Social justice in socio-hydrology—how we can integrate the two different perspectives

ABSTRACT

Socio-hydrology has become an important platform for discussion and exchange in current water research. A key challenge is the integration of the social into socio-hydrological logic. To date, most studies have excluded the potential outcomes of policy changes based on socio-hydrology models. The goal of this paper is to present a theoretical framework that extends the current debates in socio-hydrology around the concept of social justice. Social justice considers and reflects on the question of fairness in the allocation of resources, participation, and wealth across residents of a river basin. This critical review elucidates the role of social justice in flood risk management by considering the possible consequences of including social justice and by proposing how to integrate those consequences into the current socio-hydrology model. As such, the review contributes to the ongoing discussion of human–environment interaction in socio-hydrology.

KEYWORDS:

• water justice
• climate change adaptation
• water governance
• flood hazards
• socio-hydrology

Editor S. Archfield Guest Editor M. Haeffner

1 Introduction

The concept of socio-hydrology has become a prominent discussion platform in hydrological research (Sivapalan et al. 2012). In particular, the Panta Rhei research initiative of the International Association of Hydrological Sciences (IAHS) has pushed the debate to adopt a new view of the interaction between human and water systems (Montanari et al. 2013, McMillan et al. 2016). The main reasons for adopting this view are that human agents heavily influence the hydrological characteristics of many catchments across the world, such as by providing rural–urban water transfers; changing stream networks through the construction of dams, storage, or reservoirs; changing river basin characteristics through resettlement, deforestation, or agricultural practice; influencing global and regional climate; or using chemicals in agricultural production (Savenije et al. 2014). Consequently, the aim of socio-hydrology is to assess the co-evolutionary interactions and feedbacks between people and water resources (Sivapalan et al. 2012, Sivapalan and Blöschl 2015, Blair and Buytaert 2016, Konar et al. 2019) and to provide a conceptual framework to assess the consequences of individual and policy decisions in water resource management (Kandasamy et al. 2014, Blair and Buytaert 2016). As a result, socio-hydrological studies provide a modelling approach to integrate human behaviour into the human–environment system (O’Keeffe et al. 2018, Pouladi et al. 2019). Socio-hydrology has launched a new chapter within the science of hydrology, encouraging a view of coupled human–water interaction; this approach follows a long tradition within geography studies (Sauer 1925). In particular, the outcomes of current water governance arrangements and new policy encourage new human–water relationships (Linton and Budds 2014, Zwarteveen et al. 2017). Socio-hydrology has a strong political dimension in its results (Wesselink et al. 2017); however, it still often pays no attention to the social and economic consequences for individuals and communities resulting from water resource management decisions and strategies.

The social and political outcomes are, thus, still unknown, particularly the social and spatial inequalities that manifest themselves in the impacts of new policy in socio-hydrology models. Often, current water management strategies risk reproducing existing inequalities, such as the welfare gap between high- and low-income households or the urban–rural divide in the face of population decline and demographic aging. The main reason for this is that decisions to use new water management strategies are often based on political motivations and focus on a cost-benefit ideal, which often ignores deprived communities (Thaler and Priest 2014, Thaler et al. 2014, Siders 2019). Decisions are complex and include a wide range of actors, stakeholders, and citizens, who have different interests and needs (Zwarteveen et al. 2017). There is, therefore, a need for stronger integration of social and economic consequences into the current debates in socio-hydrology.

This review aims to open and extend the current debates in socio-hydrology research with the objective of integrating social justice into existing socio-hydrology frameworks. Social justice focuses on how to organize, distribute, and allocate water resources (Thaler et al. 2018) and plays an important role in life across the globe (Boelens et al. 2018, Sultana 2018). Social justice is “relational, situated, and context-sensitive rather than universalistic” (Roth et al. 2014, p. 949). The aim is to provide a systematic overview of the concept of social justice and to draft recommendations for future research. The article provides a critical review of the term social justice, presents different theoretical concepts and understandings, and outlines the benefits and limitations of various methodological approaches in conceptualizing the term.

Section 2 provides an overview of the current debates in socio-hydrology in comparison with other water-research frameworks, like the current hydro-social research discussion. Section 3 briefly presents the main discourses within the current debate on social justice. Section 4 focuses on the application of the different philosophical schools of social justice for flood risk management and their decision-making process. Section 5 presents the role of social justice in flood risk management and how to theoretically integrate the concept of social justice into current socio-hydrology modelling. Here, the review seeks to answer the questions of where social justice does or does not fit and how the current direction of socio-hydrology research can theoretically integrate the concept of social justice into current socio-hydrology modelling language. Section 6 ends with some recommendations for future research directions in socio-hydrology.

2 Current coupled human–water frameworks

2.1 Socio-hydrology research framework

Socio-hydrology makes the human an important aspect of water resource management. This research direction has enabled scholars to reach a better understanding of the selection of and decisions regarding certain long-term water management plans, such as how flood risk management can respond to the challenge of the “levee effect” or achieve the goal of the so-called “adaptation effect” (Burton and Kates 1964, Sivapalan et al. 2012, Di Baldassarre et al. 2015). The “levee effect” describes the unintended development in which the construction of flood alleviation schemes increases the number of exposed buildings (residential and non-residential) in hazard-prone areas and therefore the potential damage losses related to the low frequency of future flood events (White 1945, Tobin 1995, Hutton et al. 2019). The adaptation effect reflects how vulnerability due to frequent flood events is decreased, which is a central goal in flood risk management (Di Baldassarre et al. 2015).

The first studies within the debates in socio-hydrology were mainly based on a strong hydrological (natural science) perspective of the coupled human–water interaction; these studies succeeded in developing generic conceptual models (Di Baldassarre et al. 2013a, 2013b). These first models had a strong quantitative-hydrological research strategy, mainly using modelling language in system-dynamic or agent-based, coupled equations and with a minimal and very simplistic view of social aspects (Di Baldassarre et al. 2013a, 2013b, Chen et al. 2016, Fuchs et al. 2017, Haeffner et al. 2018, Pouladi et al. 2019, Kaiser et al. 2020). In particular, the input from social systems, such as political systems; the use of new technologies in water resource management; and the input on social memory, culture, or individual preparedness due to the implementation of property-level flood risk adaptation measures were mainly generically driven, often with simplistic yes/no or low/medium/high responses (Viglione et al. 2014, Fuchs et al. 2017, Melsen et al. 2018). However, this limited view of the social system was heavily criticized by social scientists and has improved in recent years (Wesselink et al. 2017, Melsen et al. 2018, Xu et al. 2018, Ridolfi et al. 2020).

At the same time, there are more studies that consider different aspects of current water issues, not only the “classical” themes, such as floods, droughts, irrigation, or water quality, but also the aim of linking the debates in socio-hydrology with international agreements and goals, such as the Sustainable Development Goals (SDGs) (Kuil et al. 2016, O’Keeffe et al. 2018, Di Baldassarre et al. 2019). Additionally, studies incorporate the use of empirical data to test and to develop their socio-hydrology models to move past the conceptual stage (Barendrecht et al. 2019). Of course, most of these studies use a strong quantitative perspective to understand the social components in the current co-evolutionary interaction of the coupled human–water system (Sanderson et al. 2017, Yu et al. 2017). Still, various papers use a mixed-method approach to study the co-evolutionary interactions in the socio-hydrology system, with the aim of overcoming the limitations and integrating the human component into the coupled human–water system (Ferdous et al. 2018). Nevertheless, a key omission is the consequences (positive and negative) of the outcomes of social and economic socio-hydrology models, such as the creation or enforcement of social inequality.

2.2 Hydro-social research framework

Hydro-social research has a strong focus on the questions of how governance arrangements influence the use of water, especially on distribution of injustice and access to water. Hydro-social research includes justice in its analytical framework, mainly on the assessment of who the vulnerable groups are in our society. Most studies reflect the distribution of and access to water, in terms of who is living in the floodplain areas, who has access (or no access) to drinking water, or who is living in polluted areas where water is contaminated due to point or diffuse pollutions (Boelens et al. 2018). The main research outcome is to show whether vulnerable groups are excluded from the use of water or whether vulnerable groups more commonly live in floodplain areas. The definition of vulnerability is often based on socio-economic data and methods, such as percentage of population (e.g. Hispanic), median household income, ownership, economic insecurity, and so forth (Chakraborty et al. 2014, Grineski et al. 2015, Collins et al. 2018).

A key question is “how, and therefore whose, decision making shapes the hydro-social system and what impacts this has on political and material inequity” (Wesselink et al. 2017, p. 6). In contrast to socio-hydrology, hydro-social research has a different perspective on human–water interaction in terms of ontology, epistemology, and axiology (Wesselink et al. 2017). Hydro-social research understands society as different subjects, in contrast to the positivist or post-positivist perspective of socio-hydrology. A crucial paradigm for hydro-social research is the constructivist view of human–water interaction. On the ontology side, hydro-social research understands the interaction between individualsand water as hybridity, which means that water and individuals (including their relationship) are not pre-given entitiesand are related internally. A distinction cannot be made between society and water as two discrete aspects. The two factors interact in a dialectical relation, where power, governance arrangements, infrastructure, identity, and water itself are the core research questions (Boelens 2014, Linton and Budds 2014, Mollinga 2014, Hommes et al. 2016, Cousins 2017).

Hydro-social research sees the socio-natural process through three key ideas: first, water shapes the organization of a community; second, water can structure as well as disrupt social relations; and third, water and society are related internally; water produces and reproduces social relations, and vice versa (Linton and Budds 2014). Therefore, hydro-social research combines strongly with the social dynamics of water to assess and to explain the social inequality of water (Mollinga 2014, Ranganathan and Balzas 2015).

3 Current discussions on social justice

The concept of justice has various possible interpretations and understandings among academics, decision makers, stakeholders, and citizens (Elster 1992, Mill [1863] 2010, Kaufmann et al. 2018, Thaler et al. 2018, Bennett et al. 2019, Thaler et al. 2020). At the bottom line, justice focuses on the distribution of wealth, access to resources, and participation within the decision making process (Thaler and Hartmann 2016). To answer on distribution and proceedings, there are multiple theoretical concepts of how to analyse, describe, and interpret social justice (Patterson et al. 2018, Bennett et al. 2019). Overall, three core topics may be distinguished within social justice. First, distributional justice focuses on the question of who pays, who is exposed, what is a person’s individual vulnerability, and the question of who is responsible for what. Distributional justice assesses, for example, different interpretations of fair resource allocation and distribution (such as neoclassical economic policy), that strongly focus on fair distribution and allocation because of their atomic view of society (methodological individualism and definition of general market equilibrium; Varian 1976). Second, procedural justice focuses on who has access to information, who has access to court, and who can engage in the planning process (an aspect of legitimacy). Procedural justice assesses, for example, who is engaged in current flood risk management as well as their interests and power to influence the current decision making process. Third, recognition justice is reflected in discrimination and in the respect or disrespect of individual backgrounds, such as social exclusion based on undocumented migrants, cultural background, language, minority, sexual identity, or gender (Fraser 1995, Honneth 2004). Fraser (1995) saw major injustices in individual recognition from the cultural domination of one or more groups, leading to minority groups being unrecognized or disrespected.

Usually, all three aspects go hand in hand. Socially unequal distribution reflects income, property ownership, power to engage, and lack of recognition. “People who are subject to both cultural injustice and economic injustice need both recognition and redistribution” (Fraser 1995, p. 74). Unequal economic distribution produces not only barriers in access to the labour market, education system, health care, and so forth but also natural injustices such as living in floodplain areas; these injustices are based on socioeconomic inequality. The rules of distribution (material and cultural) are mirrors of society and its institutions; however, conflicts over distribution can only be understood as “symbolic struggles over the legitimacy of the sociocultural dispositive” (Honneth 2001, p. 54). To achieve a fair distribution, political discussion and especially institutions (formal and informal) have to be changed.

4 The role of social justice in flood risk management

4.1 Different understanding of social justice

The concept of social justice in water research has slowly attracted increased attention within the research community and policy field. Since the 1980s, social justice has emerged from the questions of who was living in (and more negatively affected by) hazard-prone areas. Living in these unfavourable areas is characterized by more frequent exposure to flood hazard events or greater access to (or exclusion from) drinking water (Bullard 1990, Agyeman and Evans 2004, Walker 2009, 2012, Anguelovski 2016, Grineski et al. 2017, Boelens et al. 2018, Collins et al. 2018, Chakraborty et al. 2019, Emanuel and Wilkins 2020). In recent years, academic work on social justice (strongly supported by Fridays For Future activities) tends to focus on procedural, recognition, and distributional justice within an intergenerational perspective (Schlosberg et al. 2017, Patterson et al. 2018). Here, a crucial aspect is the long-term perspective of policies, such as climate mitigation policies or the nexus between demographic change and flood risk management, as floodplain management systems might influence the development of a certain region for centuries.

The literature distinguishes among a wide range of philosophical schools (e.g. utilitarian, libertarian, egalitarian, or prioritarian; see Table 1), which direct different interpretations of floodplain management systems in terms of who receives protection and who is excluded by publicly organized flood risk management, which measures are selected and implemented, who pays, who gains, the question of risk-sharing (Table 2), and so forth (Di Baldassarre et al. 2015, Doorn 2018, 2019, Kaufmann et al. 2018, Thaler et al. 2018, Bennett et al. 2019).

Table 1. Principles for distributing risk burdens

Principles	Who pays?
Polluter pays principle (PPP)	People who increase risk of flood events – especially those who manage land or build structures or, due to their behaviour, cause flooding to be worse – have to pay for flood protection, regardless of their socioeconomic background; this includes people who do not live in the floodplain area but influence the runoff.
Beneficiary pays principle (BPP)	People who gain (directly and indirectly) from flood reduction measures need to pay.
Ability to pay principle (APP)	People who can afford to should pay for flood protection.
Principle of equality (PE)	There should be intra- and intergenerational equality.

Table 2. Theoretical concepts of social justice and their meanings in flood risk management

Concepts of social justice	Short description	Floodplain management system
Utilitarian – e.g. Bentham ([1789] 1982), Mill ([1863] 2010)	Focus on protecting communities that will derive the highest benefit (based on cost-benefit logic). The main question is who gets protection and what type of protection.	Hard protection measures are based on those members of the community to whom the benefits offer the greatest gain to the community.
Libertarian – e.g. Hayek ([1944] 1991), Ostrom (1986), North (1990)	Focus on individual view of floodplain management system; individual responsibility with the only task for public administration being to provide a legal framework and information about hazards. The main question is how I (as a household) can protect myself (individual decision).	No hard protection measures are funded by the public administration; the main focus is on property-level flood risk adaptation (PLFRA) measures.
Egalitarian – e.g. Nussbaum (2000), Sen (2010)	Allocation of responsibility to meet resource needs equally distributed within floodplain management; the aim is to ensure equality of capability. The main question is how capability can be ensured.	The flood risk management system foresees an equal level of protection within the community.
Prioritarian – e.g. Rawls ([1971] 2005)	Focus on the most vulnerable (e.g. deprived) households within the selected communities. However, prioritarians accept inequalities within a community if the worst-off benefit from the new policy outcome. The main question is who the most vulnerable are within the community.	Hard protection measures focus on the most vulnerable within community

First, a utilitarian justice policy would base the decision making process on a cost-benefit analysis. Cost-benefit analysis in flood risk management is based on physical damages and often excludes the social-psychological effects of an event. The floodplain management system (with a focus on engineering solutions or a stronger focus on nature-based solutions) would be based only on a cost-benefit analysis. If the benefit-costs ratio of engineering solutions were higher in comparison with nature-based solutions, the floodplain management system would remain technology oriented. Often, nature-based solutions have a lower efficiency in terms of risk reduction in comparison with structural flood alleviation schemes. The responsibility to engage individuals raises serious questions about social justice and equity. As individual adaptive behaviour relies on an active, self-efficacious, financially robust household, there is a clear challenge for the large number of households that cannot afford (financially, personally, or legally) to engage in adaptive strategies. Implementing resilience strategies often causes unintended effects, such as social exclusion of certain household groups, as some households are excluded from the planning and implementation process through a lack of empowerment (e.g. resources or knowledge). The utilitarian justice policy would provide and increase technical protection if cost-benefit analysis were to show that this provides the greatest benefit for the community. Communities in low-income areas would be provided with no protection or with non-structural strategies, such as planned relocation or property-level flood risk adaptation measures.

The second type of policy—a libertarian justice policy—would envisage that individuals define what kind of floodplain management system they want; it relies on individual interests as well as their capacity to engage in the decision making process; consequently, the selection process for this kind of floodplain management is mainly based on the socio-economic background of individuals. The libertarian justice policy would select the type of management system based on the community’s socioeconomic background, as each individual has to organize, manage and finance the new system; high-income communities might select technological or nature-based solutions because they can afford them. Communities in deprived areas do not have this option as they cannot afford the costs. Deprived communities would select low-cost solutions (if they could afford them), not because of their interests but just because their options are restricted to politically inactive flood risk management policies (Mgquba and Vogel 2004, Thaler and Levin-Keitel 2016).

The third type of policy (an egalitarian justice policy outcome) would distribute the risk reduction measures or risk burdens (see also the principle of equality in Table 2) equally within the whole catchment, independently of the socio-economic background of individuals. The policy would increase the current structural protection schemes across the whole catchment (Ciullo et al. 2020). Consequently, high-risk areas get lower risk-reduction measures in comparison to low-risk areas, as everyone receives the same level of risk management strategies. This kind of floodplain management system is mainly selected on the basis of how to support deprived communities in terms of reducing risks, which would include their expression of interest. The selection of a floodplain management system (techno- or green-society) can be based on a political or stakeholder decision or the use of cost-benefit analysis. The main element is the equal distribution of the same floodplain management system. This type of policy would support communities who mainly lose out because of utilitarian or libertarian thinking (Thaler et al. 2018).

The fourth type of policy, a prioritarian justice policy, focuses its actions on the most vulnerable individuals within the community. The prioritarian justice policy allows the creation of inequalities, if the worst-off benefit from a new policy. The implementation of nature-based solutions, for example, often requires the creation of inequalities within floodplains. Examples of nature-based solutions are the use of natural water retention in the upstream part of the catchment, change of current land-use practice, wetland restoration, river renaturation, or development of flood storage areas (Dadson et al. 2017). Nature-based solutions are often implemented on private land, the owners of which do not gain from the measures. In such a case, the policy needs to compensate the private landowner. The level of compensation should depend upon the acceptance of private landowners who provide the land. Nevertheless, prioritarian policies would include those techno- or green-society floodplain management systems that mainly reduce the risk for low-income communities (Thaler et al. 2020). The justification for creating inequalities should, therefore, be based on the social status of the protected community. Including this information, social justice would highly influence the outcome and recommendations of policy within the socio-hydrology models.

4.2 Application of social justice framework to flood risk management

Social justice is an important aspect of current flood risk management policy but is often neglected by policymakers and scientific communities (Alexander et al. 2018, Collins et al. 2018, Kaufmann et al. 2018, Thaler et al. 2018). In particular, in the last decade, public administration has sought to reduce the public costs of flood risk management, and this decision has strong implications for social justice. A key goal is to promote greater engagement of private individuals in flood risk management to share risk burdens between government and non-government stakeholders (Thaler and Priest 2014, Begg et al. 2018). Social justice has become a crucial part of the debate surrounding the design of flood risk management options at local, regional, and national levels. Nevertheless, decisions on flood risk management are usually based on cost-benefit analysis with the aim of increasing social welfare instead of considering risk distributions (Ciullo et al. 2020). In terms of using cost-benefit analysis to select an “optimal” management system, the question arises of which protection scheme should be implemented; however, the literature shows various limitations in the use of cost-benefit analysis in the decision making process. The first problem that arises is the aspect of monetization of all benefits, including intangible aspects like human health which is hard to monetize (Kind 2014, Emanuel and Wilkins 2020). The second problem includes the understanding of risk, which is often defined as expected damage per year (Ciullo et al. 2020). Here, cost-benefit analysis neglects the risk aversion of individuals. Usually, individuals show stronger concern about rare (and usually more catastrophic) events than about more frequent but less severe events (Merz et al. 2009, Ciullo et al. 2020). The third problem reflects the assumption that inequalities of individuals or communities can be compensated for by others (Hansson 2007).

Based on a cost-benefit analysis, for example, planned relocation as part of floodplain management as a result of a negative cost-benefit ratio more commonly affects low-income families than high-income families in the United States (Siders 2019). The planned relocation of households is highly effective at reducing exposure; however, this strategy is often seen as the last option within flood risk management policy. It is the most controversial among current flood and coastal policies because although it is an efficient way to reduce vulnerability, it is painful and cost-intensive (Thaler 2017, 2021, Mortreux et al. 2018, Siders 2019). Planned relocation has been largely abandoned by public administration (Perry and Lindell 1997, De Sherbinin et al. 2011, Mortreux et al. 2018, Ajibade 2019, Thaler and Fuchs 2020). A key challenge is how to facilitate planned relocation to avoid additional negative impacts on households and communities given that relocation is highly contested—for example, legally, socially, and economically. People affected by planned relocation are confronted with profound changes in their lives, such as overcoming emotional attachment to the home left behind and adapting to and integrating into a new residence and community, as well as coping with financial burdens and rebuilding a new social (neighbourly) network. Planned relocation has strong social justice implications, such as the cost-benefit logic, which often leads to planned relocation strategies that disproportionately affect low-income communities (Thaler et al. 2018, Siders 2019). Consequently, planned relocation can increase the social and physical vulnerability of low-income individuals and communities (Nalau and Handmer 2018, McGhee et al. 2020). This type of policy is not the best approach to reducing inequalities within a country. Of course, a key focus is on two main aspects: (a) how to ensure that risk burdens are shared fairly in terms of distributing the costs among public authorities (national, regional, and local), private companies (e.g. insurance companies), communities, and individuals; and (b) how to share responsibility and especially encourage the responsibility of lower government agencies and non-government agents (Schuppert 2011, Thaler and Levin-Keitel 2016, Begg 2018).

5 Potential role of social justice in socio-hydrology research

Socio-hydrology originally omitted the problem of environmental justice in its assessment and analytical framework, in contrast to hydro-social research. Therefore, the question is how the theoretical concepts and political consequences of social justice can be incorporated into current socio-hydrology models. Using the understanding proposed by Di Baldassarre et al. (2015), the aim is to assess the different trajectories of flood management systems, following different social justice policy (see also Table 1). The current socio-hydrology model by Di Baldassarre et al. (2015) aims to simulate the co-evolutionary interactions and stimulations between hydro-meteorological events and societal responses to them. The model is based on four key domains: (1) hydrology (relative damage), (2) demography (population density), (3) technology (flood protection level), and (4) society (societal memory of past flood events). Furthermore, the model predicts that flood alleviation schemes will attract human activities behind the dams (Di Baldassarre et al. 2018). Therefore, the damage (F) is explained according to the actual water level, depending on the water level surpassing the technical mitigation measure; the level of damage (F) ranges between 0 and 1 and is estimated using the following equation (Di Baldassarre et al. 2015, p. 4474):

(1) $F=1-exp(-\frac{W+{\xi }_{H}H}{{\alpha }_H})ifW+{\xi }_{H}H\mathrm{\_}>H\mathrm{\_}$(1)

where W is defined as the frequency and magnitude of flood events, H is the flood protection level, ${\xi }_H$ is flood level enhancement due to the actual existing protection system, ${\alpha }_H$ is a parameter related to the stage-damage curve, and H_ includes the level of the protection scheme before the event.

Social justice has an impact on Equation (1)(1) $F=1-exp(-\frac{W+{\xi }_{H}H}{{\alpha }_H})ifW+{\xi }_{H}H\mathrm{\_}>H\mathrm{\_}$(1) . The level of physical damage and the possibility of response to a flood hazard event, for example by implementing property-level flood risk adaptation measures, depends to a great extent on the households’ socio-economic conditions, which are not equally distributed within a catchment. In many cases across the globe, low-income families show on average lower physical damage in comparison to wealthy households per flood event but are more likely to be affected by flood hazard events as, usually, these groups are more likely to live in the floodplain area. Generally, because of this, low-income householders suffer higher losses from flood hazards than wealthy households (Emrich et al. 2020). Reflecting Table 1, if the government implements a policy based on the “polluter pays” principle or “beneficiary pays” principle in terms of distributing risk burdens, low-income communities would have a lower flood protection level (H) compared to wealthy communities, because of their lack of financial resources. A different picture can be observed in the case of the implementation of the “ability to pay” principle, where communities that have the financial capacity must provide greater financial support as they can afford to do so. The government can use the financial resources from wealthy communities to fund the flood protection scheme for low-income communities.

Di Baldassarre et al. (2015) distinguish between two key prototypes of floodplain management systems: (i) techno-society, in which policy relies mainly on structural protection schemes and is assumed by the authors and (ii) green society, in which the focus is on non-structural measures and is assumed by the authors. The selection of the prototype of floodplain management is also defined by social justice policy and its implications. The selection between techno- or green-society can be based on a cost-benefit analysis in the case of a utilitarian justice policy. If the benefit-cost ratio of techno-society is higher than that of green-society floodplain management, the management selection would remain technology oriented. In practice, socio-hydrology models need to include a cost-benefit analysis within their modelling work to select a techno- or green-society approach in floodplain management. A libertarian justice policy would envisage that the protected community defines which strategies it prefers to implement, as the theory requires no state intervention except providing hazard information. In practice, socio-hydrology models would need to include a stated-preference survey from the community to select the policy needed for implementation in the floodplain. An egalitarian justice policy would entail that the selection of measures would ensure the same level of protection within the country, as within an egalitarian justice policy, the principle for selection would be the level of protection needed to be reached independently within a techno- or green-society policy of the floodplain. Heightening the flood protection scheme in one community would require that the level of protection be increased within the whole catchment. Here, a key variable for the socio-hydrology models would be the availability of the public administration’s financial resources to provide the level of protection, as heightening the level of protection would call for high financial investments. The prioritarian justice policy would envisage that the protection would address the most vulnerable groups, which can be defined through indicators that address the social vulnerability of households such as income, building value, ownership, and so forth. In practice, the socio-hydrology models would need to include an assessment of social vulnerability within their policy recommendation to select the most appropriate floodplain management policy.

Along the same lines, the level of protection (R) is defined as the relationship between the flood protection level (such as 1:100) and actual water height (Di Baldassarre et al. 2015, p. 4474).

(2) $R=\left\{\begin{array}{c}{\epsilon }_T\left(W-{\xi }_{H}H\mathrm{\_}-H\mathrm{\_}\right)withlevees\\ 0nolevees\end{array}\right.$(2)

where ${\epsilon }_T$ is the safety factor for heightening the flood protection scheme; the definition of protecting a community or heightening of the protection scheme depends on social justice policy. Within a utilitarian justice policy, R would mainly be defined by the cost-benefit analysis. Communities within a catchment would be provided a flood alleviation scheme or heightening of the current scheme if the cost-benefit ratio shows the highest outcome, which is often based on physical vulnerability (Thaler et al. 2018). In the case of a prioritarian justice policy, the decision about whom to protect or the heightening of the flood alleviation schemes should be made with regard to the interests of the most vulnerable groups within the catchment. These decisions should ideally be made with the groups concerned. The assessment of the question who is vulnerable is often based on the assessment of socio-economic status or, the value of exposed buildings in comparison to the average value of the total buildings within the community (Collins et al. 2018, Thaler et al. 2018). Egalitarian justice policy would foresee an equal distribution of R-value between the catchments, which would predict fewer risk-reduction measures in high-risk areas in comparison to low-risk areas. Finally, in a libertarian justice policy, the R-value would be defined such that each household and business has to organize its own individual flood risk management strategy (Johnson et al. 2007).

In addition, the level of protection can also be defined by the economic principle of risk sharing (see Table 2). For example, in the case of the “beneficiary pays” principle, deprived communities cannot usually afford a higher level of protection as they have to provide the additional financial resources to ensure the higher protection. In contrast, in the case of the “ability to pay” principle, the financial contribution can be distributed within the community based on their household or business income. With this strategy, the intention is that households or businesses with a higher income need to provide a higher financial contribution in comparison to low-income householders. If the “polluter pays” principle is used, the distribution of the financial contribution would be mainly managed by householders and businesses who are responsible for increasing the risk of flood events and, therefore, the need to heighten the protection scheme. This includes, on one hand, the number of exposed householders and businesses in the floodplain and, on the other, people who live upstream or outside of the floodplain area who have an influence on the runoff, such as farmers, forest-owners, and so forth.

Finally, the co-evolutionary interaction is defined as Equation (1)(1) $F=1-exp(-\frac{W+{\xi }_{H}H}{{\alpha }_H})ifW+{\xi }_{H}H\mathrm{\_}>H\mathrm{\_}$(1) coupled with the three equations that model the interplay of the human–water system (Di Baldassarre et al. 2015, p. 4475):

(3) $\left\{\begin{array}{c}\frac{dD}{dt}={\rho }_D\left[1-D\left(1+{\alpha }_{D}M\right)\right]-\mathrm{\Delta }\left(\psi \left(t\right)\right)\ast FD\mathrm{\_}\\ \frac{dH}{dt}=\mathrm{\Delta }\left(\psi \left(t\right)\right)\ast R-K_{T}H\\ \frac{dM}{dt}=\mathrm{\Delta }\left(\psi \left(t\right)\right)\ast F{D}_{-}{\mu }_{S}M\end{array}\right.$(3)

where ${\rho }_D$ is expressed as population growth, and ${\alpha }_D$ as the ratio between preparedness and awareness; $K_T$ is defined as the protection-level decay rate; and ${\mu }_S$ includes social learning and memory losses over time. In particular, the factor of ${\alpha }_D$ highly depends on the socio-economic status of householders (e.g. low-income householders usually show a lower level of preparedness due to a lack of resources; see also Fothergill and Peek (2004), Peacock et al. (2014), Muñoz and Tate (2016) or Emrich et al. (2020)). The selection of the social justice policy (Table 2) defines whether the public administration supports individuals or not. Within a utilitarian justice policy, the public administration supports individual householders according to the cost-benefit analysis. If the benefit-costs ratio of support were to show a higher ratio in comparison with no support, the public administration would act. In the case of a prioritarian justice policy, the decision about whom to support would be made with regard to the needs and interests of the most vulnerable groups within the catchment. Egalitarian justice policy would predict an equal distribution of support by public administration. Finally, a libertarian justice policy would be defined such that each household and businesses has to prepare themselves.

6 Discussion and conclusion

Today, socio-hydrology includes a wide range of disciplines with different methodological challenges, theoretical concepts, conceptual frameworks, or links to international debates, such as the SDGs or the Sendai framework (Ferdous et al. 2018, Di Baldassarre et al. 2019, Konar et al. 2019). The key aim of socio-hydrology is to understand the link between human and water systems (Sivapalan et al. 2012, Montanari et al. 2013, Yu et al. 2017); however, most studies exclude the debate about the role of social justice. This raises the key question of how to overcome the different directions of epistemological research to enhance the current direction of socio-hydrology research. In particular, socio-hydrology has an explicit political goal and expression, but excludes the aspect of the consequences their outcome might bring to individual communities, and above all the question of who gains and who loses with new justice policy. Current water management policies (such as flood risk management, water quality, water access and scarcity, or drought risk management) can reproduce, reinforce, and deepen existing inequalities through their policy and implementation (Nygren 2018, Sayers et al. 2018, Thaler et al. 2018, Siders 2019); for example, the use of property-level risk adaptation measures, water privatization, water extraction for irrigation, and contamination of rivers caused by mining projects divide high- and low-income households and communities (Stoltenborg and Boelens 2016, Nygren 2018, Attems et al. 2020). The challenge, therefore, is that

an unequal distribution of environmental goods or bads by itself may not necessarily be unjust – it is rather the ‘fairness’ of the processes through which the distribution has occurred and the possibilities which individuals and communities have to avoid or ameliorate risk, or to access environmental resources, which are important. (Walker and Bulkeley 2006, p. 656)

The next step in the current debates in socio-hydrology and modelling work is to include the themes of social justice within the analytical framework, to also stimulate a debate on the consequences of choosing new water management policies from the model. Social justice especially influences various input parameters within the socio-hydrology model, such as the selection of floodplain management systems (techno- or green-society) or the height of flood alleviation schemes (R). The selection of floodplain management systems and the variable R can be changed by the impact factors listed in Tables 1 and 2, such as the selection of floodplain management systems in Equation (2)(2) $R=\left\{\begin{array}{c}{\epsilon }_T\left(W-{\xi }_{H}H\mathrm{\_}-H\mathrm{\_}\right)withlevees\\ 0nolevees\end{array}\right.$(2) , the level of flood alleviation schemes (R) and the human response to flood events in Equation (3)(3) $\left\{\begin{array}{c}\frac{dD}{dt}={\rho }_D\left[1-D\left(1+{\alpha }_{D}M\right)\right]-\mathrm{\Delta }\left(\psi \left(t\right)\right)\ast FD\mathrm{\_}\\ \frac{dH}{dt}=\mathrm{\Delta }\left(\psi \left(t\right)\right)\ast R-K_{T}H\\ \frac{dM}{dt}=\mathrm{\Delta }\left(\psi \left(t\right)\right)\ast F{D}_{-}{\mu }_{S}M\end{array}\right.$(3) , such as choosing between flood alleviation schemes, the implementation of nature-based solutions, or using a planned relocation policy.

Socio-hydrology should include the different perspectives of social justice (distributional, procedural, and recognition justice) in future modelling work. Social justice can support the analysis of the potential social impacts of the results provided by socio-hydrology models. Here, there is a need for a stronger integration of social science and hydrology to produce better knowledge and understanding of the co-evolutionary interaction of certain water management policies and the increasing exposure and vulnerability within communities across the globe. There exist various academic techniques to overcome epistemological differences, which can also be seen recently in socio-hydrology research with the inclusion of hydro-social research methods and questions in the creation of socio-hydrological spaces (Ferdous et al. 2018). The integration of hydro-social and socio-hydrology research offers an important next question to be answered in the matter of justice: regarding the uneven distribution, procedural, and recognition justice of water (Rusca and Di Baldassarre 2019, Ross and Chang 2020).

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

This research received financial support from the Austrian Climate and Energy Fund and was carried out within the Austrian Climate Research Program [Funding No. B769942].