How do spatial factors of green spaces contribute to flood regulation in urban areas? A systematic mapping approach

Abstract

Flooding is increasing in urban areas around the world, leading to loss of life and property damage, and cities are using urban green spaces (UGS) for flood regulation. The spatial attributes of UGS have an important role in controlling and regulating urban flooding, and there is a need for a systematic map on how spatial factors of UGS, such as shape, size, location, or connectivity, impact flooding in urban areas. The objectives of this study are to analyze and synthesize published material to evaluate the impacts of the spatial dimensions of UGS on flood regulation and to identify knowledge gaps and future research directions. Pertinent literature was reviewed and synthesized using the systematic mapping method. The results of this study show that previous research on spatial configuration have examined how variables such as slope, DEM, green space coverage, and landscape shape index impact runoff reduction. Slope was found to be an important, but not determining factor in flood regulation. There is a need for further research on how the geographic context of urban regions, including climatic conditions and land use changes, impacts UGS functionality. Additionally, there is a need for further research on how the spatial configuration of UGS impacts flood vulnerability and intensity, two under-addressed yet important topics in urban flooding.

Keywords

Urban green spaces flood regulation flood hydrology spatial metrics configuration

I Introduction

Urban floods are becoming more severe, lasting longer, and occurring more frequently because of changes in climate, rapid urbanization, and population growth globally (Tellman et al., 2021; Wing et al., 2022). By 2050, the population density in flood-prone coastal and riverine areas and megacities is predicted to increase by 25% with more than a billion people exposed to such flooding (IPCC, 2021). It is estimated that the extent of metropolitan areas vulnerable to flood threats will increase by 270% between 2000 and 2030, even without accounting for climate change (Güneralp et al., 2015). As several studies show, it is estimated that the flood damages are further exacerbated by climate change and land use/land cover (LULC) changes that replace vegetated areas with impervious surfaces, such as roads, roofs, and parking lots (O’Donnell and Thorne, 2020; Pattison and Lane, 2011; Venter et al., 2020; Zuniga-Teran et al., 2020). In most regions of the world, the degree of landscape change that is associated with urbanization is significant (Haas et al., 2015; Oke et al., 2017). Although traditional structural flood mitigation infrastructure (e.g., drainage systems) are effective in urban areas, their functionalities in the face of extreme rainfall events and increased development largely depend on the capacity and location of such systems, making complementary solutions such as vegetated areas important. To tackle such urban flooding challenges, implementing different types of urban green spaces (UGS) including urban parks, undeveloped open spaces, lawns, urban forests, green stormwater infrastructure (GSI), low impact developments (LID), water best management practices (BMPs), and water sensitive urban design strategies (WSUD) can increase natural capacity to adapt to climate change impacts (Araaf Tauhid and Zawani, 2018; Prudencio and Null, 2018; Ureta et al., 2021). As the findings of multiple studies show, many different types of UGS can maintain urban ecosystems and recreational services while decreasing the impacts of extreme precipitation and stormwater runoff (Kazmierczak and Cavan, 2011).

Urban green spaces, which refers to vegetated areas with trees, shrubs, or grasses in urban settings, contribute to a wide variety of water-related ecosystem services such as flood regulation, stormwater management, and water purification (Cilliers et al., 2013; Graça et al., 2018; Song et al., 2020). UGS can be used to control and regulate urban floods especially surface runoff reduction (Liu et al., 2014; Zölch et al., 2017). For example, urban wetlands and green roofs can regulate urban runoff by capturing stormwater (McPhearson et al., 2022). The impact of UGS on urban flooding control and mitigation in urban areas has been widely investigated in previous studies from different perspectives including technical (Yang et al., 2021), planning (Afriyanie et al., 2020; Kim et al., 2021a), spatial (Baker et al., 2019), ecological (Zölch et al., 2017), social (Descher and Sinasac, 2020), economic (Sohn et al., 2021) and flood management (Ellis et al., 2021).

Additionally, the consequences of extreme weather such as storm surges, high tidal levels, and heavy rainfall in the river and coastal cities are greatly associated with an increase in impervious surface cover due to land-use change, making coastal, pluvial, and fluvial flooding more severe (Acreman and Holden, 2013; Kalantari et al., 2014). The failure of the urban drainage system also creates a unique form of flooding. Most rainfalls in cities will have to be drained using existing storm-water drainage systems because there is less land available for water attenuation and infiltration. However, when the amount of precipitation, and hence the volume of water, exceeds the area’s drainage capacity, intense precipitation events can create floods. Rainwater begins to collect in places other than the drainage system, causing water to stagnate in low-lying areas (Jegatheesan et al., 2019). UGS are becoming important components of urban flood regulation systems, which simultaneously provide multiple benefits for human health, well-being, and biodiversity in cities as well. The vegetated and previous areas in urban areas have substantive roles in runoff reduction through the infiltration process (Berland et al., 2017). Several studies have been conducted on the runoff reduction and stormwater control functionality of different types of UGS in various contexts (Li et al., 2020b; Liu et al., 2014). These studies suggest that UGS contributes to mitigating urban flood risk and stormwater management through 1) rainfall interception, 2) increasing the lag time between rainfall and runoff or water storage capacity, 3) evaporation from stored water, and 4) infiltration by soil (Kim and Park, 2016; Yang and Lee, 2021a; Sohn et al., 2021).

Increasing permeability in cities is one of the main reasons UGS can be effective in runoff regulation (Rezaei et al., 2019). The deterioration of the water cycle system is linked to an increase in impermeable surfaces in urban areas (Yang and Lee, 2021b). Imperviousness in urban catchments causes decreased infiltration capacity, increased surface runoff, increased soil erosion, and reduced base flow discharge, which increases the risk of flash floods and reduces the capacity for groundwater recharge. These repercussions are most likely where impervious surfaces are directly connected to urban streams, and they are linked to a variety of other factors such as land conversion pattern and past quality of converted land (Schuch et al., 2017). By contrast, permeable surfaces such as UGS contain some kind of spaces filled with sand, gravel, or soil, which allows runoff to permeate into the underlying soil, reducing runoff and replenishing groundwater. Since it is difficult to include a considerable amount of green space in densely populated urban areas, the spatial arrangement and structural location of green spaces are critical to make optimal use of existing land from a landscape ecology perspective (Gill et al., 2007).

Urban green spaces infiltration capacity strongly depends on their biophysical and structural characteristics. Biophysical characteristics include natural attributes of plant species, health, and soil physical properties, and structural characteristics include shape, distribution, location, and connectivity (Gill et al., 2007; Kim et al., 2017; Yang et al., 2020). UGS structure has two major components: composition (the quantity and variety of UGS kinds, regardless of their spatial structure or organization) and configuration (the spatial aspect of UGS types such as arrangement, location, size, and connectivity) (Farina, 2000). From a landscape ecology perspective, the configuration, or spatial characteristics, of UGS, has an important impact on hydrological processes and runoff regulation (Boongaling et al., 2018; Zhang et al., 2019). Yet, the majority of previous studies have largely focused on the impact of different biophysical and compositional aspects of UGS, such as testing different combinations of types or analyzing the physical characteristics of vegetation species and pavements. Previous studies have also addressed how climate conditions and precipitation events affect UGS performance against flash or pluvial urban flooding (e.g., see Quagliolo et al., 2021; Alexander et al., 2019).

Although multiple disciplines have explored spatial factors relating to UGS and flood mitigation, these perspectives are disconnected and inconsistent, limiting a comprehensive understanding of the role of spatial factors. For example, some studies indicate that smaller and dispersed patches may have more runoff reduction (Liu et al., 2022; Yang and Lee, 2021a), while others indicate the positive impact of larger connected UGS on runoff reduction in an urban watershed (Li et al., 2020). To gain a comprehensive understanding of the expansive research field of flood regulation, we seek to highlight the role of spatial factors of UGS in this study. By using a systematic mapping approach, pertinent articles were identified, broad research trends were defined, key topics in flood regulation were summarized, and strong evidence for spatial factors was identified. This study also discusses knowledge gaps and suggests areas for further research.

II Methods

This study developed a systematic map of the state of knowledge on the relationships between spatial factors of UGS and flood regulation. Inspired by Chausson et al. (2020) research on nature-based solutions, this study followed the methodology proposed by James et al. (2016). Like systematic reviews, systematic mapping is a reliable, reproducible, and open scientific method for determining the literature that is currently available on a certain subject (Zhong et al., 2018). Systematic mapping is used to represent the scope of study in an area and pinpoint knowledge gaps and identify where future review work may be possible (Clapton et al., 2009). Systematic maps are not typically used to undertake a complete data extraction or critical evaluation of the pertinent research, in contrast to a systematic review. Instead, maps show the type, scope, and characteristics of research in a given field of study (Grant and Booth, 2009). This study conducted a systematic mapping of the spatial factors of UGS to identify the most common factors and their associated characteristics in different contexts.

The study’s overarching research question and detailed sub-questions follow:

1) What is the state of knowledge on the impact of UGS’s spatial factors on flood regulation?

a) Which flood regulation topics (e.g., hydrological, or financial) does the UGS literature address?

b) Which spatial factors of UGS have been examined in the literature?

c) What methodologies and trends have been used to investigate the impact of UGS on flood regulation?

d) What are the knowledge gaps and the top priorities for future research?

1 Search strategy

To find pertinent key terms, the search’s scope was first decided. We determined that the information found using key search phrases included information on both flood regulation and spatial characteristics of UGS (location, distribution, configuration, and topography). We used two terms “flood” and “runoff” as indicators of flood regulation. In this review, the UGS refers to urban vegetated lands that fall under the category of nature-based solutions (NBS), green infrastructure (GI), low-impact development (LID), sponge cities, and sustainable urban drainage systems (SUDS). All different combinations of the above-mentioned keywords were searched using the Web of Science database as the search engine in the summer of 2022, with the search field set to 'keywords' and ‘title’ and the document type set to ‘article’. No filter for the year of publication was used, hence the sample pulled from the database includes all publication years. Since some articles might have looked at a specific type of UGS (e.g., green roof) without directly referring to UGS at all, the second round of queries included “landscape” and two commonly used UGS types were added to the list: park and green roof. Separate search queries were used, each with a different two-way combination of (keyword/title category-related) search phrases as outlined in Table 1.

Table 1.

Summary of results from scoping the study for search terms.

Search queries
Round1	Flood regulation terms UGS terms	(Runoff OR flood) AND (Green spac* OR Green infrastructure OR Nature-based solutions OR low impact development OR sustainable urban drainage system OR sponge cit*)
Round 2	Flood regulation terms UGS terms	(Runoff OR flood) AND (park OR green roof OR landscape)^a

^aSearch queries were restricted by only physical geography, forestry, and environmental studies fields in search engines.

2 Study inclusion and exclusion criteria

Three steps in succession made up the inclusion procedure. The inclusion criteria were used to evaluate titles, abstracts, and full texts. Since the focus of the study is to detect spatial factors of UGS affecting flood regulation, the methodology section of each article was reviewed first to include or exclude the publication for the final mapping process. The following elements served as inclusion criteria:

• Scale: Studies examine UGS in urban areas

• Aspect: Studies address spatial structure, including distribution, configuration, topography, and location

• Type of UGS: All urban green elements including any specific type of vegetated lands

• UGS functionality: Any type of study that examined the economic and environmental capacity of UGS to reduce flood impacts.

Although there is a relationship between UGS structure and spatial factors, we excluded all studies that focused only on the composition of UGS, including but not limited to UGS types, plant species, and the number of patches or morphological factors (e.g., height).

Articles were also excluded if they did not focus on the spatial factors of UGS in the context of urban flooding. Excluded articles included: (1) spatio-temporal investigations of the effects of flooding on landscape and UGS change; (2) studies that emphasized runoff pollutants, quality, and nutrient cycling; (3) studies that investigated the large river basins and watershed scales without including any urban areas; (4) studies with a focus on technical/engineering, social (e.g., public attitudes), structural (e.g., size and shape), biophysical aspects (e.g., plant species, media types), climatic factors (e.g., rainfall intensity) and health indicators (e.g., water quality).

3 Coding strategy

During the coding stage, information pertinent to the research questions and objectives was extracted from full texts. Since the relationship between UGS’s different spatial arrangements depends on each city’s physical and environmental characteristics, we coded the specific metrics and features of each article (Table 2). Each sampled study was categorized based on the associated codes: Bibliographic, geographic, study-specific, and topics. One row in an excel spreadsheet was set aside for each study. The same Excel spreadsheet was divided into columns for each variable, and counts were then performed to map out the general trends in the literature. Additionally, regarding the quality of articles, since we only evaluated articles from published peer-reviewed journals indexed in the Web of Science, we did not explicitly assess the quality of these articles. Additionally, critical evaluation is optional in systematic maps (James et al., 2016). We assumed that some level of quality control had already happened during the peer review process.

Table 2.

Coding strategy used for collecting data from each study (Adapted from Badullovich et al., 2020).

Code category	Data collected
Bibliographic	Title, author(s), year, journal, study methodology, departmental affiliations
Geographic	Location of sample (country)
Study-specific	Sample UGS type, spatial metrics
Topics	Topics used in flood hydrology

III Results

1 Search results and screening

Figure 1 shows the number of articles collected, screened, eliminated, and finally selected as final sampled documents for analysis during the steps of the systematic mapping. 2135 articles and abstracts were initially identified, and 1641 remained after duplicates were eliminated. 1517 articles were disqualified in the first round of screening (at the level of the title and abstract). Two runs of full-text screening were conducted; the first resulted in the exclusion of 42, and the second pass resulted in the exclusion of an additional 30. The systematic map was completed by including and coding 52 academic articles based on the key spatial factors of UGS that affect flood regulation.

2 Country of origin

A total of 52 articles contained research from a wide variety of regions. Case studies were found in 11 countries including China (18), the United States (9), South Korea (5), UK (4), Italy (3), Iran (2), Indonesia (2), Sweden (1), Netherlands (1), Germany (1), and Belgium (1). Also, five articles were conducted without considering any specific study area. Overall, more than half of the case studies were in China and the United States (27). Although these 11 countries were in different parts of the world, they were not uniformly distributed around the world. For example, the experiences of other low-to-high income nations, such as India, Vietnam, Pakistan, Bangladesh, Nepal, and Australia, which are regularly affected by flooding, were not included in these studies.

3 Dates and source titles

All articles were published in a wide variety of domains and across 31 academic journals from January 2012 to July 2022.

The earliest article was published in 2012 in the journal of Environmental Management. As shown in Figure 2, 2020 and 2021 had the highest number of articles published (12 per year). Our review of publications in 2022 includes only 6 months (January-July) and thus does not reflect the total number of publications for the year. Overall, Figure 2 indicates a rising trend in the number of published articles on this topic.

According to the analysis, 52 articles were published in 30 different peer-reviewed journals. Five journals published three or more articles during this period. These were Urban Forestry and Urban Greening, with the most articles (n = 5), Sustainability (n = 4), Landscape and Urban Planning (n = 3), Urban Water Journal (n = 3), and Water (n = 2).

4 Types of UGS

Nineteen classes of UGS were identified in the mapped literature. The different types of UGS included in sampled articles and their frequencies among studies are illustrated in Figure 3, and all UGS types included in 19 classes are displayed in Appendix 2. Urban forest was the most common UGS type that was found to be effective in flood regulation in the literature (Berland et al., 2017), followed by urban agricultural land (e.g., urban food-producing gardens and farms). Some UGS such as rain gardens and green roofs were categorized both as low impact development (LID) and green stormwater infrastructure (GSI) in some articles (e.g., see Fahy and Chang, 2019; Fiori and Volpi, 2020). According to the Environmental Protection Agency’s (EPA) definition, the term low impact development (LID) refers to “systems and practices that use or mimic natural processes that result in the infiltration, evapotranspiration or use of stormwater to protect water quality and associated aquatic habitat” and green stormwater infrastructure (GSI) defined as “an approach to managing stormwater runoff in ways that mimics the natural environment as much as possible, using plants, soil, and stone to filter and manage stormwater more effectively, reducing how much enters our sewer systems, and protecting our rivers and streams” (EPA, 2022). Following these definitions, green roofs that are beneficial for flood stormwater reduction in cities (Twohig et al., 2022) were the most studied GSI/LID type, followed by rain gardens.

5 Types of modeling methods

Based on the research methodologies, four groups of models and tools were utilized in the selected literature: process-based models (51), empirical models (30), statistical models (23), and programming tools (3). Out of 52 articles, 14 articles used only one methodological approach, while the remaining 38 articles included multiple methods.

The most common study methods were process-based or numerical models, which all articles used except one (Kim et al., 2017). ArcGIS (27), followed by EPA SWMM (10) provided by US Environmental Protection Agency’s hydrological model, were the most commonly used models in the literature. Other models included QGIS (2), multi-criteria decision analysis (1), software Infoworks ICM (1), J2000 hydrological model (1), i-Tree Hydro (1), surface-subsurface hydrological model (ParFlow) (1), Integrated urban flood modeling system (IFMS Urban) (1), ENVI (1), MIKE SHE model (1), Landscape Green Infrastructure Design (L-GrID) (1), USGS Thornthwaite Water Balance (TWB) model (1), ERDAS Imagine (1), and community-scale simulation model (1).

Empirical models were considered the second most common type of modeling approach and included regression analysis (13), FRAGSTATS (12), Soil Conservation Service Curve Number (SCS-CN) (9), correlation analysis (4), Intensity-Duration-Frequency (IDF) curve (3), hydrological equation (3), the use value economic model (value of runoff regulation) (1), radial basis function neural network (1), and Minimum Cumulative Resistance (MCR) Model (1). Statistical studies included: ANOVA analysis (1), chi-square test (1), and panel data model (3). Lastly, the programming approach involved two: Python (1) and MATLAB (2).

6 Topics

Six topics were identified during the full review. All topics reported as dependent variables in these studies were coded based on how the spatial components of the green spaces impact flooding. Figure 4 indicates the count of studies for each of the six topics. The most common topic that occurred was flood hydrology with 41 articles (79% of all topics), followed by flood costs, flood vulnerability, flood risk management, flood intensity, and flood estimation (n = 2, 4% of all topics). Most articles emphasized the hydrological responses of UGS in comparison with other dimensions, such as vulnerability. Flood hydrology studies examined discharge, total surface runoff, streamflow, and water-logging reduction in urban areas. Studies that categorized flood costs included three articles with flood damage assessment and valuing the runoff reduction resulting from UGS (Brody et al., 2017; Sohn et al., 2021). Estimating flood probability was the main topic of flood estimation studies (Kim et al., 2016; Miller and Brewer, 2018). Moreover, the authors of only two articles on flood vulnerability used the Intergovernmental Panel on Climate Change (ICPP)’s framework to analyze vulnerability based on three factors: sensitivity, capacity, and exposure (Kim et al., 2021b; Sakieh, 2017).

7 Spatial metrics of UGS per study topic

Thirty-four spatial metrics of UGS with flood-regulated services as independent variables were identified. The variables were categorized based on five major components: distribution, scale, configuration, typography, and area. Spatial configuration had the most spatial dimensions (n = 83), followed by topography (n = 46), distribution (n = 18), scale(n = 5), and area (n = 25). The most common spatial factors are known as landscape metrics, which were frequently calculated through the FRAGSTATS software, a spatial pattern analysis program for quantifying landscape structure (McGarigal and Marks 1995). Landscape pattern indices, also known as landscape metrics, quantify changes in landscape spatial patterns caused by both anthropogenic and natural disturbances such as urban flooding (Gillanders et al., 2008).

Distribution in these studies refers to at least three concepts: (1) the distribution of UGS and its influence on flood regulation, and/or (2) the spatial arrangements of UGS in terms of dispersed or centralized UGS and its effect on flood reduction, and/or (3) how the geographic location of UGS across an urban area impacts flood regulation. In this review, scale is a component of distribution unless the study (1) addressed the impact of UGS at different scales (e.g., plot, sub-catchments, and catchment) and/or (2) compared UGS in various sized cities, and/or (3) examined different scale/size of UGS (e.g., small scale GSI such as rain gardens or large scale GSI such as bioretention basin). Topography refers to the UGS’s surface elevation and slope, which were mostly measured through the digital elevation model (DEM) and digital terrain model (DTM). The frequency and meaning of 34 spatial metrics are illustrated in Table 3.

Table 3.

Spatial metrics examining the functionality of UGS for flood regulation.

Dimension	Spatial metrics	Meaning	No.
Distribution		Refers to both geographical location and dispersion of UGS in an urban area
	Spatial arrangement	Dispersed or centralized UGS and its effect	1
	Equitable distribution	Spatially homogenous placement of UGS	2
	Location	UGS’s position in different parts of the city	15
Configuration		Refers to the spatial aspect of UGS types such as arrangement, location, size, and connectivity
	Correlation length (GYRATE_AM)	Mean distance between each cell in a patch and the patch’s centroid cells	1
	Shannon’s evenness index (SHEI)	Describes the even degree of distribution of different kinds of landscape	1
	Area weighted mean shape index (AWMSI)	The average shape index (SHAPE) of patches of the corresponding patch type, weighted by patch area so that larger patches weigh more than smaller patches	2
	Effective mesh size (MESH)	Quantifies the degree of landscape fragmentation	2
	Landscape division index (DIVISION)	The landscape’s division degree by patches	2
	Total edge (TE)	An absolute measure of the total edge length of a particular patch type	2
	Clumpiness index (CLUMPY)	The proportional deviation of the proportion of like adjacencies involving the corresponding class from that expected under a spatially random distribution	3
	Edge density (ED)	The ratio of the total length of all patches’ edges to the total landscape area	3
	Fractal dimension index (FDI)	Shape complexity across a range of spatial scales	3
	Patch density (PD)	The ratio of the number of patches to the total landscape area	3
	Aggregation index (AI)	The total edge shared by each class	4
	Contagion (CON)	Describes the contagion degree of the same kind of patch	4
	Connectivity (CONNECT)	Measures each pair of patches that are either connected or not within a certain search radius	5
	Contiguity (CONTIG)	Assesses the spatial connectedness, or contiguity, of cells within a grid-cell patch to provide an index of patch boundary configuration and thus patch shape	6
	Largest patch index (LPI)	Quantifies the percentage of total landscape area comprised by the largest patch	6
	Euclidian nearest neighbor (ENN)	The straight-line distance between nearest-neighbor patches of the same type	7
	Proximity (PROX)	Each patch size is divided by the square of its distance from the focal patch	7
	Cohesion (COHESION)	Measures the physical connectedness of the corresponding patch type	8
	Landscape shape index (LSI)	Provides a simple measure of class aggregation or clumpiness and, as such, is very similar to the aggregation index	14
Topography		Refers to the arrangement of the natural and artificial physical features of an urban area
	Digital surface model (DSM)	A three-dimensional representation of the heights of the Earth’s surface, including natural or man-made objects located on it	2
	Digital terrain model (DTM)	A three-dimensional, bare-earth representation of a terrain or surface topography, consisting of an array of points with a defined height	2
	Digital elevation model (DEM)^a	A gridded raster data and a three-dimensional representation of a terrain, which filters out and excludes terrain vector features	18
	Slope^b	A rising or falling surface	18
	Concavity (depression)	A lower elevation of UGS than its surroundings and can temporarily retain storm waters	3
	Topographic wetness index (TWI)	Quantify topographic controls on hydrological processes	3
Area		Refers to the size and ration of UGS in an urban area
	Perimeter-area ratio (PARA)	A simple measure of shape complexity, but without standardization to a simple euclidean shape	1
	Class area (CA)	Percentage of UGS comprised of a patch type	1
	UGS ratio (coverage)	The ratio of the area occluded by vegetation to the entire space	16
	Percentage of landscape (PLAND)	Percentage of a particular patch type	5
	Mean patch size (area)	Average patch size	2
Scale		Refers to different levels of UGS which have been focused in studies
Scale	Scale	Refers to UGS in different scales of the city (e.g., plot, sub-catchments, and catchment) OR studied UGS in various sized cities OR studies different scale/size of UGS (e.g., small scale GSI such as rain gardens or large scale GSI such as bioretention basin)	5

^aSlope in the studied documents refers to the slope of UGS such as the slope of green roofs of grasslands.

^bDEM was used for measuring the larger UGS such as bioretention basin or wetlands.

As shown in Table 3, the most common spatial metrics in the literature were topographic metrics: digital elevation model (DEM) (n = 18) and slope (n = 18). The second most examined metric was UGS ratio/coverage (n = 16), followed by geographical location (n = 15), and landscape shape index (LSI) (n = 14). The five most frequent metrics used to assess flood regulation were UGS area, topography, and shape. The distribution of spatial metrics for each flood regulation topic showed that scale metrics were not reported in flood intensity, vulnerability, and risk management studies (Figure 5). This distribution also showed that the greatest number of configuration metrics were applied in flood hydrology studies (n = 19). No distribution metrics of UGS were examined for flood intensity and vulnerability analysis. Moreover, the number of studies that addressed scale was similar for three topics: flood hydrology, costs, and estimation.

8 The effectiveness of spatial metrics in flood studies

To synthesize the influence of reported spatial metrics reported in Table 3, we selected the most common metrics for each spatial dimension including slope, DEM, LSI, UGS coverage, and location. The COHESION index was also included due to its importance in analyzing the spatial configuration of UGS. The “scale” metric was not considered because it was infrequently used. The results of the studies indicated that the performance of UGS varied by geographic context due to different climatic and environmental conditions. Thus, UGS impacts on flood regulation might vary by city and region, and the results of one study may not be generalizable to other geographic areas.

Topography is a well-established spatial variable and is typically measured by using slope for small-scale UGS such as rain gardens and Digital Elevation Model (DEM), digital surface model (DSM), and digital terrain model (DTM) for larger UGS such as wetlands or grasslands. Slope was one of the most basic topographical factors for hydrologic analysis of different types of UGS, especially in steep zones. Studies showed that a higher estimated slope of the UGS surface could intercept more runoff (Lim and Welty, 2017; Zhou et al., 2019); however, since its effectiveness is related to other factors such as soil type, it is not a determining factor (Kim and Park, 2016). Also, as one-third of all articles reported (n = 18), DEM is a useful dataset to determine surface-water catchment boundaries and flow pathways/accumulation of UGS to examine its impact on flood regulation, especially on hydrological topics (Miller and Brewer, 2018).

The shape of UGS is also an important variable. Calculations of shape-based spatial configuration metrics based on the Landscape Shape Index (LSI) was one of the most frequently used spatial attributes. According to the findings of these studies, larger UGS with variable and patch shapes were found to help reduce stormwater runoff (Bai et al., 2018; Brody et al., 2017; Kim and Park, 2016; Zhang et al., 2015). Yet, the results of studies in Inner Mongolia (China) and the city of Ghent (Belgium) suggested that the shape of UGS alone may not have a significant impact on flood intensity or runoff reduction (Li et al., 2020b; Peng et al., 2019). These studies found that the degree of influence depends on other factors, such as distribution and distance.

Another essential spatial factor affecting the overall functionality of UGS in flooded areas is the area and percentage of vegetated lands. The studies showed that increasing UGS coverage could significantly decrease the total runoff or economic cost of flooding in, for example, Chinese cities (Kim, 2021; Li et al., 2016; Yao et al., 2020), although the degree of this reduction depends on some factors such as the precipitation duration/frequency and seasonal characteristics in Swedish cases (Sjöman and Gill, 2014). Almost one-third of articles (16) reported that UGS coverage was an influential factor to reduce the amount of runoff, flood damage, flood probability, and risk management in different case studies. However, as the results of a study in Shenzhen, China indicated, different urban areas might respond differently to scenarios of changing UGS ratio depending on topography and land use. Also, in some cases, increasing UGS coverage may not help certain water-related issues, such as waterlogging (Qian et al., 2021).

Additionally, the results of studies using COHESION as a strong measurement to examine the physical connectedness of the corresponding green patches were inconsistent across different flood topics, especially flood costs and hydrology. The findings of some articles in both virtual and real cases (e.g., Indonesia) showed that the more dispersed the green patches, the more reduction in surface runoff (Maheng et al., 2021; Yang and Lee, 2021b). Some studies found that layout features were more crucial for flood mitigation than connectivity between green spaces in China (Bai et al., 2018), meaning that the dispersion of UGS throughout the landscape was a better strategy than increasing clustering and connectivity. In addition, some studies related this type of effect to the scale of UGS, which means that the connectivity of UGS on a small-scale increases runoff and had the opposite effect over a large scale in South Korea (Kim and Park, 2016). One study examined the impact of dispersed or disconnected UGS on flood damage loss reduction in the Gulf of Mexico, US (Brody et al., 2017) and found that large green patches lost their capacity to mitigate flood damage when they were fragmented. Another study in the city of Ghent, Belgium recommended increasing the connectivity of UGS to reduce urban runoff (Li et al., 2020b). Given this inconsistency between case studies, it is still unclear whether the connectivity of UGS influences the total urban runoff.

Finally, the spatial location of UGS and its impact on flood regulation was one of the important factors that the authors reported in less than one-third of the articles (15) which is consistent with the outcome of a comprehensive review by Zhang and Chui (2018), who analyzed the spatial allocation strategies of GI. Furthermore, since diverse parts of the cities usually experience unequal impacts of climate change and urbanization due to unbalanced distribution of resources and UGS, well-sited UGS in different parts of the city can mitigate such negative impacts (Hsu et al., 2021). For example, in a study by Kim (2021), the locations of UGS and their relative positions to the current drainage system may have different flood mitigation effects across South Korea. As the findings of these studies demonstrated, the closer to urban sewerage and drainage networks, the more positive hydrological impacts in terms of runoff and streamflow reduction (Ercolani et al., 2018; Yao et al., 2020). As a result, the optimization of UGS to better conduct urban runoff to deal with the negative impacts of urban flooding has been growing, especially among green stormwater infrastructure (GSI) facilities because GSI’s effectiveness varies depending on the geographic region (Qian et al., 2021). The location of UGS concerning the downstream or upstream site is also a significant factor. The findings of a virtual case study indicated that some types of UGS (e.g., rain gardens and green roofs) may potentially make flooding more likely when retrofitting is done in downstream areas, nearer the outlet (Fiori and Volpi, 2020).

IV Discussion

This study provides the first systematic map on the most common spatial metrics of UGS that mitigate urban flooding in countries around the world. These findings identify important research areas for urban geographers, policymakers, planners, and civil engineers. This systematic mapping of 52 scientific articles is relevant to researchers and decision-makers considering common factors and metrics that affect the functionality of spatial factors of UGS for flood regulation. A summary of the key conclusions of the evidence maps as well as its limitations, including significant evidence gaps and associated directions for more synthesis and research follows.

1 Summary of key findings

Our mapping exercise revealed that a total of 52 articles reported the influence of different spatial metrics on UGS performance for flood regulation. Studies often used process-based modeling to examine the impact of different configurations and spatial characteristics of UGS on flood regulation. Most studies were conducted in countries in the industrialized Global North and Asia rather than low-income and developing countries. Flood hydrology was the most common topic addressed, and the spatial configuration of UGS was consistently identified as an important factor in flood regulation. Although the topography of UGS was not found to be a determining factor to control urban flooding, it was an important factor in stormwater management and runoff in densely built and relatively low-sloped cities where the capacity to attenuate and evacuate rainwater is limited. Runoff gathered in drainage swales or piped underground eventually finds its way into slowly draining creeks and bayous.

Our review indicated that the effectiveness of UGS varies spatially based on land use/land cover, climatic, and other contextual factors. The results of most studies indicated that the location, distribution, and arrangement of UGS depending on sewer systems or upstream/downstream positions of the site may have different impacts on runoff mitigation and flood loss. These results suggest that UGS planning must site GSI and other types of UGS according to their locational functionality against flooding.tpb

2 Evidence gaps and priorities for future research

The mapping exercise revealed several research gaps. First, studying the spatial dimension of UGS is limited to developed countries, and there is a critical need to analyze spatial factors in other countries, especially the Global South, as well as Australia, where regular flooding occurs in urban areas. Empirical research and theoretical modeling in these contexts can clarify the associations between spatial factors of UGS and flood regulation in different land uses, urban contexts, and climate conditions.

Second, some studies have investigated the impact of UGS coverage and shape on flood regulation, but there is no consensus among studies about the optimized efficiency of these criteria. Thus, we strongly recommend quantifying the potential effects of ratio, area, shape, and geometry of UGS across different scales and different flood characteristics and processes. Although some studies suggest that larger UGS patches have more potential to reduce runoff in urban areas and may reduce flood damage (Brody et al., 2017; Kim and Park, 2016), it is not yet clear exactly how much green space is needed to regulate flooding for individual cities, nor its actual impact on flood vulnerability, intensity, estimation, and risk management. Also, our systematic mapping showed that fewer studies (n = 5) examined the impact of the UGS scale on flood regulation. There is a need to investigate the tradeoffs and synergies of scale matches and mismatches.

Third, our findings indicate that there is still limited research on the impacts of siting UGS at different spatial scales in cities, including within or around cities. Further, there has been little consideration of how UGS siting can improve flood regulation in disadvantaged/marginalized communities (Fahy and Chang, 2019; Qian et al., 2021; Zellner et al., 2016). Future research should investigate the associations between various UGS locations in urban areas and their potential flood regulation consequences. Although research on the benefits of equitable distribution of UGS has increased in recent years in different contexts (see Cartier, 2021; Hunter et al., 2019; Pallathadka et al., 2022), more research to further investigate the impact of equitable distribution of UGS on reducing flood risks is needed.

Fourth, most articles focused on process-based modeling approaches. Other numerical modelings, empirical modeling, machine learning, and advanced remote sensing methods can be applied to help researchers analyze and predict flood regulation scenarios based on differential spatial factors of green spaces. Comparing the results of these models results may help identify the most optimized solutions for different urban contexts.

Fifth, there are opportunities for more spatial configuration metrics to be calculated and examined both individually and cumulatively regarding their potential to address flood regulation in various contexts. These new spatial configuration metrics can help researchers find correlations between a wide variety of metrics and better realize their functionalities in various contexts. There is a need to quantify the possible associations between these variables and all flood regulations topics to explore optimized UGS planning in cities. In addition, due to inconsistent results among studies, it is crucial to further assess the impacts of the spatial configuration of UGS, such as connectivity, on flood regulation across different climatic and soil conditions. Finally, since most of the studies have focused on the runoff reduction impact of UGS, there is a need to investigate the effect of spatial factors of UGS on other topics of flood regulation, especially vulnerability and intensity, through more empirical studies.

V Conclusion

This study analyzed the relationship between spatial factors of UGS and flood regulation through a systematic mapping of 52 peer-reviewed articles. The results of systematic mapping indicated that the general topic is gaining popularity among multiple environmental disciplines due to the growing climate and urbanization threats in urban areas.

Results from this review show that five spatial dimensions of UGS, including distribution, scale, area, configuration, and topography with different degrees can be manipulated to be effective for flood regulation services in urban settings. Previous research has identified several variables and criteria for each category that require further analysis for UGS planning and decision-making in various contexts, climate conditions, and scales. Understanding the interactions among different combinations of these variables is important to evaluate the efficacy of UGS for flood regulation in different scenarios and individual cities. The results of this review can serve geographers, environmental engineers, landscape planners, and environmental scientists who examine the impact of spatial factors of UGS on different aspects of flood regulation in different contexts and support practitioners and policymakers who develop and implement urban resiliency planning in face of increasing climate change and urban population growth. This study has some limitations. Given the variety of UGS definitions, it is possible some relevant studies were not included. Additionally, only articles published in English language journals were considered. Future studies should address these limitations.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Sina Razzaghi Asl

References

Acreman

Holden

(2013) How wetlands affect floods. Wetlands 33: 773–786.

Afriyaniea

Julianb

Riqqi

, et al. (2020) Re-framing urban green spaces planning for flood protection through socio-ecological resilience in Bandung City, Indonesia. Cities 101: 102710.

Alexander

Hettiarachchi

, et al. (2019) Can integrated green spaces and storage facilities absorb the increased risk of flooding due to climate change in developed urban environments? Journal of Hydrology 579(5): 124201.

Araaf Tauhid

Zawani

(2018) Mitigating climate change related floods in urban poor areas: green infrastructure approach. Journal of Regional and City Planning 29(2): 98–112.

Badullovich

Grant

Colvin

(2020) Framing climate change for effective communication: a systematic map. Environmental Research Letters 15: 123002.

Bai

Mayer

Shuster

, et al. (2018) The hydrologic role of urban green space in mitigating flooding (Luohe, China). Sustainability 10: 1–3584.

Baker

Brenneman

Chang

, et al. (2019) Spatial analysis of landscape and sociodemographic factors associated with green stormwater infrastructure distribution in Baltimore, Maryland and Portland, Oregon. The Science of the Total Environment 66410: 461–473.

Berland

Shiflett

Shuster

, et al. (2017) The role of trees in urban stormwater management. Landscape and Urban Planning 162: 167–177.

Boongaling

CGK

Faustino-Eslava

Lansigan

(2018) Modeling land use change impacts on hydrology and the use of landscape metrics as tools for watershed management: the case of an Ungauged catchment in the Philippines. Land Use Policy 72: 116–128.

10.

Brody

Highfield

Blessing

, et al. (2017) Evaluating the effects of open space configurations in reducing flood damage along the Gulf of Mexico coast. Landscape and Urban Planning 167: 225–231.

11.

Cartier

KMS

(2021) Growing equity in city green space, Eos, p. 102. DOI: 10.1029/2021EO158443 (accessed 20 June 2022).

, et al. (2020) Mapping the effectiveness of nature-based solutions for climate change adaptation. Global Change Biology 26: 6134–6155.

13.

Cilliers

Lubbe

, et al. (2013) Ecosystem services of urban green spaces in African countries—perspectives and challenges. Urban Ecosystems 16(4): 681–702.

14.

Clapton

Rutter

Social Care Institute for Excellence et al. (2009) SCIE Systematic Mapping Guidance April 2009. London: Social Care Institute for Excellence.

15.

Drescher

Sinasac

(2020) Social-psychological determinants of the implementation of green infrastructure for residential stormwater management. Environmental Management 67: 308–322.

(2021) Mainstreaming natural flood management: a proposed research framework derived from a critical evaluation of current knowledge. Progress in Physical Geography: Earth and Environment 45(6): 819–841.

17.

Environmental Protection Agency-EPA (2022) Urban runoff: low impact development. Available at: https://www.epa.gov/nps/urban-runoff-low-impact-development (accessed 01 October 2022).

18.

Ercolani

Chiaradia

Gandolfi

, et al. (2018) Evaluating performances of green roofs for stormwater runoff mitigation in a high flood risk urban catchment. Journal of Hydrology 566: 830–845.

19.

Fahy

Chang

(2019) Effects of stormwater green infrastructure on watershed outflow: does spatial distribution matter? International Journal of Geospatial and Environmental Research 6(1): 5–16.

20.

Farina

(2000) Landscape Ecology in Action. Netherlands: Kluwer Academic Publishers.

21.

Fiori

Volpi

(2020) On the effectiveness of LID infrastructures for the attenuation of urban flooding at the catchment scale. Water Resources Research 56: 1–17.

22.

Gill

Handley

Ennos

, et al. (2007) Adapting cities for climate change: the role of the green infrastructure. Built Environment 33(1): 115–133.

23.

Gillanders

Coops

Wulder

, et al. (2008) Multitemporal remote sensing of landscape dynamics and pattern change: describing natural and anthropogenic trends. Progress in Physical Geography: Earth and Environment 32(5): 503–528.

24.

Graça

Alves

Gonçalves

, et al. (2018) Assessing how green space types affect ecosystem services delivery in Porto, Portugal. Landscape and Urban Planning 170: 195–208.

25.

Grant

Booth

(2009) A typology of reviews: an analysis of 14 review types and associated methodologies. Health Information and Libraries Journal 26: 91–108.

26.

Güneralp

Liu

(2015) Changing global patterns of urban exposure to flood and drought hazards. Global Environmental Change 31: 217–225.

27.

Haas

Furberg

Ban

(2015) Satellite monitoring of urbanization and environmental impacts—A comparison of Stockholm and Shanghai. International Journal of Applied Earth Observation and Geoinformation 38: 138–149.

28.

Hsu

Sheriff

Chakraborty

, et al. (2021) Disproportionate exposure to urban heat island intensity across major US cities. Nature Communications 12: 2721.

29.

Hunter

Cleary

Braubach

(2019) Environmental, health and equity effects of urban green space interventions. In: Marselle

Stadler

Korn

, et al. (eds) Biodiversity and Health in the Face of Climate Change. Cham: Springer.

30.

IPCC (2021) WGII Sixth Assessment Report. Chapter 6: Cities, Settlements and Key Infrastructure. Cambridge: Cambridge University Press.

31.

James

Randall

Haddaway

(2016) A methodology for systematic mapping in environmental sciences. Environmental Evidence 5: 7.

32.

Jegatheesan

Goonetilleke

van Leeuwen

, et al. (2019) Urban Stormwater and Flood Management: Enhancing the Livability of Cities. Berlin: Springer.

33.

Kalantari

Lyon

Folkeson

, et al. (2014) Quantifying the hydrological impact of simulated changes in land use on peak discharge in a small catchment. The Science of the Total Environment 466–467: 741–754.

34.

Kazmierczak

Cavan

(2011) Surface water flooding risk to urban communities: analysis of vulnerability, hazard, and exposure. Landscape and Urban Planning 103(2): 185–197.

35.

Kim

Lee

Kuhn

, et al. (2021a) Satellite imaging reveals the increased proportion of the population exposed to floods. Nature 596: 80–86.

36.

Kim

Song

Chon

(2021b) Key coastal landscape patterns for reducing flood vulnerability. The Science of the Total Environment 759: 143454.

37.

Kim

Lee

Sung

(2016) Effect of urban green spaces and flooded area type on flooding probability. Sustainability 8(2): 134.

38.

Kim

, et al. (2017) Exploring the impact of green space health on runoff reduction using NDVI. Urban Forestry and Urban Greening 28: 81–87.

39.

Kim

Park

(2016) Urban green infrastructure and local flooding: the impact of landscape patterns on peak runoff in four Texas MSAs. Applied Geography 77: 72–81.

40.

Kim

(2021) Analyzing green space as a flooding mitigation – storm Chaba case in South Korea. Geomatics Natural Hazards and Risk 12(1): 1181–1194.

41.

(2016) SWMM-based evaluation of the effect of rain gardens on urbanized areas. Environmental Earth Sciences 75: 17.

42.

Deng

, et al (2020a) Hydrologic-environmental effects of sponge city under different spatial scales. Journal of Water Reuse and Desalination 10(1): 45–56.

43.

Van Eetvelde

Cheng

, et al. (2020b) Assessing stormwater runoff reduction capacity of existing green infrastructure in the city of Ghent. International Journal of Sustainable Development and World Ecology 27(8): 749–761.

44.

Lim

Welty

(2017) Effects of spatial configuration of imperviousness and green infrastructure networks on hydrologic response in a residential sewershed. Water Resources Research 53: 8084–8104.

45.

Liu

Chen

Peng

(2014) Assessing the effectiveness of green infrastructures on urban flooding reduction: a community-scale study. Ecological Modelling 291: 6–14.

46.

Liu

Han

Jiao

, et al. (2022) Responses of flood peaks to land use and landscape patterns under extreme rainstorms in small catchments - A case study of the rainstorm of Typhoon Lekima in Shandong, China. International Soil and Water Conservation Research 10: 228–239.

47.

Maheng

Pathirana

Zevenbergen

(2021) A preliminary study on the impact of landscape pattern changes due to urbanization: case study of Jakarta, Indonesia. Land 10: 218.

48.

McPhearson

Cook

Berbes-Blazquez

, et al. (2022) A social-ecological-technological systems framework for urban ecosystem services. One Earth 5: 505–518.

49.

McGarigal

Marks

(1995) FRAGSTATS: spatial pattern analysis program for quantifying landscape structure. Available at: https://www.umass.edu/landeco/pubs/mcgarigal.marks.1995.pdf(accessed 20 January 2022).

50.

Miller

Brewer

(2018) Refining flood estimation in urbanized catchments using landscape metrics. Landscape and Urban Planning 175: 34–49.

51.

O'Donnell

Thorne

(2020) Drivers of future urban flood risk. Philosophical Transactions of the Royal Society 378: 2168.

52.

Oke

Mills

Christen

, et al (2017) Urban Climates. London, UK: Cambridge University Press.

53.

Pallathadka

Chang

Grimm

(2022) Urban flood risk and green infrastructure: who is exposed to risk and who benefits from investment? A case study of three U.S. Cities. Landscape and Urban Planning 223: 104417.

54.

Pattison

Lane

(2011) The link between land-use management and fluvial flood risk: a chaotic conception? Progress in Physical Geography: Earth and Environment 36(1): 72–92.

55.

Peng

Wang

, et al. (2019) Does landscape pattern influence the intensity of drought and flood? Ecological Indicators 103: 173–181.

56.

Prudencio

Null

(2018) Stormwater management and ecosystem services: a review. Environmental Research Letters 13: 033002.

57.

Qian

Wang

(2021) Protecting existing urban green space versus cultivating more green infrastructures: strategies choices to alleviate urban waterlogging risks in Shenzhen. Remote Sensing 13: 4433.

58.

Quagliolo

Comino

Pezzoli

(2021) Experimental flash floods assessment through urban flood risk mitigation (UFRM) model: the case study of ligurian coastal cities. Frontiers in Water. DOI: 10.3389/frwa.2021.663378

, et al. (2019) Investigating the effective factors influencing surface runoff generation in urban catchments – A review. Desalination and Water Treatment 164: 276–292.

60.

Sakieh

(2017) Understanding the effect of spatial patterns on the vulnerability of urban areas to flooding. International Journal of Disaster Risk Reduction 25: 125–136.