1. Introduction

Today, nature-based solutions (NbS) are highly valued by a range of institutions seeking to find solutions to increasingly significant global changes, particularly those linked to the biodiversity crisis. According to the IUCN [2, 3], lNbS are "actions to protect, sustainably manage and restore natural or modified ecosystems that address societal challenges effectively and adaptively, simultaneously providing human well-being and biodiversity benefits". This definition is also in line with the one approved by the United Nations [4]. NbS incorporate or are often used as a label for approaches such as green infrastructures [5], or ecological engineering [6]. Some types of NbS that are specific to drainage basins are also referred to as natural flood management or natural water retention measures (see, for example, [7]). These NbS have also been widely used to implement climate change adaptation strategies [8], although several other areas are involved [6]. Thus, there has been an increase in the institutional use of NbS, which may partly be a rhetorical response to the changing priorities and terminology of research and practice funders. These funders' modes of action are closely linked to social and economic concerns, and they will therefore justify interventions that are often essentially techno-centred on nature [9]. NbS emerged in the 2000s alongside other, older proposals for action, such as ecological engineering and ecosystem-based adaptation. In this article, we first provide a historical overview of the genesis of ecological engineering and then of NbS. Then, we searched the abundant scientific literature discussing the merits of these modes of action for over two decades for evidence of the effectiveness of the presented experiments or work in terms of biodiversity rehabilitation for its own sake. In the articles analysed, we observed a lack of systematic interest in this question. We conclude with a discussion of the notion of socio-ecological restoration, which we argue is potentially more effective at taking biodiversity into account for its own sake. To support this discussion, we mention the emblematic French example of the rehabilitation of the Bièvre river and propose some recommendations for moving in this direction. Finally, we consider how these SfNs could ultimately become more ecocentric.

2. The first steps towards NbS: emergence and deployment of ecological engineering

The concept of ecological engineering as a mode of intervention originated in China in the 1960s. It was initially defined as the philosophy of “designing with nature”, particularly in relation to wastewater treatment [10]. For Shijun, it was “a specially designed production system in which the principles of species symbiosis and the cycling and regeneration of substances in an ecological system are applied by adopting systems engineering technology”.

This philosophical vision of “designing with nature” was to strongly influence the scientific thinking behind what would become ecological engineering in its American incarnation, developed by H. T. Odum in the 1960s. H. T. Odum's definition of ecological engineering was steeped in thermodynamics and drew on engineering sciences within a classical engineering framework. For Odum, the ecological “object” was clearly identified as the ecosystem, the cardinal unit in the grand scheme of living organisms (which are organised into organisms, populations and communities) [11]. In 1962, Odum proposed the following initial definition: “Human manipulation of the environment using small amounts of additional energy to control systems in which the main sources of energy still come from natural sources” [12]. The ecosystem would function like a machine, with each component species corresponding to a tool in a toolbox. The idea of replacing a conventional wastewater treatment plant with a forest to treat wastewater in the interests of water quality is the very foundation of ecological engineering: forest soils are rich in biodiversity and filter water, while the roots of the trees form an extensive network of drainage cavities.

Ecological engineering flourished in the United States until the 2010s. But what was happening in France? In the 1980s, ecotechnology was the favoured practice of those working in the field [14]. Ecotechnology encompasses all activities that produce goods and services aimed at measuring, preventing, limiting or correcting environmental damage, as well as problems related to waste, noise and ecosystems. At the time, practices involved the use of conventional technologies (such as filters and sensors) for effective ecological management of ecosystems. The concept of “designing with nature” was not yet a reality on the ground.

It was not until the 2000s, when two sociologists, L. Charles and B. Kalaora, launched a seminar around the idea of the emergence of “ingénierie écologique” in France, inspired by American approaches, that the proposal for a new mode of intervention slowly began to take shape (the term is a literal translation of “ecological engineering”). Two definitions were proposed by a group of researchers and associate professors between 2005 and 2007 in the wake of the Grenelle Environment Forum. This group was funded by the CNRS via a dedicated programme called INGECO [15]. INGECO's initial definition of ecological engineering was put forward in 2007. This definition was similar to H. T. Odum's. French ecological engineering was described as “the in situ manipulation of ecological systems (a few individuals, populations, communities and ecosystems) in an explicit ecosystem context (including other organisms and physical and chemical dimensions)”. Some members of the group proposed broadening this definition to create a second, broader and more ambitious definition:

In practice, the “ingénierie écologique” will not be very effective, caught between traditional engineering practices and an ecology that will, in fact, be rarely used [16]. Indeed, the combination of knowledge of the engineering sciences on the one hand, taught in dedicated French schools, and ecology on the other, taught mainly at university, will not really allow this “designing with nature” to become operational. Those involved will prefer the term “génie écologique”, since the term ‘génie’ evokes the idea of “doing” rather than the “know-how” that engineering work would denote. This new definition of ecological engineering as “a set of study and monitoring activities, project management and works that promote the resilience of ecosystems and are based on the principles of ecological engineering” [17] does not address a change in practices, at least in its definition. At the same time, research laboratories, mainly university laboratories, have done very little to develop applied research in ecological engineering [17]. In the same study, the author points out that “the influence of the French education system needs to be put into perspective; the disparity of its system and structures (engineering schools, technical schools, universities) may explain a certain confusion surrounding the definition of ecological engineering in particular” and therefore its appropriation. In the end, preference will be given to the development of “ecological restoration”, which has recently emerged from the currents of conservation biology, offering more concrete and/or tangible scientific perspectives that can be published, and a more obvious potential for implementation from the laboratory to the field. Ecological restoration is “an intentional action that initiates or accelerates the self-repair of an ecosystem that has been degraded, damaged or destroyed, while respecting its health, integrity and sustainable management” [18]. According to [14], ecological restoration, while pursuing its objectives of restoring degraded ecosystems, sets itself the priority of conserving biodiversity and favours soft techniques to reconstitute diversified and autonomous ecosystems. The idea is to achieve the reference state, i.e. the state of the ecosystem before it was degraded. This reference state is an approximation of the desirable state, a standard chosen from a number of possible alternative states that can be reached through a succession of stages called a trajectory [19].

Over 60 years have passed since the first Chinese applications of operational ecological engineering, and many field observations have been made. Has ecological engineering, or rather its operational version, ecological restoration, changed the practices of those involved? Influencing, balancing and controlling the complex abiotic and biotic interactions that govern ecological processes in order to restore one or more ecosystems remains technically difficult and costly, particularly when it has to be done on a large scale. In many cases, even restoration techniques based on scientific proposals from ecology laboratories have not been as successful as they should have been. This is often due to the fact that the area in which the work is carried out is not permanent and the projects are subject to constraints regarding completion and effectiveness that are too short to allow ecosystems to recover their own dynamics [20, 21]. In particular, after examining several meta-analyses in the literature, [22] note that: (1) of over 600 wetland restoration projects worldwide, animal and plant assemblages and biogeochemical functions recovered to only 74% of the reference level after 50–100 years; (2) of 89 lake and coastal ecosystem restoration projects, biodiversity and biogeochemical functions recovered to only 24% and 34% respectively after 16 and 12 years; and (3) in 166 forestry studies, the abundance of plants and animals recovered within a few decades, but diversity and biogeochemical functions were not clearly affected by active restoration. They conclude that “traditional approaches are oversimplified abstractions for achieving sustainable ecosystem restoration”.

How can people be convinced to incorporate more ecological processes into their practices when social and economic constraints are based on short-term performance? Planners still favour technology and control, which fully legitimises the ecotechnology presented above. “What is there to love and preserve in a chaotic universe?” [23]. In the end, the best practice is the one we know best: “We are still in the business of managing the house of biodiversity, conceived as a planetary garden” [24].

3. Genesis and trajectory of nature-based solutions

In the wake of ecological engineering as defined by Odum, the first references to what are now known as ‘nature-based solutions’ (NbS) appeared in the 1990s under the term “ecosystem-based approach” [25].These are essentially human interventions in the management of agricultural or water resources (for example, pest control or creating new habitats to reduce agricultural run-off). These actions are implemented on an ad hoc basis and are then left to run their course, based on the assumption that nature will regulate itself to achieve the initial objective.

The term NbS did not begin to appear until the early 2000s [26] when it was used to promote the use of these solutions in climate risk management and food security. At that time, nature was viewed as a resource whose flows, represented by ecosystem services, needed to be optimised for one or more predefined purposes. It was also argued that biodiversity would benefit from the use of natural elements. The term NbS is thus used as an umbrella term encompassing a number of older concepts relating to protection, restoration, management and infrastructures, such as ecological engineering, ecosystem-based adaptation, blue and green infrastructures, agroecology, etc. [27, 28].

The term NbS was then widely promoted by the IUCN (International Union for Conservation of Nature) in numerous documents (see for example [29]) dealing with land use and water resource management/protection issues. Its use was then extended more widely to issues of adaptation to climate change [2, 3, 30]. However, it was only used implicitly at the close of COP21, in the 2015 Paris Agreement, which aimed to limit global warming to 2°C above pre-industrial levels. Despite this, there are several references to the role of nature and ecosystems in adaptation measures. The role of NbS as levers for action against the effects of climate change was formally affirmed by the signing of a manifesto at the UN Climate Action Summit in New York in 2019 [4].

Since then, the term SfN has appeared in most documents dealing with adaptation to climate change and the preservation/restoration of biodiversity [31]. This is particularly true of the IPCC's 6th Assessment Report [32] which highlights solutions that the organisation considers — with a high degree of confidence — to be effective in reducing the risks induced by climate change. In particular, it highlights agroforestry, the renaturation of urban environments and the protection of wetlands. As with criticisms of the detrimental effects of adaptation measures that are prioritised over mitigation [33], les SfN peuvent aussi apparaitre comme des mesures de compensation en regard de la poursuite de la consommation d’énergies fossiles.

As these definitions have evolved, so has the definition of NbS. In hindsight, the IUCN definition mentioned in the Introduction [2, 3], seems to be the most generic. It refers to actions aimed at ‘protecting, sustainably managing and restoring natural or modified ecosystems to respond directly to societal challenges in an effective and adaptive manner, while ensuring human well-being and producing benefits for biodiversity’. The definition proposed by the European Commission [34] includes solutions ‘inspired’ by nature, and mentions the need to be ‘cost-effective’, while the more recent definition proposed by the United Nations Environment Assembly [35] specifies the different ecosystems (terrestrial, freshwater, coastal and marine, natural or modified) in which these solutions can be implemented.

In the national context, the term 'SfN' first appeared in French in scientific literature in 2016 [36] (it has been present in international scientific literature since the 2000s). Older terminology, such as ecological engineering, ecological restoration and ecosystem-based adaptation, has been in use since the early 1990s [37, 38, 39]. From an institutional point of view, this terminology first appeared in the Second National Plan for Adaptation to Climate Change [40]: “The use of nature-based solutions to improve the resilience of territories and protect the environment, for example through the greening of urban spaces, alternative sanitation techniques, and the integration of green and blue networks”. As is the case on an international scale, the growing popularity of NbS is due to its role in providing adaptation solutions. Beyond the terminology, the main objective of the European ARTISAN project, coordinated by the OFB (French Biodiversity Office), is to promote this concept (https://www.ofb.gouv.fr/le-projet-life-integre-artisan). The project temporarily introduced the term 'nature-based adaptation solution' to highlight the use of these solutions in addressing the societal challenge of adapting to climate change. On a national scale, the exploratory SOLU-BIOD PEPR (priority (national) research programmes and equipment), coordinated by CNRS and INRAE and launched in 2024, aims to promote the development of NbS as transformative responses to global change (https://www.pepr-solubiod.fr). Using a multidisciplinary approach combining ecology, biophysics, sociology and economics, the project will facilitate the design, implementation and evaluation of NbS in a variety of ecosystems, including protected areas, agricultural and natural mosaics, and urban and coastal areas. The project also aims to unify and organise the French scientific community in this field.

This new popularity (at least in certain circles) of the term NbS has enabled it to find a place in a number of methodological reports, such as those of ONERC (National Observatory on the Effects of Global Warming) [41], in regulatory planning documents such as the third National Biodiversity Strategy (SNB3), the PCAET (Territorial Climate-Air-Energy Plan), SRADDET (Regional Plan for Development, Sustainability and Territorial Equality), or in PLUs (Local Urban Plans).

In recent years, an explosion of scientific literature dedicated specifically to NbS has also been observed. A bibliometric study based on Scopus data (Figure 1) shows that the number of articles has increased from two in 2014 to 4,020 in 2024, three quarters of which focus on the urban environment. This substantial output has prompted a number of state-of-the-art and/or review articles that aim to assess this exponential production both quantitatively and qualitatively [42, 43, 44, 45, 46]. These reviews tend to focus on geographical contexts (e.g. Europe, the Mediterranean basin, China and South-East Asia) or specific themes (e.g. life cycle analysis, energy consumption, coastal risk, drought adaptation, public policies, urban planning and financing). State-of-the-art studies have even been conducted to analyse these reviews [47]. These syntheses, known as “umbrella reviews”, highlight the shortcomings of NbS in certain environments, such as forests and rural areas, and the lack of studies on social effects.

As mentioned above, it is clear from all these publications that research work dedicated to NbS focuses essentially on the urban environment. In this environment, green roofs and urban forests are the most widely studied solutions [48]. The prevalence of technostructures, such as green roofs, generates a technical perspective based on traditional engineering practices. For instance, the contribution of NbS to flood management and the mitigation of urban heat islands is emphasised, but the sustainability of these potential solutions is hardly addressed — even though global change (particularly climate change) and its various manifestations are sustainable. These publications also fail to address the link between these solutions and human and non-human benefits [6, 49, 50]. Thus, NbS seem to be effective means of addressing problems in urban environments, in line with the concept of “designing with the living”. Fieldwork has short-term objectives geared towards a rapid technical response, which does not take into account the specific ecological characteristics of the species involved nor the time and space required for their evolution. This would enable them to complete their life cycle and ensure the necessary natural renewal of individuals, in line with the idea of the sustainability of socio-ecological systems [50].

As we have already seen in the case of ecological engineering, NbS generally take the form of restoration measures that are almost exclusively for the benefit of humans and that are based on a more or less fantasised and realistic natural environment. The targeted ecosystem services are those linked to regulation (e.g. water, climate and risk management), supply (e.g. wood, energy and food production) and socio-cultural contributions (e.g. culture, leisure and heritage). The assessment of the benefits in return for biodiversity only appears as a last resort (see the IUCN definition) and is rarely put forward in practice.

Thus, NbS, which are institutionalised by many bodies (e.g. IUCN, WWF, OFB and the European Commission), are ultimately informed by vagueness, as nothing is gained in terms of biodiversity through new practices [51]. This renders NbS normative and counterproductive: the translation of ecological concepts into management issues is neither implemented nor widely used, yet institutions are already making it standard practice and advocating its widespread adoption [51]. Walczyszy [52] writes on this topic: “Promoting ‘nature-based solutions’ is tantamount to delegating to managers of environmental policy the duty of demonstrating, through concrete and visible actions, the normative principles on which a certain ideal of good governance of humans and nature is based in the 21st century”.

4. How can nature-based solutions (NbS) be implemented? Socio-ecological restoration can serve as a model for transforming practices

As we have seen, neither ecological engineering nor NbS, as interpreted and put into practice, have promoted biodiversity for its own sake. These approaches involve the integration of ecological concepts for technical use by engineers. The focus is on solving an environmental problem by implementing techniques that are deemed suitable and become standard practice. In the second half of the 20th century, decision-makers and managers believed that biodiversity loss could be managed [53]. The prevailing view on environmental issues approached protection issues as a traditional engineering problem [53]. Producing proven facts made it possible to define a classic chain, from analysing a problem to understanding it, and helped to define management rules. Engineering has always been a matter of calculating, modelling and formalising, simplifying reality to build a plausible and workable representation of the problem. This problem is linked to the world by measurement operations that serve as the basis for calculating solutions.

Within such a framework of thought, ecological engineering and NbS have remained umbrella concepts, highly institutionalised but little operational for effective implementation, particularly in terms of biodiversity. And yet, “Nature is declining globally at a rate unprecedented in human history - and the rate of species extinction is accelerating, already having serious effects on human populations worldwide”, warns the new and historic report from the Intergovernmental Platform on Biodiversity and Ecosystem Services [54].

How can we finally change our practices so that they benefit both the non-human living world and ourselves? How can we transform human activities into sources of added value for nature itself? In order to ensure that biodiversity benefits from actions on the ground too, must we not also link social and ecological objectives in the reconstruction of societies and the ecosystems that support them? Do we have the means to realise the dream of scientists who originally envisaged ecological engineering as a way of steering the dynamic co-evolution between the natural elements of the biosphere and human societies? This requires us to reconcile human beings with their milieu (Umwelt), not their environment (Umgebung): “Disasters belong to the milieu, natural phenomena to the environment.” [55]

Tidball [56] suggests that, when faced with disaster as individuals, communities or populations, humans are more likely to turn to nature to strengthen their resilience. He mentions that our affinity with the rest of nature, or the process of reconnecting with this affinity and expressing it by creating restorative environments, emerges at pivotal moments following profound crises. These opportune moments, which he calls the “back loop” of the adaptation cycle, can genuinely and certainly restore or increase ecological function because those involved are fully convinced of it. We probably need to give greater consideration to the interrelationships between natural and cultural processes and their actual consequences in the field.

In this dynamic linked to the idea of combining ecology and social issues to resolve environmental crises, past experiments carried out within a framework known as “socio-ecological restoration” provide valuable insights and clear feedback, which can help to put NbS into practice if we wish to take advantage of it.

The term “socio-ecological restoration” was first used about ten years ago, in an article discussing the ecological and social recovery of areas affected by the 2011 tsunami in Japan [58]. This article clearly explained that, after such a devastating natural disaster, it was unrealistic to expect the ecosystem and human populations to quickly return to a state perceived as normal. Local leaders, who had been working for many years to restore the diversity of the forests, played a key role in determining the desired direction for the post-disaster period. Local people rejected the construction of cement barriers, which had already been deemed unnecessary elsewhere in Japan due to the scale of the tsunami, and instead began to restore the coastal forest dunes, which naturally protect the archipelago from tsunamis. Furthermore, restoration efforts have extended to the watershed where commercial conifer plantations were abandoned decades ago, impacting the water reaching oyster farms due to its high tannin content. The restoration strategy was based on the Japanese concepts of Satoyama and Satoumi, according to which the continuity of ecotones between mountains, plains and sea is part of both economic activities and the conservation of natural landscapes. Satoyama is a mosaic of terrestrial and aquatic ecosystems, including forests, plantations, grasslands, farmlands, pastures, irrigation ponds and canals, with a focus on terrestrial ecosystems. Satoumi, on the other hand, is a mosaic of terrestrial and aquatic ecosystems comprising rocky coasts, shores and foreshores, as well as coral reefs and seagrass beds, with an emphasis on aquatic ecosystems. Biodiversity is key to the resilience and functioning of Satoyama and Satoumi landscapes. The strong links between Satoyama and Satoumi, which refer to traditional rural and coastal landscapes in Japan, have contributed to the effective restoration of the environment and strengthened the relationship between local communities and their environment. In this case, the solution was an inventive local transformation based on ecological and cultural memory, passed down through the generations by the inhabitants of these disaster-stricken areas. While the ecosystems proposed in these restorations were based on past local adaptations, they primarily served as a necessary starting point for future evolutionary trajectories that would enable new ecological dynamics. Without them, there would be no local ecological dynamics. This often forgotten starting point is a social and ecological success: populations are motivated to act together and ecosystems are restored “as they were before”; however, this installation itself is a new ecological dynamic, as it would have been before the disturbance.

The expression “socio-ecological restoration” was then taken up by [59] in an attempt to provide a practical and operational definition. In their article, socio-ecological restoration is defined as “cycles of restorative processes in which the restoration of ecosystem function is inextricably linked to the repair of cultural ecological reference points for human populations struggling to regain a normal life. In this definition, the cultural value of species, forests or landscapes lies at the heart of a process that can be seen as a healing process for people, whether following a natural disaster or a similar situation in areas devastated by war and armed conflict. In addition to the potential ecosystem services that people can obtain from restored areas, it is increasingly documented that contact with natural environments is indeed part of the therapy. Socio-ecological restoration is a broader social process than a traditional project. In this context, several actors at different levels can work in parallel or in tandem, without necessarily having a centralised lead.

This raises the question of whether ecological restoration, as a nature-based solution, actually corresponds to our current understanding of socio-ecological restoration. The IUCN lists global challenges (such as climate change, human health, food and water security, natural disasters, and biodiversity loss) that must be addressed “through the protection, sustainable management, and restoration of natural and modified ecosystems for the benefit of biodiversity and human well-being". Socio-ecological restoration is in line with this proposal and many others that implicitly or explicitly state that human well-being cannot be dissociated from healthy ecosystems [60].

Can we not draw inspiration from these successful experiments and extend them to other situations ? Cities, for example, have often had a devastating impact on the cultural and ecological history of the sites on which they are built, through the construction of roads, buildings, bridges and deep waterproofing. This has made them very anonymous, homogeneous and lifeless.

In their analysis of the city-river relationship, Carré et al. [62], retrace a history reminiscent of the experiences cited above. The authors mention three phases (phase 1: symbiosis; phase 2: rupture; phase 3: reunion) that correspond to periods of contact and rupture established on an emotional level. The reopening of the Bièvre river is a prime example of phase 3.

The Bièvre river is 36 kilometres long. It rises in the commune of Guyancourt in the Yvelines department and flows into the Seine in the 13th arrondissement of Paris [63]. From the 12th century onwards, the various activities that took place near the river (such as agriculture, industry and crafts like tanning and butchery) coupled with growing urbanisation turned the Bièvre into an “open sewer” [63, 64]. It was therefore channelled and covered, primarily in the lower reaches, notably in Paris between 1877 and 1935, and then further upstream between Antony (Hauts-de-Seine) and Verrières (Essonne) [63].The reopening of the Bièvre river, which began in the 2000s with various sections passing through several towns where it had previously been covered, has been accompanied by an initiative to promote biodiversity. In 2003, 300 metres of the Parc des Prés reach, located in Fresnes (Seine-et-Marne) ,were returned to open water, 23 kilometres from the source. The landscape was reshaped, with the creation of a natural bed to promote oxygenation and self-purification, areas where the river could flow freely, the creation of gently sloping banks, the natural sealing of pools and the development of a side stream [65]. Regarding biological monitoring, the initial aquatic state could not be diagnosed, as the watercourse was buried. However, terrestrial inventories of flora and fauna were carried out in 1994–1995 and again in 2000. Following completion of the reopening works at Fresnes, new inventories of the Parc des Prés fauna and flora, where the watercourse is located, were conducted in 2004–05 and 2007–08. These inventories revealed a significant increase in biodiversity at various levels, both specific and ecosystemic. This could be attributed to the reopening of the Bièvre river or to the development of a network of ponds and a stream [63]. From a social point of view, the reopening of the Bièvre river in the Parc des Prés has been a success because many people were heavily involved in the project, particularly local residents [65]. In addition to the Communauté d'Agglomération and the commune of Fresnes, other institutions were involved in the project: the Agence de l'Eau Seine Normandie, the region and the département provided technical support [65]. All interested parties, including local residents, schoolchildren, local associations and businesses in the area, have been involved in every stage of the project since its conception [65]. For example, they took part in writing workshops, participatory work camps, clean-up operations, educational events, site visits and consultation meetings on water-related issues [65]. In 2007, the Bièvre river finally regained its status as a watercourse [64]. Figure 2 illustrates this transformation, particularly on the stretch near Jouy en Josas (Yvelines).

On this subject, Carré et al. [62] write: "The Bièvre river seems emblematic of values torn between preserving the memory of an industrial river and a collective projection of an idealised river, in terms of its role in identity, social ties, and its exceptional urban nature". The authors emphasise the importance of storytelling. They describe the Bièvre river as the “sacrificed river that must be resurrected in a manner similar to the story of the river”, suggesting the need for a river that is as spontaneous as possible with limited human intervention. This is a river that is “alive with its wild fauna and flora, not artificial, connected to generations, inspiring artists — a river where the water has never stopped flowing”. The authors conclude that in this specific case, the proposal went beyond simply reopening a canal. Indeed, this project, which was necessary “at all costs”, was made possible because it fulfilled “symbolic, teleological and messianic functions” to bring about a Bièvre river that had finally become “natural”.

This example clearly illustrates the concept of socio-ecological restoration, showing how its principles and its effective implementation in the field are fully aligned with the objectives set out in the IUCN's definition of NbS (“protecting, sustainably managing and restoring natural or modified ecosystems to directly address societal challenges in an effective and adaptive manner, while ensuring human well-being and generating biodiversity benefits”), which have not yet been achieved. The question is whether we should change the name of a practice once again or simply draw inspiration from one practice to improve another. The urgent need is probably to better combine practices in the field that take into account both ecological and social processes with the idea of sustainability behind them.

5. What recommendations for moving forward?

Fernandez-Manjarrès et al. [59] proposed six recommendations for socio-ecological restoration, which are listed below. As their framework for action relates to post-conflict situations (e.g. war) or long-standing social and ecological degradation, we propose adapting these recommendations for broader and more varied cases. As is the case in ecological and socio-ecological restoration, these recommendations take into account the life cycle of a project whose end is unknown:

Recommandation 1: need for prior assessment and analysis of the situation. It is necessary to first define the urgent needs in order to restore minimum human standards and limit damage to ecosystems. To do this, it is necessary to understand the ecological and cultural history of the site (e.g. what biodiversity was present) as well as the social and economic practices that existed before the deterioration; this will help to restore the site's roots and habitability. It is also crucial to ensure that local populations are convinced and ready to engage in the “reunion” phase.

Recommandation 2: strategic planning. It is necessary to define which species and/or processes will be essential for the long-term rehabilitation of the site, both socially and ecologically. This involves proposing the introduction of local species that are better adapted to the context and that will be able to establish themselves spontaneously, while also taking into account issues related to climate change. An approach that preserves and promotes natural processes is essential, as these provide the conditions necessary for the developing technical solutions to human problems. Projects should be undertaken in synergy with non-human dynamics rather than against them, and the idea that ecosystem diversity offers numerous opportunities for production in terms of efficiency, resilience and adaptability should be promoted. Natural spaces must be favoured over anthropised spaces, as they are niches of opportunity, potential and resilience for other spaces. These uncontrolled spaces play a major role in the functioning of anthropised spaces and have a role in soil artificialisation, pest dynamics, pollination, etc. It is also necessary to take into account the feedback loops that intertwine many ecological functions and include delayed effects and indirect impacts on biological diversity.

Recommandation 3: mobilisation of local players. For complete success, it is also essential to mobilise local people, involving them as early as possible in the local restoration process. Can certain plant or animal species be identified as having essential cultural value, linked to ancient economic, social or spiritual practices, and which one would preferentially wish to restore/rehabilitate? Are certain places prioritised over others due to severe and specific damage? Are certain areas targeted more than others during larger-scale restoration projects (e.g. the destruction of dams or grey infrastructure to restore a riverbed, or specific natural areas before their destruction)? What are the local community's initiatives? How are local crises or conflicts around environmental degradation issues explained? What is the cultural and ecological memory of the site, according to the inhabitants? Is there a common heritage shared by all? Semi-structured interviews, followed by discussion/confrontation meetings in various forms (shared stories, role-playing, workshops focused on various actions, etc.) will be important levers for collectively mobilising local players and then allowing proposals to emerge that will always be co-constructed to ensure their full success.

Recommandation 4: mobilisation of resources. The knowledge required to restore ecological and social processes must be multifaceted and must also take into account local traditional knowledge. To this end, local surveys conducted by ecologists, historians and anthropologists may prove essential to understanding the history of sites, through consultation of archives or previous ecological diagnostic studies carried out by research consultancies (impact studies, for example).

Recommandation 5: implementation and monitoring. It is important to ensure that the rehabilitation of species, ecosystems and human groups moves in the agreed direction, and to consider the consequences of potentially different trajectories linked to hazards and uncertainties. The idea is always to allow species to follow their own trajectories, by demarcating dedicated areas whenever possible.

Recommandation 6: peer review and evaluation. It is important to ensure that restoration efforts have repaired basic overall functions, even if these have led to greater complexity in socio-ecosystems. Returning to recommendation 1, if necessary, is also a basic proposal, as a judgement on the self-management of the system must be made here.

6. Conclusions

Although ecological engineering has not enabled biodiversity to be considered for its own sake, despite its definition, the hope surrounding the deployment of NbS suggests that this new approach could facilitate such a shift. However, the normative, anthropocentric and institutional nature of NbS hinders implementation that considers biodiversity for its own sake. Socio-ecological restoration offers a more eco-centric vision in practice and may represent an inspiring new approach for NbS. Indeed, it places the cultural values of species, forests and landscapes at the very heart of the process of revitalising natural spaces, and works with a view to adding social value, but also, above all, ecological value (Figure 3). The example of the Bièvre river bears witness to this, and other experiences around the world, such as in Japan, also show that such initiatives are possible. Future work to identify past and/or ongoing socio-ecological restoration cases is necessary in order to better identify and account for this social and ecological added value. Looking ahead, the recommendations set out above can serve as a basis for initiatives rooted in a long-awaited philosophy of action for the benefit of both humans and non-humans. Solutions based on and developed with nature may then be a more accurate expression of the expectations of a genuine ecological transition.

Declaration of interests

The authors do not work for, advise, own shares in, or receive funds from any organization that could benefit from this article, and have declared no affiliations other than their research institution.