Nature-based solutions for hydro-meteorological risk reduction: a state-of-the-art review of the research area

Hydro-meteorological risks due to natural hazards such as severe floods, storm surges, landslides and droughts are causing impacts on different sectors of society. Such risks are expected to become worse given projected changes in climate, degradation of ecosystems, population growth and urbanisation. In this respect, nature-based solutions (NBSs) have emerged as effective means to respond to such challenges. A NBS is a term used for innovative solutions that are based on natural processes and ecosystems to solve different types of societal and environmental challenges. The present paper provides a critical review of the literature concerning NBSs for hydro-meteorological risk reduction and identifies current knowledge gaps and future research prospects. There has been a considerable growth of scientific publications on this topic, with a more significant rise taking place from 2007 onwards. Hence, the review process presented in this paper starts by sourcing 1608 articles from Scopus and 1431 articles from the Web of Science. The full analysis was performed on 146 articles. The analysis confirmed that numerous advancements in the area of NBSs have been achieved to date. These solutions have already proven to be valuable in providing sustainable, cost-effective, multi-purpose and flexible means for hydro-meteorological risk reduction. However, there are still many areas where further research and demonstration are needed in order to promote their upscaling and replication and to make them become mainstream solutions.

1 Introduction

There is increasing evidence that climate change and associated hydro-meteorological risk are already causing wide-ranging impacts on the global economy, human well-being and the environment. Floods, storm surges, landslides, avalanches, hail, windstorms, droughts, heat waves and forest fires are a few examples of hydro-meteorological hazards that pose a significant risk. Hydro-meteorological risk is the probability of damage due to hydro-meteorological hazards and its interplay with exposure and vulnerability of the affected humans and environments (Merz et al., 2010). Some of the main reasons for such risks are climate change, land use change, water use change and other pressures linked to population growth (Thorslund et al., 2017). The situation is likely to become worse given the projected changes in climate (see, for example, EEA, 2017). Therefore, effective climate change adaptation (CCA) and disaster risk reduction (DRR) strategies are needed to mitigate the risks of extreme events and to increase resilience to disasters, particularly among vulnerable populations (Maragno et al., 2018; McVittie et al., 2018).

Since biodiversity and ecosystem services can play an important role in responding to climate-related challenges, both mitigation and adaptation strategies should take into consideration a variety of green infrastructure (GI) and ecosystem-based adaptation (EbA) measures as effective means to respond to present and future disaster risk (see also EEA, 2015). Such approaches are already well accepted within multilateral frameworks such as the United Nations (UN) Framework Convention on Climate Change (UNFCCC), the Convention on Biological Diversity (CBD) and the Sendai Framework for Disaster Risk Reduction (SFDRR). As such, they are recognised as effective means for CCA and DRR and for the implementation of the Sustainable Development Goals (SDGs).

In view of the points above, many countries are nowadays developing adaptation and mitigation strategies based on GI and EbA to reduce their vulnerability to hydro-meteorological hazards (Rangarajan et al., 2015; EEA, 2015). Nature-based solutions (NBSs) have been introduced relatively recently. The reason for their introduction is that NBSs offer the possibility to work closely with nature in adapting to future changes, reducing the impact of climate change and improving human well-being (Cohen-Shacham et al., 2016). NBSs have been the focus of research in several EU Horizon 2020-funded projects. Horizon 2020 offers new opportunities in the focus area of “Smart and Sustainable Cities with Nature based solutions” (Faivre et al., 2017). Some of these important projects are Nature4Cites, NATURVATION, NAIAD, BiodivERsA, INSPIRATION, URBAN GreenUP, UNaLaB, URBINAT, CLEVER Cities, proGIreg, EdiCitNet, Regenerating ECOsystems with Nature-based solutions for hydro-meteorological risk rEduCTion (RECONECT), OPERANDUM, ThinkNature, EKLIPSE and PHUSICOS (nature4cities, 2019). Through these projects, knowledge of NBSs has grown rapidly and been documented in a considerable body of grey literature (project reports, for example). On the other hand, the number of scientific studies focused on NBSs to reduce hydro-meteorological risk is continuously increasing all over the world.

The aim of this article is to provide a state-of-the-art review of scientific publications on hydro-meteorological risk reduction with NBSs to indicate some directions for future research based on the current knowledge gaps. The analysis focuses on the following hydro-meteorological hazards: floods, droughts, storm surges and landslides. The review addresses both small- and large-scale interventions and explores available techniques, methods and tools for NBS assessment while also providing a snapshot of the major socio-economic factors at play in the implementation process. The key objectives and methods of this study are discussed in Sect. 3, while Sect. 2 provides a brief overview of concepts and definitions related to NBSs either in general or that are specifically linked to hydro-meteorological risk reduction. Results and conclusions are discussed in Sects. 4 and 5 respectively.

2 Overview of definitions and theoretical backgrounds

There are several terms and concepts which have been used interchangeably in the literature to date. In terms of the NBSs, the two most prominent definitions are from the International Union for Conservation of Nature (IUCN) and the European Commission. The European Commission defines nature-based solutions as “Solutions that aim to help societies address a variety of environmental, social and economic challenges in sustainable ways. They are actions inspired by, supported by or copied from nature, both using and enhancing existing solutions to challenges as well as exploring more novel solutions. Nature-based solutions use the features and complex system processes of nature, such as its ability to store carbon and regulate water flows, in order to achieve desired outcomes, such as reduced disaster risk and an environment that improves human well-being and socially inclusive green growth” (EC, 2015). The IUCN has proposed a definition of NBSs as “actions to protect, sustainably manage and restore natural and modified ecosystems that address societal challenges effectively and adaptively, simultaneously providing human well-being and biodiversity benefits” (Cohen-Shacham et al., 2016). Eggermont et al. (2015) proposed a typology characterising NBSs into three types: (i) NBSs that address a better use of natural or protected ecosystems (no or minimal intervention), which fits with how the IUCN frames NBSs, (ii) NBSs for sustainability and multi-functionality of managed ecosystems, and (iii) NBSs for the design and the management of new ecosystems, which is more representative of the definition given by the European Commission.

A NBS is a collective term for innovative solutions to solve different types of societal and environmental challenges, based on natural processes and ecosystems. Therefore, it is considered to be an “umbrella concept” covering a range of different ecosystem-related approaches and linked concepts (Cohen-Shacham et al., 2016; Nesshöver et al., 2017) that provides an integrated way to look at different issues simultaneously. Due to the diverse policy origins, NBS terminology has evolved in the literature to emphasise different aspects of natural processes or functions. In this regard, nine different terms are commonly used in the scientific literature in the context of hydro-meteorological risk reduction: low-impact developments (LIDs), best management practices (BMPs), water-sensitive urban design (WSUD), sustainable urban drainage systems (SuDs), green infrastructure (GI), blue–green infrastructure (BGI), ecosystem-based adaptation (EbA) and ecosystem-based disaster risk reduction (Eco-DRR). The timeline of each term based on their appearance in literature is shown in Fig. 1, and their definitions are given in Table 1.

Figure 1 Timeline and year of origin of each term (low-impact developments – LIDs, best management practices – BMPs, water-sensitive urban design – WSUD, green infrastructure – GI, sustainable urban drainage systems – SuDs, nature-based solutions – NBSs, ecosystem-based adaptation – EbA, ecosystem-based disaster risk reduction – Eco-DRR – and blue–green infrastructure – BGI) based on their appearance in publications.

Table 1 Glossary of terms and their geographical usage.

The commonalities between a NBS and its sister concepts (i.e. GI, BGI, EbA and Eco-DRR) are that they take participatory, holistic, integrated approaches, using nature to enhance adaptive capacity, reduce hydro-meteorological risk, increase resilience, improve water quality, increase the opportunities for recreation, improve human well-being and health, enhance vegetation growth, and connect habitat and biodiversity. More information on the history, scope, application and underlying principle of terms of SuDs, LIDs, BMPs, WSUD and GI can be found in Fletcher et al. (2015), while the relationship between a NBS, GI–BGI and EbA is described in detail by Nesshöver et al. (2017).

Although all terms are based on a common idea, which is embedded in the umbrella concept of NBSs, differences in definition reflect their historical perspectives and knowledge base that were relevant at the time of the research (Fletcher et al., 2015). The distinguishing characteristic between a NBS and its sister concepts is how they address social, economic and environmental challenges (Faivre et al., 2018). Some terms such as SuDs, LIDs and WSUD refer to NBSs that specifically address stormwater management. They use the landscape feature to transform the linear approach of conventional stormwater management into a more cyclic approach where drainage, water supply and ecosystems are treated as part of the same system, mimicking more natural water flows (Liu and Jensen, 2018). GI and BGI focus more on technology-based infrastructures by applying natural alternatives (Nesshöver et al., 2017) for solving a specific activity (i.e. urban planning or stormwater). EbA looks at long-term changes within the conservation of biodiversity, ecosystem services and climate change, while Eco-DRR is more focused on immediate and medium-term impacts from the risk of weather, climate and non-climate-related hazards. EbA is often seen as a subset of NBSs that is explicitly concerned with climate change adaptation through the use of nature (Kabisch et al., 2016). From the above discussion, it can be concluded that EbA, Eco-DRR and GI–BGI provide more specific solutions to more specific issues. One key distinction is that unlike the sister concepts, the concept of NBSs is more open to different interpretations, which can be useful in encouraging stakeholders to take part in the discussion. Moreover, features of NBSs provide an alternative to working with existing measures or grey infrastructures. Therefore, it is important to note that very often a combination between natural and traditional engineering solutions (also known as “hybrid” solutions) is likely to produce more effective results than any of these measures alone, especially when their co-benefits are taken into consideration (Alves et al., 2019).

An important advancement in the science and practice of NBSs is given by the EKLIPSE Expert Working Group, which developed the first version of a multi-dimensional impact evaluation framework to support planning and evaluation of NBS projects. The document includes a list of impacts, indicators and methods for assessing the performance of NBSs in dealing with some major societal challenges (EKLIPSE, 2017). Lafortezza et al. (2018) reviewed different case studies around the world where NBSs have been applied from the micro scale to the macro scale. Furthermore, an overview of how different NBS measures can regulate ecosystem services (i.e. soil protection, water quality, flood regulation and water provision) has been carried out by Keesstra et al. (2018).

3 Materials and methodology

The methodology consisted of two phases as schematised in Fig. 2. The first phase consisted of the identification of articles satisfying the search criteria discussed in Sect. 3.1. Next, all articles were screened and filtered based on the selection criteria discussed in Sect. 3.2.

Figure 2 Process of article selection with systematic review method on nature-based solutions for hydro-meteorological risk reduction. The final number of fully reviewed articles is 146.

3.1 Search strategy

The review analysis concerned articles from scientific journals written in English. Two main concepts were used in the search: nature-based solutions and hydro-meteorological risk reduction. As the concept of nature-based solutions appears under different names (which more or less relate to the same field of research), articles related to LIDs, BMPs, WSUD, SuDs, GI, BGI, EbA and Eco-DRR were included in the identification of relevant articles (see Table 2) The review of hydro-meteorological risk included literature on relevant terms (i.e. disasters, risks, hydrology, etc.) and different types of hazards (floods, droughts, storm surges and landslides; Table 2).

Table 2 Selected concepts and terms used to search relevant literature on NBSs for hydro-meteorological risk reduction.

During the construction of the queries, the strings were searched only within index terms and metadata “titles, abstract, and keywords” in the Scopus database. The search terms for the two concepts were linked with the Boolean operator “AND”, while the Boolean operator “OR” was used to link possible terms (Table 2). An example of a protocol is shown below:

“TITLE-ABS-KEY (“Nature-based Solutions” OR “Nature based solutions” OR “Nature Based Solutions” OR “nature-based solutions” OR “Low impact development” OR “Sustainable Urban Drainage Systems” OR “Water Sensitive Urban Design” OR “Best Management Practices” OR “Green infrastructure” OR “Green blue infrastructure” AND “flood”) AND (LIMIT-TO (DOCTYPE, “ar”) OR LIMIT-TO (DOCTYPE, “ch”) OR LIMIT-TO (DOCTYPE, “re”) OR LIMIT-TO (DOCTYPE, “bk”)) AND (LIMIT-TO (LANGUAGE, “English”))”.

The time window selected for the review process was from 1 January 2007 to 19 November 2019; 1608 articles published in scientific journals were found in the Scopus database, and 1431 were found in the Web of Science database. The articles from both databases were combined for a total of 3089 articles. Duplicate articles were removed, resulting in a total of 1439 articles to be considered for further evaluation.

3.2 Selection process

As stated in the Introduction, this study aims at reviewing the state of the art of the research on NBSs that specifically address hydro-meteorological risk reduction. In this regard, the key objectives of the present review work were carefully formulated as follows:

1. to assess the state of the art in research concerning both small- and large-scale NBSs for hydro-meteorological risk reduction;

2. to review the use of techniques, methods and tools for planning, selecting, evaluating and implementing NBSs for hydro-meteorological risk reduction;

3. to review the socio-economic influence in the implementation of NBSs for hydro-meteorological risk reduction as well as their multiple benefits, co-benefits, effectiveness and costs;

4. to identify trends, knowledge gaps and proposed future research prospects with respect to the above three objectives.

These key objectives were defined for the review, with the intention that the results could be both quantitative and qualitative.

The 1439 articles resulting from the search query were thus evaluated with respect to these objectives and those found of little or no relevance with the topic removed. This selection process involved a set of progressive steps as schematised in Fig. 2. Initially, all articles were analysed on the basis of reading titles and keywords and evaluating their relation to the search terms. Articles were discarded if their title and keywords were considered to be of little or no relevance to the key objectives. This step served to reduce the number of articles from 1439 to 433. Secondly, a more in-depth analysis was conducted, based on reading the abstract of each article selected in the previous step. The criteria at this step was that the abstract should discuss hydro-meteorological risk reduction. For example, if the abstract focused more on water quality than risk, that paper was excluded. This step served to reduce the number of articles from 433 to 205. Finally, articles were read in full to identify those that were relevant to the review objectives. Any studies appearing to meet the key objectives (dealing with subjects such as effectiveness of NBSs, techniques, method and tools for planning, and other subjects relevant to the key objectives) were included in the review. As a result, the entire selection process resulted in a total of 146 articles relevant to the objectives of the present review.

Figure 3 Illustration of large and small-scale nature-based solutions (NBSs). Large-scale NBS A illustrates NBSs in mountainous regions (e.g. afforestation, slope stabilisation, etc.), large-scale NBS B illustrates NBSs along river corridors (e.g. widening river, retention basins, etc.) and large-scale NBS C illustrates NBSs in coastal regions (e.g. sand dunes, protection dikes and walls, etc.). Typical examples of small-scale NBSs are green roofs, green walls, rain gardens, permeable pavements, swales, bio-retention, etc.

4 Findings

4.1 Lesson from research on small- and large-scale NBSs for hydro-meteorological risk reduction

In this review, NBSs for hydro-meteorological risk reduction have been divided into small- and large-scale solutions (Fig. 3). Small-scale NBSs are usually referred to as NBSs at the urban or local scale (i.e. buildings, streets, roofs or houses), while NBSs in rural areas, river basins and at the regional scale are referred to as large-scale NBSs (Fig. 3).

4.1.1 Research on small-scale NBSs for hydro-meteorological risk reduction

Small-scale NBSs are usually applied to a specific location such as a single building or a street. However, for some cases, a single NBS is not sufficient to control a large amount of run-off. Therefore, this review discusses the application and effectiveness of both individual NBSs and multiple-NBS combinations. There are 41 articles that have been reviewed on the effectiveness of small-scale NBSs (Table 3). A majority of these (31 articles) discuss the effectiveness of a single or individual NBS site, while only 13 articles discuss the effectiveness of multiple-NBS sites (around 31 %). A summary of the effectiveness, co-benefits and cost of NBS measures at small scale is shown in Table 3.

To date, various types of single-NBS sites have been studied with objectives such as reduction of the flood peak (Carpenter and Kaluvakolanu, 2011; Ercolani et al., 2018; Liao et al., 2015; Mei et al., 2018; Yang et al., 2018), delay and attenuation of the flood peak (Ishimatsu et al., 2017), reduction of volume of combined sewer overflows (Burszta-Adamiak and Mrowiec, 2013), and reduction of surface run-off volume (Lee et al., 2013; Shafique and Kim, 2018). The review found just three articles that discuss the reduction of drought risk by using NBSs. Lottering et al. (2015) used NBSs to reduce water consumption in suburban areas, while Radonic (2019) showed that rainwater harvesting can help reduce household water consumption. Finally, Wang et al. (2019) demonstrated that forests can significantly mitigate drought impacts and protect water supplies for crop irrigation.

The most common NBS measures in urban areas appear to be intensive green roofs (Burszta-Adamiak and Mrowiec, 2013; Carpenter and Kaluvakolanu, 2011; Ercolani et al., 2018), extensive green roofs (Cipolla et al., 2016; Lee et al., 2013), rain gardens (Ishimatsu et al., 2017), rainwater harvesting (Khastagir and Jayasuriya, 2010), dry detention ponds (Liew et al., 2012), permeable pavements (Shafique et al., 2018), bio-retention (Khan et al., 2013; Olszewski and Allen, 2013), vegetated swales (Woznicki et al., 2018) and trees (Mills et al., 2016). However, the authors of these studies investigated the performance of such measures individually (i.e. at the specific, local and single site) without evaluating them in combination with other NBS sites or in hybrid combinations.

The literature to date acknowledges that the effectiveness of NBSs greatly depends on the magnitude and frequency of rainfall events. Green roofs are recognised in reducing peak flows more effectively for smaller-magnitude frequent storms than for larger-magnitude infrequent storms (see, for example, Ercolani et al., 2018). There are also reports that rain gardens are more effective in dealing with small discharges of rainwater (Ishimatsu et al., 2017). Swales and permeable pavements are more effective for flood reduction during heavier and shorter rainfall events. Zölch et al. (2017) suggested that the effectiveness of NBSs should be directly linked to their ability to increase (as much as possible) the storage capacities within the area of interest while using open spaces that have not been used previously and/or while providing benefits to other areas for urban planning.

Several studies evaluated the performance of multiple-NBS (or combined-NBS) measures (i.e. a train of NBSs; see, for example, Damodaram et al., 2010; Dong et al., 2017; Huang et al., 2014; Luan et al., 2017). One of the most successful international projects in combining several NBS measures at the urban scale is the Sponge City Programme (SCP) in China. The SCP project was commissioned in 2014 with the aim of implementing both concepts and practices of LIDs and NBSs as well as various comprehensive urban water management strategies (Chan et al., 2018). Nowadays, the concept (Sponge City) is widely used when a city increases resilience to climate change. It also combines several systems, such as the source control system, urban drainage system and emergency discharge system.

Table 3 Summary of run-off volume and peak flow reduction effectiveness, co-benefits and costs of small-scale NBS measures.

^* Cost of each measure is based on CNT (2009), Nordman et al. (2018) and De Risi et al. (2018).

Porous pavement appears to be one of the most popular measures suitable to be combined with other NBSs for urban run-off management. Examples of this are described in Hu et al. (2017), who used inundation modelling to evaluate the effectiveness of rainwater harvesting and pervious pavement as retrofitting technologies for flood inundation mitigation of an urbanised watershed. Damodaram et al. (2010) concluded that the combination of rainwater harvesting and permeable pavements is likely to be more effective than pond storage for small storms, while the pond is likely to be more effective to manage run-off from the more intense storms.

Several studies argue that multiple-NBS measures can lead to a more significant change in run-off regime and more effective long-term strategies than single-NBS measures (Webber et al., 2018). For example, Wu et al. (2018) simulated eight scenarios changing the percentage of combined green roof and permeable pavement in an urban setting. The results show that when green roofs and permeable pavements are applied at all possible locations, a 28 % reduction in maximum inundation can be obtained. In comparison, scenarios implementing either green roofs or permeable pavements alone at all possible areas experienced a reduction of 14 %. One of the main reasons for the superior performance of combined NBSs is that they work in parallel, each treating a different portion of run-off generated from the sub-catchment (Pappalardo et al., 2017). For these combinations, the spatial distribution should be carefully considered because it can improve the run-off regime better when compared to centralised NBSs (Loperfido et al., 2014).

Further research on the use of combined NBSs and grey infrastructure (i.e. hybrid measures) is desirable, as only three contributions were found in the review. Alves et al. (2016) presented a novel method to select, evaluate and place different hybrid measures for retrofitting urban drainage systems. However, only fundamental aspects were touched upon in the methodology, and they suggested that future work should include the possibility of considering stakeholders' preferences or flexibility within the method. In the work of Vojinovic et al. (2017), a methodological framework that combines ecosystem services (flood protection, education, art and culture, recreation, and tourism) with economic analysis for the selection of multi-functional measures and consideration of small- and large-scale NBSs has been discussed for the case of Ayutthaya in Thailand. Onuma and Tsuge (2018) compared the cost and benefits and performance of NBSs and grey infrastructures, concluding that NBSs are likely to be more effective when implemented through cooperation with local people, whereas hybrid solutions are more effective than a single NBS in terms of performance.

The first limitation of the above studies is that they only assess the effectiveness of NBSs at urban scales. This may not be sufficient for large events, as climate change is likely to increase the frequency and intensity of future events (Qin et al., 2013). A large-scale NBS could be a solution for storm events with large magnitude and long duration, which is usually the case for disaster risk reduction applications, and therefore research in this direction is highly desirable (Giacomoni et al., 2012). Although Fu et al. (2018) analysed variations in run-off for different scales and land-uses, the impact of NBSs was only examined for the small urban scale. Another limitation is that none of these contributions incorporated cost–benefit analyses (CBAs). CBAs can be used as a tool to support the decision-making process, as they serve the feasibility of implementation costs and the potential benefits of NBSs.

4.1.2 Research on large-scale NBSs for hydro-meteorological risk reduction

Large-scale water balance, water fluxes, water management and ecosystem services are affected by future changes such as climate change, land use changes, water use changes and population growth. To address such challenges, large-scale NBSs are needed to make more space for water to retain, decelerate, infiltrate, bypass and discharge (Cheng et al., 2017; Thorslund et al., 2017). Generally, a large-scale NBS combines different NBSs within a larger system to achieve better long-term strategies. There are some examples of NBS measures for hydro-meteorological risk reduction summarised in McVittie et al. (2018) and Sahani et al. (2019). A summary of effectiveness, co-benefits and cost of large-scale NBS measures is shown in Table 4.

Table 4 Summary of effectiveness, co-benefits and costs of large-scale NBS measures.

Only few articles have addressed the combined behaviour of NBSs at large scales. One of the possible reasons is that large-scale systems are more complex than small-scale systems. The most common large-scale NBSs are flood storage basins (De Risi et al., 2018) and preservation and regeneration of forests in flood-prone areas (Bhattacharjee and Behera, 2018), making more room for the river (Klijn et al., 2013), river restoration (Chou, 2016), wetlands (Thorslund et al., 2017) and mountain forestation (Casteller et al., 2018).

A classic example of a large-scale NBS implementation is the Room for the River Programme implemented along the Rhine and Meuse rivers in the Netherlands. The Room for the River Programme consisted of 39 local projects based on nine different types of measures (Klijn et al., 2013). These measures are floodplain lowering, dike relocation, groyne lowering, summer bed deepening, water storage, bypassed and floodways, high-water channels, obstacle removal, and dike strengthening. The benefits that the programme achieved are more than just reducing flooding, also increasing opportunities for recreation, habitat and biodiversity in the area (Klijn et al., 2013). Another case study of a large-scale NBS is the Laojie River project in Taoyuan in Taiwan. The study focused on changing the channelised, culverted, flood-control watercourse into an accessible green infrastructure corridor for the public (Chou, 2016). The landscape changes resulting from this project have increased recreation activities and improved the aesthetic value in the area.

NBSs may benefit people in coastal areas by reducing risk from storm surges, wave energy, coastal flooding and erosion, as documented by several authors (see, for example, Van Coppenolle, 2018; Joyce et al., 2017; Ruckelshaus et al., 2016; Sutton-Grier et al., 2018). NBSs for coastal areas can be implemented either at large or small scales. They include dunes, beaches, oyster and coral reefs, mangroves, seagrass beds, and marshes. These measures can also provide habitats for different species such as fish, birds and other wildlife (Ruckelshaus et al., 2016). Schoonees et al. (2019) provided lists of general recommendations, technical guidelines and policies, and design considerations for NBSs in coastal areas. However, only a few articles of the 146 reviewed focused on the potential benefits of NBSs in coastal areas.

Casteller et al. (2018) concluded that native mountain forests could be used to reduce hydro-meteorological risk such as flash floods and landslides. Moreover, the use of NBSs in different forest ecosystems to reduce shallow landslide impacts should be addressed (de Jesús Arce-Mojica et al., 2019). To reduce the impact of large-scale hydro-meteorological events, more research is needed on large-scale NBSs and their hybrid combinations designed to attenuate flows and improve drainage. They should be implemented to include improvements in solid waste management, community-based river cleaning programmes and reforestation (De Risi et al., 2018).

4.2 Techniques, methods and tools for planning, selecting, evaluating and implementing NBSs

Figure 4 illustrates a typical process for the selection and evaluation of NBSs. The process starts by selecting possible measures that correspond to the local characteristics and project's target. The next step is concerned with evaluating the measures' performance using numerical models, cost–benefit analysis and/or multi-criteria analysis. For more complex systems with a large number of scenarios and parameters, optimisation can be used to maximise the benefits and minimise the costs. The techniques, methods and tools for planning, selecting, evaluating and implementing NBSs are reviewed in the following section.

Figure 4 Evaluation process of nature-based solutions. The process includes selecting possible measures and evaluating and optimising measures' performance using available tools.

4.2.1 Selection of NBSs

It has been a well-accepted fact that not all NBSs are suitable for all conditions. Therefore, it is important to consider the feasibility and constraints at the site at an early stage in the selection process. The first consideration in selecting NBSs is to define the objective such as the target area (i.e. urban or rural) and performance requirements such as quantity and/or quality (Romnée and De Herde, 2015; Zhang and Chui, 2018). For example, Pappalardo et al. (2017) chose permeable pavements and green roofs because they can detain run-off or enhance infiltration to the subsoil. Another approach is to consider both primary benefits and key co-benefits. For instance, Majidi et al. (2019) developed a framework to select NBSs to reduce flood risk and enhance human thermal comfort (reducing heat stress). Many authors suggest restricting the choice of appropriate NBSs based on common site constraints such as land use, soil type, groundwater depth, catchment characteristics, political and financial regulations, amenities, environmental requirements, and space available (Eaton, 2018; Joyce et al., 2017; Nordman et al., 2018; Oraei Zare et al., 2012). For example, Eaton (2018) selected bio-retention measures because they are more suitable in low-density residential land use. Moreover, the study of Reynaud et al. (2017) describes how the type of NBS has an impact on individuals' preference for ecosystem services. Therefore, a screening analysis is necessary for selecting the NBS measures that are best suited to local constraints and objectives, providing decision makers with valuable information. The way forward in the selection of NBSs is to consider spatial planning principles to locate the position for measures. Spatial planning principles can facilitate and stimulate discussion among local communities, researchers, policymakers and government authorities.

4.2.2 Frameworks and methods for evaluation of NBSs

There are several frameworks and methods that can be used to evaluate the performance indicators of NBSs discussed in this review. One of the most popular evaluation approaches is to analyse, simulate and model hydrology, hydraulics and water balance processes. This information is then used to support decision makers, planners and stakeholders in their evaluation of performance and potential of NBSs by comparing modelled results against the current situation, baseline scenario or targets (Jia et al., 2015).

In addition to hydrological and hydraulic analyses, cost–benefit analyses are often used to select and evaluate NBSs (Huang et al., 2018; Nordman et al., 2018; Watson et al., 2016; Webber et al., 2018). The common benefits considered include prevented damage costs, omitted infrastructures and prevented agricultural losses. One cost–benefit approach is to evaluate NBSs by applying the whole life cycle costing (LCC) approach, including construction, operation, maintenance and opportunity costs (Nordman et al., 2018), and return on investment (ROI; De Risi et al., 2018).

Another method for the evaluation of NBSs is multi-criteria analysis (MCA), which has the potential to integrate and overcome the differences between social and technical approaches (Loc et al., 2017). It can be used to structure complex issues and help find a better understanding of costs and benefits. Such analysis is useful for decision makers when there are multiple and conflicting criteria to be considered (Alves et al., 2018b; Loos and Rogers, 2016). The MCA takes different criteria into account and assigns weights to each criterion. This process can produce a ranking of the different measures that can be implemented on the site (Chow et al., 2014; Jia et al., 2015). For example, Loc et al. (2017) integrated the results from numerical modelling and social surveys into a MCA and ranked the alternatives based on the evaluation criteria of flood mitigation, pollutant removal and aesthetics. Loos and Rogers (2016) applied multi-attribute utility theory (MAUT) to assess utility values for each alternative by assuming that preference and utility are independent of each other. Petit-Boix et al. (2017) recommended that future research combine the economic value of the predicted material and ecological damage, risk assessment models and environmental impacts of NBSs.

Since not all assessments can be done with modelling alone, interviews and fieldwork are often necessary. For instance, Chou (2016) used 18 open questions from six topics, namely accessibility, activities, public facilities, environmental quality, ecological value and flood prevention. These questions are used to evaluate the qualitative performance of river restoration. However, some of the methods are only appropriate for small-scale applications and cannot be applied in large catchments. Yang et al. (2018) proposed relative performance evaluation (RPE) methods, which use a score to calculate the performance for all alternatives. This score is calculated as the weighted sum of the scores of individual indicators.

From the discussion above, it can be observed that there are still challenges in evaluating intangible benefits of NBSs and incorporating stakeholders' preferences into the process. For complex systems with a large number of scenarios and parameters, simple trial-and-error methods may not be the feasible approach. In such cases, an automated optimisation method could be effectively applied to handle these tasks and to combine the above-mentioned methods. There is also a challenge in combining a range of aspects that can and cannot be expressed in monetary terms into the same framework of analysis.

4.2.3 Optimal configuration of NBSs

In order to implement NBSs, typical selection factors include the number of NBS measures, size, location and potential combinations of NBSs. Optimisation of NBS strategies has been increasingly used in the context of urban stormwater management. Most of the studies focus on minimising water quantity and improving water quality by selecting the type, design, size and location of NBSs (Behroozi et al., 2018; Gao et al., 2015; Giacomoni and Joseph, 2017; Zhang and Chui, 2018). Zhang and Chui (2018) systematically reviewed optimisation models that have different structures, objectives and allocation components. This section reviews some examples of using optimisation to assess NBSs.

A comprehensive modelling system typically refers to an optimisation package tool that integrates an “easy-to-use” user interface with physically based deterministic models. Examples include SUSTAIN (the System for Urban Stormwater Treatment and Analysis IntegratioN; Zhang and Chui, 2018) and best management practice decision support (BMPDSS; Gao et al., 2015). The SUSTAIN model was developed by the US Environmental Protection Agency (US EPA) and aims to provide decision makers with support in the process of selection and placement of NBS measures and to optimise the hydrological performance and cost-effectiveness of NBSs in the urban watershed (Leslie et al., 2009; C. Li et al., 2018). There are several studies that apply SUSTAIN with the aim of minimising the cost of NBSs for both run-off quantity (flow volume and peak flow) and run-off quality (pollutant removal; Gao et al., 2015; N. Li et al., 2018). It is, however, important to note that comprehensive modelling systems are not always easily modified to fit with the specific needs of users.

Another optimisation tool approach is integrated model–algorithm tools, which combine numerical (hydrological–hydrodynamic) models with optimisation algorithms. A popular optimisation method used to evaluate NBS performance is a multialgorithm, genetically adaptive multiobjective (AMALGAM) method using the multilevel spatial optimisation (MLSOP) framework (Liu et al., 2016).

In the reviewed articles, the non-dominated sorting genetic algorithm II (NSGA-II) is used in most of the studies to date. Wang et al. (2015) concluded that NSGA-II is one of the most popular multiobjective evolutionary algorithms (MOEAs) despite limited parameter tuning features and generally outperformed the other MOEAs in relation to the set of solutions generated. There are several examples of the use of NSGA-II. Oraei Zare et al. (2012) minimised run-off quantity while maximising the improvement of water quality and maximising reliability. Karamouz and Nazif (2013) minimised the cost of flood damage while minimising NBS cost in order to improve system performance in dealing with the emerging future conditions under climate change. Yazdi and Salehi Neyshabouri (2014) optimised cost-effectiveness, which focused on land use change strategies including orchard, brush and seeding measures in different parts of the watershed. All of the above-mentioned studies coupled NSGA-II with the Storm Water Management Model (SWMM) developed by US EPA (Cipolla et al., 2016; J. Li et al., 2018; Mei et al., 2018; Tao et al., 2017; Wu et al., 2018; Yang et al., 2018; Zhu and Chen, 2017) to address the optimisation problems.

There are two different optimisation methods of particle swarm optimisation (PSO) which have been found in the course of this review. The modified particle swarm optimisation (MPSO) is used by Duan et al. (2016) to solve the multi-objective optimal (MOO) of the cost-effectiveness of NBS-based detention tank design. Similarly, Behroozi et al. (2018) used the multi-objective particle swarm optimisation (MOPSO) by coupling it with SWMM to optimise the peak flow and mean total suspended solid (TSS) concentration reduction by changing the combinations of NBSs.

Another algorithm that is used for optimising the performance of NBSs is simulated annealing (SA). SA is a general probability optimisation algorithm that applies thermodynamic theories in statistics. An example of a study with SA is given by Huang et al. (2018), who automatically linked SA with SWMM to maximise the cost and benefit for flood mitigation and layout design. The cost–benefit analysis is computed using annual cost, which includes both annual fixed cost and annual maintenance cost. Another study that applied SA is Chen et al. (2017), who combined SA with SWMM to locate NBSs in Hsinchu County in northern Taiwan by considering three objective functions. These were minimising depths, durations and the number of inundation points in the watershed.

It can be observed that most of the optimisation models to date (both the comprehensive modelling system and model algorithms) are coupled with SWMM for urban storm management. There is still a lack of research that uses optimisation to maximise the efficiency of NBSs on a large scale as well as combining other co-benefits in optimisation (Table 3). Furthermore, there is a lack of research that employs two-dimensional models in the optimisation analysis. This is particularly important when considering estimation of flood damages and other flood propagation-related impacts.

4.2.4 Tools for selection, evaluation and operation of NBSs

Recently, several selection and evaluation tools have been developed in order to assist stakeholders in screening, selecting and visualising NBS measures. Examples of web-based applications developed to screen urban NBS measures are the green–blue design tool (atelier GROENBLAUW, 2019), PEARL KB (PEARL, 2019), climate adaptation app (Bosch Slabbers et al., 2019) and Naturally Resilient Communities solutions (Naturally Resilient Communities, 2019). These web-based tools allow the user to filter NBSs in relation to their problem type, measure, land use, scale and location.

In addition to the above, there are also tools that combine both the selection and evaluation processes together to use as planning support systems tool. An example is the SuD selection and location (SUDSLOC) tool, which is a GIS tool linked to an integrated 1-D hydraulic sewer model and a 2-D surface model. UrbanBEATS (the Urban Biophysical Environments and Technologies Simulator) aims to support the planning and implementation of WSUD infrastructure in urban environments (Kuller et al., 2018). Other tools that can be used to select and evaluate potential NBS interventions are Long-Term Hydrologic Impact Assessment-Low Impact Development (L-THIA-LID; Ahiablame et al., 2012; Liu et al., 2015) and the GIS-based tool called the Adaptation Support Tool (AST; Voskamp and Van de Ven, 2015). Although these tools could be useful in assisting decision makers, some of them may not be suitable for every location and scale. For example, source data required into L-THIA-LID cover only the US, and QUADEAU (Romnée and De Herde, 2015) is only suitable for urban stormwater management on a public-space scale.

In addition to the above, other models such as MIKE packages developed by DHI (Semadeni-Davies et al., 2008), the soil and water assessment tool (SWAT; Cheng et al., 2017), IHMORS (Herrera et al., 2017), and the urban water optioneering tool (UWOT; Rozos et al., 2013) can be effectively used in the analysis effectiveness of NBSs.

To date, very few tools have been developed to calculate multiple benefits of NBSs in monetary terms as well as to address their qualitative benefits. Some examples are the Benefits of SuDs Tool (BeST), which provides a structured approach to evaluating potential benefits of NBSs (Digman et al., 2017; Donnell et al., 2018; Fenner, 2017), and the MUSIC (Model for Urban Stormwater Improvement Conceptualization) tool, which is a conceptual planning and design tool that also contains a life cycle costing module for different NBSs that are implemented in Australia (Khastagir and Jayasuriya, 2010; Schubert et al., 2017).

There are also other tools that can be used for modelling stormwater management options and/or to assess economic aspects of NBSs in urban areas. These are documented in the work of Jayasooriya and Ng (2014). However, most of these tools only focus on small-scale NBSs such as bio-retentions, pervious pavements, green roofs, swales, retention ponds, biofiltration and rainwater harvesting. There are only a few tools that can address river and coastal flood protection measures and droughts, while none of the tools can be used to reduce the risk from landslides and storm surges. A lack of information systems, information clusters, and platforms for information exchange between authorities and practitioners has been recognised by Kabisch et al. (2016).

There is also the need to explore the use of sensors, regulators, telemetry, and supervisory control and data acquisition (SCADA) systems for efficient and effective operation and real-time control of NBSs. Such configurations, which are based on the use of real-time control technology for operation of NBSs, can be referred to as SMART NBSs. The value of exploring SMART NBS configurations may be particularly beneficial for hybrid systems, where NBS sites need to be configured to work closely with different kinds of measures.

4.3 Socio-economic influence on implementation of NBSs

Investing in NBSs for hydro-meteorological risk reduction is essential to ensure the capability for future socio-economic development (Faivre et al., 2018). In this respect, the European Commission has been investing considerably in the research and innovation of NBSs or EbA, and some recent efforts have been placed on practical demonstration of NBSs for climate change adaptation and risk prevention (Faivre et al., 2017). The European Commission is dedicated to bringing about innovative “sciences–policy–society” mechanisms, open consultations and knowledge-exchange platforms to engage society in improving the condition for implementation of NBSs (Faivre et al., 2017). There are some inventories of web portals, networks and initiatives that address NBSs at European, national and sub-national levels (Table 6).

Denjean et al. (2017) noted that the people who propose NBSs are in many cases ecologists and biologists who have been trained within a very different scientific paradigm and thus speak a “different language” to the key decision makers, who are often civil and financial engineers, contractors and financing officers. Hence, this may limit the feasibility of implementation of NBSs.

Very few articles study actions or processes in relation to stakeholder participation. However, those that do so stress the importance of involving stakeholders in the evaluation and implementation of NBSs and the current practical limitations of implementing NBSs. One of the important reasons is to ensure that stakeholders and local government are fully aware of the multiple benefits of NBSs so that they can integrate them better into planning for sustainable cities (Ishimatsu et al., 2017). For example, Liu and Jensen (2018) and Chou (2016) claim that the implementation of NBSs with visible benefits in the landscape and the liveability of the city (in terms of amenities, recreation, green growth and microclimate) can create positive attitudes among stakeholders towards applying NBSs. Moreover, as the implementation of NBSs is often a costly investment for local communities, and the facilities are expected to be in place for a decade, it is essential for stakeholders to know the effectiveness of NBSs (Semadeni-Davies et al., 2008). Involving the community with authorities in both the planning and implementing process can be a very useful strategy (Dalimunthe, 2018). In a case study of the Great Plains in the US, Vogel et al. (2015) addressed how local perceptions of NBS effectiveness and applicability limit its adoption. One of the factors was a lack of awareness of NBSs and support from stakeholders and authorities. Another case in Portland, Oregon, US, Thorne et al. (2018) concluded that the limited adoption of NBSs is caused by the lack of confidence in public preferences and sociopolitical structures as well as the uncertainty regarding scientific evidence related to physical processes. To solve this, they suggested that both sociopolitical and biophysical uncertainties must be identified and managed within the framework for designing and delivering sustainable urban flood risk management.

Schifman et al. (2017) proposed a framework for adaptive socio-hydrology (FrASH) that can be used in NBS planning and implementation by bringing ideas together from socio-hydrology; the capacity for adaptation, participation and inclusiveness; and organised action. The framework also helps in creating a connected network between municipalities, public-works departments, organisations and people in the community. This potentially allows for the management of resilience in the system at multiple scales.

Often, it is not as easy to address socio-economic issues as technical questions. These socio-economic issues include perception and acceptance, policies, interdisciplinary nature, education, and documenting the economic benefit of NBS implementation (Alves et al., 2018a; Santoro et al., 2019; Vogel et al., 2015). Nevertheless, social-science research (i.e. surveys, interviews and focus groups) helps to review and gain insights into the obstacles and motivations for implementing NBSs as well as to understand a community's resilience and stakeholders' risk perception (Matthews et al., 2015; Santoro et al., 2019), for instance, bringing the findings to stakeholders and community members to discuss what level of flood hazards is acceptable and what level of climate change adaptation capacity the community plans to achieve (Brown et al., 2012). Moreover, sociopolitical dynamics in NBSs is still lacking; there are few case studies available that critically evaluate the politics of NBSs in the role of community mobilisation (Triyanti and Chu, 2018). Not only it is essential to involve stakeholders in the selection, planning, design and implementation of NBSs, but it is also important for bridging gaps between researchers, engineers, politicians, managers and stakeholders. This may help to improve our capacity for using both small- and large-scale NBSs. There are good documentations of policy arrangements, scientific niches and the current status of governance studies of NBSs that were reviewed by Scarano (2017) and Triyanti and Chu (2018).

4.4 Multiple benefits of NBSs

The literature on a NBS and its sister concepts increasingly refers to multiple benefits of social, economic and environmental enhancements. The reason for this is that NBSs are regarded as sustainable solutions that use ecosystem services to provide multiple benefits for human well-being and the environment, which differs from grey infrastructure. Moreover, these multiple benefits of NBSs can help to achieve many of the goals of the 2030 Agenda for Sustainable Development. The recent publication shows how NBSs can contribute to achieving the SDGs (Seifollahi-Aghmiuni et al., 2019). This publication reports that wetland ecosystem services in Sweden positively interact with SDG 1 (no poverty), SDG 2 (zero hunger), SDG 3 (good health and well-being), SDG 6 (clean water and sanitation), SDG 7 (affordable and clean energy), SDG 11 (sustainable cities and communities), SDG 12 (responsible consumption and production), SDG 13 (climate action), SDG 14 (life below water) and SDG 15 (life on land). One of the processes that could provide these benefits is to give more significant consideration to landscape function, adaptive and multi-functionality design (Lennon et al., 2014; Vojinovic et al., 2017), restoring naturally occurring ecosystems, and promoting desirable soil (Keesstra et al., 2018).

The literature to date shows that multiple challenges can be continually addressed through NBSs. These include reducing flood risk (Song et al., 2018), storing and infiltrating rainfall run-off, delaying and reducing surface run-off, reducing erosion and particulate transport (Loperfido et al., 2014), recharging groundwater discharge, reducing pollution from surface water (Donnell et al., 2018), increasing nutrient retention and removal (Loperfido et al., 2014), maintaining soil moisture, and enhancing vegetation growth. Such benefits help in reaching SDG 6 – ensuring sustainable water management.

Beyond water management, the case for NBSs includes their ability to provide additional benefits in improving socio-economic aspects (SDG 11) and human well-being (SDG 3) through recreational areas and aesthetic value (Song et al., 2018) as well as encouraging tourism through the access to nature (Sutton-Grier et al., 2018). Wheeler et al. (2010) quantified the volume and intensity of children's physical activity in green space and found that time in green space is more likely to lead to greater activity intensity amongst children. The use of NBSs can bring economic benefits (SDG 1 and SDG 8) in different ways, such as reduced or prevented damage costs from hydro-meteorological events, energy savings from the reduction of stormwater that typically needs to be treated in a public sewerage system, and carbon savings from reduced building energy consumption (heating and cooling; Soares et al., 2011). Such energy and carbon savings will help contribute to SDG 13.

The environmental benefits of NBS measures can have various positive impacts. Some of the most important are the ability to enhance environmental and ecosystem services by connecting habitat and biodiversity (Hoang et al., 2018; Reguero et al., 2018; Thorslund et al., 2017), increasing carbon consequences, reducing air and noise pollution (Donnell et al., 2018), and improving urban heat island effect mitigation (Majidi et al., 2019; Raymond et al., 2017). Zhang and Chui (2019) reviewed the hydrological and bio-ecological benefits of NBSs across spatial scales and suggested that there should be more research at the catchment scale to consider the full benefits of NBSs.

The hydrological and water quality benefits of NBSs have been widely reviewed and discussed, but there are few articles that focus on evaluating the multiple benefits of NBSs. Doing so could help raise awareness and enhance the institutional and social acceptance of these measures (Pagano et al., 2019). Hoang et al. (2018) proposed a new integrated methodology using a GIS approach to assess benefits and disadvantages of NBSs, which include habitat connectivity, recreational accessibility, traffic movement, noise propagation, carbon sequestration, pollutant trapping and water quality. Donnell et al. (2018) used BeST and the Blue-Green Cities toolbox to assess benefits, and Mills et al. (2016) assessed air pollution reduction based on tree canopy cover. Alves et al. (2019) presented a novel methodology for valuing co-benefits for NBS applications in urban contexts. Fenner (2017) recommended that their spatial distribution should be assessed through multi-functional design, making it possible to identify how this is valuable to stakeholders and where the overall aggregated benefits occur. However, there is still a need for deeper understanding of assessment of the multiple benefits of NBSs (Liu et al., 2017). A challenge is the lack of information on the values of ecosystem and multi-related ecosystem economic valuation.

Figure 5 Number and trend of published articles on nature-based solutions (NBSs) for hydro-meteorological risk reduction and their sister terms: low-impact developments (LIDs), best management practices (BMPs), water-sensitive urban design (WSUD), green infrastructure (GI), sustainable urban drainage systems (SuDs), nature-based solutions (NBSs), ecosystem-based adaptation (EbA), ecosystem-based disaster risk reduction (Eco-DRR) and blue–green infrastructure (BGI).

4.5 Trends, knowledge gaps and future research prospects

The literature material reviewed in this study showed that NBSs have not been equally applied to all hydro-meteorological risk reduction contexts. The review identified, in total, 1204 journal articles from 2007 to the end of 2019. The analysis of publications from 2007 to 2019 shows that only 121 out of 1439 articles (i.e. 11 %) explicitly used the term nature-based solution for hydro-meteorological risk reduction (Fig. 5). This can be explained due to the term NBSs being used only after 2008 (MacKinnon et al., 2008), while other terms were used earlier in different countries (Fig. 1). However, the significant increase in published articles in recent years shows how NBSs are a rapidly growing research area (Fig. 5).

One hundred forty-six publications of NBSs for hydro-meteorological risk reduction were reviewed. Most of the literature to date is about NBSs in urban areas, whereas the contexts concerning river and coastal floods, droughts, and landslides are the least addressed; 82 % of all articles were concerned with run-off reduction or flood risk reduction in urban areas (Fig. 6). Even when there were 2 search terms that include “Urban” out of the 10 search terms, it contributed to only 2.7 % of the total 82 % urban cases. The large-scale NBSs is only 16 articles, mostly focusing on river and coastal flooding (Table 5).

Figure 6 Percentage of published articles that have been studied on nature-based solutions (NBSs) for different types of hydro-meteorological risks.

Table 5 Overview of knowledge gaps and potential future research prospects.

Table 6 An overview of web portals, networks and initiatives that address nature-based solutions.

An overview of quantitative results, some research gaps and future research prospects is given in Table 6. This table indicates subjects or areas in which knowledge is missing or insufficient. The knowledge gaps have been divided into eight subjects, which are the effectiveness of small-scale NBSs, the effectiveness of large-scale NBSs, selection and assessment of NBSs with focus on risk reduction, multiple benefits of NBSs, application of tools, multi-functional design, stakeholder participation, and financial governance and policy. Some of the key challenges' concluding remarks are summarised below.

There is a clear gap between the amount of research on small-scale NBSs in urban areas and large-scale NBSs at the catchment (river basin), rural and regional scale. The reason for this is that a large-scale system is more complex than a small system. Therefore, research and frameworks that deal with the problem of reducing hydro-meteorological risk with upscaling NBSs from the urban scale to the catchment (river basin) scale would be beneficial, and it would also be beneficial to understand both the natural processes of large-scale NBSs and how they change over time. Furthermore, there are only a few studies that combine NBSs at both the small and large scale, and further research in this direction is highly desirable.

Obviously, there is no single NBS that can solve all problems. Every project needs to be designed to address a particular challenge in its local contexts and in its respective community. Therefore, an understanding of site conditions is necessary for NBSs to achieve the target of the project.

Based on the findings of the literature review, there are still challenges in relation to methods and tools for planning and implementing NBSs. These include improving and developing methods for assessing co-benefits (especially social and ecological benefits, i.e. aesthetical values, community liveability and human health), frameworks and methods for evaluating large-scale NBSs, and hybrid measures (i.e. combinations of grey infrastructure and small- and large-scale NBSs).

There are also challenges in incorporating local stakeholder participation within the framework and models and within the assessment and implementation process. Other challenges regarding governance are to develop guidance on effective models of governance and provide information on actors, institutions and legal instruments, and other requirements that are relevant for implementing NBSs. The reason for this is the lack of workable frameworks that can bring together a variety of stakeholder groups. Moreover, there is still a lack of financial studies and guidelines for cost-effective implementation, maintenance and operation of NBS projects, and mechanisms that can be used to promote new business and financial models for successful implementation of NBSs.

There should also be more efforts in the development of assessment tools that incorporate new technologies such as real-time control systems, forecast models and coupled models to provide more active and integrated operational solutions (i.e. SMART NBSs). There is a need for the development of databases that include functions, benefits and costs of large and small-scale NBSs to facilitate future research.

5 Conclusions

The present paper provides a critical review of the literature and identifies future research prospects based on the current knowledge gaps in the area of nature-based solutions for hydro-meteorological risk reduction by using a systematic review. The systematic review method helps to limit the scope of the work and also provides useful direction for defining research gaps, as articles can be collected from a board range of sources. However, there are some disadvantages of systematic reviews. For example, a finite selection of keywords will introduce gaps into the list of articles to be reviewed. Also, important grey literature (e.g. reports and books) could be overlooked. Finally, poorly written abstracts may cause an article to be excluded from the review.

The review process started by analysing 1608 articles sourced from Scopus and 1431 articles from the Web of Science from 1 January 2007 to 19 November 2019. The final full analysis was performed on 146 articles. The systematic review has shown that considerable achievements have been made to date. However, there are still many challenges and opportunities that will play an important role in extending the knowledge of NBSs in the coming years. Some examples of research gaps are combining small-scale and large-scale NBSs; the effectiveness of NBSs in reducing risk at the regional and catchment scale; the frameworks, methods and tools for assessing co-benefits; involvement of local stakeholders in the selection; assessment and implementation process; integration of NBSs with new technologies; and development of NBS databases.

The effectiveness, benefits and acceptance of NBSs are dependent on the implementation purpose, local context and cultural setting. For example, small-scale NBSs (e.g. swales, green roofs or porous pavements) are more suitable for urban flooding, while large-scale NBSs (river restoration, dunes or wetlands) are more suitable for river floods, coastal floods, droughts and landslides. Small-scale NBSs are more effective in reducing flood peaks for smaller-magnitude frequent storms (e.g. 2-year return period) than larger-magnitude infrequent storms (e.g. 10-year return period). Large-scale NBSs can provide more benefits compared to small-scale NBSs because they encompass larger spaces; thus more functions can be included in the design process. For example, the Laojie River project in Taoyuan in Taiwan changed the channel into an accessible green corridor. This project helps in reducing flood risk, improving riverside landscapes, increasing recreation area, increasing the aesthetic value and improving river water quality. On the other hand, small-scale NBSs need less area because most of the measures can be implemented in the free space. For example, green roofs can be implemented on the roofs of buildings, and permeable pavements can be implemented in car parks. Investments in NBSs will benefit society by providing cost-effective measures and adaptive strategies that protect communities and achieve a range of co-benefits. Moreover, bridging the gaps between researchers, engineers and stakeholders will help to improve the capacity of NBSs in reducing hydro-meteorological risk as well as increasing their benefits. Strengthening these aspects may be beneficial for improving acceptance of NBSs at the local level.

Horizon 2020 projects, namely RECONECT, PHUSICOS and OPERANDUM, were initiated in 2018 to help bridge the gaps in the innovation of NBSs and to test their efficacy in rural, mountain and transition land environments. Development of techniques, methods and tools for planning, selecting, evaluating and implementing NBSs are among the common products of RECONECT, PHUSICOS and OPERANDUM.

Appendix A: Abbreviations

AST	Adaptation Support Tool
BeST	Benefits of SuDs Tool
BGI	Blue–green infrastructure
BMPDSS	Best management practice decision support
BMPs	Best management practices
CBA	Cost–benefit analyses
CBD	Convention on Biological Diversity
CCA	Climate change adaptation
CEM	Commission on Ecosystem Management
DRR	Disaster risk reduction
EbA	Ecosystem-based adaptation
Eco-DRR	Ecosystem-based disaster risk reduction
EC	European Commission
FrASH	Framework for adaptive socio-hydrology
GI	Green infrastructure
IIED	International Institute for Environment and Development
IUCN	International Union for Conservation of Nature
LCC	Life cycle costing
LID	Low impact development
L-THIA-LID	Long-Term Hydrologic Impact Assessment-Low Impact Development
MAUT	Multi-attribute utility theory
MCA	Multi-criteria analysis
MLSOP	Multilevel spatial optimisation
MOEAs	Most popular multiobjective evolutionary algorithms
MOO	Multi-objective optimal
MOPSO	Multi-objective particle swarm optimisation
MUSIC	Model for Urban Stormwater Improvement Conceptualization
NBSs	Nature-based solutions
NSGA-II	Non-dominated sorting genetic algorithm II
PSO	Particle swarm optimisation
RECONECT	Regenerating ECOsystems with Nature-based solutions for hydro-meteorological risk rEduCTion
ROI	Return on investment
RPE	Relative performance evaluation
SA	Simulated annealing
SCADA	Supervisory control and data acquisition
SCP	Sponge City Programme
SDGs	Sustainable Development Goals
SEI	Stockholm Environment Institute
SFDRR	Sendai Framework for Disaster Risk Reduction
SUDCLOC	SuD selection and location
SuDs	Sustainable urban drainage systems
SUSTAIN	System for Urban Stormwater Treatment and Analysis IntegratioN
SWAT	Soil and water assessment
SWMM	Storm Water Management Model
TSS	Total suspended solids
UN	United Nations
UNFCCC	UN Framework Convention on Climate Change
US EPA	United States Environmental Protection Agency
UrbanBEATS	The Urban Biophysical Environments and Technologies Simulator
UWOT	Urban water optioneering tool
WSUD	Water-sensitive urban design