Dam and reservoir removal projects: a mix of social-ecological trends and cost-cutting attitudes

The removal of dams and reservoirs may seem to be an unforeseen and sometimes controversial step in water management. The removal of barriers may be different for each country or region, as each differs greatly in terms of politics, economy and social and cultural awareness. This paper addresses the complex problem of removing dams on rivers and their connected reservoirs. We demonstrate the scales of the changes, including their major ecological, economic, and social impacts. Arguments and approaches to this problem vary across states and regions, depending on the political system, economy and culture, as confirmed by the qualitative and quantitative intensities of the dam removal process and its global geographical variation. The results indicate that the removal of dams on rivers and their connected reservoirs applies predominantly to smaller structures (< 2.5 m). The existing examples provide an important conclusion that dams and reservoirs should be considered with regard to the interrelations between people and the environment. Decisions to deconstruct hydraulic engineering structures (or, likewise, to construct them) have to be applied with scrutiny. Furthermore, all decision-making processes have to be consistent and unified and thus developed to improve the lack of strategies currently implemented across world.

. Since the 1980s, dam removal has become an issue among the Lower Elwha Klallam Tribe and environmental organizations. In 1992, Congress passed the Elwha River Ecosystem and Fisheries Restoration Act 24 listing the fish populations impacted by two dams 25 and The US Congress decided to allow the federal government to purchase the privately owned dams from a pulp and paper mill company, and a study on the potential impacts of their removal was initiated 20,24,26,27 . Similar issues occurred with a middle-sized concrete dam of the arch type, called San Clemente on the Carmel River in California, subsequently leading to its removal in 2015 (Table 1). In 2008, its capacity was only 86,000 m 3 , which constituted 5% of its original volume 15 .
The loss of original volume was observed in reservoirs more than 40 years old in the US, whose cost of restoration would amount to approximately 90% of the price of new objects. Consequently, at the beginning of the 1960s, decisions were made to eliminate some of the medium-sized and large dams 21 .
One of the reasons for the removal of small dams is a concern for public safety 28 . In particular, these low barriers pose a serious threat to river users. Tens of thousands of these dams were built in the US after 1800 to enable the operation of mills, sawmills, and to collect potable and industrial water 29 . From 2000 to 2015, the Association of State Dam Safety Officials (ASDSO) k documented 241 fatalities and 98 injuries in 282 incidents related to individuals crossing small dams of the low-head-dam type (data for 42 states). www.nature.com/scientificreports/ European countries lack a uniform system of inventory and monitoring of river dams' status, and data access is therefore handled within each individual state. Based on the data from the current report of the DRE f and other collected data from governmental institutions h,i,j,l,m,n,o,p,r,s , our own analysis was carried out in terms of the removed dam's height and the trend in the time of the removals, as well as the intensity of removals. In the years 1996-2019, a total of 342 objects were dismantled, approximately 95% of which are so-called low barriers-similar to the US-54.7% are dams up to 2.5 m high, 40.6% are 2.5-7.5 m high, 2.3% are objects 7.5-15 m high, and 2.0% are higher than 15 m (Fig. 1). Only one removed dam exceeded 30 m; the demolition of the dam on the Sélune River in France began in 2019 ( Table 2). The intensified removal of dams on European rivers began in approximately 2006th (Fig. 2) and continued for less than 10 years, with regards to low structures (< 7.5 m). For larger dams, the trend remains at a similar level continuously (Fig. 2).
The data collected by the DRE and used in this study usually include the location of each removed dam but information about its height or the date of removal is often unavailable (e.g., in Sweden, Finland, and the UK).
The mass implementation of low artificial river barrier removal is associated with the start of the Water Framework Directive (WFD) (2006/118/EC), which was implemented in 2006 10,30,31 . The WFD has significantly reinforced the drivers for restoration, thus encouraging the improvement of the ecological status of water bodies. To comply with WFD requirements, the Spanish Ministry of Environmental Affairs (MAPAMA) m developed a National Strategy for River Restoration in 2006, including some of the projects described in this document 32 . The www.nature.com/scientificreports/ French Ministry of Environment and the Swedish government supported various river restoration projects-for the first time in the EU. The WFD's pioneering water resource management projects, which took place between 2009 and 2015, aimed to increase the importance of a progressive integrative restoration suit 30 . For the UK, the national database is divided between four independent jurisdictions (Scotland, Wales, Northern Ireland, and England) with individual agencies operating within these four jurisdictions. Data for Northern Ireland were not available for this study. In Scotland, the body responsible for maintaining reservoirs is the Scottish Environment Protection Agency (SEPA). Scotland reservoirs are regulated under the Reservoirs Act 33 . Before the act's enactment, local councils were responsible for collecting data on and maintaining reservoirs and dams. Based on the data from the SEPA covering the 2011-2020 period, four reservoirs were designated discontinued sites: two in 2017, one in 2018, and one in 2019. The height of the dams ranges from 1.2 to 3.0 m. The cubic capacity at the top water level ranges from 40,000 to 95,000,000 m 3 . The oldest dams were constructed in 1863, and the newest dams were constructed in the 1970s (Appendix , Table A1).
Natural Resource Wales (NRW) is the institution that collects and maintains data on all reservoirs designed or capable of holding more than 25,000 m 3 of water above the natural level of any part of the land adjoining them defined as "large raised reservoirs" under the Reservoirs Act 34 . Two types of reservoirs are maintained within the register: impounding (dammed) or non-impounding (pumped/unimpeded). The analysed data indicate that the first dams were decommissioned in 1986 and the most recent in 2017. The oldest dam was constructed in 1830, and the newest dam was constructed in 1977. The reservoirs' capacity ranges from 32,000 to 2,000,000 m 3 . The dam height varies from the smallest dams of approximately 2.0 m to the tallest measured at 20.0 m (Appendix, Table A2). The Llaeron 20-m high dam, built in the mid-1860s, was decommissioned in 2019 for safety reasons following the closure of the nearby quarry, emptying the reservoir and leaving the dam structure intact for cultural heritage purposes. Furthermore, the same approach was utilized in the removal of the Ratcoed dam and reservoir (8 m high).
The data available for England, provided by the Environment Agency (EA) h , include only reservoirs with volumes exceeding 25,000 m 3 . Consequently, in certain cases, the implemented actions entail only reducing the amount of retained water to below 25,000 m 3 , thus avoiding the need to comply with regulations on completely dismantling any dams connected to the reservoir, the reduction of barriers, or the reservoir itself. This system of registry therefore does not refer to the height of the dams. According to the acquired data, 251 reservoirs have been reduced since 1984. The oldest of these reservoirs was commissioned in 1758, while the newest was commissioned in 2014. The average age of a reservoir at the time of removing it from the register exceeded 95 years, ranging from 0 to 232 years (Appendix , Table A3).
Safety is considered the main reason for dam removal or decommissioning in the UK due to the dam locations in densely populated areas 35 . Other common factors include ecosystem recovery and channel restoration. Additionally, ecosystem services are considered highly important when reasoning over the process of decommissioning/removing dams in the UK 36 .
According to the data for Sweden, received from the Swedish Meteorological and Hydrological Institute (SMHI) i and accessed in 2013, out of 5,280 dams recorded in the register, 557 were dismantled or demolished (Appendix 1, Table A4), of which 190 had available data on their height. Out of the 190 with defined heights, only 2 exceeded a value of 7.5 m (10 and 8 m, respectively), which amounts to 1% of the dams in total. Thirteen dams fell within the range of 5-7.5 m, which constitutes less than 7%. Almost half (49%) of the dismantled dams were 2.5-4.9 m high. The remaining 82 dams (43%) did not exceed 2.5 m. As analysed 37,38 , the most dams dismantled or considered for dismantling in Sweden are low dams. In this case, safety, law and policy, economy, and ecology are considered major reasons for dam removal.
Swedish findings share similarities with neighbouring Norway, which has 4,758 registered dams in the official database at the Norwegian Water Resources and Energy Directorate (NVE) j . Among them, 61 dams have been decommissioned, removed, or modernized as of 2019 (Appendix, Table A5). The dam size varies in lengthfrom approximately 3-743 m-and height-from approximately 1-25 m. These larger dams (> 5 m) have been decommissioned through a process of sinking or modernized by raising them, such as Inntakskanal Kykelsrud  Table A4). The reasoning for the decommissioning or removal process is available for approximately one-third of the cases registered for dam decommissioning or removal 40 . Several considerations are made as the dams are removed, i.e. effects on biodiversity, the public's use of structures, hydrology, and the cultural heritage associated with the structures. However, whether this is for the purpose of environmental consideration or for securing better public use of the area is not stated clearly in most cases 40,j .
The French Ministry of Environment has been working to keep a complete inventory of dams on French rivers. The most recent update in 2017 shows that there are over approx. 90,000 obstacles (all types), and approx. 70,000 of them are dams with weirs. The removal of three dams in the Loire tributaries in 1996-1998 was the first major dam removal operation in France 41 . Saint-Étienne-du-Vigan (12.0 m high), Maisons-Rouges (3. 8 m) and Kernansquillec (14.0 m) were demolished and shared common features: poor technical condition, advanced age of the structures, and positive prognosis for rebuilding fish migration.
Poland has 32,972 registered dams in the official database of the State Water Holding Polish Waters (PGW Wody Polskie) n . The OTKZ is the institution that collects and maintains data on all large and large dams. However, the OTKZ o database does not contain any data on demolished reservoirs or dams. Three dams suffered from construction difficulties but were rebuilt. The 10 m high Wilkówka dam with a capacity of 26,200 m 3 ( Table 2) is being prepared for demolition in 2020. This dam was damaged by a small spring flood in 2019 due to problems with constructional defect. There are several decommissioned dams awaiting an action plan (see Appendix, Table A6).

Scientific Reports
| (2020) 10:19210 | https://doi.org/10.1038/s41598-020-76158-3 www.nature.com/scientificreports/ Russia offers a special case of dam removal. Here, during the transition period from the USSR to the Russian Federation and change from state ownership of all hydrotechnical objects to private ownership, many dams lost their status and were thus left unregistered by the authorities. Therefore, the absence of ownership is the main problem with existing dam maintenance, leading to a specific type of dam classification: abandoned (meaning not belonging to an owner). The situation led to a lack of controlled maintenance of such dams and a loss of safety standards. Since the Water Code of the Russian Federation p was adopted, the problem is currently addressed either by registering the ownership rights of the dams or by removing the dams. Additionally, a federal act q formulated the main approaches to abandoned dam removal. All the existing abandoned dams are low dams (< 10 m height) with a capacity of approximately 1-3 million m 3 . No larger dams, to our knowledge, were ever removed within Russia. A recent overview of these approaches has been published t . According to official statistics by the Federal Service for Environmental, Technological and Nuclear Oversight of Russia (FSETNOR), there were 6,816 abandoned low dams in Russia in 2008, and between 2010-2014, 319 to 945 dams were removed annually (Appendix 1, A7).

The social-economic issues of dismantling dams: case studies and examples. It is important to
emphasize that dam removal projects should consider the interests of different stakeholders.
For 1,100 dams removed before 2016 in the US, only 130 of these removals had any ecological or geomorphic assessments, and less than half of those included before-removal and after-removal studies 43 . As emphasized by Duda et al. 24 , although many dams have been removed in the US, studies assessing ecosystem changes in the physical, biological, and chemical properties of rivers and their final impact on the potential for restoration are limited. After numerous experiences with small dam removal projects in France, new analytical methods were recommended to help understand and interpret this controversy through the use of two complementary approache 44 . The first approach is a geo-historical approach. The second method is based on political ecology. It is based on the assumption, to better understand and interpret the controversy related to the demolition of dams, these two complementary approaches are necessary. It is also important to create optional scenarios by considering short-and long-term effects and presenting the possible course of events both in the case of leaving the dam intact as well as in the case of its removal. Comprehensive plans may present local communities with possibilities related to new forms of development for areas formerly occupied by reservoirs, which may effectively and successfully provide greater social and economic benefits 45,46 . Examples of projects involving the dismantling of dams on rivers in the US, Sweden, Finland, Netherlands and France show the significance of societal participation in the decision-making process (Fig. 3), although projects become more suited to the general public's needs 4 .
Research conducted in the Netherlands discerned three types of approaches to projects involving the restoration of water systems: commitment, the appeal of nature, and the rurality of the landscape. The communities representing the commitment and rurality types more noticeably express concerns and opposition against restoration projects or renaturalization 47 . In the US, in New England, local communities make a commitment to the heritage of dams, similar to the European cases 4 , while in the Native American territories, for example, the river Klamath at the border of California and Oregon, there is a more visible difference between indigenous peoples, economically and culturally dependent and spiritually connected to a largely untransformed environment, and settlers pursuing contemporary agriculture 48 . In this case, the decision regarding the demolition of four dams resulted from a consensus found among over twenty groups of stakeholders. In New England, excluding the indigenous peoples, local communities exhibited a considerable commitment to a transformed landscape, often perceived by the general public as natural as well as cultural heritage, in which dams largely shaped an understanding of history and the economy of the region. This phenomenon is reflected, for instance, in the use of dams as symbols in city heraldry 4 . However, the New England region has a number of indigenous people and federally recognized Indian Tribes. One of them has been involved in a significant dam removal project (on the Penobscot River in 2012-2013) 49 .The situation unfolding in the state of Wisconsin was similar to that in New England 50 . Eighty objects with an average height of 4.3 m have been dismantled since 1960. All the dams considered for dismantling no longer served their economic functions, and the costs of their repairs were considerably higher than the costs of demolition 51 . Regardless, there was considerable public opposition to this project. The residents expressed their doubts, such as the value of adjacent real estate after removing the reservoir, proprietary issues from the uncovered land, the loss of recreational functions, or the appearance of the land, fearing the creation of an unappealing wetland 50 . However, as stated by Wyrick et al. 52 , whose research was performed in New Jersey, residents living close to dams considered for dismantling often had high expectations in terms of the biophysical changes to watercourses, as well as an increase in the value of properties and the recreational potential. Another example is the research referring to the social perception of the Mactaquac Dam in New Brunswick (Canada) 53 . The First Nation Tribe called for the removal of the dam. The end result was that it did not happen. Residents desire to keep the structure in place, even after discontinuing energy production.
There are also examples of resistance to demolition, i.e., in France and Sweden. According to the European River Network organization (ERN) s , an example of this phenomenon is the Poutès dam on the Allier River in France. The 20-year fight for the removal of the dam ended at the end of 2011. A compromise was made; the dam will be maintained but will be lowered and extensively modified. Additionally, the Blois dam of the Loire,  www.nature.com/scientificreports/ (e.g., type of fuels, explosives, etc.) used in the demolition process 45 . Nitrogen flux and eutrophication in coastal watersheds can have a possible negative environmental impact especially for small estuaries 60 . In the case of ichthyofauna and benthos, the removal of the dam led to a major transformation of fish communities. At the same time, due to the activation of debris accumulated in the reservoir, a temporary deterioration of the living conditions of species inhabiting river segments downstream of the dismantled dam should be considered 61 . Long-term research performed in Denmark indicates a considerable increase in the population of sea trout, both upstream and downstream of the removed dam, regardless of minor changes in the quality of the habitat. In most cases, removing the barrier on the river has an impact on how quickly it can be colonized by fish communities 43 . Examples show that recolonization by migratory fish was observed in the first year after dismantling the structures 20,61 . Noble fish species appeared, such as sea trout, salmon, and cyprinids endemic species 14 . However, research has proven that the removal of two barriers on the Wolf River (Wisconsin, US) did not result in a substantial increase in fish movement or the immediate colonization of newly accessible habitat 62 . In Sweden, dam removal reduced some macroinvertebrate taxa at the downstream site and found a reduction of taxonomic richness and that same dam removal effects persisted or even increased over time 63 . Three reaches of the Olentangy River (Ohio, US) noticed an initial drop in macroinvertebrates between ~9 and ~15 months after dam removal, and all variables consistently increased thereafter 64 . For example, in the Great Lakes region (US), artificial barriers such as dams can limit the dispersal of exotic species, and here, removing dams could harm native fish 65 . In this context, a holistic approach was suggested (not just a barrier decommissioning) between flow regulation and an active eradication of exotic fish in Arizona streams (US) for the successful conservation of native species 66 .
The recovery in terms of longitudinal connectivity allows new dam permeability along the fluvial system in terms of species movements and dispersion 67 . Especially interesting in this context is the case studies in Spain, i.e. performed along the Segura basin (SE Spain) 31 and in Northern Spain (Enobieta dam, Navarra), a promising experiment studying the effects of emptying a reservoir completely on the aquatic communities and water quality before the planned dismantling has recently been completed 68 .
The case of restoring the abiotic environment seems, in general, particularly challenging, with contrasting experiences worldwide. In most cases, analysis of dams dismantled so far indicated that there was a quick initiation of the erosion process of the reservoir's sediments 23 . Depending on the structure of accumulated sediments, the dam dismantling options and the spatially diverse reactions of the environment, the river system was rapidly restored each time. However, each case should be investigated separately due to the geographic context, the nature of the river, and the development of nearby land 43 . Mechanical removal of sediments has the smallest impact on the downstream ecosystem, but it is the most expensive option. On the other hand, the spontaneous erosion/ removal of reservoir sediments by a restored fluvial system has a negative environmental impact downstream of the removed structure, but it is the least expensive option 23 . A properly selected option for the removal of hydrotechnical objects (partial removal, slow, fast) limits the influence (manner) of the eroded sediments on the contamination of the environment downstream 23,69 . For example, to retain polluted sediment in the reservoir, not complete demolition of the Enobieta dam (Spain) 68 . Concerning the removal of materials, different behaviours have been described. In some cases, quick removal has been observed, such as in the Grangeville and Lewiston dams on the Idaho Clearwater River (US), where the bottom material was removed from the reservoir trough within a week e . Otherwise, a very slow sediment emptying can also be observed, e.g., after dismantling the Newaygo Dam-Muskegon River (Michigan, US), the emptying of the debris may last 50-80 years 70 . In some cases, where sediment in the reservoir is coarse-grained and minimal, and downstream areas are resistant to erosion, there is little channel morphology responses. This effect was achieved after the removal of two dams on the Penobscot River in Maine in the US (Great Works and Veazie dams-6 and 10 m high) 71 . Although in general, there are some negative ecological effects of the demolition of dams, it has been observed that these impacts on river ecosystems are tendentially short-lived.
An example is the Elwha River, where, as presented by Duda et al. 20 , "restoration has seen both early successes and setbacks, with the ultimate outcomes and lessons to unfolding in the coming decades". During the first five years after the dams were removed, 65% of the sediment (approximately 15.5-19.3 million tonnes) was released 22 and transported down the river 69 . The time period of negative impacts from sedimentation in the Elwha due to dam removal appears to have passed 23 .
Decision-making processes: highlighting the differences between Europe and the US. In our assessment, we show that there are noticeable differences in social and economic trends in the US and Europe in the removal of barriers on rivers and reservoirs. These differences become even more noticeable when larger water objects are removed. Low barriers have been removed both in US and Europe due to a long time of low economic benefits. Besides, their removal can be achieved at a low cost while providing significant environmental benefits 58  The provisions of the EU WFD indicated, inter alia, that by 2015, it was necessary to achieve a good water status. In addition, the two geographic regions differ in terms of dam ownership. In the US, most of the large dams are privately owned (according to NID, NABD and American Rivers) and are "ageing", and the trend in the number of structures dismantled is steadily increasing (Fig. 2). In Europe, it is mostly national governments that control dams and reservoirs or share the facility in public-private partnerships. In Europe, in the case of the www.nature.com/scientificreports/ EU member states, the maintenance of the hydroelectric structure took an approach to be achieved "at all costs". Investments in the water sector are subsidized with cheap loans by the European Commission and European Investment Bank (EIB) 72 . In Europe, reservoirs have been present since the medieval cultural landscape 73 ; in the US, the history of documenting reservoirs began in the first half of the nineteenth century 17 . In the US, for example, in indigenous territory, artificial reservoirs are not historical objects; hence, the participation of stakeholders (indigenous peoples) in the discussion of dam removal is prevalent 4,20 . A different example on the American continent is New England, where a more European approach to dam and reservoir maintenance is represented 4 . Based on the literature review, we revised arguments for and against in the public debate on the demolition of dams and the removal of reservoirs in Europe and the US (Table 3). In both the US and European countries, the most common criteria for removal is in the case of small and large dams are the loss of their original function and the loss of utility (functional purpose). There are arguments for the reconstruction of the fish migration path and poor technical condition, which may result in potential future failure. This is especially the case when considering complex facilities in urban areas, where security issues are considered the main reason for the removal or dismantling of dams 36 . In countries undergoing continuous economic transformation, problems arise with abandoned post-industrial water facilities. This problem affects Russia to a large extent, where removal of the abandoned "wrecks" of communism began in 2006. In the case of European countries, the strong economic dependence on existing large dams is apparent. Often, demolition is considered an unnecessary cost; instead, a new dam is built directly below or above, as in the case of Norwegian dams (e.g., Kykelsrud, Store Vargevatn, Namsvatn Hoveddam, Skjerkevatn) and German dams (e.g., Herbringhauser). In the US, there are examples of the removal of over a hundred barriers on rivers in New England 4 , eighty in Wisconsin 50 , and planned to remove four large dams on the Klamath River, on the border of Oregon and California 48 ; one of the decisive criteria for removal was the high cost of modernization (Table 3). In both the US and Europe, indigenous peoples support the removal of barriers on rivers. In the case of the projects to dismantle the dams on the Elwha and Klamath rivers, Native Americans participated from the beginning of the process, raising the argument for recovering the lands, the possibility of salmon fishing and the importance of culture and beliefs 29,48 . In Northern Europe, Sweden's and Norwegian's indigenous Sámi people, in turn, have insisted on the economic benefit of removing the dams in the form of regaining valuable pasture lands and the possibility of removing barriers to the seasonal migration of reindeer 74 . The main arguments against dam removal are the loss of cultural heritage, the sentimental and emotional attachment to the dam and reservoir, concerns about pollution and landscape deterioration, the fear of river disappearance and the emergence of unattractive wetlands, and the associated decline in land value. Concerns about the deterioration of the quality of the environment are justified, example of high pollutant concentrations below of the decommissioned the Enobieta dam 68 . Research in New England indicates the need for a better estimate of pollutant release following demolition 60 . Only where projects have undergone thorough scientific research does criticism dissipate from the discussion, e.g., the project on the Elwha River or the Tikkurila dam in Finland. In the case of the Elwha River, one approach was to collect as many basic studies as possible prior to removal 20 . The process before the removal of the Tikkurila dam was certainly shorter than that on the Elwha River, but the similarity of actions undertaken is clear 75 . Other arguments against dam removal include the high costs of river demolition and restoration or opposition to the monopolization of the river's functions as a migration route for selected fish species at the expense of the utility functions of storage reservoirs. In particular, the argument that the river should serve the wider community and not only selected fish species was raised during projects on the Selune River 76 and the Allier River in France (European River Network report) s and during projects in Sweden 37,38 . Experience from New England shows that in some cases, it is worthwhile to undertake alternatives to dam removal that can maintain the reservoir while improving fish flow and safety 60 . For example, a compromise was reached with the Poutes dam on the River Allier in France, where instead of being removed, the dam was modernized. We have demonstrated the course of the decision-making process in Fig. 3. We found the main reasons for the formal discussion to be the devaluation of functions, cost-cutting attitude, technical conditions, and ecological issues. The rank of the function depends on whether a large dam with a multifunctional reservoir or a low barrier is to be removed for stream metabolism improvement and stream ecosystem productivity. The main stakeholders participating throughout the process and their attitudes are as follows: (1) administration (national-regional-local level), politicians, scientific experts and businesspeople, who represent neutral/mixed attitudes, especially businesspeople and politicians, depending on their location; (2) environmental organizations and indigenous peoples, who are consistently concerned with removing barriers from rivers; and (3) local communities, usually those in the vicinity of dams and reservoirs, which are opposed to their removal (Fig. 3). A clear division in the regions into characteristic groups of countries representing the attitudes of their stakeholders is noticeable. The US is the only region in which all stakeholders participate in the process. However, it cannot be said that this is an optimal option for removing dams. It was highlighted by Fox et al. 4 , Germaine and Lespez 76 that the involvement of too many stakeholders extends the process, and conflicts growing over time often shift decision-making towards public administration and political actors.
Predominantly, public administration has considerable decision-making power in all the countries and regions considered in this study, mainly due to its control of the legal and financial instruments to carry out the relevant projects. An interesting example of this occurs in Central and Eastern European countries, including Poland. Despite integration with the European Union in 2005, the number of stakeholders in the decisionmaking process surrounding dam removals remains limited, and the entire responsibility for the decision lies with the public administration. Therefore, it can be concluded that the decision-making mechanisms and the level of ecological awareness have changed only slightly even though 30 years have passed since the political transformation in Poland. The analysis of the attitudes of stakeholders in individual countries in Europe also shows that there is no uniform implementation of the procedures in water management and protection of the aquatic environment developed in the EU, and the pattern of the decision-making process in removing dams involving wider stakeholder participation, such as that in the US, has yet to be achieved. An important element www.nature.com/scientificreports/ would therefore be developing the rules (procedures) for public participation in the process of creating, modernizing, or liquidating water reservoirs from the concept stage to the implementation stage. Currently, in the EU, public participation in this area is marginalized 47 and is most often limited to public consultations when obtaining decisions on environmental conditions-a formal requirement of the Water Framework Directive. It should be emphasized that the model for the decision-making process in the US should be used for future activities in this area of expertise. Therefore, holistic approaches considering the entire river system with a deep and detailed understanding of local features are recommended (e.g., the presence of invasive species upstream and the potential consequences on other downstream infrastructure indirectly affected). An example of this is the catchment area of the Willamette River in Oregon (US), where active management would enable the restoration of the continuity of 52% of the watercourses, with a drop in the production of electric energy and stored water by only 1.6% 78 . Another example highlighting the negative effects is the selective removal of dams in the Allier River basin in France, where the removal of a single dam did not solve the problem of the lack of a river continuum (FNRRC) 55 .

Conclusion
This review has shown that there are no complete statistical databases for removed dams on rivers. The research revealed that data may be sparse, even on the national level. In the UK, Norway, and Sweden, some dams have been decommissioned, not physically removed; rather, their height has been lowered to a level where they no longer fit the safety standards set for dams and lose their classifications as dams. Additionally, the poor technical condition of some dams in these countries will result in these dams being abandoned. Nevertheless, they are registered as decommissioned despite only being abandoned. This is the case with post-Soviet dams in Russia, where the removal of such structures is on-going but has affected only small structures so far. So-called small www.nature.com/scientificreports/ object dams are still being built in Russia, Poland and Norway, and these countries are also characterized by a very strong commitment to the maintenance of obsolete dams through refurbishment. Two accessible information sources are American Rivers and the DRE. These organizations store data about the name and location of the dam, the name of the river, sometimes the height of the dam, and what the dam was made of. The DRE dam removal list is not really a database, but simply a map-based resource. However, the presentation of general data (a mix of information on the removal of culverts, thresholds, small barriers and large dams) may certainly drive the boom to shorten the lifespan of structures on rivers.
Large dams in the US are still in operation, and those that were removed had suffered technical problems or were abandoned. However, none of the dismantled constructions had been located on main navigational waterways. Only 14 large dams have been removed of the 91,486 registered in the US. Examples include decommissioned dams and reservoirs full of sediments that were unable to provide the population with sufficient water volumes; thus, they had ceased to fulfil their original function, or their function had depreciated over time. The situation in Europe is comparable, as 12 large dams have been removed so far, and the scheduled deconstructions of larger facilities cover only those that are completely worn out. Certain EU countries, such as Poland, and the Russian Federation still develop programmes aimed at constructing large dams.
The identification of various groups of interest, a multiple-criterion analysis of social needs and options for their satisfaction, and the use of decision support tools facilitate indications of strategic priorities and a final decision to remove or spare a dam, river barrier, or associated reservoirs. These actions should be preceded by comprehensive familiarity with natural and anthropogenic conditions, the size and type of the structures, and their intended use and impact, all of which show significant geographical variability across the globe and regionally (e.g., the functions of the structures, their cultural and historical context, their safety and their technical condition). In terms of water management, dam removal poses a challenge for river management plans.

Methods
We identified dam removal studies published through February 28, 2020, using available scientific journal databases, Google Scholar, and Researchgate. However, the most important sources for this study were governmental and nongovernmental databases. In this work, we depended on four types of databases maintained by governments with free access to data: unpublished government data with sectorial consent for the use of data, such as the Open Government License; nonprofit organizations; and scientific research projects. The first group of databases is as follows: AMBER. We applied the information from the databases to graphically analyse the number of removed dams, the cumulative number of removals by year, and the distribution of dam heights for removal. We also identified the set of determinants responsible for the implementation of disposal projects. Scientific journal data were applied to determine the main social, economic, and environmental impacts.

Data availability
All data generated or analysed during this study are included in this published article. www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.