Introduction

A growing global population and related demand for freshwater have increased freshwater abstraction. Dam structures built across a river or estuary are constructed to secure freshwater resources for irrigation, domestic use, flood control, generation of hydroelectricity, and navigation (Altinbilek 2002; Figueroa et al. 2022). Their accompanying irrigation systems, diversions, flood attenuation, and increases in freshwater use have fragmented and transformed the world’s rivers and are impacting the functioning and health of associated estuaries (Olsen et al. 2006). The majority of dams are constructed for irrigation (~ 48%), while a smaller number are used to generate electricity (~ 20%) or for flood control (8%) (World Commission on Dams 2000). Globally, major dam construction commenced in the 1930s with more than half of the world’s large river systems now affected by dams (Chen 2005; Nilsson et al. 2005). Only 23% of rivers flow uninterrupted to the sea, and it is predicted that by 2030, natural flows will be altered for 93% of river volume worldwide (Thieme et al. 2020; Kotzé 2022). This decrease in freshwater inflow will be aggravated by climate change (IPCC 2022). This increasing demand and abstraction of freshwater is threatening the health and functioning of aquatic ecosystems, such as estuaries.

Low-inflow estuaries (LIEs) are particularly vulnerable to changes in freshwater inflow (Table 1). They occur in arid and semi-arid regions and are subject to periods of little to no freshwater inflow, with either seasonal or episodic high discharge peaks (Largier et al. 1997; Largier 2010; Walter et al. 2018). River inflow is often erratic and highly variable. Hypersaline conditions are typically present during dry periods, during which low tidal exchange, high temperatures, and long residence times result in evaporation rates exceeding rates of freshwater inflow (Largier 2010; Cira et al. 2021). Smaller estuaries that are temporarily closed/intermittently open to the sea typically occur in low inflow areas (Adams and Van Niekerk 2020). In arid and semi-arid parts of the world, where flow is often unpredictable and highly variable, irrigation and domestic water use tend to be the primary reason for dam construction. Such dams are generally not designed to release floods, or even in some cases baseflow, as water is a scarce resource in these areas where LIEs are generally located. LIEs are thus sensitive to human impacts and will be further stressed by climate-induced reductions and extremes in rainfall (Adams and Van Niekerk 2020).

Table 1 Importance of freshwater inflow to low inflow estuaries
Full size table

Freshwater inflow to estuaries is critically important to sustain ecosystem processes and ecological function (Alber 2002; Estevez 2002; Montagna et al. 2013; Adams and Van Niekerk 2020; Chilton et al. 2021). Healthy estuaries provide the ecosystem services that we depend on, with their importance as biodiversity hotspots and migration corridors for biota being well described. Freshwater species with estuarine or marine life-cycle phases (e.g., freshwater prawns and shrimps, catadromous crabs, and eels) can be cut off from estuaries and their catchments if there is inadequate freshwater inflow (Van Niekerk et al. 2019a). It is well known that dams, weirs, levees, and other forms of constrictions affect longitudinal connectivity between catchments, rivers, and the sea. This poses a barrier to the transport of water, sediment, organic matter and nutrients, and the movements of organisms and impacts ecosystem functioning and productivity (Drinkwater and Frank 1994; Chen et al. 2016; Opperman et al. 2019; Weng et al. 2020). Dams also disrupt lateral connectivity with estuary floodplains (Bornman et al. 2002; Clark et al. 2022) and influence turbidity regimes (Figueroa et al. 2022) in both directions. For example, the Burdekin Falls Dam (Australia) caused a permanently turbid estuary downstream due to fine silt that remained suspended in the reservoir which impacted all aquatic life (Wolanski and Hopper 2022).

Further, dams have induced a series of broader environmental consequences that may not have been anticipated such as reductions in sediment delivery to the ocean, significant global erosion of deltas and coasts, and losses of coastal forests and mangroves (Giosan et al. 2014; Ezcurra et al. 2019). Construction of the Aswan High Dam on the Nile River reduced flow by over 90% which collapsed the coastal fisheries (Nixon 2003). This early case study showed the importance of freshwater inflow to the marine environment. In arid and semi-arid regions, upstream dams and freshwater abstraction lead to reduced scouring, downstream sedimentation and closure of estuary mouths to the sea (Adams and Van Niekerk 2020).

Because river inflows patterns influence the health, functioning and productivity of estuarine and coastal ecosystems (Loneragan and Bunn 1999), the planning and operation of dams and other flow regulation infrastructure should take into account the consequences of changing the timing and magnitude of flows into these systems (Sharma et al. 2022). To mitigate the impacts of dams on downstream aquatic ecosystems, water releases are becoming an important practice around the world. In this context, a water release from a dam refers to the action of actively releasing a planned volume of water through a sluice gate or dam outlet. This is in contrast to a spill which occurs when a dam reaches a set capacity and then overflows, normally induced by a flood. Ideally, such dam releases should be incorporated in environmental flow (also referred to as EFlows) determinations for affected catchments, defined as the quantity and quality of freshwater flows—in terms of timing, duration, frequency, and intensity—necessary to sustain aquatic ecosystems to support cultures, economies, sustainable livelihoods, and human well-being (Arthington et al. 2018; Adams and Van Niekerk 2020).

The implementation of dam releases as part of EFlows and restoration programs is becoming a critical intervention to restore and manage estuaries to ensure continued provision of ecosystem services and associated societal benefits. However, to date, a critical evaluation of dam release studies on LIEs has not been undertaken to determine their effectiveness in achieving pre-defined purposes. Therefore, the objective of this study was to critically review the responses of estuaries to dam releases—as documented in the literature—and to identify key learning. Case studies were used to identify ecological health and societal benefits associated with flow releases. This assessment focuses on LIEs considered to be most sensitive to impacts from reduced freshwater inflows. As a result the study did not address high inflow systems such as the Yellow River (China), where dam construction and water abstraction have, for example, resulted in downstream erosion of deltas (Wang et al. 2017). We applied the learning gained from the study, to compose a socio-ecological systems (SES) framework for monitoring the release of freshwater inflows from dams. Finally, based on the findings of this study, we posed key considerations for effective planning and implementation of future dam release practice as part of EFlows or estuary restoration programmes. Although this investigation focuses on dam releases as mitigation for downstream environmental impacts, dam demolitions are increasingly being implemented to eliminate impacts.

Methods

A total of 11 case studies were reviewed to identify the response of LIEs to planned dam releases (Fig. 1, Table 2). The best described studies with details relating to water releases from dams and downstream estuary benefits were included in this assessment. Most of the case studies did not provide information on dam infrastructure or the operational practices pertaining to the dam releases. Additional data searches were conducted to provide such context. The location of the case studies is shown in relation to the site’s annual precipitation (mm) using information from WorldClim (Fig. 1). The image used in Fig. 1 was obtained from http://i.imgur.com/6lvzmXZ.jpg on Reddit, Inc. This presentation of global annual precipitation (mm) was generated from WorldClim that uses data layers which are generated through interpolation of average monthly climate data from weather stations (WorldClim 2022). Mean annual precipitation represented as a range (minimum, maximum) was provided for each case study from the sources indicated in Table 2.

Fig. 1
figure 1

Location of case studies where water has been released from dams for downstream estuary benefits in relation to the site’s annual precipitation (mm) using information from WorldClim

Full size image
Table 2 Summary of case studies including annual precipitation, dam purpose, storage capacity, wall height and dam length (where information available), distance from coast and associated data sources
Full size table

The Elwha Estuary (USA) cannot typically be classified as a LIE as it occurs in a catchment with annual precipitation above 1000 mm per annum. However, this case study was included for comparative purposes to highlight an extreme management intervention, i.e., dam removal and the unexpected responses from this. Further details for each case study are included in the Supplementary Material.

The case studies illustrate that in semi-arid areas (e.g., South Africa, Australia, and Texas), dams were mainly constructed for domestic and agricultural use (Table 2), while the hydroelectricity dam schemes generally occurred in catchments where rainfall was 900 mm or greater. The Diama Dam was constructed to prevent saltwater intrusion upstream and supply water for irrigation—it is situated on the Senegal River with a mean annual runoff (MAR) greater than 1200 mm. However, as indicated details were not readily available on whether this and other dams had the capacity to release baseflow and floods to the downstream estuary. Dams on the Groot Brak, Kromme (South Africa) and Elwha systems (USA) occur close to the coast and head of the estuary. As a result, there is little opportunity for river tributaries downstream of the dams to mitigate impacts and contribute to natural variability in flow. Dams can store a large percentage of the mean annual runoff thus having a severe impact on the downstream estuarine environment.

To explore the effectiveness of dam releases we analysed the purpose, the release practice, and associated ecological and societal outcomes of the selected case studies. From these findings, we recommended key considerations for planning and implementation of future dam releases to LIEs. These can be integrated in EFlow or estuary restoration planning towards improving ecological health, as well as associated ecosystem services and societal benefits.

Case study outcomes then informed the composition of a socio-ecological systems (SES) framework for monitoring the release of freshwater inflows from dams (adapted from Adams et al. 2020). The framework links estuary state and well being of people through ecosystem services and is based on Ostrom (2009) and the millennium ecosystem assessment approach (MEA 2005). Examples of estuary and societal indicators, as well as ecosystem services influenced by releases of freshwater from dams, were provided. This included fisheries and nursery habitats, nutrient cycling and water quality maintenance. The concept of socio-ecological systems is an important approach for managing natural resources as it emphasizes that human populations and coastal ecosystems are interlinked. Knowledge exchange between scientists from different disciplines, decision makers, and stakeholders can take place through a shared understanding of terms, such as “sustainability” and “ecosystem services” (Hossain et al. 2017).

Results and Discussion

Evaluation and Learning from Case Studies

The purpose of dam releases spanned an array of interventions and included: improvement of sedimentary processes (e.g., Nueces Estuary), supply of freshwater to restore ecosystem and habitat (e.g., Colorado, Nueces, Volta, and Coorong), improved water quality and removing excess macrophytes (e.g., Groot Brak and Nakdong), prevention of saline intrusion and hypersalinity (e.g., Senegal, Groot Brak and Kromme estuaries), facilitation of artificial mouth breaching (e.g., Groot Brak), managing floods (e.g., Senegal, Volta), and improving fish migration and coastal fisheries (e.g., Shoalhaven). Ecological and societal outcomes for each of the selected case studies was summarized (Table 3 and Supplementary Material).

Table 3 Summary of case studies highlighting key purpose of dam release, overview of release practice, ecological and societal outcomes (further detail in Supplementary Material)
Full size table

From the evaluation of the case studies, it was evident that planned freshwater releases were generally successful in moving sediment, keeping the mouths of estuaries open, allowing mixing, maintaining salinity and turbidity gradients, and triggering spawning and/or migration of fauna (Table 3). Studies showed that estuaries are resilient and, in many cases, their ecological health can improve following freshwater releases from dams. However, strong regulatory frameworks and ongoing commitment were needed to deliver environmental water over periods of prolonged droughts. In most cases, holistic adaptive management approaches required the involvement of multiple agencies to achieve environmental flow objectives as well as inclusion, and ongoing engagement, of communities and support of aquatic ecosystem users (e.g., river boat operators).

An assessment of the case studies showed that it was important to understand the whole system to ensure that dam releases were beneficial. In the Colorado system, if water had simply been released from the dam, the targeted lower ecosystems would not have benefited from the environmental flow as ‘surplus’ and would have been taken up by other users. Water was supplied through irrigation canals which bypassed dry areas (Table 3). Dam removal is an important restoration action but can result in rapid and unexpected responses and needs careful whole system planning (Table 3, Elwha Estuary study). Ecosystems and communities respond in unexpected ways frequently negating potential intended benefits. The spatial–temporal scale of impacts from dam removal and persistence of these impacts need careful consideration through dedicated monitoring and reporting particularly for LIEs.

Biota have evolved life histories based on natural cycles of flooding and drought. In long-lived species, such as riparian vegetation, desired outcomes in relation to environmental flow allocations only occur over multiple years or decades (Tonkin et al. 2021). Thus, it is key to understand life cycles when planning releases from dams. At the Shoalhaven Estuary frequent pulse releases were more effective than one large dam release in stimulating Australian bass to commence pre- and post-spawning migrations (Reinfelds et al. 2013; Al-Nasrawi et al. 2016; Ji et al. 2022). Baseflow establishes salinity gradients and improves biological productivity in estuaries (Bate and Adams 2000; Snow et al. 2000; Strydom and Whitfield 2000). Freshwater inflow and a reduction in salinity in the Nueces Delta increased biological productivity (Table 3). Floods are needed to move sediment and organics, and prevent macrophyte encroachment (Batalla et al. 2006; Gómez et al. 2014; Ibáñez et al. 2012, 2020; Human et al. 2016).

Water releases from dams improved downstream ecosystem health and ecosystem services provided by estuaries such as the nursery function and fishing opportunities (Table 3, Senegal Estuary). Releases provided societal benefits and improved livelihoods through job opportunities. An improvement in water quality increased recreational use (Table 3, Groot Brak Estuary).

Dam releases should always attempt to mimic the natural flow patterns as far as possible. “Novel” flow regimes often prioritize the flow characteristics that supply the most ecological benefit (Richter and Thomas 2007; Chilton et al. 2021). However, not all dams have spillways or outlets that are designed for controlled releases of water and are thus not able to mimic natural flow patterns. For example, the capacity is often there to release base flows but not the floods necessary for the effective flushing of waters and sediments from downstream estuaries. Releases are mostly made to address migratory fish populations. In impassable dams where river connectivity is completely blocked, upstream river courses are often devoid of migratory species (Fernandez et al. 2022). Some dams have fishways/passages to mitigate these impacts and although there has been an increase in the rate of construction of fishways, the performance of passing fish through these structures continues to be low in many regions (Silva et al. 2018). Legitimate constraints to environmental flow implementation include dam purpose and design, ageing water infrastructure, and encroaching development in floodplains below dams that prevent mimicry of natural variability, especially floods (Warner et al. 2014).

Innovative methods and best practice were identified from the case studies presented such as improved irrigation practices that increased flow to the Colorado Estuary. At this system, the required environmental flows had to be attained from existing users to reduce water demand pressure. At the Coorong Estuary, water also became available from improved infrastructure and on-farm irrigation technology. Legislation that controlled the licenses/permits for set volumes of water extracted and number of irrigators in a catchment was also important (Gippel et al. 2009; Chilton et al. 2021). Purchasing water rights is a key mechanism to ensure downstream baseflow inputs to estuaries in over allocated catchments. For example, California (USA) has recently begun considering the establishment of ‘ecosystem water budgets’ based on the total volume of water required to satisfy environmental flow needs that consider ecosystem management objectives, current water uses, and institutional arrangements (Grantham et al. 2020; Stein et al. 2021). A similar approach is followed in South Africa where water allocated to aquatic ecosystems is “Reserved” as a water right and allocated in law (Van Niekerk et al. 2019a, b).

Socio-ecological Framework for Dam Release Management

Freshwater releases from dams are an important management or restoration action that can improve ecological health and functionality to sustain the continued provision of ecosystem services and societal benefits associated with LIEs. However, such undertakings need to take place within a socio-ecological systems framework using an adaptive management approach (Fig. 2). A SES approach can be used to track freshwater releases from dams as a restoration action in LIEs. Objectives are set, actions are implemented, and then monitored and adapted using a learning-by-doing approach. The goal/action would be to release freshwater from dams to downstream estuaries to improve estuary health and societal benefits that can be measured using a range of indicators (Fig. 2). A SES approach ensures communication and co-ordination amongst all stakeholders.

Fig. 2
figure 2

A strategic adaptive management socio-ecological systems framework for the release of freshwater from dams to downstream estuaries to improve estuary health, ecosystem services and state of the societal system. Examples of estuary health and societal indicators given (adapted from Adams and Van Niekerk 2020)

Full size image

The social and cultural values associated with EFlows, including dam releases, are increasingly considered and require an understanding of ecology-culture relationships, as well as direct flow-culture relationships (Stein et al. 2021) to inform effective EFlows. These are described as ecohydrological principles that Wolanski and Hopper (2022) call for in the future management of the river basins (e.g., Burdekin River, Australia) to avoid duplicating the mistakes of the Murray-Darling River basin where water resources were not managed at the basin-scale. Notwithstanding the limitations, dam releases should be optimized as best as possible for ecological and societal benefits. Environmental flow assessments allow for the evaluation of a range of ecological and social outcomes as part of a series of scenarios/management options (Brown et al. 2020; Van Niekerk et al. 2019b). Governments and stakeholders can then assess options and negotiate the future they want. Engagement and inclusion of multiple agencies, ongoing community engagement and support from all users are essential to ensure the successful implementation of water releases from dams as indicated from the analysis of the case studies (Table 3).

Richter and Thomas (2007) proposed a similar framework for implementing dam reoperation that considers both ecological and social consequences. Addressing non-flow related impacts (e.g., land-use change, dredging, pollution, artificial breaching, over-fishing, invasive species) is also a critical consideration in holistic catchment to coast management, as well as restoration planning that aims to mitigate the risks associated with drivers of change (Arthington et al. 2018; Van Niekerk et al. 2019b; Chilton et al. 2021). New reservoir management strategies including targeted control of dam storage and flushing sediment operations, banning fishing activities, and removing unnecessary dams (obsolete or small dams) are becoming crucial tools for ecosystem restoration (Zhang et al. 2022). Water demand management through water reuse, recycling, rainwater harvesting and desalination can play a key role in the protection of baseflows to LIEs. Protection of groundwater resources against over-abstraction and/or poorly planned forestry activities are also essential for the persistence of baseflows to LIEs.

Long-term research programs are also needed to track the ecological and societal benefits of flow releases from dams to estuaries, as shifting baselines in response to climate change effects are expected. Mediterranean climates where most LIEs occur are becoming warmer and drier (Drobinski et al. 2020), potentially resulting in ecological regime shifts. Research and monitoring will improve our understanding of the role of extreme events (e.g., floods, coastal storms, and storm surges) and decadal oscillations on estuaries and how these events will be affected by climate change (Stein et al. 2021). Due to increased storminess, many estuaries along exposed, sediment-rich, microtidal coastlines will close to the sea more frequently (Adams and Van Niekerk 2020). This will increase the need for environmental flow allocations, including dam releases, to ensure downstream estuary health and ecosystem services. Similarly, restoration of blue carbon habitats includes the provision of EFlows flows (Adams et al. 2021). An investment in long-term monitoring of catchment signals and estuary responses is essential for proactive adaptive management. Climate change is predicted to have severe impact in many regions of the world that support LIEs (e.g., Van Niekerk et al. 2022). Of key concern is the predicted decrease in river flow and the likely increase in the frequency and duration of droughts. Flow releases from dams may mask climate change impacts to some degree resulting in water resource managers becoming complacent and not taking proactive measures to secure additional flows need to contend with drought conditions in a hotter and often drier climate.

Releases from dams can provide resilience and buffer against the impacts of current and future global change pressures (Sun et al. 2013; Chilton et al. 2021). However, releases to maintain downstream estuary health are threatened by increasing abstraction of water, competing water uses, over allocation and increasing droughts in response to climate change in some regions. Although estuaries have resilience towards natural droughts and floods, changes induced by increased water use and flow regulation can artificially extend or intensify droughts putting these systems under increased stress (Oliver and Webster 2011). During droughts, the reality is that dam releases are often not made due to competing water users as described for the Groot Brak Estuary and other case studies (Supplementary Material). To ensure the provision of baseflow during droughts a strong regulatory environment is needed. EFlow and associated dam releases, and the monitoring thereof, are mostly only implemented if requirements have been embedded in a strong regulatory framework that compels water resource managers and users to consider such environmental matters (Brown et al. 2020). Such regulatory frameworks should be backed by investment in monitoring and auditing to track compliance (Van Niekerk et al. 2022). Appropriate dam release practices need to be formally incorporated into environmental flow planning, dam design and operational implementation, and aligned with climate change adaptation strategies (Chilton et al. 2021).

Conclusion and Recommendations

There are a few cases globally that demonstrate the efficacy of environmental flow releases from dams in maintaining downstream estuary health and societal benefits. This assessment has shown that a holistic systems approach is needed to improve downstream health of estuaries particularly in LIEs. It requires not only flow releases from dams to increase freshwater inflow to estuaries, but also the management of water infrastructure (e.g., canals and off-channel storage facilities) to provide downstream inflow.

Releases of freshwater inflow from dams have many ecological benefits; however, in most cases, current dam releases can usually only provide baseflow input for LIEs and not flood releases that are important for maintaining long-term estuary health and function. The sustained implementation of dam releases forms an important component of environmental flow implementation that should be based on adaptive management and social engagement. Flow releases should follow the natural flow regime as far as possible, mimicking floods and droughts. This will require trade-offs between different water users and stakeholders. In most cases, such releases only occur if embedded within a legislative framework and planned from the start of dam construction. Societal commitment is needed to deliver environmental water over periods of drought.

From the findings of this study, we recommended key considerations for planning and implementation of future dam releases to LIEs. These can be integrated in EFlow or estuary restoration planning towards improved ecological health, as well as associated ecosystem services and societal benefits. The recommendations for future estuary dam release studies were:

  1. 1.

    Understand the whole socio-ecological system when restoring flows as both ecosystems and the communities can respond in unexpected ways and negate potential intended benefits

  2. 2.

    Release water from dams to mimic natural flow patterns in relation to biotic life cycles.

  3. 3.

    Supply baseflow to ensure salinity and other physicochemical gradients that improves biological productivity in estuaries.

  4. 4.

    Supply baseflow to maintain marine–estuary connectivity, while floods are important for catchment-estuary connectivity and estuary-floodplain connectivity.

  5. 5.

    Implement flood releases to remove sediment and accumulated organic matter, as well as prevent macrophyte encroachment.

  6. 6.

    Release frequent flow pulses as this is more effective than one large dam release in stimulating ecological response, e.g., biotic pre- and post-spawning migrations.

  7. 7.

    Use a SES framework to track freshwater releases from dams as a restoration action using an adaptive management approach.

  8. 8.

    Include multiple agencies to achieve outcomes; ongoing community engagement and support of users is critical for successful implementation.

  9. 9.

    Implement long-term monitoring and dedicated research programs to understand responses and inform a science-based management approach.

  10. 10.

    Strengthen the regulatory environment, especially in ensuring the provision of baseflows during droughts.

LIEs are particularly influenced by drought that will increase in frequency and duration with climate change. For many case studies, severe drought placed attention on the need for EFlows and funding was made available for monitoring and in some cases implementation of required EFlows. Unfortunately, this focus and funding are often withdrawn when the system goes back into a wet phase. In other examples, allocated and even legislated drought flows were not released from dams. It is important to monitor and report on EFlows continuously to understand responses. This study is considered timely as there are few published works integrating the responses of estuaries to dam releases so that lessons learned could be shared.