A citespace-assisted overview of citizen science research in the built environment field Open Access

Citizen Science (CS) broadly refers to “the active engagement of the general public in scientific research tasks” (Vohland et al., 2021). This inclusion of non-experts in scientific data collection and analysis creates strategic opportunities for diverse stakeholders, including citizens, government agencies, industry, academia, and community organisations, to jointly contribute to knowledge production (Ciasullo et al., 2019; Whitelaw et al., 2003). In particular, recent technological advancements in the internet and mobile platforms have driven a proliferation of CS research projects, thereby enhancing public engagement in environmental management and elevating public awareness about the importance of research and monitoring (Dickinson et al., 2013).

CS has increasingly been recognised as a powerful mechanism for democratising knowledge and advancing public collaboration in scientific research related to the Built Environment (BE) (Wood et al., 2022b). According to Renalds et al. (2010), the BE encompasses spaces that have been modified by human activity, including residential areas and neighbourhoods or community settings, with which individuals interact in their daily lives. The relationship between humans and the BE is bidirectional and dynamic—people shape the environment, and in turn, the environment shapes human behaviour and well-being, creating a feedback loop of mutual influence (Caan, 2011; Hutchison, 2018; Shahruddin et al., 2024). Therefore, incorporating CS in the BE research promotes participatory and responsive design practices and also strengthens community resilience through lived experiences and local knowledge (Delpino-Chamy and Pérez Albert, 2022; Liu et al., 2022). This highlights the urgent need for systematic exploration and application of CS to ensure its broader, more impactful integration into BE planning and policy.

Previous studies have demonstrated the suitability of using CS in the natural and BE fields of research, including but not limited to ecology, geography, agriculture, and water resources (Conrad and Hilchey, 2011; Ebitu et al., 2021; Ramírez et al., 2023; Trojan et al., 2019; Vasiliades et al., 2021). However, despite the growing interest in CS, the field lacks a systematic and critical review that evaluates how CS is specifically applied in BE contexts. While there has been an increasing number of studies using CS to investigate BE-related topics, such as the role of urban environments in promoting healthy aging (Wood et al., 2022b), the impact of community settings on individuals (Roger and Motion, 2022), and the influence of community environments on physical activity (Hinckson et al., 2017), it lacks a comprehensive nor critical review on the application of CS in the broad BE field.

To address this research gap, this paper utilises CiteSpace, a powerful bibliometric analysis tool, and content analysis to present a holistic picture of the current state of CS research in the BE field. CiteSpace was utilised to generate co-occurrence network maps for geographical areas, authors, and keywords, illustrating the collaboration networks of countries and authors and identifying research hotspots. Additionally, a keywords-clustering timeline map was created to demonstrate the evolution of research themes over time. This was conducted alongside content analysis to examine the distribution of articles across various journals and publication years, as well as to extract detailed information about citizen scientists, including the number, age, and level of engagement. The challenges associated with implementing the CS approach were also identified and categorised. By integrating these two methods, this paper explores the overall development trajectory of CS research in the BE field, providing researchers with an extensive and in-depth analytical framework to better conduct future research.

The concept of citizen science began to take shape in the 1990s, when scholars and practitioners sought to describe and formalise the role of the public in scientific research. Irwin (1995) introduced the term from a sociological perspective, highlighting its potential for democratising science and addressing sustainability challenges. Around the same time, Bonney et al. (2009) defined citizen science more narrowly in 1996 in the context of ecological research, focusing on large-scale data collection projects involving volunteers. Building on this foundation, citizen science today actively involves members of the public, referred to as “Citizen Scientists,” as collaborators in scientific research, engaging them in various stages, such as formulating research questions, data collection or interpretation of findings, and implementing actions based on these findings (Bonney et al., 2009; Irwin, 1995).

Although the definition of citizen science is still evolving, it encompasses a wide range of practices, initiatives, and activities designed to involve the public in scientific research in various meaningful ways (Vohland et al., 2021). Specifically, CS is an integral part of a broader movement towards open science, open government, and participatory democracy, emphasising transparency, inclusiveness, and collaboration (Eitzel et al., 2017). The aim is to make scientific research and governmental processes more accessible to the public and ensure they are accountable to the people they serve (King et al., 2016). It generates an open dialogue where research is no longer just done for the public but with them or by them, recognising that community members have valuable voices, lived experiences, and expertise that can significantly contribute to the research process (Eitzel et al., 2017). Engaging the public in this way helps to ensure that research addresses real-world concerns while benefiting from diverse perspectives (Engage, 2020).

Based on citizen scientists' level of involvement in the research process, Marks et al. (2022) classified the CS approach into four categories, namely contributory (Level 1), collaborative (Level 2), co-created (Level 3), and citizen-led (Level 4). Contributory CS research (level 1) primarily involves citizen scientists in data collection and basic data analysis, with researchers designing the projects, formulating research questions, and determining methods while citizen scientists execute these methods. Collaborative CS research (Level 2) extends the role of citizen scientists beyond data collection and basic analysis to include adjusting research plans, drawing conclusions, and proposing new research directions. Co-created CS research (Level 3) involves citizen scientists in more project phases, including collaborating with researchers to formulate research questions, select methods, and share outcomes. Citizen-led CS research (Level 4) represents the highest level of citizen scientist involvement, where they define problems, design research, collect data, analyse and interpret results, and disseminate and advocate for the research outcomes. Citizen scientists participating in these projects have accumulated substantial experience from co-created projects.

This review adopts a scoping and descriptive approach, aiming to provide a structured synthesis of the CS literature in the BE field. Rather than testing hypotheses or building new theories, the study focuses on mapping the current research landscape, which offers researchers and practitioners a clearer understanding of participation patterns, regional disparities, and implementation challenges—thereby informing future methodological design and policy development. The current study systematically identified, screened and reviewed literature that used the CS approach in the BE field across three databases: Taylor & Francis Online, Scopus, and Web of Science (WoS). These databases were selected to ensure comprehensive coverage and inclusion of a wide range of high-quality research articles to achieve a thorough and balanced review, ensuring the reliability and validity of the literature review on applying the CS approach in the BE field. The research procedure can be seen in Figure 1.

The Taylor & Francis Online database was selected as the first platform for the screening process because it includes the Built Environment as a separate discipline and two screening criteria were established: (1) the articles must explicitly state that they employed the CS approach; (2) the articles must be related to the BE field.

The first search was conducted using the keywords “citizen science” and “citizen scientist*”. However, upon reviewing the articles, it was observed that some studies also employed the keywords “community participatory” and “community-based research” to describe public involvement in CS research. Then, a second search was performed using only these two keywords. After carefully examining the articles from this search, it became clear that, apart from the articles that explicitly stated the utilisation of the CS approach and also referenced “community participatory” or “community-based research” (these articles were already captured in the first search), the remaining articles primarily discussed public involvement in data collection without identifying them as CS research. Therefore, using “citizen science” and “citizen scientist*” as search keywords is sufficient to identify relevant articles.

Further screening also revealed that some articles briefly mentioned the term “citizen science” without actually applying its principles—such as participatory data collection, shared research design, or institutional coordination. These were also excluded. As such, this review includes only formal CS research projects: studies that explicitly label themselves as “citizen science” research in the title, abstract, or full text and are typically initiated or coordinated by academic, governmental, or institutional actors. This definition ensures conceptual consistency and aligns with established distinctions in the literature (Eitzel et al., 2017; Vohland et al., 2021).

As of April 29, 2024, a total of 191 articles have been searched in the Taylor & Francis Online database. Then, a manual review was conducted, examining the titles, abstracts, introductions, and methodology sections of these articles to select articles that meet established criteria. Afterwards, individual journals containing selected articles were searched to confirm that no relevant articles were overlooked. Ultimately, 8 articles met the established criteria. These articles were pertinent to urban planning, architecture, and construction management within the BE. This initial screening process yielded a focused collection of relevant articles and established clear criteria for article selection in subsequent database searches.

Next, a comprehensive article screening was conducted in the Scopus database using the keywords and two criteria established during the Taylor & Francis Online database screening. Given that Scopus does not have a specific “BE” discipline category, an exclusion method was applied to filter out clearly unrelated fields (including Agricultural and Biological Sciences, Computer Science, Physics and Astronomy, Medicine, Biochemistry, Genetics and Molecular Biology, Business, Management and Accounting, Mathematics, Economics, Econometrics and Finance, Chemistry, Chemical Engineering, Materials Science, Immunology and Microbiology, Pharmacology, Toxicology and Pharmaceutics, and Nursing), and focus on articles published in English-language journals across the remaining relevant subject areas to the built environment, including Environmental Science, Social Sciences, Earth and Planetary Sciences, Multidisciplinary, Engineering, Arts and Humanities, Energy, Psychology, Decision Sciences, and Health Professions. As of April 29, 2024, a total of 5,708 articles were searched. After the process of manually reviewing these articles and checking individual journal databases, an additional 97 articles met the inclusion criteria (with 105 articles in total at this stage).

Finally, the same criteria and exclusion method were used in the Web of Science (WoS) database. As of April 29, 2024, 4,120 relevant articles were found, and most relevant articles overlapped with those in the Scopus database during the screening process. After excluding these overlapping articles, 2 additional relevant articles were identified, making a total of 107 related CS articles for data analysis.

The screening process ensured that this study covered the major databases and enhanced the comprehensiveness and accuracy of the article selection through multi-level manual reviews and cross-validation across databases. While the final sample size is relatively modest (n = 107), it reflects a targeted focus on Citizen Science (CS) research within the Built Environment (BE) field based on clearly defined inclusion criteria.

After identifying relevant articles, a combined content analysis method and the bibliometric tool CiteSpace were used to comprehensively analyse these articles. Specifically, when conducting the analysis using CiteSpace software, the co-occurrence network maps for geographical areas, authors, and keywords were generated, demonstrating collaboration among countries and authors, as well as identifying research hotspots. Subsequently, the keywords clustering timeline map was generated to illustrate the evolution of research themes over time. While performing content analysis, the distribution of articles across various journals and publication years was meticulously recorded to observe research trends. Additionally, detailed information about citizen scientists was extracted, including their numbers, ages, and levels of engagement. Finally, the main challenges associated with implementing the CS approach in the BE field were identified and categorised.

Table 1 shows the distribution of 107 articles across 52 journals, with the majority of these journals from the urban planning and health-related domain. Among these journals, four contained five or more articles. The International Journal of Environmental Research and Public Health contained the most with 16 articles, followed by Citizen Science: Theory and Practice with 6 articles, Health & Place with 6 articles, and Cities & Health with 5 articles. Notably, the journal Citizen Science: Theory and Practice is a peer-reviewed, multidisciplinary publication focused on the global field of CS and other participatory sciences, emphasising the recognition and exploration of the significance of the CS approach.

The line chart (see Figure 2) shows the annual publication volume trend from 2011 to 2023, as there were only 9 papers in 2024 (not a whole year), which may affect the trend line.

It is clearly shown that there is a significant upward trend in the number of articles, particularly after 2020, which indicates a growing interest and emphasis on applying the CS approach in the BE field. Specifically, from 2011 to 2017, the annual number of articles remained relatively low, ranging from 1 to 4. From 2018 to 2019, this number of articles increased to 7 to 8. By 2020, the number of articles nearly doubled, reaching 17. Although there was a slight decline to 15 articles in 2021, the overall upward trend continued, with 19 articles in 2022 and 25 articles in 2023.

Figure 3, generated using CiteSpace, visualises the geographical distribution and international collaboration of countries. Table 2 provides corresponding quantitative data, where Frequency indicates the number of publications and Centrality reflects a country’s influence and connectivity within the global research network.

The United States ranks first in publication count (54) and shows high centrality (0.40), indicating both strong research productivity and active international collaboration. The United Kingdom follows with fewer publications (21) but slightly higher centrality (0.44), suggesting a more central role in global collaborative networks. Other countries with moderate centrality include Sweden (0.23), Colombia (0.18), the Netherlands (0.14), and Germany (0.12). Notably, Colombia is the only Latin American country with measurable centrality, and South Africa (0.07) and Malaysia (0.07) are among the few countries from the Global South with observable network connectivity. Australia (0.09) and Canada (0.07) also show moderate levels of international linkage.

In contrast, a large number of countries—including China, Brazil, South Korea, Argentina, and several African and Southeast Asian nations—appear with zero centrality, indicating a lack of integration into international CS research networks despite some level of publication activity. This suggests that many of these countries operate in relative isolation from the global discourse.

These patterns may reflect underlying structural disparities across regions. Countries with high centrality often appear to benefit from stronger research infrastructure, sustained funding and established academic partnerships. In contrast, underrepresented regions may face challenges such as limited institutional capacity, restricted access to digital tools, and weaker policy support for participatory science, which may collectively constrain their integration into the global CS research network. While these interpretations are grounded in the observed distribution of publications and network metrics, they remain exploratory and should be further examined through empirical studies focusing on regional research ecosystems and collaboration dynamics (e.g. Vohland et al., 2021).

The author’s co-occurrence network map reveals the collaborative relationships among researchers (see Figure 4). The map shows Abby C. King as a central figure within the network, demonstrating significant collaboration with numerous other authors such as Ann W. Banchoff, Sandra J. Winter, Lisa G. Rosas, Marianne Granbom, Susanne Iwarsson, Deborah Salvo, Jylana L. Sheats, and Olga L. Sarmiento.

Additionally, Price’s Law, proposed by Derek J. de Solla Price, suggests that in any given field, a small proportion of contributors account for a large proportion of the total output (Price, 1963). Wang et al. (2022), based on Price’s Law, pointed out that half of the papers on a given topic are written by a group of highly productive authors, and this group of authors, i.e. core authors, is numerically equal to the square root of the total number of all authors. Price’s Law formula shows as follows (Price, 1963):

In the formula (1), where $n(x)$ is the number of authors with $x$ papers, $I$ is the number of papers by the most prolific authors (​$nmax$⁠), $N$ is the total number of authors, and $m$ is the minimum number of publications by core authors. According to Price’s Law formula (2):

In this paper, with $nmax$ = 27, then $m$ = 0.749 $×$ √27 $≈$ 3.8892. Therefore, publishing more than 3 papers (including 3papers) qualifies an author as a core author (see Table 3). Further analysis shows that 11 core authors have collectively published 66 articles out of 107, accounting for approximately 61.68% of the total publications. This exceeds Price’s proposed 50% standard, indicating that the field has formed a stable group of collaborating authors.

Furthermore, after reviewing the articles of these core authors, their four key research areas were identified: (1) Applying the CS approach in the BE field to promote physical activity and health, (2) Utilising mobile and information technology in CS research projects, (3) Advancing age-friendly communities through the CS approach, and (4) Enhancing environmental and health outcomes through community engagement.

Figure 5 shows the distribution of the number of citizen scientists reported in the 107 CS research papers. Unfortunately, 22.43% of citizen science papers did not provide specific participant numbers. Of the papers that did report participant numbers, 31.78% had between 1 and 25 participants, and 16.82% had between 26 and 50 participants, accounting for a total of 48.6% combined (nearly half). Some papers report larger-scale CS projects, although their proportion is relatively low. Specifically, 9.35% of research projects/papers had between 51 and 100 participants, 4.67% had between 101 and 200 participants, and 7.48% had between 201 and 500 participants. CS research projects with 501–10,000 citizen scientists accounted for only 7.47% combined. This distribution indicates that most CS research tends to be small-scale (i.e. involving less than 50 citizen scientists), primarily because smaller research projects are easier to manage and have higher participant engagement. Lower participant numbers in CS research could result from higher levels of engagement required by individuals, dependent on their availability to voluntarily collect data over a period of time. CS methods can also result in more complex project management to support participants, alongside allowances for wide-ranging data collation that can be challenging to analyse.

Figure 6 illustrates the age distribution of citizen scientists reported in 107 articles. Specifically, 11% of the articles did not provide information on the ages of the citizen scientists. Among the articles that reported age information (95 out of 107), 46% engaged adult individuals from diverse age groups (i.e. 18 years and above) to acquire various perspectives and expertise for various BE-related topics. Notably, 36% of CS research projects targeted specifically older adults (65 years and above), primarily aiming to improve the living environment of older adults. Only 7% of CS research projects worked exclusively with participants under 18 years old.

The articles demonstrated varying degrees of citizen scientist participation across four levels: Contributory (Level 1), Collaborative (Level 2), Co-Created (Level 3), and Citizen-Led (Level 4). This classification is based on the framework proposed by Marks et al. (2022), which builds on earlier typologies developed by Den Broeder et al. (2016) and English et al. (2018). It distinguishes participation levels based on the extent of citizen scientist involvement across different stages of the research process—such as problem definition, data collection, data analysis, and decision-making. These varying levels of engagement, from simple data collection to comprehensive leadership, reflect citizen scientists' diverse roles and the degree of contributions to the CS approach within the BE field.

Table 4 shows that Contributory (Level 1) is the most common, accounting for 45.79% (49 projects). For example, citizen scientists used standardised noise measurement devices to collect noise values in an environmental noise monitoring study (Othman et al., 2024). In another urban greening system study, citizen scientists used mobile applications to photograph and upload images of plants while recording environmental data (Vo et al., 2019). In a climate monitoring study, citizen scientists also used meteorological sensors to collect temperature and humidity data (Best et al., 2023).

Collaborative (Level 2), which accounts for 28.04% (30 projects), involves a deeper level of engagement from citizen scientists compared to Contributory (Level 1). For instance, citizen scientists collected data in a community forestry study and provided their insights into local forest dynamics (Eames and Egmose, 2011). Similarly, another example is a sustainability study, where citizen scientists used various environmental sensors to monitor local conditions and collaborated with professional researchers to interpret the findings (Muhamad Khair et al., 2020). In an urban planning study, citizen scientists used geographic information system (GIS) tools to map neighbourhood features and participated in workshops to give their opinions on urban development plans (Loh and Kim, 2021).

Co-created (Level 3), representing an even greater level of engagement, accounts for 26.17% (28 projects). For example, in a study on environmental impacts in informal settlements, community members (citizen scientists) helped identify issues and co-designed data collection methods, such as focus group discussions and surveys (Corburn et al., 2022). Another instance is a study where citizen scientists used air quality sensors and worked with professional researchers to analyse data and develop solutions for urban air pollution (Kumar et al., 2023). Additionally, citizen scientists collaborated with professional researchers in a human-computer interaction study to design and test new digital tools for community use (Cooney and Raghavan, 2022). Notably, although this level of involvement generally leads to higher social acceptance and impact, it also demands significant time, resources, and commitment.

Citizen-led (Level 4) was not recorded in the BE field. This is primarily because, in built environment research, it is unrealistic for citizen scientists to take full control of the research process, from posing research questions to designing methods and analysing data, and it also presents challenges in maintaining scientific rigour and resource allocation, often requiring substantial support and training for citizen scientists to succeed.

Figure 7 displays a network map of co-occurring keywords, which is generated by CiteSpace software based on the frequency of keyword appearances in the selected literature, indicating focal areas, thereby providing a comprehensive overview of the current research hotspots. The co-occurrence of these keywords demonstrates the interrelation of the urban built environment’s physical, social, and health aspects, highlighting the integrated applications of the CS approach in the broad BE field.

The frequent appearance of keywords such as “residential characteristics”, “community participation”, “neighbourhood”, “urban areas”, “cities”, “environmental design”, and “environmental planning” underscores that CS research in the BE field mainly focuses on community-centred planning and development within the urban context. This hotspot also reveals the CS research’s attention on the intrinsic connection between the physical built environment and community involvement in shaping liveable urban environments. Specifically, “residential characteristics” refers to the physical attributes of the built environment, emphasising tangible aspects of residential areas such as housing, infrastructure, public spaces, etc. “Community engagement” emphasises the role of residents in decision-making and research processes, while “neighbourhood” refers to the social structure and collective identity within specific geographical areas, both of which pertain to the social and participatory dimensions of urban development. Furthermore, “urban areas” and “cities” indicate that CS research was mainly conducted within urban contexts. Finally, “environmental design” and “environmental planning” highlight the importance of using CS research for scientific planning and design in optimising urban spaces, improving residents' quality of life, promoting sustainable community development, and creating environments that are conducive to long-term liveability.

Keywords including “physical activity,” “health promotion,” “public health,” and “exercise” reflect a significant focus of CS research on health-related hotspots within the BE field. These keywords underscore the importance of understanding how urban design and residential characteristics influence physical health and well-being. The keywords “aging” and “aged” focus specifically on older adults. By expanding on the keyword’s nodes in CiteSpace, it is evident that there is a close relationship with the broader goal of creating age-friendly communities, which are designed to support the well-being and engagement of older adults, ensuring they can live independently and fully participate in society.

Keywords clustering timeline groups keywords from literature based on their co-occurrence relationships using Log-Likelihood Ratio (LLR) clustering analysis and arranges them along a time axis (see Figure 8). LLR clustering is chosen for its ability to identify significant keyword associations, ensuring that the resulting clusters represent meaningful and statistically relevant themes while visually displaying changes in research themes over time.

Figure 8 illustrates the evolution of various research themes in CS research within the broad BE field from 2011 to 2024, highlighting the prevalence of each theme over different time periods and their interconnections. Specifically, the theme of #physical activity has remained consistently prevalent from 2014 to 2024, where most of the relevant CS research focused on developing effective planning and architectural design to enhance residents' physical activity levels, thereby improving their overall health.

From 2014 to 2022, #minority health has been a prevalent theme, reflecting a more inclusive and equitable research perspective on marginalised communities' health and living conditions. In more recent years, the prevalence of similar themes, such as #environmental justice and #disadvantaged neighbourhoods, have further marked a shift towards developing a more inclusive and equitable living environment using CS research, emphasising the health and well-being of these marginalised populations.

Over the last decade there was a notable prevalence of #digital health (2014–2024), #computer vision (2021–2022), #discovery tools (2016–2024), and #sensitivity analysis (2017–2020), in addition to #citizen science (2011–2020). The themes of #digital health, #computer vision, and #discovery tools reflect the increasing integration of modern information communication technologies and mobile devices in data collection and processing, which have been widely applied in CS research. Themes like #sensitivity analysis underscore that understanding and predicting the potential impacts of various variables in CS research grabs great attention from researchers, such as data collection accuracy, citizen scientists’ behaviour diversity, and environmental condition fluctuations, etc. The #citizen science theme has consistently remained prevalent, emphasising the crucial role of public participation in the scientific research process.

Lastly, the theme of #noise pollution has gained attention since around 2017, with research focusing increasingly on the impact of urban noise on residents. In urbanisation, effective management and control of noise pollution have become significant issues.

Despite the growing interest in CS applications in the BE field, several persistent challenges continue to hinder the effectiveness and scalability of such initiatives, as summarised in Table 5. Among the 107 reviewed articles, the most frequently reported challenge—mentioned in 53 studies—was maintaining the long-term engagement of citizen scientists. BE-related CS projects often span weeks or months and require sustained involvement in tasks like environmental monitoring and participatory planning, which can lead to fatigue and dropout. Key causes include vague project goals, limited feedback, and lack of recognition.

To improve retention, future CS projects should adopt adaptive engagement strategies. Gamification features (e.g. points, badges) can boost motivation, and real-time dashboards that visualise contributions can reinforce participants’ sense of impact. Public recognition or certificates from local institutions may add social value while embedding CS activities into community programs (e.g. planning workshops) can foster belonging—both critical to sustained participation.

A second key challenge—reported in 33 articles—is citizen scientists' lack of technical skills, especially in tasks such as environmental sensing, mobile data entry, and spatial mapping. These skill gaps can result in inaccurate or incomplete data, compromising the reliability and usefulness of research outcomes. To address this, projects should provide scalable training through in-app tutorials, microlearning modules, or hands-on workshops—methods that are low-cost and adaptable to diverse groups. Pairing novices with trained facilitators (e.g. research assistants, NGO staff, and experienced volunteers) can also offer practical guidance. In addition, AI-assisted validation tools—such as anomaly detection or photo classification, available on many open-source CS platforms—can improve accuracy and reduce manual review. A related issue, mentioned in 19 articles, is data variability. Even with training, inconsistencies in understanding and execution can lead to incomplete or unreliable data. This can be mitigated through standardised tools, multi-layered quality control (e.g. automated checks plus human oversight), and routine data reviews.

Sample representativeness was noted as a concern in 14 articles. Many CS projects disproportionately engage participants from more educated, tech-literate, and urban populations, leading to sample bias and limited generalisability. To address this, projects should adopt inclusive recruitment strategies targeting underrepresented groups—such as older adults, ethnic minorities, and low-income communities. Effective methods include partnering with local organisations, offering multilingual and mobile-friendly platforms, and providing appropriate incentives to reduce participation barriers.

Ethical and privacy concerns—raised in 9 articles—pose additional challenges, particularly in projects involving health data or geolocation. Ensuring informed consent, protecting anonymity, and securing data are essential for maintaining trust in citizen engagement and data practices. To that end, CS projects should adopt transparent governance frameworks, embed ethics review from the outset, and consider secure technologies such as blockchain or differential privacy.

Addressing these technical and inclusion-related challenges can enhance data quality and participant experience. Beyond technical fixes, broader structural changes are also needed. These include multi-stakeholder partnerships with municipalities, academia, and communities as well as policy frameworks that formally recognise CS. Measures like government-backed micro-grants, streamlined data-sharing mechanisms, and integrating citizen-generated data into official planning workflows can help institutionalise CS as a routine element of urban governance. These strategic responses address immediate implementation issues and point toward broader technological and institutional shifts, which are further discussed in the next section.

The technological and institutional implications of CS in the Built Environment can be better understood by reflecting on the engagement barriers, data quality issues, and representativeness concerns highlighted in the preceding analysis.

First, emerging technologies present new opportunities to transform the practice of CS in the BE context. Tools such as mobile sensing platforms, Internet of Things (IoT) devices, and AI-based systems will enable more accurate, real-time, and large-scale data collection while reducing participant burden and manual errors. For example, sensor-based monitoring can automate data capture and improve temporal resolution. AI techniques can assist in validating or classifying participant data, improving consistency. Mobile apps and dashboards also offer immediate feedback, encouraging sustained engagement. However, these technologies also raise concerns about privacy, accessibility, and algorithmic bias. As most CS projects remain in the early stage of adoption, further research is needed to evaluate their practical value across diverse geographical, socio-economic, and institutional settings.

Second, regional disparities continue to shape the global CS landscape. As the geographical distribution shows, countries like the United States and the United Kingdom dominate global networks, while many regions—especially in the Global South—remain underrepresented. This likely reflects disparities in research infrastructure, funding, digital access, and supportive policy environments for participatory methods. These barriers deserve deeper investigation, particularly to identify practical obstacles to access and policy implementation in underrepresented regions and to inform future capacity-building efforts.

Finally, embedding CS into institutional systems demands proactive policy interventions rather than ad hoc pilot programs. These shall include dedicated funding for long-term CS projects, the formal integration of CS-generated data into official monitoring and planning systems, and the establishment of structured platforms for collaboration among municipalities, academia, and local communities. Clear regulatory frameworks must define data ownership, enforce privacy protections, and ensure accountability mechanisms are in place for ethical use. Embedding CS into planning and environmental governance can help integrate citizen contributions into formal decision-making processes, enhance transparency, and strengthen public trust in urban governance.

This review provides a structured overview of Citizen Science (CS) research in the Built Environment (BE) field based on 107 peer-reviewed articles published across 52 journals. Drawing on both bibliometric and content analysis, it outlines key characteristics of CS research in BE, including publication trends, author collaboration, geographical distribution, citizen scientists' engagement level, etc. Most CS projects focus on environmental monitoring, urban planning, and public health. Based on the current distribution of topics and project designs, the adoption of the CS approach in the built environment field is at an early and exploratory stage. In addition, most of these CS projects involve citizen scientists for the primary purpose of data collection. Collaborative and Co-created involvement for CS participants are less common, especially in earlier or more analytical phases of research, which is mainly due to time constraints, lack of technical skills, and institutional reluctance to decentralise decision-making.

Several persistent challenges hinder the effective implementation of citizen science in BE, including limited participant retention, data quality issues, sample bias, and ethical concerns. While emerging technologies such as IoT, AI, and mobile sensing offer promising tools to improve data reliability and engagement, their adoption remains limited. Moreover, regional disparities are evident: countries like the US and the UK dominate the global CS landscape, while regions in the Global South are underrepresented. These imbalances are often driven by unequal access to funding, infrastructure, and policy support. Addressing these challenges requires sustained support and structural integration of CS initiatives. Key measures include dedicated funding for CS projects, the incorporation of citizen-generated data into official decision-making systems, and the development of collaborative partnerships among municipalities, researchers, and local communities. Regulatory frameworks are also essential to clarify data ownership, protect privacy, and ensure accountability to foster ethical and responsible participation.

Future research shall explore how emerging technologies can be responsibly integrated into CS to improve scalability and trust. Cross-regional comparisons can help uncover structural barriers. Studies may also examine links between participation levels and research impact. Finally, topic modelling and other computational tools can be adapted to reveal deeper thematic trends of CS adoption across disciplines.

This study has limitations. As it focuses on English-language publications and includes only three databases, it may exclude some regional or grey literature. In addition, the engagement-level analysis is descriptive and does not infer statistical relationships. Despite these constraints, the review offers a clear picture of the current CS landscape in BE and provides a foundation for future research and practice.

We would like to thank the Australian Research Council for funding this research under grant number DP230101313. We would like to express our deepest gratitude to Professor Bo Xia, from the School of Architecture and Built Environment, Queensland University of Technology, for his invaluable guidance and support throughout this research. We also thank Professor Laurie Buys, Dr Qing Chen, and Dr Kirsty Volz for their insightful feedback, contributions, and support.