should be resistant, particularly hurricanes and typhoon winds.

2.7.3. Resilience Qualities

Figure 11 shows the eight topics in the Resilience qualities category, ranked by volume. This category was created based on the research at Hamad bin Khalifa University, Qatar, that published a review paper presenting a taxonomy for the most common resilience qualities and their interdependencies [10]:

• Reflectivity (Rf)—Reflective, preparedness, adaptation, planning.
• Robustness (Rb)—Robust, absorptive capacity, absorption, scalability.
• Redundancy (Rd)—Redundant, reserve capacity, absorption, diversity.
• Flexibility (Fx)—Flexible, adaptability, strategies.
• Resourcefulness (Rs)—Resourcefulness, emergency management, disaster preparedness, resource utilization, mitigating the losses by the community.
• Rapidity (Rp)—Rapidity, responsiveness, restorative capacity, recoverability, recovery, adaptation, recovery activity.
• Inclusivity (Ic)—Inclusiveness, community engagement.
• Integration (It)—Integration, adaptive capacity, adaptability, adaptation

The researchers concluded that integrated resilience indicators, planning, and design methodology are crucial for incorporating resilience qualities into a framework that influences the built environment [10]. Reflectivity is the top Resiliency qualities sub-category, with 25% of the dataset. Integration and Resourcefulness rank second and third with 14% and 12% of the dataset, respectively.

Figure 12 presents the R&D momentum filtered to show only those topics in the Resilience qualities category.

Reflectivity, Resourcefulness, and Rapidity all fall in the Established quadrant and are well-studied in the resilient residential retrofit literature. There is only one topic in the Hot quadrant (Integration), as Flexibility has almost completed its transition from Hot to Established. Integration (569 publications) is most likely to be discussed along with Urban settings (57%), Costs (44%), and Effectiveness (37%). In terms of topics of interest, Integration has a 22% co-occurrence with Flooding, 8% with Overheating, 5% with Wind, and 2% with Wildfires. Integration can have different meanings; for example, the integration of national strategies into practice and the integration of photovoltaic thermal systems in a building [48,49].

There are two Emerging topics in the Resiliency qualities category. Redundancy has the highest growth rate (39 publications). Half (51%) of the Redundancy publications also discuss Urban settings, 41% discuss Costs, and 24% discuss Risks. In terms of topics of interest, Redundancy has a 15% co-occurrence with Flooding, 7% with Wind, shares one publication with Overheating, and none with Wildfires. A publication from the University of Hertfordshire, UK, mentioning Redundancy uses a network model to study resilience design, including the building, site, and region, and finds system redundancy as a common denominator between the three aspects [50].

Robustness (199 publications) also plots in the Emerging quadrant and includes the concepts of absorptive capacity and scalability. Robustness is most likely to be discussed along with Urban settings (47%), Effectiveness (37%), Costs (35%), and Reflectivity (34%). In terms of topics of interest, Robustness publications co-occur with Flooding (19%), 13% with Overheating, and 7% with Wind. They also share three publications with Wildfires. An international collaboration between the Swiss Federal Institute of Technology, Sweden’s Chalmers University of Technology, and Lund University presents a statistical method to assess and quantify the relative robustness of retrofitting measures in the long term, taking uncertainties of climate change and extreme conditions into account [51]. The same researchers also published a paper about evaluating the performance of different retrofitting measures to the building envelope for future climatic conditions [52].

2.7.4. Solutions

Thirty-six Solutions topics were identified in the dataset, covering 70% of the resilient residential retrofit publications. Figure 13 shows the top 15 topics. The top three are: Energy efficiency, with 24% of the dataset, Energy usage (21%), and Technologies (14%). A third of the top 15 relates to energy (Energy efficiency, Energy usage, Energy retrofits, Energy conservation, and Solar energy). Despite not being the focus of this study, the word “energy” shows up in 40% of the dataset. An interesting oddity found in the dataset is Amphibious architecture. With only nine publications, Amphibious architecture is neither among the top topics nor included in the R&D momentum, but it is interesting to note the existence of research looking into making amphibious homes that are resistant to flooding, hurricane winds, and storm surges, or presenting the benefits of amphibious foundation retrofits to preserve historic buildings [53,54,55,56].

Figure 14 presents the R&D momentum indicator, filtered to show the Solutions category. Technologies and Energy retrofit are the only two topics in the Hot quadrant. Despite the fact that the general Technologies group plots in the Hot quadrant, no specific technology plots in either the Hot or Established quadrants. Some more specific technologies appear in the Emerging quadrant, including AI, Living roofs, Nature-based solutions, Sensors, and Smart homes.

There are 13 Solutions topics in the Emerging quadrant. Of these, seven have a relatively high growth rate: Nature-based solutions, Thermal retrofit, Net-zero buildings, Energy storage, Holistic approaches, AI, and Drainage. The Nature-based solutions topic has 35 resilient residential retrofit publications and is most likely to be discussed along with Urban settings (74%), Infrastructures (54%), and Reflectivity (51%). In terms of topics of interest, the Nature-based solutions topic has 34% co-occurrence with Flooding, 6% with Overheating and Winds, and none with Wildfires. The nature-based solutions publications with the highest FWCI are about sponge cities. A publication from the University of Oxford investigates the potential of nature-based solutions for historic buildings to improve health, sequester carbon, enhance biodiversity, provide acoustic comfort, reduce urban heat islands, enhance water management, facilitate urban agriculture, improve air quality contributing to economic vitality through job creation, and enhance social cohesion [57]. Combating sea level rise with landscape architecture is also an interesting retrofit option using nature-based solutions [58].

Thermal retrofit is another topic with a high growth rate in the Emerging quadrant. Thermal retrofit has 111 publications and is more likely to be discussed along with Comfort (61%), Thermal comfort (55%), and Heating (49%). Interestingly, Heating is more often discussed with Thermal retrofit than Cooling (26%). In terms of topics of interest, Thermal retrofit has 34% co-occurrence with Overheating, shares one publication with Flooding, and none with Winds and Wildfires. Recent residential terraced houses built to be energy efficient are often characterized by high indoor temperatures and are not designed to account for future climate and warmer summers and are therefore at risk of overheating. Careful retrofitting through the implementation of passive cooling design systems should be considered to increase occupant comfort [59]. Bio-based products such as straw-based technologies also have potential in thermal retrofit beyond the immediate impacts on the occupants since their life cycle assessments (LCAs) score better than retrofitting with traditional materials that yield higher greenhouse gas (GHG) emissions [60].

The Disappearing/Brand New quadrant includes 14 Solutions for resilient residential retrofit, including Water management, Landscaping, Co-benefits, Trees, and Maintenance and Adaptive measures. However, these topics do not contain a significant number of publications. Nevertheless, Restoration/Modernization, Solar Energy, and Passive houses may eventually reach the Established quadrant.

2.7.5. Tools

Tools are mentioned in 73% of the dataset. Figure 15 shows the top 15 of the 20 Tools identified. The top topic is Investments (25%), followed by Simulations (18%) in second place and Policies (16%) in third. The presence of topics implying the involvement of organized authorities (Policies, Building regulations, and Governance) in the top 15 suggests an acknowledgment of the importance of the role governments play in resilient residential retrofits. A more predominant role is given to planning and evaluation activities overseen by local authorities, such as Simulations, Resiliency strategies, Risk assessments, Frameworks, Decision-making, Cost effectiveness, Performance assessments, and Climate models. The Tools with the highest FWCI are the Sendai framework (3.07, 15 publications), followed by Governance (1.62), Policies, and Performance assessments (both 1.53).

Figure 16 presents the R&D momentum filtered to show only topics in the Tools category. The most well-covered Tools in the resilient residential retrofit literature, as shown in the Established quadrant, are Policies, Resiliency strategies, Risk assessments, Optimizations, and Frameworks. Taken together, they are mentioned in 60% of the Tools publications. It must be noted that, overall, no topic in the Tools category is considered to have a high growth rate (threshold of 1.5).

There are only three Tools in the Hot quadrant, all with a small growth rate: Investments, Simulations, and Resilience Impact. Resilience Impact, with its slow and steadily growing publication count, could have moved recently from the Emerging to the Hot quadrant. About half of the publications on Investments relate to Urban settings (49%), 41% to Costs, 35% to Risks, and 37% to Houses. In terms of topics of interest, the Investments topic has a co-occurrence of 14% with Flooding, 9% with Overheating, 7% with Wind, and 2% with Wildfires. Investments and Costs share significant ties, as they are often mentioned together to discuss the “investment cost” of a retrofit, such as a heat pumping system or phase change materials, or in flood protection measures like implementing multilayered safety or reconstructing houses using the elevating coastal houses method [61,62,63].

There are only three Tools topics in the Emerging quadrant: Decision-making, Thermal performance, and Climate models. The Climate models (110 publications) topic has the highest growth rate of the category, but it is relatively low at 1.1. The topic is most frequently discussed along with Urban settings (45%), Simulations (43%), Reflectivity (41%), and Energy usage (37%). In terms of topics of interest, Climate models has a co-occurrence of 14% with Overheating, 6% with Flooding, and 4% with Wildfires, and also shares one publication with Wind. Climate models are used to calculate the impact of a retrofit, for example, how cool roofs outperform green roofs in reducing the temperature in urban settings, particularly at night [64,65]. The focus can also be on improving climate modeling through artificial intelligence by, for example, using building stock data to estimate the risk of overheating or to generate a dynamic intelligence building retrofit decision-making model [66].

Thermal performance has 182 publications and mostly correlates with Energy efficiency (63%), Building envelopes (58%), Comfort (54%), and Energy usage (46%). In terms of topics of interest, Thermal performance has a co-occurrence of 30% with Overheating, 8% with Wind, shares one publication with Flooding, and none with Wildfires. A publication in Thermal comfort stands out with a very high FWCI of 14.49. In 2022, an international collaboration between India, Germany, and the USA published a review of the phase change materials in the building envelope and their potential to optimize the end-use of energy in buildings, where space cooling is a major source of demand. Phase change materials are described as materials that can absorb or let go of heat at certain temperatures and which can go from solid to liquid (or vice-versa) in the process [67,68].

Finally, most topics in Tools fall under the Disappearing/Brand New quadrant: Building regulations, Cost effectiveness, Performance assessments, Governance, Energy policy, Incentives, Building information modeling, Zoning, and Sendai framework. A deeper analysis of each would be necessary to show if the interest is growing or fading in the resilient residential retrofit research community.

2.8. Climatic Events Topics

To better understand which climatic events are driving resilient residential retrofit research, climatic events topics were extracted from the Problems and Issues category. These topics were plotted in a matrix with the Emerging topics from the other categories to provide insights about growing research interests (defined as a high percentage within topic), as well as gaps in the resilient residential retrofit literature. Empty spots in the figures mean that there are few publications discussing two topics alongside and are, therefore, highlighting potential gaps in the resilient residential retrofit literature. There are four climatic events topics of particular interest to this study: Flooding, Overheating, Wildfires, and Wind.

Figure 17 and Figure 18 show the percentage of the climatic events in which Emerging topics are discussed together or overlap. The figures are filtered to a minimum of 5% of overlap to better identify trends. The graph should be read vertically. For example, 5% of the Overheating publications mention Lighting. Empty cells do not mean there is no overlap between the topics, but if there is overlap, it is lower than 5% of the publications of the climatic event. Since showing all the Emerging topics in a single figure was too dense, Figure 17 shows only topics for the Housing, Resilience qualities, and Tools categories and Figure 18 shows the emerging topics from Problems and Issues and Solutions.

The strongest overlap in Figure 17 is between Overheating and Thermal performance (18%). It has the highest percentage in the topics of interest and, overall, in Housing, Resilience qualities, and Tools topics. Other significant overlaps are Winter and Thermal performance (15%), Hurricanes and Decision-making (15%), Heatwaves and Thermal performance (13%), and Sea level and Decision-making (13%). It is not a surprise to see a strong correlation between Thermal performance and the climatic events related to heat and cold perception by the home occupant.

The highest overlaps in Flooding (669 publications) are with the following:

• Decision-making (10%);
• Settlements (7%);
• Robustness (6%).

The highest overlaps in Overheating (298 publications) are with the following:

• Windows and doors (8%);
• Robustness (8%).

The highest overlaps with Wildfires (56 publications) are with the following:

• Decision-making (11%);
• Settlements (5%);
• Robustness (5%).

Other interesting overlaps include the following:

• Decision-making and wildfires (11%);
• Winter and windows and doors (10%);
• Overheating and windows and doors (9%);
• Snow and beams (9%);
• Wind and beams (7%).

The strongest overlap in Figure 18 shows that slightly more than one-third of the publications on Flooding discuss Flood control (34%). The second strongest overlap is between Heatwaves and Heat islands (28%), followed by Tsunamis and Post-disasters (23%).

The other strongest overlaps with Flooding are the following:

• Drainage (9%);
• Risk reduction (9%).

The strongest overlaps with Overheating are the following:

• Thermal retrofit (13%);
• Heat islands (6%);
• Net-zero buildings (5%).

The strongest overlaps with Wildfires are the following:

• Risk reduction (7%);
• Permafrost (4%);
• Sensors (4%).

The strongest overlap with Wind is with Heat islands (7%).

Other interesting overlaps include the following:

• Stormwater and living roofs (6%);
• Snow and sensors (9%);
• Heatwaves and Living roofs (12%);
• Erosion and Flood control (7%).

2.8.1. Flooding

Flooding is the most frequent type of natural disaster globally. Floods are characterized by an overflowing of water where the land is usually dry, be it from heavy rainfalls, rapid snowmelts, or storm surges. In the publication dataset, Flooding is the leading natural disaster topic, with 669 publications [69,70].

Flooding is discussed most frequently, along with Risks (64%), Urban settings (55%), Disasters (50%), and Reflectivity (300). Among the most-cited publications on the Flooding topic are papers about sponge cities (eight publications, half with an FWCI from 3.41 to 21.70). Sponge cities, as discussed earlier, are an urban planning concept that uses natural-based solutions such as trees, lands, and parks to absorb excess water and prevent flooding. This concept is gaining traction, particularly in China, and has the potential to help older residential districts meet stormwater management standards through renovations using permeable pavement, bio-retention cells, grassed pitches, and rain gardens [45,71,72]. By adding living roofs and rainwater storage tanks to the strategies mentioned, there are also gains to be made in flood control and drainage capacity in colder climate communities [73]. Similarly, blue and green components and infrastructure are consistently mentioned in the literature as having potential benefits for flood resilience. Green infrastructure uses living vegetative elements, whereas blue infrastructure, a more economical solution, is designed to store water and reduce stormwater peak loads. Moreover, as mentioned previously, with green roofs, blue roofs have benefits that include reducing energy consumption for heating and cooling because of the insulating effect of the stored water [74,75,76,77,78]. About 13% of the publications on Flooding mention blue–green solutions, and 40% of these publications have a higher FWCI than the dataset average (1.14).

Other flood resistance strategies can be applied directly to the building. Fiber-reinforced polymers (FRPs) have the potential to strengthen and repair deterioration in buildings to help them resist natural disasters [19,79,80]. While typical elevated slabs can be brittle with low ductility, a single layer of carbon fiber-reinforced polymer (CRFP) laminate can increase the floor load capacity by about 30% [80]. CRPF can also be used to strengthen existing concrete girders to give them adequate flexural capacity for flood protection [79]. More economical solutions, like reinforcing renders, can improve sectional flexural capacity by over 11 times and help vernacular masonry buildings resist lateral flood loads [81]. By taking life cycle analysis (LCA) into account when selecting materials in a flood zone, it is concluded that timber and steel frames are the worst materials to use, while brick is considered to be the most sustainable and functional choice [82]. Other retrofit measures against flooding include backflow of drainage pipes, submersion of equipment, and buoyancy to reduce and delay damage to existing homes [83]. Finally, elevating homes is a helpful retrofit solution for flood resilience, although not necessarily economically viable [84,85]. In fact, amphibious homes represent a more cost-effective solution for new buildings, and amphibious retrofits are also a possibility [53,54].

2.8.2. Overheating

Overheating is related to indoor thermal comfort and is described as a cumulative effect of heat stress and the health of the occupant exposed to continuous indoor heating [86]. Since thermal comfort is an individual experience with personal preferences, there have been issues with determining how to assess overheating due to the lack of an evidence-based methodology, particularly for residential buildings [87]. There are three main factors for an indoor environment to overheat, starting with the external temperature. During a heatwave, occupants are told to stay inside. However, if their indoor environment is prone to overheating, it may not be the best recommendation. Another source of overheating is solar gain. The shading, time of day, season, and type of glazing will have an impact on the indoor temperature. Occupant lifestyle also influences indoor temperature, with heat coming from electrical appliances and occupants themselves [87]. As mentioned earlier, newer homes tend to have better airtightness than older houses, which puts newer buildings at more risk of overheating [16].

Overheating has 298 publications in the dataset. It co-occurs most frequently with Heating (64%), Comfort (59%), Simulations (48%), and Energy efficiency (46%). The UK is the clear leading country in Overheating research for resilient residential retrofit with 92 publications, with the second leading country, Italy, having only 39 publications.

Modifications can be brought to a home to reduce the risk of overheating. External wall or roof insulation is an example and has proved preferable to interior wall insulation since the latter is less effective and can also increase the risk of overheating. Interestingly, retrofitting internal walls and floors has an increasing temperature effect as well [88]. The control of solar gains through glazing with shutters and fixed shading for south, east, and west-facing rooms is also a known effective solution [41,89]. Moreover, shading systems maximize their impact if located outside of homes [90]. An automated roof–window control system can also significantly decrease the risk of overheating without compromising the indoor air quality [91].

The data seen in the literature lead to the idea that external modifications to a home have a better impact on indoor overheating than modifications to the interior of the building. Interestingly, despite the numerous articles in the dataset discussing the indoor cooling effect of blue–green walls and roofs and the use of trees for shading, little is directly tied to the concept of Overheating in the dataset, with 3% for Living roofs, two publications for Nature-based solutions, and one publication for Trees [76,92,93,94,95,96,97,98,99,100,101,102,103].

2.8.3. Wind

Wind is an everyday occurrence but can become a threat depending on its intensity. High winds can be due to a storm, hurricane, tornado, or other weather events. Severe wind can be the cause of injuries and damage to property and infrastructure [104]. Wind has 231 publications in the dataset.

Wind overlaps most frequently with Urban settings (42%), Hurricanes (42%), Risks (40%), Building envelopes (39%), and Stormwater (39%). This topic is led by the USA, which has 82 publications, followed by China, with 32 publications.

Not surprisingly, there are strong correlations among Wind, Hurricanes, and Stormwater because hurricanes and other tropical storms produce excessive winds and rainfall. Therefore, many publications in Wind focus on proofing homes and designing against the impact of hurricanes [105,106,107]. Modeling, simulations, probabilistic quantifications, and other analyses of wind loads are frequently used to test retrofit strategies in residential buildings [107,108,109,110,111,112,113]. For example, this may include creating scaled models of fiber-reinforced polymer sheets as roof-to-wall connections and exposing them to various wind speeds [114].

Wind can also be used as a nature-based solution ventilation strategy to cool indoor temperatures and decrease the risks of overheating. Smart windows that automatically close in the case of strong winds are an example of the combination of modern technologies and nature-based solutions in resilient residential retrofit [111,115,116,117].

2.8.4. Wildfires

Wildfires are unplanned fires of human or natural causes that burn a natural area like a forest, grassland, or prairie. Wildfires are destructive and can lead to the loss of resources, crops, and property. Wildfires can also lead to fatalities, injuries, and deterioration of mental health. Moreover, wildfires are a cause of morbidity due to air pollutants in wildfire smoke and long-term impact on health, such as impairment of speech, hearing, muscle weakness, and vision problems because of the mercury released in the air [118].

Wildfires has 56 publications in the dataset and co-occurs most often with Risk (71%), Urban settings (64%), Disasters (43%), and Reflectivity (34%). The low number of articles related to Wildfires in this dataset could be due to resilient residential retrofit literature being oriented toward Urban settings (46% of publications in the dataset), where wildfires may be a lesser risk compared to other types of fires, as urban settings may not be surrounded by a natural area (unless in a wildland–urban area). Most of the publications are from the USA (70%).

The protection of houses against wildfires depends on the regulations in place at the time the building was made and on the willingness of the owners to make their homes resilient to wildfire when planning an addition to their existing home [119,120,121]. The latter appears to have a lower uptake. Despite recent guidelines recommending/requiring both new and existing buildings to be built with traditional local materials, the costs involved in fireproofing a home remain a significant barrier, as mentioned in a collaboration between the University of Melbourne and the University of Wollongong (Australia) [122]. The study used five homes as case studies, none of which met any building standards for wildfire resilience despite scoring the highest level of wildfire exposure [123]. Similar results were found in Canada, where most of the respondents to a survey about FireSmart mitigation practices said they perceived a low to moderate wildfire risk to their properties. Moreover, the survey participants preferred implementing low-cost and low-effort mitigation measures such as cutting grasses and cleaning debris to retrofitting their homes, citing costs as the single and biggest factor in their decision [47]. Belonging to a lower socio-economic class can also put occupants at risk in other ways, such as living in subsidized housing units in California, in which case the vulnerable households are dependent on policy interventions [124].

2.9. Top Countries

Country information was derived from author affiliation metadata, which was available for 98% of the dataset. A total of 124 countries have publications in the dataset, and 30 have at least 40 publications in the field (i.e., a minimum of 1% of the dataset). Hard numbers are reported in this section without normalization, although the interpretation of the results should take into account that each country does not have an equal chance of publishing. Each country varies in terms of population size, funding practices, number of active researchers, odds of being affected by specific climate events, and more. Figure 19 presents the top 15 countries publishing the most in resilient residential retrofit with their number of publications, their FWCI, and their international collaboration rate (i.e., share of country publications co-authored with a partner from another country). The United States is the country that is the most active in the field, with 20% of the publications. China (11%) and the UK (10%) rank second and third, respectively, by number of publications.

Ordinarily, the USA and China dominate most fields of research, making up 40–50+% of the dataset; here, they collectively only make up close to 30%. The Netherlands has the highest FWCI (2.49), followed by Iran (2.29) and Belgium (2.25).

The USA has research excellence (e.g., a significantly higher average FWCI in a topic compared to the country’s average FWCI in the whole dataset) primarily in Heat islands and also in Extreme events, Wildfires, and Decision-making. China shows research excellence in Heat islands, Emission control, Post-disasters, Robustness, Decision-making, and Flood control. The UK has research excellence in Risk reduction, Extreme events, and Thermal retrofit.

3. Conclusions

The purpose of this study was to conduct a comprehensive scientometric analysis of the current research landscape in resilient residential retrofits from 2012 to the present. References were retrieved and analyzed using text mining software and online searching. In all, 4011 publications and 2623 patents were identified for this study.

The research trends analysis indicates that emerging interests in resilient residential retrofit include nature-based solutions, energy efficiency, and thermal comfort. Nearly half of the publications mention urban context and over one-third address costs. The building envelope is the most frequently discussed housing component. Although energy retrofit was not the primary focus of this study and was not specifically searched for, energy concerns were still prevalent in the dataset, with 40% of the publications mentioning “energy”. Even without planning for energy efficiency, energy consumption or backup power is likely to be part of the retrofit planning process for issues like overheating or prolonged power outages [59]. The USA leads the resilient residential retrofit research field in volume, followed by China and the UK. In the patent landscape, China is the dominant leader. Concerns about water, roofing, urban settings, houses, and interior walls each account for a significant portion of the patent dataset. The top topics all relate to housing or addressing climate-related issues in construction.

Like the challenges it is tackling, resilient residential retrofit research is multifold. There is effectively no “one size fits all” when it comes to resilient residential retrofit, and the “best” solution will depend on several factors, such as the type of housing, its location, occupant preferences, and socio-economic factors. Resilient retrofits cannot address all climate risks simultaneously, and while thorough and holistic planning is recommended, implementation is likely to be incremental at best [125]. Research on wildfire resilience highlights the importance of the occupants’ risk perception and financial means in the decision-making process [47,119,122]. Bottom-up community-based measures are crucial for post-disaster reconstruction and overcoming limits to climate resilience [45]. Community engagement can be facilitated through education and awareness [126]. However, occupants’ desire to upgrade existing homes will likely be a reaction to risk exposure, underscoring the role of governmental entities in providing sound policies and incentives to protect their populations, especially those most vulnerable.

This analysis revealed several research gaps in resilient residential retrofits, including low publication rates on durability, holistic approaches, microclimates, nature-based solutions, and traditional homes. There is also a lack of cross-themed research specific to rural and suburban settings, and few studies address combinations of themes like overheating in high-rise buildings, wildfires in Nordic climates, and flooding risks in smart homes.

By identifying key trends, gaps, and leading contributors, scientometric studies guide future research, funding decisions, and policy-making. In the context of resilient residential retrofits, this analysis highlights areas of focus and emerging interests, helping stakeholders prioritize efforts and resources effectively. Looking forward, it is essential to address these identified research gaps and foster interdisciplinary collaborations to develop holistic and innovative solutions. By doing so, we can better prepare residential buildings to withstand the impacts of climate change, ensuring safer and more resilient communities for the future.