Sustainable urban stormwater management in developing countries: integrating strategies to overcome Brazilian barriers

ABSTRACT Sustainable urban stormwater management (SUSM) is essential to urban sustainability. However, barriers to adopting it are observed even in places where SUSM is more widespread. Recent studies have evaluated strategies for overcoming some barriers. However, no study has systematically analyzed the strategies available for overcoming the most common barriers in developing countries. Thus, this article aimed to provide a literature review on these strategies. Ninety-two documents were evaluated, resulting in eight solution strategies, detailed by 80 implementation measures. The interrelationships among the solution strategies and their applicability to overcome the SUSM-related barriers were evaluated. This analysis showed that the solution strategies are interdependent, so it would be inefficient to adopt them in isolation. On the other hand, adopting a strategy can help overcome several barriers, also enhancing other strategies and urban issues. The availability of this systematized information can help to break through common barriers, optimizing efforts to adopt SUSM.


Introduction
Sustainable urban stormwater management (SUSM) is a comprehensive and integrated approach to manage urban water, which addresses not only traditional sanitary aspects, but also environmental and quality of life issues (Fletcher et al. 2015).SUSM is not a new concept proposal, as many advanced and well known technical solutions already exist, but comprises all them (e.g.compensatory techniques, alternative techniques, best management practices (BMP), integrated urban water management (IUWM), sustainable urban water management (SUWM), low impact urban design and development (LIUDD), low impact development (LID), water sensitive urban design (WSUD), green infrastructure (GI), nature based solutions (NBS), and the sustainable urban drainage system (SUDS)).Regardless of the adopted concept, the goals are similar: managing the urban stormwater in a more sustainable way, mimicking the natural hydrological cycle (Zevenbergen, Fu, and Pathirana 2018;Qi et al. 2020).This approach has several benefits when compared to conventional urban drainage (Ballard et al. 2015), however it has not yet been adopted worldwide.
The SUSM implementation concepts are well known and many guidelines are available, e.g. from Maryland (Chesapeake Bay Foundation 2004) in the US, and from the UK (Ballard et al. 2015).Despite the technical knowledge, the paradigm shift from conventional urban drainage to SUSM faces several barriers, which have already been observed in some places where SUSM is more widespread -mostly developed countries (e.g. Brown and Farrelly 2009;Marlow et al. 2013;Chang et al. 2018b).Nevertheless, Vasconcelos et al. (2020) carried out a literature review of the barriers to widely adopting SUSM worldwide and validated them to Brazila developing country with their common urban management issues (Gusmão 2016).
The SUSM-related barriers have several sources, such as local government issues, strategic vision, laws and regulations, financial resources, community engagement, and urban drainage knowledge.There are some studies that discuss overcoming some barrier sources, especially related to technical aspects and governance (e.g.Petrucci et al. 2013;Sage, Berthier, and Gromaire 2015;Dhakal and Chevalier 2017;Qiao, Kristoffersson, and Randrup 2018).These works are essential to the SUSM wider adoption, because the barriers are observed worldwide and they threat the SUSM progress.In developing countries, the gap between the potential and the real SUSM implementation is even greater (Gusmão 2016), due to their specificities, e.g.scarce financial resources, limited technical and administrative capacities in the municipalities, turbulent and discontinuous economical investments and politics, including wastewater and stormwater systems, fast and unplanned urban expansion, commonly in informal settlements with a great population density and imperviousness rate, precarious urban infrastructure, climate differences when compared to the countries were SUSM concepts were developed, great population vulnerability, and complex social systems (Jha, Bloch, and Lamond 2012;Gusmão 2016).These differences can be noticed on the Sponge City Program, launched by China's federal government to advance in SUSM, which is focused on financing and water quantity control (Zevenbergen, Fu, and Pathirana 2018), and not the most common water quality control.Nevertheless, even this recent program is facing many implementation obstacles (e.g.Liang 2018;Zevenbergen, Fu, and Pathirana 2018) and no studies on how to overcome the developing countries specific barriers were already conducted.The need for these studies are also reinforced by Girma et al. (2019) e Macedo et al. (2021).
Moreover, the barriers are interrelated and, therefore, their solution strategies must also be integrated to have greater chances of being successfully adopted (Brown and Farrelly 2009).Compiling the solution strategies available to overcome the SUSM-related barriers is important both for countries that are more advanced in SUSM, as well as for countries that are beginning their transition, which is the case of the most part of the developing countries (Vasconcelos et al. 2020).The last ones are specially benefited when they are the study focus enabling them to learn from the developed countries experiences, but optimizing their efforts for their own realities.
In this context, this article aims to fill a gap of integrated and updated information on strategies for overcoming barriers and effectively implementing SUSM in developing countries.This study builds upon the barriers pointed out by Vasconcelos et al. (2020) for Brazil, as a case study.Thus, the specific objectives of the paper are: to review the literature about strategies to overcome barriers and implement SUSM; to critically analyse these strategies applicability in the developing countries context; and to evaluate the interrelationship among the strategies and their applicability to overcome the SUSM-related barriers, offering a practical source of information to other countries with similar challenges to plan and optimize their SUSM implementation programs.

Methods
The basis of the research was a literature review on the existing strategies for adopting SUSM that have the potential to collaborate with overcoming SUSM-related barriers in developing countries.Thus, they were analyzed in this specific context to evaluate their potential contribution to this complex issue.A flowchart showing the methodology is illustrated in Figure S1 (supplementary material).
The literature search was carried out in the Web of Science and Google Scholar databases in the first semester of 2019 and updated in the first semester of 2021.The following keywords were searched: sustainable stormwater management, compensatory techniques, alternative techniques, BMP, source control, stormwater control measures, IUWM, SUWM, LIUDD, LID, WSUD, GI, SUDS, sponge cities, sustainable drainage, water, and urban.Variations in word spelling were considered (e.g.stormwater, storm water, and storm-water) and also combinations between the terms, aiming to refine the results.Documents in English and in Portuguese were considered, aiming to comprise the up-to-date proposals and, simultaneously, the studies related to developing countries, especially Brazil, adopted as a case study.The documents' reference lists and those that cited them were also evaluated, aiming to identify any relevant documents that were not found in the searches.Thus, based on their titles and abstracts, their potential contribution to the strategies to overcome the SUSM-related barriers in developing countries was evaluated and a significant number of documents were pre-selected.
The pre-selected documents included articles from peerreviewed scientific journals, official documents, such as legislation and manuals, critical reviews of these documents, and academic theses.After reading their abstracts and conclusions, the documents with ideas to implement SUSM that could help to overcome the most common SUSM-related barriers in developing countries contexts were selected.It is important to notice that the goal of this literature review was finding solutions for SUSM effective adoption in developing countries, which are not necessarily the most advanced proposals in the topic, as they usually are not feasible in this context.Thus, 92 documents, presented in the list of references, were selected to compose this literature review.Documents with similar ideas identified through the literature search were discarded.For each selected document, the following items were identified: location of the study; if SUSM-related barriers were analyzed; for which barriers it presented solutions (among the 31 identified by Vasconcelos et al. (2020)); the proposed solutions; if the proposed solutions have already been applied; if the solutions have already been applied, what results have been obtained.The broader solutions were called solution strategies.The details of how to implement them, based on the ideas found in the literature review, resulted in the implementation measures.The strategies were ordered in a potential implementation sequence, based on sectorial studies (Roy et al. 2008;McManus 2009) and their adoption interdependences.It supported an analysis of the importance of each strategy to make the subsequent ones feasible to be adopted.Hence, the solution strategies and their respective implementation measures were described and discussed in the developing countries context to provide information to support adopting them where similar issues are observed.Then, each solution strategy was related to the barriers that could be overcome by adopting it, based on the information presented throughout the discussion.

Strategies for SUSM adoption and overcoming barriers
The literature review enabled us to identify eight solution strategies and 80 implementation measures, which were summarized in Table S1 (supplementary material) and discussed in the following sections.The strategies were presented in a potential order of implementation, even though this sequence is not straight linear, as some strategies success implementation are dependent on other strategies implementation, which are presented later in this order.Figure 1, built upon some guidelines to SUSM implementation (e.g.Roy et al. 2008;McManus 2009;Jha, Bloch, and Lamond 2012;Ballard et al. 2015;IWA 2016) and upon the discussions of the Sections 3.1 to 3.8, presents this interdependence among the solution strategies and can help to plan and prioritize the strategies to be adopted to overcome particular barriers.
The adoption of a strategy may be necessary or enhance the effectiveness of others, and its implementation planning must consider these relationships.Furthermore, accordingly to Brown and Farrelly (2009) the barriers are inter-dependent implying that they are likely to be less responsive to mutually exclusive programs of change, what is reinforced through the analysis of Figure 1.It is observed, for example, that the strategy '1.Information generation' collaborates directly or indirectly with all the other seven strategies and, therefore, must be included in an effective program for SUSM adoption.In general, the importance of each strategy to the implementation of the others can be estimated through the numbers under each strategy name in Figure 1's 'Influencing strategies' column.The greater these numbers, the more strategies are dependent on this one.Overall, the 'Influencing strategies' column numbers decrease from top to bottom, while the numbers of the 'Influenced strategies' increase.The greater the numbers of the influenced strategies, the more dependent of a wide planning program to their success implementation.These interdependences should be considered in the planning for the adoption of each strategy and implementation measure.
It can also be observed in Figure 1 that the arrows are mostly downward, which reinforces the coherence of the proposed sequence.Nevertheless, some arrows are upward, representing the feedback needs and the integrated planning necessity.This highlights the importance of considering wider aspects when planning SUSM policies, embracing all the urban sectors related to drainage, as stated by Bohman, Glaas, and Karlson (2020).Special attention should be given to the '6.Community engagement', '7.Financing', and '8.Inspection' strategies.The '7. Financing' comes in the seventh position of the strategies sequence because it is usually required that all the previous strategies are already minimally developed to a program to be financed.Nevertheless, to develop each of the previous strategies, some financial investments are also required, which are usually arise from governmental programs incentives.The financial issues are even more challenging in the developing countries context, due to its scarcity and the high quantity of urgent demands related to the primary needs of the population, which are commonly not properly attended.
The '6. Community engagement' is getting an increasing attention, due to the results of the monitoring of some SUSM adoption attempts (Li et al. 2020), where the role of the community to its success is being noticed.SUSM is not feasible while the civil society is a merely recipient, which believes that the stormwater management is a public responsibility (Zevenbergen, Fu, and Pathirana 2018).In this sense, Dai et al. (2018) recommended that the Chinese Government should slow down the whole Sponge Cities Program implementation and, instead, help the society and private sector to catch up and participate.Li et al. (2020) highlight the differences between the adoption of GI in UK and China and attributes the success of the UK's case to their bottom-up approach, involving more stakeholders in the project.In developing countries, the distrust in government due to corruption and inefficiency of public agencies in using resources makes the bottom-up approach even more useful to the public reliability and engagement in the SUSM projects.This focus also can help to finance the SUSM implementation and to guarantee the long-term performance of the infrastructure, through the collaboration of the community in monitoring, financing, and inspection, especially relevant where human and financial resources are scarce.Furthermore, the '8.Inspection' is also essential to the long term success of SUSM, because it has a main role in guaranteeing the proper operation and maintenance of all the implementation measures adopted.

Strategy 1: information generation
SUSM implementation depends on knowledge about the current drainage conditions and the existing SUSM structures, applied to the local context (Subramanian 2016), which is less available in developing countries due to their lack of experience in the theme.Thus, it is important to have a register of existing drainage systems (Subramanian 2016) and to monitor their behavior (Li et al. 2017).Zevenbergen, Fu, and Pathirana (2018) reinforces the importance of monitoring to the effectiveness of SUSM and present some technological approaches (e.g.crowdsourced data, early warning, real-time control systems, and smart-phone-based sensors) to enhance this implementation measure involving the community and reducing the costs, which are key issues in developing countries.
Local and reliable knowledge about SUSM is also essential to be widely adopted and lacks especially in developing countries (Macedo et al. 2021).Therefore, this research field must be fostered (Qiao, Kristoffersson, and Randrup 2018).It should focus on developing broad (national) and specific (local) guidelines for the design, construction and maintenance of SUSM structures (Dhakal and Chevalier 2017) in the local language (Brasil 2005).A good example of this type of material is 'The SuDS Manual', from the UK, which includes technical details for different devices for the local context (Ballard et al. 2015).This measure is even more important in developing countries, where SUSM experience is scarce and the context (e.g.climate, social systems, political and public administrative praxis, urban infrastructure, and financial resources availability) is much different from the countries that developed the existing SUSM knowledge.Creating a regional agency for research and education -for professional stakeholders and population -can collaborate with these objectives (Dhakal and Chevalier 2017).
Besides the guidelines, the barriers presented by Vasconcelos et al. (2020) point to the need of researching costs and benefits, long-term performance and maintenance.These data are essential to justify the SUSM adoption and are scarce for many contexts, including the developing countries.For example, it is believed that SUSM costs less in the long term than conventional drainage (US EPA 2008a; Chang et al. 2018b), but lacking costs data, difficults it's proofing.There are studies, such as those by Forgiarini (2010), Lengler (2012), Lisbôa, Barp, and Duarte (2012), and Zhang et al. (2015), which present some BMP costs data.A publication by the UK's environmental agency provides a large database of costs for this country (Keating et al. 2015), and could be followed as an example by other countries.
Computational modeling can be used to simulate the longterm performance (Li et al. 2017), but the ideal would be to collect real information, based on monitoring activities.Longterm performance and proper maintenance are directly related and, hence, should be properly addressed to guarantee the expected benefits of SUSM adoption.Nevertheless, studies in this regard are scarce and applied to a specific local.A US environmental agency document presents a study on water quality control in BMPs over time (Tetra Tech,Inc 2010).Studies from the Chesapeake Bay Foundation (2004), Lim and Lu (2016), and Mehring and Bridgeman (2019) provide some information on the necessary maintenance.
As mentioned in the last paragraphs, the existing studies about the key issues to justify the SUSM adoption come from developed countries and, hence, do not properly address the developing countries specificities.To tackle this challenge, pilot projects are very useful tools.They generate information about incipient technologies and still help to approach the community (technique and population) to these initiatives and, therefore can significantly enhance the SUSM adoption process (Van Roon, Dixon, and Van Roon 2005;Cruz, Souza, and Tucci 2007;Buehler et al. 2011).In this sense, the Chinese Government, in the Sponge Cities Program, financed pilot projects in 30 cities.They were mainly developed in green areas, what is useful for cities where these areas are available, but do not explore the SUSM implementation with retrofitting, essential for stablished and densely urbanized zones (Zevenbergen, Fu, and Pathirana 2018), a common situation in developing countries big cities.
Because SUSM is still under explored in some places and surrounded by uncertainties, public policies can be adopted to encourage experimentation and support possible failures (Farrelly and Brown 2011).As a starting point, SUSM strategies could be adopted in public buildings (Cruz, Souza, and Tucci 2007;Bitting and Kloss 2008;Chang et al. 2018b) or in local government projects (McManus 2009).Chaffin et al. (2016) propose adopting a strategy called adaptative management, which includes the adoption of monitored demonstration projects and the implementation of broader projects in stages, with incremental local learning.In the same direction, some projects in Europe developed methods for co-creating of SUSM devices' design through implementing 'Living Laboratories', which are demonstration areas in cities aiming to improve water resilience and enhance the stakeholders' participation (Qi et al. 2020).These measures would help to overcome several barriers related to SUSM adoption.Overall, it can be observed that generating local information and knowing the existing drainage system is essential for progress in SUSM and supports the proper planning.

Strategy 2: capacity building
In conjunction with the information needed to safely and effectively adopt SUSM, technical professionals (institutional and private) must be trained (Brown and Farrelly 2009;McManus 2009).SUSM should be included in higher education courses ( Van Roon, Dixon, and Van Roon 2005; Haruna et al. 2018).Furthermore, entrepreneurs, designers, constructors, maintenance teams, and drainage system managers must be properly trained (Cruz, Souza, and Tucci 2007;Heal et al. 2009).The training program could be based on the local barriers (Brown and Farrelly 2009) and be promoted through workshops (Roy et al. 2008) or courses that lead to certification (Lim and Lu 2016).
At the institutional level, capacity building should consider the organizational structure.Inter-sectorial training should be promoted (Brown and Farrelly 2009), covering technical, political, and institutional skills related to SUSM's daily life.A multidisciplinary institutional team (Van Roon, Dixon, and Van Roon 2005) with a drainage specialist in the sector responsible for implementing SUSM (Cruz, Souza, and Tucci 2007) is also important.In the case of small cities, a regional team can be structured for their support, so that they are assisted by trained professionals, despite their small technical team (Parkinson et al. 2003).Training should be planned at all levels, as it is necessary and different for the various stakeholders, such as designers, maintenance workers, and managers, who need political and institutional skills.Clemens et al. (2016) connects the professional training with the community engagement through the social learning, which is based in the interaction of stakeholders to facilitate the production and use of knowledge, working together on problems and development of solutions.Nevertheless, the authors point out that the top-down governance, typical of the developing countries, are not used to the bottom-up raising of solutions and, hence, would need a step-by-step progress and further research focused in social learning in developing countries.In this context, Zevenbergen, Fu, and Pathirana (2018) suggest the leapfrogging focused in new governance frameworks to enhance the cities capacity to deal with the urban water challenges.These frameworks should comprise the experiences exchange between the cities with pilot projects and other SUSM projects implemented.

Strategy 3: decision support tools
Technical support for decision-making is required for effective SUSM implementation.It is provided by decision support tools and could help local governments in planning and developing their SUSM projects (Subramanian 2016).It could consist of a framework with tools and guidelines to support the planning of new development (McManus 2009).Thus, an example of a decision support tool would be the sustainability indicators for urban stormwater management (Silva 2016).
The tool must also consider the local characteristics, such as its physical restrictions, current use, and the area's slope (Galderisi and Treccozzi 2017).Some drainage manuals present suitability matrices from different SUSM structures that serve as decision support tools, such as the São Paulo (São Paulo 2012) and Curitiba (SUDERHSA,CH2M Hill 2002) manuals in Brazil.Using available spaces should also be considered, such as vacant areas (Qiao, Kristoffersson, and Randrup 2018), and the adoption of SUSM in multifunctional public infrastructures, without changing their original function (Ballard et al. 2015).Combining the use of existing spaces with their skills, the US Green Infrastructure Municipal Handbook (US EPA 2008b) indicates that as streets are the main sources of diffuse pollution in urban areas and they have great potential for adopting green infrastructure, one should take advantage of this potential and adopt SUSM in these locations to mitigate impacts at the source.In addition, vegetation and the narrowing of streets result in many benefits for the population (Van Roon, Dixon, and Van Roon 2005).
The dialogue among the stakeholders, directly or indirectly involved in the urban stormwater management, is essential to implement SUSM successfully.Therefore, it must also be used as a support tool in the decision-making process.Hence, functional communication mechanisms at each level of governance and also with other stakeholders should be established (Dhakal and Chevalier 2017).The community must actively participate in choosing measures to be adopted (Van Roon, Dixon, and Van Roon 2005;Ballard et al. 2015;Loc et al. 2017), which may increase the approval and engagement.For example, Loc et al. (2017) found out that their interviewees in Vietnam would prefer public SUDS (e.g.urban green space, and pervious pavement) to household solutions.Moreover, low-income respondents preferred the rainwater harvesting, because of the additional supply of water.These preferences depend on the local culture and context, and, hence, should be locally surveyed to support the projects.Furthermore, these results reinforce the importance of multifunctional structures focused on the interests of the local people.
One way of putting this dialogue into practice, which has already been shown to be ineffective, is through participatory budgeting (Parkinson et al. 2003).Other option, presented by Trogrlić et al. (2018), is the rebuild by design competition, used in Hoboken -US to plan the reconstruction of the city after the Hurricane Sandy.Focusing the competition in SUSM adoption made it an opportunity to a fast local shift of paradigm, bringing a great change in the governance processes and creating a narrative for long-term change.It was also observed an intense public participation, research policy interface, capacity building, and improvement of long-term planning, all favourable conditions to the effective SUSM adoption.The collaboration between authorities, private sector, and philanthropists created conditions for innovations and the critical situation made feasible to raise funds to the first steps.Nevertheless, innovative financial mechanisms and involvement of private sector were needed to continue the GI implementation and bring up the necessity for continuous financing.Furthermore, rebuilding necessity after a natural disaster is not the ideal way to plan a city paradigm shift, but these kinds of events are not rare in developing countries and, therefore, could be used as an opportunity to the desired change.
The SUSM implementation planning in developing countries needs to consider their financial constraints and highly urbanized areas context.Participation can increase the active involvement of the community, including the financing of structures and the possibility of using private land to implement SUSM (Adegun 2017;Qiao, Kristoffersson, and Randrup 2018), helping to solve both issues.It would also enhance the fair division of responsibilities (Costa Junior and Barbassa 2006;Ballard et al. 2015).Also concerning the financial and space issues, Mei et al. (2018) present a framework to flood mitigation in highly urbanized areas considering the life cycle costs of the proposed GI.
This framework can be useful to justify the adoption of SUSM in developing countries in a cost-effective way.Other decisionmaking frameworks were proposed by several authors, considering different goals and contexts, some of them focused in developing countries (e.g.Ariza et al. 2019;McClymont et al. 2020).Besides the community wishes, the needs of developers must also be considered (McManus 2009).As an integrated solution for all these aspects, Qiao, Kristoffersson, and Randrup (2018) propose a form of governance called open cogovernance which is based on dialogue and active participation of different stakeholders, and would be more suitable for the effective implementation of SUSM.In this scenario, researchers would mediate communication between public and private sectors, acting as knowledge brokers, and tackling the challenge of putting the theory into practice.This measure would be very comprehensive with regard to overcoming SUSM-related barriers, therefore its documentation and sharing, when implemented, would be of great value.In short, decision support tools optimize SUSM, and community participation legitimizes choices and enhances their effectiveness.

Strategy 4: technical guidelines
Technical norms and manuals must comprise the SUSM guidelines.Each location has its specific characteristics, hence, it requires targeted guidance.The SUSM guidelines must clearly explain its objectives, and methods that should be adopted to overcome each impact (Tucci and Meller 2007;McManus 2009).For example, the objectives may be to reduce impervious surfaces in densely urbanized areas to prevent further soil imperviousness, and to restore the river ecosystem (Galderisi and Treccozzi 2017).The objectives might be in consonance with the local water issues.Therefore, aiming at these objectives, some impacts should be mitigated.To minimize the impact of increasing the pollutants load in the water, the suspended solids load must be reduced.For floods, it is recommended to maintain the pre-development peak flow (Tucci and Meller 2007).
Quality control is one of the main objectives of SUSM because urban areas contaminate the runoff and, consequently, the surface and groundwater bodies.The type and level of contamination are directly related to the land use.Hence, in North Rhine Westphalia (Germany) the level of treatment required for runoff is defined according to the land use of the contribution area (Dierkes, Lucke, and Helmreich 2015), what seems to be feasible to incorporate in the urban plans of developing countries.
However, most guidelines for quality control recommend the retention of frequent rain.This is the case of Maryland and Los Angeles, in the US, which request the retention of the first inch of rain (Chesapeake Bay Foundation 2004), and of 0.75 inch or 85% of a 24-hour rain (County of Los Angeles 2014), respectively.In the UK, the interception of the first 5 mm of rain is required, which actually comprises more than 50% of the country's precipitation events (Ballard et al. 2015).Most part of the SUSM guidelines are developed in places where the water is scarce and, hence, the main goals are to enhance and guarantee the water supply, harvest the rainwater, and avoid the contamination of the water bodies.Nevertheless, these guidelines are not appropriate to places with greater rainfall indexes, which is the case of many of the developing countries.Thus, Chang et al. (2018b) recommend that the objectives of SUSM for arid and humid climates should be different.In drier climates, the control of water quality becomes more important than that of quantity and the potential to use rainwater as an alternative and economically viable source of water resources must be emphasized, as pointed out by Gerolin, Kellagher, and Faram (2010) and Steffen et al. (2013).
As the goals of the SUSM guidelines must comply with the local water issues, in places with high precipitation volumes, the quantity control (volume and flow) should be prioritized.This is the case of, for example, Brazil, China, India, and Singapore, which suffer constantly with floods.Hence, guidelines with this focus are required.Interventions to control the water quantity must be planned at the local basin scale.The combined effect of all the planned interventions should also be verified through hydraulic-hydrological modeling, to avoid possible adverse effects, such as the combination of delayed flows (Tucci and Meller 2007;Petrucci et al. 2013).The Chesapeake Bay Foundation ( 2004) and Petrucci et al. (2013) advise that priority should be given to infiltration structures, which enable the permanent interception of water and also collaborate with the improvement of its quality (Sage, Berthier, and Gromaire 2015), what even if the water quality is not the main local SUSM goal, should not be neglected.
The strategies proposed for the runoff volume control must adequately address the established objectives.For example, Martins (2017) states that compensatory techniques are efficient for events of up to five years of return period and, for storms greater than 10 years of return period, the performance of medium and large detention basins is better.In addition, Tucci and Meller (2007) states that distributed structures are effective in areas smaller than 10 ha, and for larger areas, a specific study is necessary to determine which strategies should be adopted.Detention basins could be built in squares, as multifunctional structures, through their lowering and preparation for water infiltration (Ballard et al. 2015;Miguez, Rezende, and Veról 2015).Ballard et al. (2015) also suggests using marginal roads areas and parking lots, separated by elevated guides, to retain the water.These strategies are useful for developing countries, with their densely urbanized cities, where space availability to SUSM devices may be an issue.
Another commonly adopted strategy to control the quantity of water is its harvesting, which can be beneficial in places where the volume of precipitated water is less than the consumption of non-potable water.Gerolin, Kellagher, and Faram (2010) concluded that the relation between the average annual rainfall and the annual demand for non-potable water should be considered.If this ratio is less than 1.0, the benefits of reuse are significant.However, if this value is greater than 1.5, the effects of controlling the runoff volume are negligible.In addition, if the rainfalls occur concentrated at certain times of the year, it is more hydrologically and economically viable to adopt two reservoirs, one for controlling runoff and the other for storage to use the water (Gerolin, Kellagher, and Faram 2010).The authors also suggest the joint adoption of cisterns and green roofs in places with high rainfall indexes.
Regarding quantity control, the ideal goal should be to keep the pre-development flows and volumes (Cruz, Souza, and Tucci 2007;Chang et al. 2018b).Thus, the UK recommends, in addition to the interception of the first 5 mm of precipitation, the control of a 100-year return period and six-hour duration rainfalls at predevelopment levels (Ballard et al. 2015).In this case, the volume of extreme rainfalls can be retained and released later, at controlled flow.In the US, however, the recommendation is to control peak flows to pre-development levels from a minimum return period of 10 years (National Research Council 2008).However, this requirement applies differently according to the location.In Maryland, for example, the same volume of water must be infiltrated into the soil that it did before the occupation (Chesapeake Bay Foundation 2004).In Pennsylvania, it is established that the peak flows of high return periods must be kept equal to those of pre-development, and the volumes of twoyear return periods and 24-hour duration rainfalls must be retained and not released (i.e.infiltration, evapotranspiration, or water use), aiming at the joint control of water quality (National Research Council 2008).
The rainfall volumes for the same design criteria vary, therefore it is necessary to establish the control guidance according to the local rainfall characteristics, land use, and economic viability.In regions with higher rainfall indexes, if considered the same design criteria as regions with lower rates, control structures would have impracticable sizes in terms of economic and urban space use (Lim and Lu 2016).These criteria differences are also evident when observing that in Singapore the SUSM structures are required to support only three-month return period rainfalls.The rest of the flow must be driven to the drainage pipes, which are dimensioned for a 10-year return period (Lim and Lu 2016).In China, the Sponge Cities Program aim to control 70% of the rainwater (Zevenbergen, Fu, and Pathirana 2018), but Yang et al. (2020) point out two key challenges in stablishing proper guidelines to this goal: the determination of the volume capture ratio of annual rainfall, and the estimation of a proper rainfall threshold.The authors suggest that the balance between the investments and the potential economic benefits should be considered to design these guidelines locally.Van Roon, Dixon, and Van Roon (2005) affirm that the best way to maintain flows and volumes at pre-development levels is through urban design and the appropriate relation between impervious and vegetated areas.However, in established urban areas, it is difficult to apply it.In these cases, strategies that fit the existing land cover could be adopted.Martins (2017) suggests the combined adoption of linear and centralized devices.Palla and Gnecco (2015) point out that a reduction of at least 5% of the impervious area is necessary to obtain hydrological benefits.This could be achieved by recovering vegetation in strategic places (Miguez, Rezende, and Veról 2015), such as flood plain areas, which can be combined with linear parks and multifunctional structures.
In addition to water quantity and quality, the guidelines must comprise the maintenance procedures and routine, necessary for the proper performance of SUSM structures.Research institutions and companies could provide maintenance guidelines and technical support (Li et al. 2019).A maintenance manual for the population, with appropriate language and content, should also be developed (Chesapeake Bay Foundation 2004; Mehring and Bridgeman 2019).
SUSM regulation must also consider optimization of the structures, addressing several aspects, such as water quality, quantity, economics, ecosystem services and the population's quality of life (Ballard et al. 2015).Thus, Birgani, Yazdandoost, and Moghadam (2013) performed a multicriteria analysis considering quality, quantity and economics, and recommended the adoption of detention basins and permeable pavements.However, Galderisi and Treccozzi (2017) pointed out that GI have advantages over gray infrastructure, even though they are both SUSM structures.By this bias, bio-retention systems, green roofs, and rain gardens would be more advantageous, in addition to being multifunctional structures, which use spaces that are normally available in urban areas (Lim and Lu 2016).Depending on the devices catalog and the criteria used in the decision analysis, these results would change (e.g.Fenner 2017;McClymont et al. 2020).This integrated analysis is especially interesting in developing countries because the financial constraints require that the projects are optimal, contributing to the solution of more urban issues.
Optimization should be considered in the design phase.One example could be the use of wetlands instead of detention basins, and the division of their storage volume in two parts, to improve the water quality (Chang et al. 2018b).Other more technological option is monitoring and operating in real time, what can optimize the SUSM devices performance.Nevertheless, the real-time control is still few applied to SUSM devices and has a high implementation cost, what hinders its implementation in developing countries.Ease of maintenance (Heal et al. 2009), safety, aesthetics, and integration into the urban landscape (Ballard et al. 2015) must also be part of the design criteria.Projects with filter strips, that facilitate the sediment removal (Heal et al. 2009), and that use spontaneous vegetation (Chang et al. 2018b), for example, are more likely to receive adequate maintenance, as it is simpler.
An overview of the project context should also be considered, especially in developing countries, for which the focused guidelines are still scarce.Loperfido et al. (2014) pointed out that distributed SUSM structures reproduce pre-development conditions better than centralized ones.They also conclude that the increase in the forested area and the reduction of impervious areas are more important than the spatial distribution of structures.Restoration of rivers, as stated by the Active, Beautiful, Clean (ABC) Waters Program, in Singapore, also has a very broad positive impact (Lim and Lu 2016), despite its practical implementation challenges.In addition, combining optimization and an overview of the project, the structures' failure probability and relative impacts could be assessed, and a redundancy for the most sensitive structures -with the greatest negative impact in the event of failure -could be included in the project (Bahrami, Bozorg-Haddad, and Loáiciga 2019).In general, the standardization guides the use of SUSM through organization and technical communication and consists of a link between stakeholders.

Strategy 5: legislation
The existence of laws -including policies, plans, and programsand the appropriate regulations to the SUSM adoption is essential for its effective implementation.Many places already have laws on SUSM and, in these cases, an effective and continuous communication of them is necessary (Buehler et al. 2011).The regulation is necessary for the law's enforcement, by detailing it enough to be applicable, and it is directly dependent on the existence of technical guidelines to the local context.When existing laws are not properly regulated, their adoption is not feasible, and this is a current challenge in developing countries due to the lack of technical guidelines designed considering their specificities, as discussed in Section 3.4.The regulation must be based on local hydrological criteria (Dhakal and Chevalier 2017).In a scenario of widespread adoption of SUSM, stormwater would transit between the properties, because, analogously to the natural cycle, water does not follow political limits.To make this dynamic possible, the Los Angeles County in the US establishes that stormwater management plans should include formal agreements between neighbors, as a property commitment, which comprise receiving runoff from adjacent properties (County of Los Angeles 2014).
In addition to the regulation of existing laws, it is necessary to incorporate SUSM into relevant laws that do not yet address it, which is a reality in most places where SUSM is incipient.Therefore, support for the proposal of these laws and regulations is necessary (Roy et al. 2008), both in the political sphere, and among professionals and the community.It is also important to review the SUSM-related laws, so that eventual conflicts are corrected.SUSM adoption should be mandatory in new development and reforms, aiming to keep the predevelopment conditions, for new developments, and to fit the storm-sewer pipes capacity, in case of reforms in fully urbanized areas.In this sense, Zevenbergen, Fu, and Pathirana (2018) suggest different SUSM implementation strategies to the typical urban zones (e.g.historical city center -'accept and accommodate', established urban zone -'adapt and retrofit', and new build-up areas -'water-sensitive developments').Legislation must be flexible, to allow the adoption of simple projects, which address the SUSM requirements, and according to the interests of the stakeholders (Van Roon, Dixon, and Van Roon 2005).It is important to ensure fair liability sharing between the runoff generators and that the laws consider the combined effect of the requirements in the basin scale.Attention should also be given to local specificities (e.g.physical, climatic, and governance) (McManus 2009).An example of this is reported by Chang et al. (2018b), by pointing out that Western countries usually prioritize water quality control in their SUSM strategies, while Eastern countries give more importance to quantity control and the use of rainwater.
To foster the wide adoption of SUSM policies, its long term validity must be ensured (Dhakal and Chevalier 2017).They also need to be attached to a stable source of financial resources (National Research Council 2008), as is case of some policies adopted in Europe (European Commission 2013).An example of the importance of this measure can be observed in the Chinese Sponge City Program, where the Federal Government financed the first phase of the program, paying for the pilot projects implementation in 30 cities, but the funds to continue its adoption were expected to be raised by public-private partnerships, which are not being obtained as planned and, therefore, are hindering the continuity of the program (Liang 2018).Resources for operation and maintenance must be specially planned by the government (Li et al. 2019), since the good performance of the structures depends on them.
There are some programs, laws and regulations in place that serve as a reference.On a continental scale, there is the European Union's Water Framework (2014/101/EU) (Council on European Community 2014), which is a framework for stormwater quality management in Europe.In this continent, Germany stands out in the request and incentives for SUSM adoption (Chang et al. 2018b).This country has federal laws dealing with the quality of water infiltrated into the soil, and which require compensation for the occupation of natural areas (Dierkes, Lucke, and Helmreich 2015).Its states also have landscape plans, which require the adoption of GI (Buehler et al. 2011).
In Australia, the states are responsible for their policies and regulations -establishing SUSM guidelines -and local authorities determine project-specific provisions for development (Chang et al. 2018b).China has recently adopted the Sponge City Program, which aim to retain 70% of the stormwater volume in situ (Chang et al. 2018b). Since 2006(regulated in 2011), Singapore has the ABC Waters Program, aiming to improve urban water management and the population's quality of life through the SUSM adoption.This program requires developments from 0.2 ha to control its peak effluent flow for a 10-year return period and a 4-hour rain duration (Chang et al. 2018b) and foresees, in addition to the adoption of BMPs, restoration of rivers and the use of rainwater (Lim and Lu 2016).
In the US, several cities already have their own SUSM laws.Los Angeles and Maryland, for example, require a stormwater management plan for constructions and reforms (Chesapeake Bay Foundation 2004; County of Los Angeles 2014).In Seattle, it is required to maintain natural drainage patterns, care for floodprone areas, minimize their floods risk, protect water bodies, and install BMPs for all projects (Seattle 2016).
In Porto Alegre, Brazil, legislation requires hydrological studies for projects larger than 100 ha (Porto Alegre 2005), however its applicability is low, given the scarcity of projects with such dimensions in urban areas.Los Angeles, in the US, solve this issue by requiring stormwater control for all projects, but with different complexities: a volume of water to be retained is established for projects with more than four residences; for smaller projects, a list of BMPs is available, of which at least two must be implemented, not requiring a specific study (County of Los Angeles 2014).This simplification idea could be useful in developing countries to tackle the need for studies for each project, what would certainly hinder the SUSM adoption in these countries.
Water resources management must occur at the hydrographic basin scale.However, due to political arrangements, it is often subdivided by political boundaries, such as municipalities.Thus, the legislation might integrate the municipality and the hydrographic basin water resources management -including urban stormwater -to avoid the transfer of impacts downstream (Tucci and Meller 2007).Therefore, to tackle this issue, the US has a watershed permitting framework for managing stormwater (National Research Council 2008).Another proposal for the integration of management is by establishing drainage districts, which consist of the intersection between the limits of the municipality and those of the hydrographic basin (SUDERHSA,CH2M Hill 2002).Dhakal and Chevalier (2016) propose the concept of a two-tier governance model that includes an integrated government between the municipality and the drainage district, in addition to being more decentralized and involving stakeholders, characteristics favorable to the management of SUSM.Despite being a measure that collaborates with the overcoming of several SUSM-related barriers, it has not yet been implemented on a full scale and, hence, requires more studies to evaluate its feasibility in developing countries.To avoid the transfer of impacts downstream, hydrological modeling can also be used as an urban planning tool (Bahrami, Bozorg-Haddad, and Loáiciga 2019).This can be feasible through partnerships between researchers and the drainage sector (Farrelly and Brown 2011).
The planning and development of the cities must consider hydrological criteria (Tucci and Meller 2007), what is also a target of the Sponge City Program, in China (Chang et al. 2018a), incorporating the natural processes of the rivers and hydrological features into the urban design and providing explanations to the population (Liao, Le, and Nguyen 2016).Land use regulation must preview SUSM (Baptista et al. 2013), ensuring its adoption in new developments (SUDERHSA,CH2M Hill 2002).Thereby, the areas to be kept vegetated must be planned in order to preserve the infiltration capacity in strategic places (Van Roon, Dixon, and Van Roon 2005).Parkinson et al. (2003) recommend allocating about 2% of the public area of new developments to SUSM.In settled urbanized areas, it is only possible to propose mitigating devices (SUDERHSA,CH2M Hill 2002).
In order for integrated management to occur more effectively, it is recommended that the rainwater sector be integrated with the others operating in the urban space (Cruz, Souza, and Tucci 2007).Carneiro (2008) suggests a single institutional management body for related issues, for example a regional agency composed by a public consortium for infrastructure and sanitation management in metropolitan regionswhich could also serve to generate local information on SUSM.This would also enable regional investments (Tucci 2004).However, other configurations, such as the establishment of a stormwater management committee (Li et al. 2019), are also possible.Singapore created a water agency, the Public Utilities Board (PUB), to implement the ABC Waters Program (Lim and Lu 2016).Beyond the municipal scale, SUSM should be integrated at different government levels (Buehler et al. 2011), where each level complements the one before, but honoring the local specificities (McManus 2009).This measure also may comprise the collaboration among cities with similar challenges to tackle their issues together through a collaborative management (Li et al. 2019b).
The long term performance of the devices must be ensured to compute their hydrological contribution in a scenario of widespread adoption of SUSM.Thus, adequate planning and support for the operation and maintenance of the structures must be carried out (National Research Council 2008).A maintenance and inspection plan -which establishes the owner's responsibility to maintain the structures, and the public authority's role to inspect them -can be developed (Seattle 2016).The plan's maintenance activities must be aligned with the local government (McManus 2009).
Inspection of the devices' operation can also be carried out by the community -which must be trained to do so -and not only by the public authorities (Chesapeake Bay Foundation 2004).The local population has daily contact with the structures, and does not depend on public institutions' employees, which are commonly scarce (also pointed out as one of the SUSM-related barriers by Vasconcelos et al. (2020)).To train the population, a maintenance support program must be created, providing information, reminders and instructions (Bitting and Kloss 2008;Li et al. 2019).These measures are key to the SUSM programs success, as the effectiveness of the devices is directly dependent of their proper operation.
A way to foster community collaboration is through SUSM adoption incentives, which should be part of the legislation.They could address the local barriers to SUSM, monitoring and adjusting them, when necessary (Bitting and Kloss 2008).At an early implementation stage, incentives should be prioritized over punitive measures (Goulden et al. 2018).Thus, it is believed that the wide adoption of SUSM in Germany is due to a policy of overlapping incentives, which deal with various urban issues in an integrated manner (Buehler et al. 2011).After widespread use of incentives, the adoption of SUSM may become mandatory, because there will probably be less rejection and greater commitment from the population (Buehler et al. 2011).Other community engagement measures, which are not related to the legislation, are discussed in Section 3.6.
Economic incentives are the most common, which can be direct, such as subsidies, or indirect, such as reduced rates (Chang et al. 2018b;Li et al. 2019).In Japan, for example, there are subsidies for stormwater harvesting (Chang et al. 2018b).Rates according to the impervious area are a common and effective way of indirect economic incentive, adopted in Germany (Chang et al. 2018b).Positive incentives consist of financial compensation for the population to adopt and pay for SUSM devices (Forgiarini 2010).Hence, credits or discounts on materials and services, or taxes reductions, for example, can be provided for those who adopt mitigating measures (Cruz, Souza, and Tucci 2007;National Research Council 2008;US EPA 2009;Dierkes, Lucke, and Helmreich 2015;Dhakal and Chevalier 2017).Market-based incentives calculate the prices of the externalities related to the adopted design, so that SUSM would be encouraged (Miguez, Rezende, and Veról 2015).Nevertheless, Adegun (2017) states that marked-based mechanisms fail in providing for the needs of the disadvantaged groups (e.g.residents of informal settlements, typical of developing countries).Thus, models of economic benefits for the catchment can be created to guide collaborative investments in SUSM (Subramanian 2016), or to commercialize the controlled effluent (water volume or contaminants retained) (National Research Council 2008).Other viable incentives to be proposed in developing countries are the recognition and award programs (Bitting and Kloss 2008;Dhakal and Chevalier 2017), such as the certification of constructions (Cruz, Souza, and Tucci 2007), and the priority in obtaining licenses (Bitting and Kloss 2008).As financial resources are critical and hinder SUSM adoption in developing countries, the economic benefits and financing measures (Section 3.7) should be planned together.Therefore, legislation -combined with guidelinesrequires and guides the SUSM adoption, and the combined use of incentives enhances its effectiveness.These strategies are complex and should consider the local specificities.Successful examples should be used as inspiration, but, as presented through Section 3.5, they are scarce in developing countries.Hence, there is a knowledge gap about effective legislation and governance to SUSM adoption in developing countries.

Strategy 6: community engagement
The population is one of the main stakeholder groups who need to be engaged with SUSM for it to perform well, because the devices need to be built, maintained, and inspected.Stakeholder engagement and knowledge exchange facilitate and enhance SUSM opportunities in cities (Qi et al. 2020).As already written in Section 3, the lack of community engagement is hindering the sponge cities implementation in China.Hence, Dai et al. (2018) suggest the slowdown of the pilot projects implementation to focus on the society and private sector engagement.Thus, policies and plans for community engagement must be established (Ballard et al. 2015;Pradhananga and Davenport 2017).According to Ballard et al. (2015), there are several engagement possibilities, namely: information, consultation, involvement, collaboration and empowerment.
To facilitate participation, the community must have easy interaction with government agencies responsible for adopting SUSM, as already mentioned in Section 3.3 (Ballard et al. 2015;Dhakal and Chevalier 2017).The community's reliability on the government can enhance their collaboration to SUSM projects led by them, as reported by Adegun (2017) for South Africa and by Haruna et al. (2018) for Ghana.The government and specialized companies must provide technical support and guidance for design, construction, operation, and maintenance of the devices.In new developments, the SUSM structures and each owner's responsibilities must be clear and formalized in the contracts (Ballard et al. 2015).The government also should mediate the land acquisition to implement SUSM projects (Haruna et al. 2018).
To foster community engagement, it is necessary to understand, through targeted research, the public attitudes about stormwater management and the existing SUSM structures (Goulden et al. 2018).Overall, Wang et al. (2021) found out that the commercial and community stakeholder groups' willingness to participate is directly driven by their perceived contribution to their daily lives, while the government is motivated by the public interest.Therefore, education and awareness programs, considering the stakeholders' interests, may be proposed and come into force (Ballard et al. 2015;Maeda et al. 2018), using information campaigns (Almeida 2014), population targeted guidelines (Chesapeake Bay Foundation 2004), signs next to the existing SUSM devices, awareness and instruction activities when new structures are installed (Santos et al. 2016), teaching SUSM in schools (Dhakal and Chevalier 2017;Haruna et al. 2018), and effective communication of the policies (Buehler et al. 2011) (also useful for implementing the existing legislation -Section 3.5).The creation of a regional agency to support research and education, suggested as a way to encourage the generation of information (Section 3.1), could help implement all these activities (Dhakal and Chevalier 2017).The awareness programs should also consider the local specificities.For example, Haruna et al. (2018) propose that the lack of population's awareness could be tackled though a national campaign, especially targeting the residents of the most vulnerable cities.The authors also suggest that in Ghana this campaign could be done using the local radio stations, which reach great part of the vulnerable population, and through them the importance of the backyard gardens should be reinforced.
Depending on context and stakeholders' preferences, the strategies to raise awareness on SUSM in the population should be different, as shown by Maeda et al. (2018).The authors concluded that residents of the neighborhoods studied in the US prefer that the dissemination of educational content is done without needing to go somewhere or receive someone at home, by distributing leaflets in homes or videos on the internet, for example.Education programs must include empowerment of the community to participate in decision-making processes (Brown and Farrelly 2009), and to inspect the operation of existing SUSM structures in their neighborhood (Chesapeake Bay Foundation 2004;Ballard et al. 2015).The Learning and Action Alliance framework is a way to implement the social learning, already cited in the '2.Capacity building' strategy, which boosters the effective community engagement through the decision making process.The formal and informal relationships among the participants create a reliable and productive atmosphere, where innovative ideas to solve the sociotechnical problems of the community can be created.Thus, these ideas can progress through formal decision-making channels, bringing institutional changes and allowing the adoption of innovative solutions, which are more suitable to the community wishes (O'Donnell, Lamond, and Thorne 2018).The Learning and Action Alliance framework is also a way to put the academic research in practice, contributing to the 'Gap between theoretical and practical' barrier overcome.Nevertheless, the top-down typical governance in developing countries hinder this kind of measure successful implementation.The effective stakeholder participation in the decision making process demands time and efforts investments, but it enable that the designed infrastructure delivers multiple cobenefits, which are valued by the beneficiaries, thus fostering its success (Qi et al. 2020).
In addition to awareness and training in SUSM, its dissemination to all stakeholders is also essential for its widespread.The adoption of SUSM in public buildings (measure already presented in Section 3.1) (Cruz, Souza, and Tucci 2007;Bitting and Kloss 2008;Chang et al. 2018b) can collaborate with the dissemination.SUSM guidelines to specific groups (e.g.population, entrepreneurs, technical professionals, and institutional teams), with appropriate language and knowledge level, may also be developed.Multifunctional structures can be used to foster the propagation of SUSM and community engagement, overcoming barriers related to the lack of priority of urban drainage, and lack of space and money for SUSM structures (Cruz, Souza, and Tucci 2007;Vojinovic and Huang 2014;Subramanian 2016).Zhang et al. (2018) reinforces the relation between the community's willingness to pay for SUSM infrastructure with the perceived surrounding benefits, as improved livability or a better living environment, instead of flood safety alone.Similar results were obtained by Adegun (2017) in South Africa.Considering the multifunctional structures adoption, the needs of different sectors can be met (Buehler et al. 2011) and it is possible to contemplate the expectations of multiple uses and benefits of the population.As well as collaborating in decision making (strategy 3), community engagement and participation also increase the effectiveness of SUSM measures that depend on the collaboration of the population, such as construction, maintenance and inspection of structures.It has several correlated benefits, and therefore should receive persistent trials, especially in developing countries, where there is a big gap of information on how to effectively implement SUSM, and, hence, the community can help to create their personalized path to effective SUSM adoption, also helping to tackle many implementation barriers.

Strategy 7: financing
Financing is essential for implementing SUSM, from generating the necessary information to the hiring of trained teams and the construction of structures.Thus, the solution of financial resources issues can contribute to the overcoming of several barriers and is essential to the implementation of all the strategies.In this sense, it is necessary to establish financial sources, incentives, or assistance programs, for all activities related to the SUSM adoption, including its operation and long-term maintenance (US EPA 2008a; Li et al. 2019).
The money source depends on the organizational and financial structure in force at each location.Carneiro (2008) proposes, for example, that the funds for implementing SUSM should come from federal and state governments to regional agencies, which are responsible for infrastructure and sanitation in metropolitan regions.Baptista et al. (2013) suggest using the municipal fund and the Chesapeake Bay Foundation (2004) guides the use of the fund for maintenance and renovation of existing drainage systems.
A local government investment strategy experiment for SUSM structures was carried out in Cincinnati, US, and obtained a good result.A survey on the acceptance of residents to install cisterns and rain gardens in their homes, without construction costs for them, was conducted in the project area.They could also request some money for this installation.However, the owners would be responsible for maintaining the devices, with technical support and reminders from the responsible governmental agency.Based on the survey's answers, the best properties for installing the structures were selected, considering costs and hydrological efficiency, and they were installed by the local government.This experiment resulted in the construction of a large number of SUSM structures at a low cost (most owners did not even ask for an extra amount for the installation) (National Research Council 2008).
Another way of financing commonly used is collection fees (National Research Council 2008;Dierkes, Lucke, and Helmreich 2015), which may be linked to the use of the public drainage system (Cruz, Souza, and Tucci 2007), to the impervious area of the property (National Research Council 2008;Chang et al. 2018b), the inspection of the drainage system (National Research Council 2008) or the granting of licenses (Dhakal and Chevalier 2017).In the US, an average fee is charged for all houses in the same category, which eases the levying system, but does not encourage the owners to adopt SUSM devices (Chang et al. 2018b).In Germany, as already reported in Section 3.5, the charge is proportional to the impervious area of each property (Buehler et al. 2011;Chang et al. 2018b), which also incentives the community to adopt SUSM structures.However, it is important to note that the population is often opposed to collection fees due to corruption and the common inefficiency of public agencies in using resources properly.Thus, if this measure is to be adopted, it must be combined with an effective and transparent policy in terms of using the resources and boosting the population's confidence in the local government (Forgiarini 2010), especially in developing countries.Additionally, Haruna et al. (2018) suggest that the collected fines from environmental laws could serve as a financial resource for SUSM devices maintenance.
Public financing mechanisms and private financing are other possibilities.They can occur through loans from development banks or by companies interested in adopting SUSM to preserve the quality or good functioning of their activities, respectively.The European Commission (2013) reported a case of a mineral water company that financed the adoption of SUSM in a hydrographic basin upstream to its water source to guarantee the quality of its products.A combined publicprivate financing mechanism was proposed by China's Government to implement the Sponge City Program.China's financing strategy consisted in funding pilot projects in some cities and obtaining public-private partnerships funding for the continuity of the program.The government also encouraged financial institutions to set up credit support for the sponge city projects, through priority and long-term loans, and the sponge city projects should be prioritized when reviewing and approving the city's planning.Similar proposals were presented by Haruna et al. (2018) to Ghana's context.The authors considered national and foreign Non-Governmental Organizations and religious groups as potential private partners, which could receive governmental incentives to be attracted to invest in SUSM development in the country.Nevertheless, in practice, China is facing several difficulties in obtaining the private capital to the sponge cities, due to conflicting incentive policies, distrust of the local governments with the private investors, and the long-term commitments involved, as the program is seen as a risk investment, because it is not clear how they can obtain returns on their investments (Dai et al. 2018;Liang 2018).Thus, Dai et al. (2018) suggest that the government slow the pace in implementing the pilot projects and remedy the deficiencies observed until now, which includes the strategies '2.Capacity building', '4.Technical guidelines', '5.Legislation', '6.Community engagement', and '7.Financing'.Li et al. (2020) state that UK's SUDS funding strategy could be efficient to tackle this public-private partnerships problem, as this country applies funds from many areas, which are related to the SUSM multiple benefits.But they also point out that this financing source is only shortterm and, hence, would not be enough to maintain the programs.
Searching for other fund sources to the sponge cities, Wang, Sun, and Song (2017) evaluated the public's willingness to pay, through increasing water prices and buying government-issued credit securities.The respondents accepted a raise of 17% of the water prices and a payment of 55% of the average annual capital surplus for the sponge cities construction.Nevertheless, they also pointed out that the government grants and the publicprivate partnerships should be the main financial source, what reinforces the importance of an adequate public-private partnership model.The authors, as well as Adegun (2017) in South Africa, also observed that the income of the respondents affected their willingness to pay.Other study conducted by Wang et al. (2020) found that the population of Guyuan city in China would accept an 8.3% surcharge of domestic water tariff, less than half of the obtained in the previous cited study.Furthermore, this study showed that the perceived value is a key antecedent to the willingness to pay, what reinforces the importance of the community engagement.
Overall, the financing in developing countries is a special challenge, due to their scarce financial resources and many population's basic needs unattended.Browder et al. (2019) launched a publication focused in financing to the greygreen shift in developing countries.Despite the authors did not tackle many of the specific developing countries barriers, if the country know their own challenges, the publication can be useful to guide for some implementation measures related to financing.In short, a wise financing plan is required to SUSM implementation, as it is necessary for all strategies.

Strategy 8: inspection
To ensure the effectiveness of SUSM, the inspection of devices is necessary.It guarantees the proper construction, and the long-term operation and maintenance, according to the technical guidelines (strategy 4), and the local legislation (strategy 5), what is critical to the devices' effectiveness.The community can play an important role in the inspection, as discussed in Section 3.6.In the case of violation of SUSM strategies, when other less offensive measures were unsuccessful, some penalty must be applied, such as formal complaints, fines, or civil and criminal penalties (Chesapeake Bay Foundation 2004).The collected fines could be used as financial resource for inspection and maintenance, as also presented in Section 3.7.The inspection, combined with the incentives, presented in Section 3.5, enhances the SUSM adoption.

Applicability of the solution strategies to the barriers
Identified the local SUSM-related barriers, the next step is to know the solution strategies available to overcome them.The applicability of the eight solution strategies to overcome the barriers identified by Vasconcelos et al. (2020) is shown in Table 1, built on the information presented in Section 3. The correlation between barriers and solution strategies provides a practical source of information for managers to seek solutions to overcome their local barriers.As the barriers do not necessarily belong all to the same barrier type, it might hinder resolution efforts and request an integrated solution analysis and purposing, as already discussed in Section 3. Note that to overcome any of the barriers, it would be necessary to adopt more than one strategy.For example, to overcome the barrier 'Population would play a role in the maintenance', the adoption of the strategies 4 to 8 would be required.It is necessary to establish the maintenance activities routine guidelines (strategy 4), which shall be enforced by the laws (strategy 5).However, if the community is not engaged in playing its role in the maintenance, this will compromise the hydrological and environmental performance of the device.One way to verify this lack of community collaboration would be through inspection (strategy 8) and, eventually, with financing (strategy 7).This reinforces the need for a broad and integrated approach to advance in an effective adoption of SUSM.The implementation measures to be adopted to overcome a barrier depend on the local context and the preferences of the stakeholders.If the approach is not broad and integrated, effort and money can be wasted, as SUSM measures will not bring the expected benefits.Nevertheless, these systematized data can contribute to the optimization of SUSM adoption efforts in any place, especially in incipient ones, such as the developing countries.Reinforcing the importance of these compiled data, Chang et al. (2018a) suggests that China increases its international cooperation, aiming to learn with other countries experience and to advance more efficiently the Sponge City Program.

Conclusion
This article presents systematized and updated information on strategies to overcome barriers and widely adopt SUSM in developing countries.Eight solution strategies and 80 implementation measures were presented and their applicability was discussed for developing countries specific challenges.Therefore, a long list of alternatives is available and can be used in various contexts, according to the interests and demands of the local stakeholders.
Each strategy adopted can collaborate with the overcoming of more than one barrier and with the improvement of the urban stormwater management sustainability scenario as a whole.Furthermore, adopting isolated solution strategies would be inefficient, since they are interdependent.Thus, an effective proposal to overcome barriers and effectively implement SUSM must include several strategies.Besides the city governments, some strategies apply to other stakeholders, such as developers, property owners, and SUSM-related product suppliers, which have a co-responsibility on their implementation and maintenance.
The correlation between barriers and solution strategies provides a practical source of information for managers to seek solutions to overcome their local barriers.These systematized data also contribute to the optimization of SUSM adoption efforts in any place, but are especially important in developing countries, to minimize the costs and efforts in SUSM effective implementation planning.The successful SUSM implementation in developing countries has another important role to play: diminishing the population's vulnerability to floods and water quality issues, besides increasing their quality of live in cities lacking the basic urban and health infrastructure.
Research on barriers is recent, as it is usually identified with effectively implementing SUSM.Thus, as it advances, other barriers and solutions will emerge and it is important that they continue to be documented.This includes theoretical implementation measures, such as open co-governance and two-tier governance model.