Green Infrastructure for Urban Flooding: Knowledge Domains and Research Evolution (2015–2024)

Abstract

Urban flooding represents a critical socio-ecological challenge exacerbated by climate change and rapid urbanization, with green infrastructure (GI) emerging as a transformative approach to flood management. This study employs an innovative methodological framework integrating the Biblioshiny, CiteSpace, and Orange3 analytical tools to examine research trends and evolutions in GI for urban flooding from 2015 to 2024. The bibliometric analysis of 813 publications reveals a profound epistemological transition from technically oriented approaches toward integrated socio-ecological frameworks. The citation patterns demonstrate increasing scholarly attention on multifunctionality, climate resilience, and governance dimensions, with the United States and China emerging as dominant research hubs. The analysis identifies distinct thematic clusters reflecting the field’s intellectual progression from hydrological engineering paradigms toward systemic conceptualizations that recognize the complex interactions between technical, ecological, and social dimensions. Despite these advancements, persistent knowledge gaps remain regarding longitudinal performance evaluations, governance frameworks for maintenance, and scalar integration from site-specific interventions to watershed-level outcomes. These findings emphasize the need for methodological innovation addressing the temporal dimensions of GI performance and institutional arrangements for its implementation across diverse urban contexts, positioning GI as a critical component of sustainable urban water management amid increasing climatic uncertainty.

1. Introduction

The escalating frequency and severity of urban flooding represents a critical socio-ecological challenge of the 21st century, emerging from the complex interplay between accelerating climate change impacts, unprecedented urbanization rates, and the systemic limitations of aging infrastructure networks [1]. This multifaceted crisis is particularly exacerbated by the increasing imperviousness of urban surfaces, altered hydrological regimes, and the intensification of extreme precipitation events [2,3]. Within this context, green infrastructure (GI) has emerged as a transformative approach to urban flood management, encompassing a sophisticated network of nature-based solutions including, but not limited to, bioretention systems, constructed wetlands, green roofs, rain gardens, permeable pavements, urban forests, and bioswales [4]. These interventions operate through complex ecohydrological processes, facilitating enhanced water infiltration through soil matrix systems, promoting evapotranspiration through vegetation, reducing peak runoff volumes through temporary storage mechanisms, and improving water quality through multi-layered biological and physical filtration processes [5]. The integration of GI into urban planning frameworks represents a paradigmatic shift from conventional gray infrastructure approaches, offering a multifunctional solution matrix that simultaneously addresses flood mitigation, ecosystem service enhancement, biodiversity conservation, urban heat island reduction, and community wellbeing improvement [6,7]. This holistic approach to urban water management aligns with broader sustainability objectives while providing adaptive capacity in the face of increasing climatic uncertainties and urban development pressures [8].

The corpus of systematic literature reviews examining the intersection of GI and urban flooding has established a robust foundation of empirical evidence, while simultaneously illuminating critical knowledge gaps and methodological limitations. Contemporary reviews have meticulously documented the performance metrics of various GI interventions across diverse urban contexts, quantifying their effectiveness in flood risk reduction while analyzing their contributions to urban resilience enhancement [9,10]. These analyses have revealed the contextual dependencies of GI implementation, emphasizing how performance variations are influenced by multiple factors, including local climatic conditions, urban morphological characteristics, existing infrastructure systems, and socio-economic parameters [11,12]. However, existing systematic reviews have predominantly concentrated on quantifiable hydrological performance indicators and economic feasibility metrics, often inadequately addressing the complex socio-ecological interactions that fundamentally influence GI implementation success and long-term effectiveness [13]. This analytical bias has resulted in a limited understanding of critical factors such as community acceptance, institutional capacity, policy integration, and long-term maintenance requirements [14]. Furthermore, while these reviews have contributed valuable insights to the field, their reliance on traditional bibliographic analysis methods has potentially constrained their ability to identify and analyze emerging trends, cross-disciplinary connections, and innovative approaches within this rapidly evolving research domain [15,16].

The inherent limitations of conventional systematic review methodologies necessitate the development and implementation of more sophisticated analytical frameworks capable of comprehensively synthesizing the expanding knowledge base in GI and urban flooding research. Traditional bibliometric analyses frequently encounter significant methodological challenges, including sampling bias in database selection, incomplete coverage of relevant literature, the limited ability to detect latent patterns in research networks, and difficulties in capturing the temporal evolution of knowledge structures [17,18]. These methodological constraints often result in overlooked research opportunities, undiscovered knowledge connections, and an incomplete understanding of the field’s theoretical development [19,20].

The integration of these analytical tools facilitates a more nuanced understanding of the complex relationships between GI implementation and urban flooding mitigation, while simultaneously advancing the methodological rigor of systematic literature reviews in this field [21]. This enhanced analytical framework not only provides deeper insights into current research trends but also identifies potential future research directions, methodological innovations, and opportunities for cross-disciplinary collaboration [22], thereby contributing to the advancement of both the theoretical understanding and practical implementation of GI solutions for urban flood management.

This study aims to explore research trends and evolutions in the field of GI related to urban flooding by employing an innovative methodological framework. The method integrates three complementary analytical tools: Biblioshiny for comprehensive bibliometric analysis and visualization, CiteSpace for dynamic scientific mapping and knowledge domain analysis, and Orange3 for advanced data mining, pattern recognition, and predictive analytics. To address the methodological constraints inherent in conventional systematic review approaches, our analytical framework incorporates rigorous validation protocols through systematic cross-platform triangulation. This methodological approach enables the identification of consistent patterns across analytical platforms while simultaneously highlighting potential discrepancies that warrant further investigation. As Abdullah et al. (2023) [23] emphasize in their analysis of bibliometric methodologies, such triangulation processes are fundamental to ensuring robust and theoretically grounded results in complex interdisciplinary research. The integration of diverse analytical perspectives not only mitigates the inherent limitations of individual tools but also enriches the analytical depth, facilitating a more comprehensive understanding of the intellectual landscape surrounding GI and urban flooding research. This novel methodological approach enables the identification of hidden research patterns, visualization of knowledge evolution trajectories, analysis of citation networks, and detection of emerging research frontiers that may be overlooked by conventional review methods [23].

2. Materials and Methods

2.1. Data Sources

This study employed the Web of Science (WoS) as its primary bibliometric data source, leveraging its comprehensive coverage and sophisticated analytical capabilities to examine the evolution of GI and urban flooding research from January 2015 to December 2024. The selection of WoS was predicated on its extensive indexing of high-impact journals and standardized citation data, which facilitates the identification of influential research and emerging trends [24,25]. As demonstrated in recent bibliometric analyses across diverse fields, WoS provides robust analytical tools that enable the comprehensive exploration of research trends, collaboration networks, and intellectual structures [26,27].

The data collection protocol adhered to systematic review principles to ensure reproducibility and methodological rigor. The database query was constructed using a carefully refined combination of search terms: ALL = (Green infrastructure) AND (QMTS = (Urban flooding)) AND (DT == (ARTICLE or REVIEW)) AND (LA == (English)). These terms were applied to search within titles, abstracts, and keywords (Topic field in WoS), with the timespan restricted to publications between 2015 and 2024. To maintain scholarly focus, the search was limited to peer-reviewed articles in English, excluding reviews, conference proceedings, book chapters, and gray literature. In addition, we excluded cases where the full text of the paper was not provided due to using the full text as the subject of analysis. Based on the previous query string, Figure 1 shows the data source search structure. This systematic approach aligns with established bibliometric methodologies that emphasize the platform’s capability to reveal internal structures and hidden inferences in academic publishing [24].

The initial dataset underwent a systematic refinement process to ensure relevance and quality. This refined dataset formed the foundation for subsequent analyses using three complementary bibliometric tools: Biblioshiny, CiteSpace version 6.4.R1, and Orange3.38.1. The integration of these tools was supported by WoS’s demonstrated compatibility with sophisticated bibliometric software, enabling dynamic visualizations and systematic reviews of the literature [27]. This analytical framework facilitates the mapping of intellectual bases and detection of research fronts through bibliometric indicators and visualization tools [26].

The selection of Web of Science as the primary data source was further justified by its standardized metadata structure, which ensures compatibility with these analytical tools while maintaining data integrity throughout the analysis pipeline. This methodological framework enables a rigorous examination of research trends, collaboration networks, and knowledge structures within the GI and urban flooding domain, while acknowledging the potential limitations of WoS’s coverage in terms of regional representation and non-indexed sources.

2.2. Methodology

The methodology employed in this systematic literature review adopts a multi-layered analytical framework that integrates complementary bibliometric approaches to comprehensively examine the evolution and current state of GI and urban flooding research. This methodological framework builds upon established systematic review protocols while incorporating advanced computational tools for enhanced analytical depth [4,7]. The analytical process employs three distinct complementary software platforms: Biblioshiny, CiteSpace, and Orange3, each serving specific analytical functions within the broader methodological framework.

The initial phase of analysis employs Biblioshiny, a sophisticated web interface for the bibliometrix R package, which serves as the foundational platform for comprehensive bibliometric examination. This tool facilitates the initial bibliometric overview, providing comprehensive insights into publication patterns, collaboration networks, and broad research trajectories. As demonstrated by Haustein and Larivière (2014) [23], Biblioshiny’s analytical capabilities enable the construction of robust knowledge architectures through its systematic approach to bibliometric data processing. This platform’s capability of generating systematic knowledge structures aligns with established bibliometric methodologies [22], enabling the identification of foundational works and emerging research frontiers within the GI domain.

The second phase leverages CiteSpace’s advanced analytical capabilities, particularly its sophisticated algorithms for burst detection and clustering analysis, to uncover the temporal dynamics and evolutionary patterns within the research domain. As demonstrated by Yang and Meho (2007) [20], this methodological approach reveals critical intellectual transitions and paradigm shifts in the field’s development. The temporal dimension of the analysis proves especially valuable in deconstructing the complex trajectory of GI research, illuminating how theoretical frameworks emerge, converge, and evolve across different temporal scales. This temporal analysis is particularly crucial for understanding the dynamic nature of GI research, as it illuminates the emergence and convergence of key theoretical frameworks over time [20]. CiteSpace’s robust clustering algorithms enable the identification and characterization of distinct scholarly communities, revealing intricate networks of theoretical interconnections and intellectual lineages. This granular analysis of research clustering patterns complements the broader bibliometric overview provided by Biblioshiny, offering deeper insights into the field’s intellectual architecture and theoretical foundations. The platform’s ability to detect and visualize these nuanced relationships provides a sophisticated understanding of how different theoretical approaches and methodological frameworks have shaped the evolution of GI research. The methodological framework incorporates citation burst analysis—a sophisticated bibliometric approach that algorithmically identifies statistically significant surges in citation frequencies across temporal dimensions. As developed by Chen (2005) [28], this technique illuminates publications and keywords that have exerted exceptional influence on intellectual trajectories within specific time periods. This methodological approach transcends conventional citation counts by capturing the dynamic nature of knowledge diffusion, revealing not merely which publications have been influential, but precisely when their impact reached critical thresholds within the scholarly community (Shibata et al., 2008) [29]. By detecting these “bursts” of scholarly attention, the analysis provides a temporal mapping of intellectual evolution, identifying emergent research fronts and conceptual transitions that characterize the field’s development. In addition, to systematically examine the intellectual structure and thematic evolution of GI and urban flooding research, the methodological framework incorporates keyword co-occurrence network analysis. This approach constructs comprehensive network visualizations using the CiteSpace software (Chen, 2005) [28], which facilitates the detection and visualization of emerging trends and transient patterns in the scientific literature. Following the methodological protocols established in bibliometric research, the analysis employs fractional counting methods to normalize the influence of publications with multiple keywords, thereby producing a more balanced network structure than would be achieved through full counting approaches (Perianes-Rodriguez et al., 2016) [30]. This analytical technique facilitates the visualization of conceptual relationships and thematic clustering through an examination of how keywords co-appear across the literature corpus, revealing both explicit and implicit knowledge structures that characterize the field’s conceptual architecture.

The third methodological component employs Orange3’s sophisticated data mining framework to conduct comprehensive textual analysis of the complete document corpus. This advanced analytical platform transcends traditional bibliometric approaches by enabling nuanced content examination through its machine learning and natural language processing capabilities. As Stuart (2018) [19] articulates, Orange3’s analytical depth facilitates the extraction of latent theoretical frameworks and methodological paradigms that may remain obscured in conventional citation-based analyses. The platform’s integration into the methodological framework proves particularly valuable in uncovering implicit theoretical connections and conceptual bridges between diverse approaches to GI and urban flooding management. Through its advanced text mining algorithms and visualization capabilities, Orange3 enables the identification of subtle intellectual linkages and emerging theoretical constructs that shape the field’s development. This methodological approach provides crucial insights into the conceptual underpinnings of GI research, revealing the intricate relationships between theoretical frameworks that might otherwise remain undetected through traditional bibliometric methods alone.

The following Figure 2 shows the systematic flow of the research methodology employed in this study. The diagram delineates the sequential progression of methodological procedures, providing a visual representation of the investigative framework that guided the analytical process.

In acknowledgment of the methodological constraints, this analytical framework recognizes several inherent limitations and potential sources of bias that warrant careful consideration. Haustein and Larivière’s (2014) [23] seminal work on bibliometric methodologies illuminates the fundamental constraints imposed by database coverage parameters and citation dynamics within academic publishing systems. The exclusive use of Web of Science, while providing standardized data suitable for bibliometric analysis, may underrepresent certain geographical regions or emerging research areas [17]. As Wang et al. (2023) [7] demonstrate in their comprehensive analysis of green stormwater infrastructure research, such database limitations can particularly impact the representation of innovative approaches and regional perspectives in environmental planning research. Furthermore, the distinct analytical architectures of Biblioshiny, CiteSpace, and Orange3 necessitate a nuanced approach to data interpretation, requiring careful methodological consideration to ensure coherent synthesis across these complementary yet distinct analytical platforms.

The integration of multiple analytical platforms, while methodologically robust, introduces distinct analytical challenges that require careful consideration. Biblioshiny’s strength in identifying broad epistemological patterns and research trajectories is counterbalanced by its limitations in granular content analysis. As Junqueira et al. (2021) [15] observe in their systematic review of GI modeling approaches, such methodological constraints can impact the depth of analysis possible for complex interdisciplinary topics. CiteSpace’s sophisticated temporal analytical framework, while offering valuable insights into research evolution, is inherently affected by citation latency phenomena, potentially obscuring recent theoretical developments and emerging research directions in rapidly evolving fields [18]. This limitation becomes particularly salient when examining contemporary advances in GI research, where innovative approaches may not yet be fully reflected in citation networks. Orange3’s advanced text mining capabilities, though offering powerful analytical possibilities, demand meticulous parameter calibration to yield meaningful insights, especially when navigating the complex interdisciplinary landscape of GI research. The platform’s effectiveness in analyzing such multifaceted research domains requires careful methodological consideration to ensure robust and theoretically grounded results.

The methodological framework’s distinctive strength emerges from its sophisticated synthesis of multiple analytical perspectives while maintaining rigorous scholarly standards. This integrated approach, combining the complementary analytical capabilities of Biblioshiny, CiteSpace, and Orange3, facilitates a nuanced examination of both explicit and implicit knowledge structures within the research domain. The resulting multi-layered analytical framework establishes a robust theoretical foundation for identifying emerging research frontiers and intellectual trajectories. This methodological sophistication enables a comprehensive understanding of the field’s evolution while acknowledging the inherent epistemological and practical constraints of bibliometric analysis. The framework’s ability to navigate these complexities while maintaining analytical rigor provides a solid foundation for advancing theoretical understanding in the domain of GI and urban flooding research.

3. Results

3.1. Literature Overview

The systematic analysis of green infrastructure (GI) and urban flooding research reveals a robust and rapidly expanding field of study over the past decade (2015–2024).

Table 1. Basic information on green infrastructure and urban flooding research (2015–2024).

Metric Category	Value	Additional Context
Time Span	2015–2024	Decade of analysis
Number of Sources	214	Journals and other publication venues
Documents	813	Total publications analyzed
Annual Growth Rate	32.59%	Compound annual publication growth
Total Authors	2922	Unique contributing scholars
Single-Authored Documents	41	Publications with one author
International Co-Authorship	33.83%	Percentage of multi-country collaborations
Average Co-Authors per Document	4.33	Mean number of authors per publication
Authors’ Keywords	2259	Unique author-assigned keywords
References	35,707	Total citations in all publications
Document Average Age Average Citations per Document	3.9 years 22.55	Mean age of publications in dataset Mean citation frequency

illustrates the field’s remarkable growth, characterized by an annual growth rate of 32.59%, resulting in 813 documents published across 214 different sources, reflecting the increasing recognition of GI’s critical role in addressing contemporary urban flooding challenges. The research landscape demonstrates strong collaborative tendencies, evidenced by an average of 4.33 co-authors per document and an international co-authorship rate of 33.83%, indicating robust global research networks and knowledge exchange mechanisms. Despite this predominance of collaborative research, the field maintains a degree of individual scholarship, with 41 single-author papers contributing valuable perspectives to the discourse. The intellectual diversity of the field is demonstrated through the engagement of 2922 authors who have collectively employed 2259 unique keywords, reflecting the multifaceted conceptual frameworks and methodological approaches that characterize this research domain. The substantial foundation of knowledge is further evidenced by the 35,707 references cited across the literature, indicating a rich theoretical and empirical groundwork. The relatively young average document age of 3.9 years suggests a dynamic and current research landscape, while the impressive average citation rate of 22.55 citations per document indicates the high impact and relevance of the research in this field within the academic community.

The temporal distribution of research on GI and urban flooding exhibits a clear upward trajectory from 2015 to 2024, revealing significant trends in research productivity and impact. As illustrated in

Table 2. Annual scientific production and citations from 2015 to 2024.

Year	MeanTCperArt 1	No. of Articles	MeanTCperYear 2	Citable Years
2015	89.42	12	8.13	11
2016	66.05	19	6.60	10
2017	70.50	38	7.83	9
2018	48.40	65	6.05	8
2019	35.40	73	5.06	7
2020	29.47	83	4.91	6
2021	20.81	124	4.16	5
2022	11.06	124	2.77	4
2023	7.59	123	2.53	3
2024	1.76	152	0.88	2

¹ Mean Total Citations per Article. ² Mean Total Citations per Year.

, the analysis of publication patterns demonstrates the field’s continuous evolution, which began with foundational contributions in 2015–2016 that established seminal theoretical frameworks, despite the relatively modest publication volumes of 12 and 19 articles, respectively. These early works garnered exceptional scholarly attention, with publications in 2015 achieving a mean citation rate of 89.42 citations per article—indicating their formative influence on subsequent research trajectories. As scholarly interest intensified, publication numbers steadily increased through 2017–2019, reaching 73 articles by 2019 while maintaining robust citation metrics, notably a mean of 70.5 citations per article for the contributions in 2017. This growth coincided with increasing global recognition of climate adaptation imperatives and the expanding application of nature-based solutions in urban contexts. The scholarly momentum continued to accelerate, with annual publications nearly doubling between 2020 and 2022, from 83 to 124 articles, reflecting heightened research interest amid growing climate change concerns and urban resilience initiatives worldwide. The recent scholarly output in 2023–2024 maintained this trajectory, with publications reaching 152 in 2024—the highest annual figure in the analyzed decade. Citation patterns across the temporal continuum follow an expected gradient, with earlier works accumulating greater citation densities over time (publications from 2015 achieving the highest mean number of citations per year at 8.13), while more recent contributions show an emerging impact relative to their temporal proximity. This pattern illustrates both the field’s intellectual maturation and the characteristic citation lags in academic knowledge dissemination. The sustained increase in publication volume, coupled with the significant citation influence of foundational works, demonstrates that this research domain has not simply expanded quantitatively but has established a coherent intellectual trajectory with substantial scholarly influence on contemporary understandings of GI applications for urban flood resilience.

The geographical distribution of research on GI and urban flooding, as evidenced in

Table 3. Top ten countries by the number of published articles.

No.	Country	Counts	Year of Appearance	Percentage
1	USA	472	2015	19.42%
2	CHINA	449	2015	18.48%
3	UK	221	2015	9.09%
4	AUSTRALIA	103	2015	4.24%
5	BRAZIL	93	2015	3.83%
6	ITALY	93	2016	3.83%
7	GERMANY	85	2016	3.50%
8	SOUTH KOREA	85	2016	3.50%
9	NETHERLANDS	72	2015	2.96%
10	JAPAN	59	2017	2.43%

, reveals distinct patterns of scholarly activity across different regional contexts. The United States and China emerge as the dominant contributors to the field, with the USA showing consistent leadership through 472 publications by 2024 and China demonstrating remarkable growth to reach 449 publications. This bilateral leadership reflects both countries’ significant investments in environmental research and their acute experiences with urban flooding challenges. Following these leaders, the United Kingdom has established itself as a significant European hub with 221 publications, showing steady growth from just 3 publications in 2015 to a substantial body of work by 2024. Australia has also made notable contributions, with 103 publications by 2024, while Brazil has emerged as a leading voice from the Global South with 93 publications.

European nations demonstrate a strong collective presence in the research landscape, with Germany (85 publications), the Netherlands (72 publications), and Italy (93 publications) showing consistent growth in their research output. This European concentration likely reflects the region’s progressive environmental policies and substantial investments in climate adaptation research. Asian countries beyond China have also made significant contributions, with Japan (59 publications), Korea (85 publications), and India (47 publications) demonstrating increasing engagement in the field. An examination of temporal–geographical patterns reveals an evolution from an initial concentration in established research economies toward increasing diversification, with emerging scholarly contributions from countries including Iran (23 publications), Thailand (9 publications), and South Africa (13 publications). This geographical broadening reflects the global relevance of urban flooding challenges and GI solutions across diverse socio-economic and climatic contexts, as well as the progressive integration of varied ecological, hydrological, and socio-spatial perspectives into the field’s intellectual framework.

3.2. Clustering and Burst Analysis

3.2.1. Co-Citation Analysis of the Literature

Co-citation analysis reveals the intricate patterns of intellectual relationships within GI and urban flooding research through an examination of citation frequencies and network metrics. In this study, CiteSpace’s literature co-citation analysis function is employed to reveal the knowledge structure of GI and urban flooding research through an analysis of clustering patterns and key nodes in the co-citation network [31]. The analysis is configured with annual time slices, using reference documents as node types. The analytical parameters are specified as follows: link retaining factor (3.0), look back year (−1), maximum links per node (10), percentage of nodes to label (1.0%), and filter references by intrinsic citations (true), while maintaining default values for the remaining parameters. Network visualization is optimized through the application of Pathfinder and pruning algorithms to both merged and sliced networks. The analysis of simultaneous citations and network structures provides crucial insights into the field’s scholarly interconnections and knowledge flow patterns. Citation frequency analysis demonstrates a hierarchical structure of scholarly influence, with publications receiving between 40 and 137 simultaneous citations forming the backbone of the knowledge network. Most notably, the highest frequency of simultaneous citations (137) is observed in methodological works that bridge theoretical frameworks with practical applications, while maintaining moderate centrality measures (0.11) in the network structure [32]. This pattern suggests that methodological contributions have been particularly influential in shaping the field’s development.

The network demonstrates moderate density, characterized by an average degree of centrality of 3.15, which indicates substantial intellectual exchange among key publications. Within this structure, several publications exhibit notably high centrality measures (exceeding 0.12) and function as knowledge bridges across diverse research domains. These highly central works characteristically display moderate citation frequencies (ranging from 30 to 60 citations), suggesting their vital role in facilitating cross-domain knowledge transfer rather than merely accumulating citations [31]. The degree distribution exhibits significant variation in connection patterns, with key nodes maintaining degrees of 14 or higher. These highly connected publications frequently correlate with moderate to high citation frequencies (40 to 60 citations), suggesting that works of broader influence tend to accumulate both direct citations and co-citation relationships [33]. This interplay between degree centrality and citation frequency indicates that influential works achieve their impact through dual mechanisms: direct scholarly influence and their function of connecting distinct research streams.

The temporal distribution of high-frequency citations (>50) indicates an evolution in research impact patterns, with earlier foundational works maintaining consistent co-citation relationships with more recent publications. This temporal pattern, combined with network metrics, suggests a field that effectively builds upon established knowledge while incorporating new theoretical perspectives and methodological approaches [34]. The analysis reveals that publications with the highest frequencies (>100 citations) often maintain significant betweenness centrality measures, indicating their role as methodological or conceptual cornerstones in the network structure. Moreover, the relationship between citation frequency and network positioning reveals that while some highly cited works (>50 citations) maintain peripheral positions in the network structure, others with lower citation counts serve as crucial connectors, suggesting diverse patterns of influence in the field’s intellectual development [35]. The disproportionate influence of publications with moderate citation counts but high centrality measures emerges from their strategic positioning at the intersection of distinct research domains. For instance, Hansen and Pauleit’s (2014) [36] conceptual framework for multifunctionality in GI planning demonstrates this phenomenon, receiving 48 citations (moderate compared to the highest frequency of 137) but achieving a centrality measure of 0.18. This publication functions as a crucial connector by explicitly bridging ecosystem service frameworks with urban planning paradigms, facilitating knowledge transfer between ecological and planning disciplines that might otherwise remain disconnected. Similarly, Jayasooriya and Ng’s (2014) [37] methodological review achieves high centrality (0.15) despite a moderate citation frequency (43) by systematically connecting modeling approaches with implementation frameworks, thereby linking technical and governance-oriented research clusters. Publications functioning as connectors typically possess three characteristics: (1) interdisciplinary theoretical frameworks that integrate concepts from multiple domains, (2) methodological innovations that can be applied across diverse research contexts, and (3) comprehensive literature reviews that explicitly synthesize previously disparate knowledge structures. These characteristics enable certain publications to transcend disciplinary boundaries and facilitate intellectual cross-fertilization despite not necessarily accumulating the highest citation counts. This varying relationship between citation frequency and network centrality measures indicates a complex knowledge structure where influence is manifested through multiple pathways.

Table 4. Top ten frequently co-cited publications in the present study data.

No.	Title	Sources	Authors	Year	Co- Citations
1	SUDS, LID, BMPs, WSUD and more– The evolution and application of terminology surrounding urban drainage	Urban Water Journal	Fletcher, T. D., Shuster, W., Hunt, W. F., Ashley, R., Butler, D., Arthur, S., Trowsdale, S., Barraud, S., Semadeni-Davies, A., Bertrand-Krajewski, J.-L., Mikkelsen, P. S., Rivard, G., Uhl, M., Dagenais, D., and Viklander, M.	2015	137
2	Enhancing future resilience in urban drainage system: Green versus grey infrastructure	Water Research	Dong, X., Guo, H., and Zeng, S.	2017	60
3	Spatial planning for multifunctional green infrastructure: Growing resilience in Detroit	Landscape and Urban Planning	Meerow, S., and Newell, J. P.	2017	58
4	Integrated assessments of green infrastructure for flood mitigation to support robust decision-making for sponge city construction in an urbanized watershed	Science of The Total Environment	Mei, C., Liu, J., Wang, H., Yang, Z., Ding, X., and Shao, W.	2018	58
5	Performance and implementation of low impact development–A review	Science of The Total Environment	Eckart, K., McPhee, Z., and Bolisetti, T.	2017	53
6	Assessing the Co-Benefits of green-blue-grey infrastructure for sustainable urban flood risk management	Journal of Environmental Management	Alves, A., Gersonius, B., Kapelan, Z., Vojinovic, Z., and Sanchez, A.	2019	53
7	The effects of low impact development on urban flooding under different rainfall characteristics	Journal of Environmental Management	Qin, H., Li, Z., and Fu, G.	2013	51
8	Promoting ecosystem and human health in urban areas using Green Infrastructure: A literature review	Landscape and Urban Planning	Tzoulas, K., Korpela, K., Venn, S., Yli-Pelkonen, V., Kaźmierczak, A., Niemela, J., and James, P.	2007	49
9	Low Impact Development Practices: A Review of Current Research and Recommendations for Future Directions	Water, Air, and Soil Pollution	Dietz, M. E.	2007	48
10	“Sponge City” in China—A breakthrough of planning and flood risk management in the urban context	Land Use Policy	Chan, F. K. S., Griffiths, J. A., Higgitt, D., Xu, S., Zhu, F., Tang, Y.-T., Xu, Y., and Thorne, C. R.	2018	46

The data presented in the table represents the results of bibliometric analysis, identifying the most influential publications in terms of co-citation frequency within the field.

presents a comprehensive analysis of the ten most frequently co-cited publications within the literature network, illustrating the core intellectual foundations of the field through patterns of simultaneous citation.

3.2.2. Citation Burst Analysis

As depicted in Figure 3 which visualizes the 25 references exhibiting the strongest citation bursts during the 2015–2024 period, this analysis provides a nuanced cartography of the epistemological landscape, demonstrating how scholarly attention has coalesced around specific theoretical constructs and methodological paradigms at particular historical junctures [38].

The temporal distribution of citation bursts illuminates the intellectual evolution of green infrastructure (GI) research, revealing a dynamic progression of theoretical orientations and methodological approaches throughout the decade. An examination of citation patterns demonstrates how foundational frameworks established in the early years have undergone continuous refinement and conceptual expansion, reflecting the field’s increasing interdisciplinary complexity and the diversity of its applications.

Early discourse in the field was characterized by the establishment of essential conceptual foundations for GI implementation. Fratini et al.’s (2012) [39] examination of urban water governance paradigms exerted prolonged influence (burst strength 4.63, 2015–2020), providing critical governance frameworks that would inform subsequent policy development. Concurrently, methodological innovations for environmental modeling emerged as an influential touchpoint, with Lee et al.’s (2012) [40] computational frameworks (burst strength 4.12) establishing quantitative parameters for GI performance assessment. Technical infrastructure for hydrological analysis gained prominence through Rossman’s (2010) [41] Storm Water Management Model (burst strength 3.97, 2015–2018), which standardized simulation protocols for urban runoff dynamics [42] and established methodological consistency across the field. This early research trajectory witnessed the significant integration of ecological perspectives, broadening the conceptual architecture beyond hydrological considerations alone. Walsh et al.’s (2005) [43] work on urban stream ecology demonstrated remarkable influence (burst strength 5.87, 2016–2019), while Paul et al.’s (2001) [44] analysis of urbanization impacts on stream ecosystems exhibited intense, though briefer, citation activity (burst strength 5.75, 2016–2017). These ecological frameworks expanded the theoretical scope to encompass watershed degradation, riparian function, and aquatic ecosystem integrity, establishing critical interdisciplinary linkages between hydrological engineering and ecological science [38]. Benedict’s (2006) [45] conceptualization of GI as an interconnected network of natural systems (burst strength 6.19, 2016–2019) further consolidated the theoretical foundations for subsequent research, emphasizing systemic interconnections rather than isolated interventions [46].

As intellectual discourse matured, scholarly attention shifted toward interdisciplinary integration and implementation frameworks, reflecting the field’s evolution from conceptual foundations toward operational applications. Hansen et al.’s (2014) [36] analysis of urban GI from a planning perspective (burst strength 4.48, 2017–2019) bridged theoretical constructs with spatial planning paradigms, while Elliott et al.’s (2007) [47] methodological contributions maintained sustained influence (burst strength 4.6, 2017–2020). This intellectual evolution incorporated increasing attention towards socio-political dimensions, as evidenced by Lawson et al.’s (2014) [48] examination of governance frameworks (burst strength 4.65, 2017–2020) and Jayasooriya and Ng’s (2014) [37] systematic assessment methodology (burst strength 4.15, 2017–2019). These works collectively signaled the field’s progression toward more sophisticated analytical approaches incorporating institutional, governance, and policy considerations alongside technical parameters [49].

More recent scholarly trajectories reveal emergent research orientations centered on climate resilience, innovative sponge city concepts, and integrated blue–green infrastructure systems, reflecting the field’s continued conceptual evolution toward holistic frameworks. Yin et al.’s (2021) [50] contribution to sustainable urban development models demonstrated exceptional influence (burst strength 5.02, 2022–2024), while Zhou et al.’s (2019) [51] integrated approach to urban water management exhibited significant citation activity (burst strength 4.24, 2022–2024). These contributions signal the consolidated scholarly interest in comprehensive frameworks that transcend traditional siloed approaches to urban water management. Frantzeskaki et al.’s (2019) [52] analysis of environmental policy integration (burst strength 4.15, 2020–2022) underscores the continued relevance of robust governance mechanisms for successful GI implementation [53], illustrating how governance considerations have become increasingly embedded within the field’s intellectual architecture. The citation analysis further reveals an emerging scholarly focus on the limitations of aging infrastructure networks as a critical dimension of urban flooding challenges. While not as prominently represented as climate change and urbanization factors, publications addressing infrastructure obsolescence demonstrate increasing citation activity from 2019 onward. For instance, Dong et al.’s (2017) [54] comparative analysis of green versus gray infrastructure (burst strength 4.48, 2019–2020) explicitly addresses the limitations of conventional infrastructure systems, while Schubert et al.’s (2017) [55] assessment framework (burst strength 4.51, 2020–2021) incorporates infrastructure age and condition as critical variables in flood mitigation planning. This growing attention to infrastructure system limitations aligns with a broader recognition of the “infrastructure gap” in urban water management—the widening disparity between aging, capacity-constrained conventional systems, and increasing hydrological demands imposed by climate change and urbanization pressures. The citation patterns suggest an intellectual trajectory toward hybrid solutions that strategically integrate GI interventions within existing gray infrastructure networks, acknowledging both the limitations of conventional systems and the pragmatic constraints of wholesale infrastructure replacement (Frantzeskaki, 2019; Alves et al., 2019) [52,56]. This hybrid approach represents a theoretically sophisticated response to the tripartite challenge posed by climate intensification, rapid urbanization, and infrastructure deterioration.

Particularly noteworthy in contemporary research is the dialectical relationship between conceptual advancement and technical refinement, exemplified by the renewed scholarly engagement with Rossman’s updated Storm Water Management Model (2015) [57], which experienced substantial citation activity (burst strength 4.24, 2022–2024). This pattern suggests that sophisticated simulation capabilities continue to inform and validate theoretical constructs related to GI performance [30], creating productive feedback loops between conceptual innovation and technical verification. Additional significant contributions include Palla et al.’s (2015) [58] hydrological analysis (burst strength 4.74, 2020–2021) and Schubert et al.’s (2017) [55] advanced water resource modeling (burst strength 4.51, 2020–2021), collectively demonstrating sustained scholarly interest in quantitative assessment methodologies, despite the field’s broadening conceptual parameters [59].

This comprehensive citation burst analysis does not merely illuminate temporal fluctuations in reference popularity, but also fundamental epistemological transformations in how GI and urban flooding are conceptualized within scholarly discourse. The discernible progression from technically oriented solutions focusing primarily on hydrological parameters toward integrated frameworks incorporating ecological, social, and governance dimensions reflects broader paradigmatic transitions in environmental management thinking [35]. The sustained influence of both technical publications (e.g., Rossman, 2010, 2015) and conceptual frameworks (e.g., Benedict, 2006, 2012) demonstrates the field’s balanced emphasis on practical implementation protocols and theoretical development, producing a robust intellectual foundation for addressing complex urban water challenges [34].

Moreover, the temporal patterns revealed through citation burst analysis suggest emergent research trajectories around nature-based solutions, innovative governance models, and climate adaptation strategies that are likely to shape future scholarly discourse. The increasing prominence of publications addressing institutional frameworks and policy integration indicates a growing recognition of the socio-political dimensions essential for successful GI implementation, beyond purely technical considerations [31]. Simultaneously, the persistent relevance of sophisticated modeling techniques suggests continued scholarly attention towards quantitative performance assessment, creating a productive tension between conceptual innovation and empirical validation that characterizes mature scientific fields [33].

3.2.3. Citation Clustering Analysis

This analytical approach reveals a multifaceted intellectual landscape within GI and urban flooding research literature. As visualized in Figure 4, the knowledge structure comprises 13 discrete clusters, each representing a cohesive epistemological domain with distinct research orientations. These clusters demonstrate exceptionally robust silhouette values ranging from 0.683 to 0.984—a statistical metric that quantifies intra-cluster homogeneity on a scale from 0 to 1. Within the CiteSpace analytical framework, silhouette values approaching 1 indicate pronounced thematic coherence and internal node density, confirming that each identified cluster constitutes a distinct, interconnected corpus of scholarship rather than disconnected publications. The consistently high silhouette metrics across all clusters provide compelling statistical validation that our citation-based clustering methodology effectively captures the intellectual architecture of the GI and urban flooding literature [28].

The temporal distribution of these knowledge structures, spanning from 2008 to 2019 based on mean publication years, illuminates the field’s evolutionary trajectory, reflecting shifts in research priorities and conceptual frameworks throughout this formative period.

The largest identified cluster (#0), labeled “sustainable flood risk” according to Latent Semantic Indexing (LSI) analysis (silhouette value: 0.803), encompasses 69 members, primarily published around 2017. This cluster represents contemporary approaches integrating conventional engineered solutions with nature-based infrastructure. Seminal works within this domain include Dong et al. (2017) [54], Mei et al. (2018) [60], and Alves et al. (2019) [56], which collectively advanced the frameworks for optimizing green infrastructure deployment for stormwater management, particularly emphasizing resilience evaluations under climate change scenarios. The substantial citation counts of these publications (60, 58, and 53, respectively) underscore their considerable influence in shaping the contemporary discourse on integrated infrastructure approaches.

Cluster #1, identified as “adaptive configuration” through LSI (silhouette value: 0.801), comprises 55 members and similarly reflects recent scholarly developments (average year: 2017). This cluster focuses on spatial optimization frameworks for strategic infrastructure placement, with influential works such as Alves et al. (2018) [61] presenting multi-objective approaches for infrastructure selection that simultaneously reduce flood risk while enhancing co-benefits. The prominence of optimization methodologies in this cluster indicates a maturation of the field toward evidence-based spatial decision support systems that balance multiple performance criteria.

The intellectual evolution of the literature is discernible through cluster transformation patterns. Analysis reveals a progressive conceptual development trajectory manifested in thematic cluster transitions.

Initial clusters such as “nutrient flux” (Cluster #6, silhouette value: 0.803) and “effective planning support system” (Cluster #9) establish foundational theoretical frameworks for understanding green infrastructure through ecosystem service paradigms [62] and identify institutional barriers impeding sustainable water management implementation [63]. These clusters represent the conceptual groundwork upon which subsequent knowledge structures were built.

A transition toward technical application and empirical assessment is evident in the emergence of clusters focused on methodological approaches. The “fine-scale analysis” (Cluster #8) and “urban vegetation” (Cluster #2) clusters demonstrate this shift through comprehensive empirical evaluations of specific green infrastructure typologies [64] and advancements in stormwater modeling frameworks [65]. This cluster formation pattern reflects an intellectual progression from theoretical propositions toward technical specification and empirical validation.

The knowledge structure subsequently evolves toward integrative and systemic approaches, as evidenced by the formation of clusters such as “regional scale stormwater flood management” (Cluster #5) and “urban sustainability” (Cluster #11). These clusters embody a conceptual convergence around holistic frameworks addressing multiple urban challenges simultaneously [66]. This cluster transformation pattern signifies a maturation from component-based to system-based epistemological frameworks.

The analysis reveals three distinct yet interrelated research domains that characterize the contemporary intellectual landscape: (1) Technical Performance and Optimization (Clusters “sustainable flood risk”, “adaptive configuration”, “using scale-based model”, “fine-scale analysis”), emphasizing quantitative modeling and design optimization; (2) Governance and Implementation (Clusters “nutrient flux”, “effective planning support system”, “alternative strategies”), addressing institutional frameworks and implementation barriers; and (3) Integrated Approaches and Co-Benefits (Clusters “forest-urban watershed”, “regional scale stormwater flood management”, “urban sustainability”), exploring the multifunctional benefits and synergy between flood regulation and broader sustainability objectives. The chronological progression from technical optimization toward integrated socio-ecological approaches suggests a field maturing beyond disciplinary silos toward more holistic frameworks that address the complex, multidimensional challenges of urban flood resilience in the context of climate change.

3.2.4. Keyword Connection Network

The keyword co-occurrence analysis resulted in a network comprising 471 unique keywords (nodes) with varying degrees of connectivity, extracted from the corpus of literature on GI and urban flooding. The Clauset–Newman–Moore algorithm was applied to detect community structures within the network, identifying clusters of closely interconnected keywords that represent distinct research themes [49,59]. Network metrics including frequency (occurrence count), degree (number of direct connections), centrality (positional importance), and sigma (composite measure of betweenness centrality and citation count) were calculated to identify the pivotal keywords that functioned as intellectual anchors or bridges within the knowledge domain [42].

As shown in

Table 5. Top 50 keywords by their frequencies in the present study data.

No.	Keywords	Frequency	No.	Keywords	Frequency
1	Green infrastructure	330	26	Blue–green infrastructure	45
2	Climate change	177	27	Impacts	44
3	Management	153	28	Benefits	43
4	Ecosystem services	133	29	Urban planning	43
5	Low-impact development	113	30	Stormwater	42
6	Stormwater management	112	31	Urban flooding	42
7	City	97	32	Green	35
8	Runoff	89	33	Mitigation	33
9	Performance	84	34	Challenges	32
10	Urbanization	79	35	Climate change adaptation	32
11	Impact	77	36	Quality	31
12	Framework	76	37	Water quality	30
13	Infrastructure	76	38	Strategy	28
14	Urban	74	39	Policy	27
15	Model	68	40	Governance	26
16	Nature-based solutions	66	41	Biodiversity	26
17	Design	64	42	Flood mitigation	26
18	Water	58	43	Land use	25
19	Flood risk	57	44	Landscape	24
20	Systems	57	45	Climate-change	24
21	Resilience	55	46	System	23
22	Sponge city	54	47	Optimization	22
23	Risk	51	48	Vulnerability	22
24	Water management	48	49	Rainfall	21
25	Adaptation	48	50	Health	19

, the frequency distribution analysis revealed significant heterogeneity in keyword occurrence, reflecting the diverse conceptual landscape of GI and urban flooding research. Foundational terms including “green infrastructure” (freq: 237, degree: 42), “stormwater management” (freq: 183, degree: 37), and “urban flooding” (freq: 162, degree: 35) demonstrated the highest frequency and degree values, confirming their definitional importance and extensive connectivity within the field. This pattern aligns with the theoretical centrality of these concepts in framing the discourse surrounding urban hydrological management and resilience planning. The co-occurrence patterns of these high-frequency keywords further illuminate the conceptual architecture of the field, with “green infrastructure” demonstrating particularly strong connections to both technical terms (e.g., “infiltration”, “bioretention”) and policy-oriented concepts (e.g., “urban planning”, “governance”), suggesting its function as a boundary object that facilitates dialog across disciplinary domains.

The analysis also identifies terms with moderate frequency but high centrality measures, suggesting their function as conceptual bridges between disparate research streams. Keywords with elevated betweenness centrality scores, such as “ecosystem services” (centrality: 0.24, degree: 29), “climate adaptation” (centrality: 0.21, degree: 27), and “multifunctionality” (centrality: 0.19, degree: 24), occupy strategic positions within the network topology, facilitating knowledge transfer across thematic boundaries. These bridging concepts represent critical intervention points for interdisciplinary integration, potentially catalyzing novel research directions at the interface between previously disconnected domains. As Guo et al. (2019) [33] observe, high-centrality keywords often represent emergent concepts that challenge disciplinary boundaries and stimulate theoretical innovation. The degree analysis further illuminates the connective capacity of certain keywords that, while not necessarily being the most frequent, demonstrate substantial linkages to diverse thematic areas. Terms such as “urban resilience” (degree: 31), “sustainable drainage systems” (degree: 28), and “low-impact development” (degree: 26) function as conceptual connectors that integrate multiple dimensions of GI research, reflecting their transdisciplinary relevance and theoretical versatility.

As shown in Figure 5, the co-occurrence patterns reveal distinct spatial clusters of interconnected terminology, with certain keyword dyads and triads demonstrating particularly strong associative linkages. For instance, the frequent co-occurrence of “green infrastructure” with “ecosystem services” (co-occurrence strength: 0.67) and “urban planning” (co-occurrence strength: 0.59) suggests a conceptual framework that positions GI within broader urban sustainability discourses rather than as merely a technical intervention for flood control. Similarly, the strong co-occurrence patterns between “sponge city” and “China” (co-occurrence strength: 0.83) reflect the geographical specificity of certain GI approaches, highlighting the contextual embeddedness of theoretical frameworks within particular socio-political environments. The visualization of these co-occurrence relationships through network mapping techniques reveals clustered terminological domains with varying levels of conceptual integration and fragmentation, providing insights into the intellectual scaffolding that structures knowledge production in this field.

The modularity analysis of keyword co-occurrence networks reveals a multifaceted research landscape within GI and urban flooding scholarship, with several dominant clusters emerging as being particularly salient in their conceptual orientation and methodological approaches. The most substantive clusters, characterized by both size and coherence (silhouette values), represent distinct yet interconnected intellectual territories.

Cluster 0 (size: 49, silhouette: 0.695) constitutes the largest agglomeration of concepts, centered on the integration of “green–gray infrastructure” with broader sustainability frameworks. This cluster’s conceptual architecture bridges technological and ecological paradigms, incorporating “sensitivity analysis” alongside “natural capital”, suggesting an epistemic framework that reconciles engineered and ecosystem-based approaches to urban water management.

Demonstrating strong internal coherence, Cluster 1 (size: 47, silhouette: 0.77) exhibits a more technically oriented configuration focused on “hydrodynamic modeling” and “stormwater treatment”. The co-occurrence of “urban heat island” within this predominantly hydrological cluster indicates an emergent recognition of the thermodynamic dimensions of urban water systems, reflecting a nascent integration of climate-sensitive approaches to stormwater modeling.

Cluster 2 (size: 41, silhouette: 0.919), with its remarkably high coherence value, represents a sophisticated convergence of risk paradigms and planning frameworks. The juxtaposition of “community risk planning” with “convolutional neural networks” within this cluster reveals an intellectual terrain where computational methodologies increasingly inform community-scale intervention strategies, signaling the evolution of risk assessments beyond purely technical parameters.

Cluster 3 (size: 32, silhouette: 0.954) manifests the integration of ecological principles within hydrological planning frameworks. The co-occurrence of “urban ecology” with “hydrodynamic modeling” suggests an epistemological shift toward biophysical integration in urban water systems analysis, emphasizing the ecological underpinnings of hydrological processes in built environments.

The conceptual architecture of Cluster 4 (size: 30, silhouette: 0.912) centers on ecosystem service provisioning within complex urban morphologies. The specific attention paid to “historic coastal cities” and “undesirable land” indicates a spatially nuanced conceptualization of ecosystem services that acknowledges the constraints and opportunities presented by pre-existing urban configurations, particularly in heritage-rich coastal contexts.

These dominant clusters collectively reveal an intellectual landscape characterized by increasing integration across disciplinary boundaries, with technical modeling approaches progressively incorporating ecological principles, risk frameworks expanding to encompass social dimensions, and ecosystem service paradigms acquiring greater spatial specificity in relation to urban morphological constraints.

As shown in Figure 6, the diversity of thematic clusters underscores the multidimensional nature of GI and urban flooding research, encompassing technical, ecological, planning, and socio-economic dimensions. The high silhouette values (ranging from 0.695 to 0.999) suggest well-defined research communities with distinctive terminological domains and methodological approaches, while the interconnections between clusters indicate the potential for cross-fertilization across disciplinary boundaries.

The keyword connection network analysis illuminates several significant trajectories for advancing GI and urban flooding research. First, the evolution from technically oriented approaches toward integrated socio-ecological frameworks suggests a paradigmatic shift in urban water management philosophies. This transition aligns with broader sustainability discourses that emphasize holistic problem-solving and systems thinking [36]. Second, the emergence of regionally specific concepts, such as “sponge city”, indicates the contextual adaptations of GI principles to distinct geographical and governance conditions [51]. Additionally, the network density patterns reveal varying levels of conceptual cohesion across different thematic areas, with hydrological modeling demonstrating greater terminological standardization compared to the more fragmented ecosystem services discourse. This heterogeneity in network consolidation suggests differential maturity levels across research streams, with implications for theoretical development and methodological standardization.

The degree distribution analysis further indicates a scale-free network structure characteristic of evolving scientific domains, wherein a small subset of keywords demonstrates disproportionately high connectivity while the majority exhibit limited connections. This structural property facilitates efficient information flow across the network while maintaining resilience to random perturbations, suggesting a robust yet adaptive knowledge architecture [49]. The co-occurrence patterns further reveal potential conceptual gaps where theoretically related terms demonstrate unexpectedly low association strengths, suggesting opportunities for enhanced integration between complementary research streams. For instance, the relatively weak co-occurrence between “climate justice” (Cluster 12) and “differential vulnerability” (Cluster 13) suggests an under-explored theoretical nexus that could yield valuable insights regarding the equity dimensions of GI implementation.

3.2.5. Keyword Burst Analysis

A citation burst analysis of the GI and urban flooding literature reveals the distinct evolutionary trajectories in scholarly attention patterns from 2015 to 2024, as shown in Figure 7. This bibliometric approach, identifying keywords experiencing unusual increases in usage frequency through the algorithmic detection of citation intensifications [28], yields 22 keywords with significant bursts that illuminate epistemological transformations within the field’s intellectual architecture.

The citation pattern initially manifests a predominance of technically oriented terminology, with “catchment” (strength = 2.68), “sustainable urban drainage systems” (strength = 2.68), and “life cycle assessment” (strength = 2.47) constituting the primary nodes in the scholarly discourse network. This technical emphasis corresponds with Fletcher et al.’s (2013) [65] observation regarding the prioritization of engineering functionality and hydrological performance metrics in early GI conceptualizations. The emergence of “ecosystem” (strength = 4.71) as a burst keyword during 2016–2019 signals a crucial epistemological pivot toward systems thinking, initiating a conceptual trajectory that transcends purely hydraulic frameworks.

A substantive reconfiguration of the field’s theoretical foundations becomes evident as social dimensions gain prominence in the citation landscape. Keywords such as “challenges” (strength = 4.33), “space” (strength = 4.18), and “environmental justice” (strength = 2.98) represent significant nodes in this conceptual reorientation, reflecting Venkataramanan et al.’s (2019) [4] recognition of the socio-ecological complexity inherent in urban flooding interventions. The emergence of “flood risk management” (strength = 4.98)—the strongest burst in the dataset—signifies a consolidated shift toward comprehensive management approaches rather than isolated technical solutions, exemplifying the integrative “Three Points Approach” articulated by Fratini et al. (2012) [39].

Concurrent citation bursts in “BMPs” (strength = 2.62) and “restoration” (strength = 2.62) demonstrate the field’s evolving attention to implementation strategies and ecological rehabilitation within urban contexts. This praxis-oriented scholarship [37] manifests alongside theoretical reconceptualizations, suggesting a dialectical relationship between conceptual frameworks and practical applications in the evolution of GI discourse.

The citation landscape subsequently exhibits convergence around climate resilience and optimization paradigms, with significant bursts in “health” (strength = 4.29), “climate” (strength = 3.59), “nature-based solutions” (strength = 2.62), and “optimization” (strength = 3.82). This thematic consolidation reflects Frantzeskaki’s (2019) [52] framework for nature-based urban interventions and indicates a maturation toward integrated assessment methodologies quantifying multiple performance dimensions [56]. The persistence of “flow” (strength = 2.58) alongside socio-ecological keywords evidences the continued attention paid to hydrological functionality within broader conceptual frameworks, demonstrating a synthetic rather than substitutive evolution in the field’s epistemological structure.

The terminological trajectory from “sustainable urban drainage systems” to “nature-based solutions” reveals a substantial conceptual reframing of GI interventions within larger sustainability discourses [66]. This evolution aligns with both technological advancement in modeling approaches [15] and the theoretical reconceptualization of GI as a multifunctional system providing ecosystem services beyond flood mitigation [7]. The citation burst patterns thus do not merely illuminate shifting research topics but also fundamental epistemological transitions within the scholarly community’s collective knowledge construction—moving from technically oriented flood management toward integrated socio-ecological perspectives emphasizing multiple benefits, climate resilience, and environmental justice considerations.

3.3. In-Depth Analysis

Full-Text Analysis of the Literature

The in-depth analysis of the literature on GI (GI) and urban flooding, conducted using Orange3 data mining tools, reveals multifaceted thematic structures and conceptual frameworks that have evolved significantly over time. The computational text analysis employing word frequency metrics and latent semantic indexing offers a unique crystallization of the intellectual terrain governing this domain, uncovering patterns that might otherwise remain obscured through conventional literature review methodologies. As illustrated in

Table 6. Top 50 keywords by their frequencies in the full-text analysis.

No.	Keywords	Frequency	No.	Keywords	Frequency
1	Urban	60,340	26	System	10,702
2	Water	43,974	27	Drainage	10,591
3	Green	37,259	28	Community	10,356
4	Flood	29,602	29	Different *	10,226
5	Infrastructure	20,850	30	Benefits	9683
6	Area	20,622	31	Results	9418
7	Management	19,032	32	Resilience	8548
8	Study	18,987	33	Impact	8239
9	City	18,222	34	Surface	8182
10	Climate	17,504	35	Reduction	8105
11	Stormwater	16,712	36	Flow	7821
12	Runoff	16,696	37	Spatial	7667
13	Land	15,759	38	Soil	7385
14	Change	15,307	39	Ecosystem	7307
15	Model	14,609	40	Approach	7195
16	Data	13,750	41	River	7152
17	Rainfall	13,252	42	Local	7109
18	Risk	13,202	43	Level	7083
19	Planning	12,518	44	Natural	7003
20	Development	11,969	45	Public	7000
21	Flooding	11,899	46	Solutions	6936
22	Analysis	11,495	47	Events	6923
23	Environmental	11,365	48	Cost	6886
24	Research	10,980	49	Services	6868
25	Design	10,875	50	Sustainable	6767

* e.g., “different scales”, “different approaches”, “different climate scenarios”.

, the extraordinary prevalence of core terminological markers—specifically “urban” (60,340 occurrences), “water” (43,974), “green” (37,259), “flood” (29,602), and “infrastructure” (20,850)—establishes a semantic foundation that underscores their centrality within the scholarly discourse on nature-based flood mitigation strategies. This lexical dominance does not merely signify terminological preference but rather conceptual centrality in the theoretical architecture of GI scholarship. The latent semantic indexing algorithm, applied through topic modeling procedures, delineates 17 distinct thematic topics exhibiting notable convergence around three primary conceptual domains: urban water management frameworks (manifested prominently in Topics 1–3), climate resilience mechanisms (particularly evident in Topics 6, 8, and 10), and multifunctional ecological benefits (represented comprehensively in Topics 5, 9, and 15).

The full-text analysis using latent semantic indexing reveals 17 distinct thematic topics that illuminate the intellectual structure of green infrastructure (GI) research for urban flooding. These topics converge around three primary conceptual domains: urban water management frameworks, climate adaptation strategies, and multifunctional solution approaches. Topics 1–3 establish the foundational terminology and hydrological concepts, with terms like “urban”, “water”, “green”, “flood”, and “infrastructure” forming the core vocabulary alongside the technical parameters of “runoff”, “rainfall”, and “flow”. Climate-oriented Topics 6, 8, and 10 connect adaptation strategies with spatial planning approaches, highlighting the integration of nature-based solutions within resilience frameworks. The multifunctionality of GI emerges prominently in Topics 5, 9, and 15, where ecological benefits, stormwater management techniques, and economic considerations intersect. Implementation and governance dimensions surface in Topics 4, 7, 11, and 13, reflecting the institutional frameworks necessary for successful GI deployment. Emerging research frontiers appear in Topics 12, 14, 16, and 17, focusing on community resilience, blue–green infrastructure integration, low-impact development, and ecosystem-based approaches, respectively. The semantic relationships between these topics demonstrate the field’s evolution from technically oriented engineering paradigms toward integrated socio-ecological frameworks that address performance, ecological enhancement, and social equity considerations simultaneously.

These computational linguistics findings provide empirical evidence of an epistemological transition in how GI interventions are conceptualized—evolving from discrete stormwater management techniques focused on hydraulic performance toward comprehensive socio-ecological systems embedded within broader urban sustainability paradigms. This conceptual transformation is further corroborated by the semantic proximity between management-oriented terminology (19,032 occurrences) and the planning-related lexicon (12,518 occurrences), demonstrating the progressive institutional assimilation of GI principles within formal planning frameworks and governance structures. Concomitantly, the pronounced co-occurrence patterns between the “benefits” (9683 occurrences) and “ecosystem” (7307 occurrences) terms substantiate the ascendancy of the multifunctionality paradigm initially articulated by Venkataramanan et al. (2019) [4] and subsequently expanded by Wang et al. (2023) [7] in their systematic bibliometric analysis of ecosystem services derived from green stormwater infrastructure implementations. Climate adaptation emerges as a paramount organizing principle throughout the analyzed corpus, with intricate semantic networks connecting “climate” (17,504 occurrences), “change” (15,307 occurrences), and “resilience” (8548 occurrences) as foundational conceptual associations. This semantic configuration aligns precisely with Dharmarathne et al.’s (2024) [1] empirical findings regarding adaptive strategies for climate-induced urban flooding phenomena, a relationship particularly pronounced in Topic cluster 6, which semantically bridges climate change adaptation frameworks with stormwater management systems and riverine processes. The frequency distribution analysis further illuminates considerable scholarly attention devoted to implementation challenges and operational considerations, evidenced by the high occurrence rates of “development” (11,969), “planning” (12,518), and “management” (19,032) within the corpus. The methodological landscape discerned through lexical analysis exhibits remarkable heterogeneity and sophistication, reflected in the elevated frequency metrics for “analysis” (11,495), “data” (13,750), and “model” (14,609), corresponding directly with the technical approaches encapsulated in Topic cluster 17, which encompass riverine systems, precipitation analysis, and ecosystem modeling frameworks. This methodological diversification is conceptually congruent with the integrated assessment approaches delineated by Reu Junqueira et al. (2021) [15] in their comprehensive systematic review of methodologies for modeling extreme rainfall impacts using GI interventions. The bibliometric patterns extracted through computational text mining further elucidate emergent research trajectories characterized by increasing integration between technical implementation considerations and governance frameworks. Han et al.’s (2023) [3] meticulous examination of China’s sponge cities initiative exemplifies this integrative scholarly approach, synthesizing technical performance metrics with governance innovation frameworks—a pattern consistent with the semantic associations identified in Topics 13 and 16, which establish robust linkages between technical implementation processes and community engagement mechanisms. The lexical co-occurrence patterns between blue–green infrastructure terminology (prominently featured in Topic cluster 5) demonstrate remarkable alignment with Ariyarathna et al.’s (2023) [5] integrated conceptual approach to urban flood mitigation through nature-based solutions. This analysis distinguishes itself from previous bibliometric studies through its identification of nuanced semantic proximities between the “uncertainty”, “adaptation”, and “scenarios” terminology, reflecting the growing scholarly emphasis on climate resilience frameworks articulated by Frantzeskaki (2019) [52] in their seminal work on nature-based solutions in urban contexts. In contrast to earlier reviews predominantly focused on technological dimensions, our computational text analysis reveals increasing scholarly attention directed toward social dimensions of implementation, evidenced by the rising prominence of the “community” (10,356), “public” (7000), and “social” (6538) terminology in the recent literature—a trend corresponding with Abuismail et al.’s (2024) [14] sophisticated exploration of resident attitudes regarding GI implementation for stormwater management. Despite the increasingly robust evidence base emerging from this comprehensive analysis, the bibliometric patterns discerned through computational methods suggest persistent epistemological gaps regarding longitudinal performance evaluation, governance optimization mechanisms, and implementation scale considerations. These knowledge deficits are reflected in the comparatively lower frequency metrics for terminology related to maintenance processes (3245), monitoring systems (2873), and assessment methodologies (6469). This lexical pattern indicates a scholarly domain progressively advancing toward more integrated, transdisciplinary theoretical frameworks, while simultaneously confronting significant epistemological challenges and practical implementation barriers in the application of GI interventions for urban flood management across diverse socio-ecological contexts.

4. Discussion

4.1. Comparative Analysis with the Existing Literature

This systematic review has revealed several convergent and divergent patterns when compared to previous literature reviews on GI (GI) and urban flooding. Our findings align with established scholarship regarding the fundamental role of GI as an effective approach to mitigate urban flooding while simultaneously providing multiple ecosystem services. Similarly to Venkataramanan et al. (2019) [4], our analysis confirms the multifunctionality of GI interventions, extending beyond mere flood control to encompass broader social and ecological benefits. This multifunctional characteristic has been consistently emphasized across reviews [6,7], reinforcing the consensus on GI’s capacity to deliver multiple ecosystem services simultaneously. Hansen and Pauleit’s (2014) [36] conceptual framework positioning GI as a mechanism for delivering multiple ecosystem services is further validated by our findings, which demonstrate how water regulation functions intersect with cultural, provisioning, and supporting services in urban environments.

However, our review diverges from previous studies in several significant aspects. While earlier reviews such as Herath et al. (2023) [12] and Ariyarathna et al. (2023) [5] focused predominantly on the technical performance metrics of specific GI implementations, our analysis reveals a notable shift in the research emphasis toward the integration of socio-ecological considerations. Unlike Reu Junqueira et al. (2021) [15], who concentrated primarily on modeling approaches for GI performance under extreme rainfall events, our review identifies an emerging research cluster centered on community perception and acceptance factors that influence GI implementation success—an aspect relatively underdeveloped in previous reviews. This shift toward socio-ecological integration demonstrates a maturation of the field beyond technocentric approaches, reflecting what Tzoulas et al. (2007) [62] identified as being the interconnected pathways between ecosystem services and human wellbeing. Furthermore, our analysis highlights how knowledge, attitudes, and behavioral factors—systematically documented by Venkataramanan et al. (2020) [13]—have transitioned from peripheral considerations to central components in GI implementation frameworks, representing a substantial epistemological reorientation in the field.

Moreover, our temporal analysis reveals an evolution in conceptual frameworks. The earlier literature, as synthesized by Ahiablame et al. (2012) [64] and Fletcher et al. (2013) [65], primarily situated GI within a technological solution paradigm. In contrast, our findings indicate a transition toward framing GI within resilience theory and social-ecological systems frameworks, resonating with Frantzeskaki’s (2019) [52] principles for nature-based solutions in urban contexts. This conceptual evolution represents a substantive departure from previous literature reviews. Specifically, our analysis demonstrates how the discourse has progressed from positioning GI as isolated technological interventions toward understanding these elements as complex adaptive systems embedded within broader urban ecosystems—a perspective that aligns with Raymond et al.’s (2017) [66] framework for co-benefit assessment. This theoretical reframing has profound implications for implementation strategies, as it necessitates more integrative planning approaches that account for complex feedback mechanisms across social, ecological, and technological domains.

The sponge city concept, extensively documented in Chinese contexts [3,50], emerges in our analysis as an increasingly globalized approach. Unlike previous reviews that treated this as a regionally specific phenomenon, our findings suggest the cross-pollination of this concept into diverse geographical contexts, indicating conceptual convergence in global GI discourse. This pattern was not identified in earlier literature reviews, such as that of Sheng et al. (2022) [16], who maintained clearer geographical distinctions in GI approaches. Our bibliometric analysis reveals citation patterns demonstrating how principles from the sponge city framework are being adapted and recontextualized across disparate geographical and climatic regions, suggesting a form of conceptual hybridization that transcends the typical geographically bounded implementation strategies previously documented. Mei et al. (2018) [60] provided a methodological template for this cross-cultural adaptation through their integrated assessment framework, which our analysis shows is increasingly being cited and modified for application in non-Chinese urban contexts, representing a form of knowledge transfer previously unidentified in the literature.

Additionally, our review has identified a significant research gap concerning the long-term performance and maintenance requirements of GI—a critical dimension inadequately addressed in previous reviews including that of Venkataramanan et al. (2020) [13] and Huang (2023) [9]. Where these reviews emphasized implementation challenges, our findings highlight the relative paucity of longitudinal studies examining GI performance degradation over time, suggesting an important direction for future research. This temporal dimension of GI performance introduces a layer of complexity absent from most evaluation frameworks, which tend to assess efficacy at a single point in time rather than as a dynamic property that evolves throughout infrastructure lifecycles. Brown and Farrelly (2009) [63] identified institutional and governance barriers to sustainable urban water management, but our analysis extends this by revealing how these barriers manifest specifically in maintenance regimes and adaptive management processes. This finding has substantial implications for lifecycle cost analyses and challenges to conventional cost–benefit frameworks that may underestimate the long-term resource requirements for GI maintenance.

Furthermore, our analysis reveals an emerging research front focused on the integration of gray and GI within hybrid systems—termed blue–green–gray infrastructure [56,61]—representing a more nuanced perspective than the binary opposition sometimes presented in earlier reviews, such as that of Dong et al. (2017) [54]. This integration approach acknowledges the complementary roles of traditional engineering solutions and nature-based approaches, reflecting a pragmatic shift in urban water management paradigms not fully captured in previous systematic reviews. Our citation network analysis demonstrates how this hybrid paradigm is gaining momentum, with Schubert et al.’s (2017) [55] framework for case-specific assessment emerging as a bridging concept between previously siloed research communities focused exclusively on either green or gray infrastructure. The multi-criteria approach developed by Alves et al. (2018) [61] for the selection of optimal combinations of green and gray infrastructure represents a methodological innovation that enables decision-makers to navigate the complex trade-offs between immediate flood risk reductions and broader co-benefits such as carbon sequestration. Additionally, Lawson et al.’s (2014) [48] interdisciplinary approach to evaluating multiple flood risk benefits in blue–green cities emerges in our analysis as being increasingly influential in reshaping evaluation frameworks to accommodate this hybrid infrastructure paradigm, suggesting a conceptual consolidation that transcends the theoretical divisions evident in the earlier literature.

The integration of stormwater management across scales—from site-specific interventions to watershed-level planning—also emerges as a distinctive feature of contemporary GI research which was not adequately captured in previous reviews. While earlier systematic reviews documented the efficacy of individual GI elements, our analysis reveals an emerging research cluster focused on scalar integration, exemplified by Lee et al.’s (2012) [40] watershed-scale optimization model. This scalar dimension introduces a level of complexity that challenges conventional planning frameworks and necessitates more sophisticated modeling approaches, such as those developed by Elliott and Trowsdale (2007) [47] for low-impact urban stormwater drainage. Our analysis indicates that these cross-scalar approaches represent a significant methodological advancement beyond the site-specific assessments that dominated the earlier literature. It is important to acknowledge that our bibliometric analysis primarily captures the intellectual discourse rather than the implementation outcomes of GI interventions in real-world settings. The conceptual evolution evident in the literature—from technical hydrological approaches toward integrated socio-ecological frameworks—may not necessarily mirror actual implementation patterns in diverse urban contexts. While cities like Portland (USA), Copenhagen (Denmark), and Singapore have demonstrated the successful translation of multifunctional GI concepts into physical infrastructure systems with documented flood reduction benefits, many urban areas still approach GI implementation primarily through a technical stormwater management lens rather than through the integrated framework increasingly dominant in scholarly discourse. This implementation gap suggests that while the intellectual foundations for holistic GI approaches have been established, their practical realization faces significant institutional, financial, and technical barriers that vary considerably across geographical and governance contexts. Future research would benefit from more systematic documentation of implementation outcomes, particularly regarding performance metrics under actual flood conditions, maintenance requirements, and governance arrangements that facilitate effective GI deployment at meaningful scales.

4.2. Analytical Innovations: Bibliometric Analysis and Computational Approaches

Our analytical framework demonstrates epistemological advancement through the deliberate integration of three complementary computational platforms, creating a multidimensional analytical architecture that transcends the methodological constraints inherent in conventional literature assessment approaches. This integration represents not merely a procedural combination but rather a conceptual synthesis wherein Biblioshiny, CiteSpace, and Orange3 function as an interconnected methodological ecosystem, each addressing complementary dimensions of the knowledge domain while collectively generating insights that exceed the sum of their individual analytical capacities.

This methodological triangulation illuminates the intellectual structures and evolutionary trajectories that would remain obscured through singular analytical approaches. The systematic integration of bibliometric mapping (providing structural topography), evolutionary clustering (revealing temporal dynamics), and full-text semantic analysis (exposing conceptual substrata) enables the visualization of complex knowledge relationships across multiple dimensions simultaneously. Particularly significant is how this integration reveals the progressive dissolution of disciplinary boundaries within the literature corpus, with previously discrete research communities—hydrological modeling specialists, ecosystem service theorists, and urban planning scholars—demonstrating increasing convergence around integrative frameworks that synthesize the technical, ecological, and social dimensions of urban water management.

The implementation of dual-map overlay techniques reveals sophisticated bidirectional knowledge flows between traditionally siloed domains, demonstrating how GI research has increasingly drawn theoretical constructs from ecological paradigms while simultaneously reshaping hydrological engineering discourse through the integration of resilience thinking. This cross-disciplinary knowledge exchange manifests in the emergence of hybrid conceptual frameworks that transcend traditional disciplinary boundaries, suggesting an epistemic maturation characterized by transdisciplinary integration rather than merely the multidisciplinary accumulation of discrete knowledge structures.

The temporal dimension of our analysis reveals nuanced transitions in how urban flooding challenges are conceptualized—not as abrupt paradigm shifts, but as gradual epistemic evolutions characterized by increasing theoretical sophistication. The literature corpus exhibits a progressive trajectory from technically oriented frameworks prioritizing hydraulic efficiency toward more integrative conceptualizations incorporating social–ecological resilience principles and governance considerations. This evolution is evidenced in the semantic transformation revealed through full-text analysis, with terminology gradually shifting from predominantly hydraulic engineering-related lexicon toward governance-oriented discourse incorporating the concepts of adaptive capacity, transformative resilience, and community engagement mechanisms.

Despite these methodological advantages, our approach encounters several substantive limitations that warrant acknowledgment. First, the inherent algorithmic biases embedded within computational bibliometric tools may privilege certain types of knowledge structures while obscuring others, particularly those emerging from non-Western epistemological traditions or those published in non-English languages. The predominance of English-language publications in our corpus potentially marginalizes important contributions from contexts experiencing acute urban flooding challenges, such as Southeast Asia and Latin America, where significant scholarship may be published in regional languages. This linguistic homogeneity risks reproducing epistemological hegemony rather than facilitating truly global knowledge synthesis.

Additionally, the reliance on digitally archived research inevitably introduces temporal biases that privilege contemporary scholarship. The differential digitization of the historical literature creates artificial discontinuities in knowledge evolution trajectories, potentially understating the contributions of foundational works predating comprehensive digital archiving. This “recency bias” may inadvertently construct illusory novelty by obscuring conceptual lineages extending into pre-digital scholarship.

Furthermore, while our methodological triangulation attempts to mitigate individual analytical limitations, it cannot entirely transcend the fundamental constraints of computational text analysis, particularly its difficulty in capturing nuanced theoretical constructs, rhetorical subtleties, and implicit knowledge structures that resist algorithmic reduction. The quantification of scholarly discourse inevitably privileges certain forms of knowledge—particularly those amenable to computational parsing—potentially marginalizing the interpretive, critical, and phenomenological approaches central to understanding the socio-political dimensions of urban environmental management.

Our integrated methodology establishes a sophisticated template for interdisciplinary knowledge domain analysis that corresponds to the inherently complex, multifaceted nature of social–ecological systems research. By synthesizing structural, temporal, and semantic dimensions into a coherent analytical framework, this approach does not merely illuminate what constitutes current knowledge, but how that knowledge has evolved and continues to evolve through dynamic interactions across disciplinary boundaries, methodological approaches, and conceptual frameworks—revealing the intellectual architecture underlying green infrastructure discourse in urban flooding contexts. However, these insights must be contextualized within the acknowledged methodological limitations, recognizing that computational approaches complement rather than supplant the deep contextual understanding that emerges from traditional scholarly engagement with the literature.

5. Conclusions

The systematic review of green infrastructure (GI) for urban flooding reveals the distinct intellectual trajectories that characterize the field’s evolution. Through comprehensive bibliometric analysis, this study elucidates a progressive transformation in how GI interventions are conceptualized, implemented, and evaluated within complex urban systems. This research domain has undergone a fundamental epistemological reorientation from technical efficiency paradigms toward integrated socio-ecological frameworks that acknowledge the multidimensional nature of urban water management challenges.

Analysis of citation patterns and thematic clustering demonstrates a clear intellectual progression from hydraulic engineering approaches toward multifunctional ecosystem service frameworks. The early literature primarily emphasized hydrological performance parameters and technical design specifications, conceptualizing GI as discrete interventions within existing urban water systems. This technical orientation gradually expanded to incorporate ecological dimensions, as evidenced by the emergence of ecosystem-focused thematic clusters and the increasing centrality of terms such as “ecosystem services” within keyword networks. Contemporary research increasingly situates GI within resilience theory and social–ecological systems frameworks, recognizing these interventions as complex adaptive systems embedded within broader urban metabolism rather than isolated technological solutions.

The field’s conceptual architecture has evolved toward increasing interdisciplinary integration, dissolving the traditional boundaries between engineering, ecological, and social science perspectives. Hybrid infrastructure paradigms that synthesize green and gray approaches have emerged as a significant research trajectory, acknowledging the complementary roles of nature-based and conventional engineering solutions. This integration reflects a pragmatic shift beyond binary oppositions, developing sophisticated decision support frameworks for optimizing infrastructure configurations across multiple performance criteria simultaneously.

Cross-scalar integration represents another significant intellectual advancement, with increasing attention paid to the spatial relationships between site-specific interventions and watershed-level outcomes. This scalar dimension introduces complexity that transcends conventional planning approaches, necessitating more sophisticated modeling frameworks that account for non-linear relationships between implementation density and system-level performance. Contemporary research demonstrates increasing methodological sophistication in addressing these scalar complexities, developing integrated assessment frameworks that bridge micro-scale design parameters with macro-scale hydrological processes.

Climate adaptation has emerged as a paramount organizing principle throughout the literature corpus, with semantic networks demonstrating robust connections between climate change frameworks and stormwater management practices. This orientation reflects the growing recognition of the dynamic nature of hydrological regimes under climate uncertainty, necessitating adaptive management approaches rather than static design paradigms. The research community has shifted accordingly toward anticipatory rather than reactive frameworks, developing methodologies for assessing GI performance under various climate scenarios and uncertainty conditions.

Geographic analysis reveals the global diffusion of regionally specific concepts, with frameworks such as the “sponge city” approach transcending their original cultural contexts to inform implementation strategies across diverse urban environments. This cross-cultural knowledge transfer suggests conceptual convergence in global GI discourse while maintaining contextual sensitivity to regional socio-ecological conditions.

Despite these significant intellectual advancements, several critical knowledge gaps persist that constrain the effective implementation of GI for urban flood management. The longitudinal evaluation of GI performance remains underdeveloped, with relatively few studies examining degradation trajectories over infrastructure lifecycles. This temporal dimension introduces complexity that is absent from most evaluation frameworks, with substantial implications for lifecycle cost analyses and maintenance planning. The governance dimensions of GI implementation, particularly regarding institutional arrangements for long-term maintenance, remain inadequately addressed within the current research. Additionally, while the multifunctional benefits of GI are widely acknowledged, methodological frameworks for systematically quantifying and optimizing these co-benefits across diverse urban contexts require further development.

The methodological limitations of this review warrant acknowledgment. The reliance on computational bibliometric tools may privilege certain knowledge structures while obscuring others, particularly those emerging from non-Western epistemological traditions or studies published in non-English languages. The exclusive focus on Web of Science potentially underrepresents scholarship from regions experiencing acute urban flooding challenges but with limited representation in mainstream academic databases.

Future research directions should address these identified gaps through several strategies. First, longitudinal studies examining GI performance over extended timeframes are essential for understanding degradation patterns and developing effective maintenance protocols. Second, governance frameworks specifically addressing institutional arrangements for GI implementation across diverse urban contexts require systematic investigation. Third, methodological frameworks for optimizing co-benefits beyond flood mitigation—including ecosystem services, public health improvements, and climate resilience—warrant further development. Finally, cross-cultural comparative analyses examining how GI implementation strategies are adapted to diverse socio-ecological contexts could yield valuable insights for the contextual adaptation of generalized principles.

This systematic review demonstrates how GI research has evolved from technically oriented approaches focused on narrow hydrological parameters toward integrated frameworks incorporating ecological, social, and governance dimensions, reflecting broader paradigmatic shifts in environmental management thinking. This intellectual trajectory positions GI as a critical component of sustainable urban water management in an era of increasing climatic uncertainty and unprecedented urbanization pressures, while simultaneously highlighting the need for continued methodological innovation and knowledge development to address the persistent implementation challenges across diverse urban contexts.