Addressing fish-passage issues at hydropower and irrigation infrastructure projects in Indonesia

Additional keywords: freshwater, sustainable development, tropical systems.

Large tropical river-floodplain systems, being increasingly developed for irrigation and energy requirements, are experiencing rapid fishery declines (Oldani and Baigún 2002; Ziv et al. 2012). Tropical river-floodplain systems have productive and diverse fish communities that provide environmental, social and economic benefits to humans (Dudgeon 2000). However, the extensive construction of water-management structures prevents fish from accessing spawning, feeding and nursery habitat, which can affect multiple life-history stages (Rosenberg et al. 2000; Dugan et al. 2010) and reduce spawning and recruitment (Pringle et al. 1988). Over time, fishery productivity falls, and development-project benefits may become negatively offset (Baumgartner et al. 2012).

Indonesia is an oceanic nation comprising 10 individual islands in the southern hemisphere and is situated north of Australia (Fig. 1). Like all tropical rivers, waterways in Indonesia face challenges from increasing populations and urbanisation (Amri et al. 2014). The current population of Indonesia exceeds 250 million people and is placing increasing pressure on water resources and local aquatic fauna. Furthermore, significant land-use changes arising from increased cropping, deforestation and urbanisation have two significant impacts (Remondi et al. 2016). First, they increase overland runoff in the wet season, leading to increased flooding frequency. Second, they draw down groundwater reserves in the dry season, leading to significantly decreased flows. Thus, human expansion is significantly affecting river hydrology across many parts of Indonesia. Floodplain Rivers in Indonesia are typical monsoonal streams characterised by high wet-season flows and low-discharge dry seasons. All rivers in Indonesia flow from headwaters to the sea.

Fig. 1.  Map of Indonesia. Locations of the two only operating fishways at Perjaya Dam (Sumatra Island) and Poso Dam (Sulawesi Island) are shown. All Indonesian rivers flow to the ocean and diadromy is a dominant life-history trait.

The inland-fishery sector is valued at US$266 million and supports thousands of jobs in subsequent supply chains (Coates 2002); protecting the fishery resource base from the impacts of irrigation infrastructure is of paramount importance in Indonesia. Inland capture fisheries in Indonesia deliver food security and income for rural households and also serve as a valuable source of protein and important micro-nutrients (Kottelat et al. 1993). In all, 1218 species belonging to 84 families have been reported from Indonesian rivers, including 1172 native species from 79 families; 630 species are endemic and 28 are exotic species (Wirjoatmadjo 1987; Christensen 1992). Exotic species are present and, in the context of fish passage, it is important to ensure that fish-passage solutions do not expand distributions and transfer additional stresses on native fish (Hubert et al. 2015). With a density of 0.6 species per 1000 km², Indonesia hosts one of the world’s highest density of fish species, ahead of Brazil (0.37 species per 1000 km²) and the Democratic Republic of Congo (0.48 species per 1000 km²), which are two countries that host some of the world largest and speciose tropical rivers.

Indonesian freshwater fish fauna are either obligate freshwater species (potamodromous) or diadromous species. Potamodromous fishes can complete all aspects of their life history in freshwater. These species generally do not require fish-passage solutions to maintain distributions, but they may be affected if barriers block migrations to important spawning sites or alter habitat to the point where species may no longer be supported. Very little data exist on potamodromous species in Indonesia. There is an urgent need to categorise these species, and understand habitat preferences and distribution to determine the potential impacts of river development.

Diadromous fish spend portions of their life histories in the ocean and are further categorised as anadromous, i.e. fish that migrate from the sea to freshwater to spawn, and catadromous, i.e. those that migrate from freshwater to the ocean (Augspurger et al. 2017). Examples of important anadromous fish in Indonesia include species of Polynemidae (e.g. Polynemus melanochir), Pangasiidae (eq. Pangasius krempti), Ariidae (e.q Arius sumatranus) and Sciaenidae (Boesemania microlepis). Common catadromous fishes include Anguillidae (e.g. Anguilla bengalensis, A. bicolor, A. marmorata, A. interioris, A. celebesensis, A. borneensis). The strong need for these species to connect between freshwater and ocean environments demands that any river development occurs in a way which preserves these important migrations (Miles et al. 2013).

Catadromous fish, such as eels, have important commercial and cultural values in Indonesia. In 2009, the international trade for eels exceeded 270 000 tonnes (Mg) (Bruinsma 2003). These were dominated by glass eel exports, which can fetch up to US$5000 kg^–1. However, resources of temperate anguillid eels are declining globally. The numbers of species such as Japanese eel (Anguilla japonica), European eel (A. anguilla) and American eel (A. rostrata) have decreased rapidly in recent years, and these species are now listed on IUCN red List (Jacoby and Gollock 2014). Compensating for losses in temperate eel fisheries, the harvest of tropical eel species has increased significantly. Trade statistics (see http://unstats.un.org/unsd/trade/, accessed 28 August 2018) noted that ~3593 Mg of tropical eel harvest originates from Indonesia, with an estimated market value of US$194 million (Brasor and Tsubuku 2012). Increasing fishing pressure on these species will require strict management.

Irrespective of the migration strategy, it is important that both future and existing riverine development projects consider potential impacts on fish. Construction of cross-river obstacles, such as dams, weirs and roads, as means for rapid development in response to increasing population and demand for agriculture products, hydropower generation or urbanisation, are major threats to the long-term sustainability of inland fisheries as any changes in migration, reproduction and biodiversity of aquatic populations has the potential to decrease productivity. Management interventions may be required to protect these important fisheries.

The energy sector of Indonesia is undergoing significant changes that are creating both challenges and opportunities (Hasan et al. 2012). Energy demand continues to expand with the nation’s economic growth. The energy industry is undergoing a profound change and more than 86 million people in Indonesia still lack basic access to electricity (Erinofiardi et al. 2017). Total energy demand in Indonesia is expected to increase by 8.7% each year up to 2024 (Erinofiardi et al. 2017). The challenge is to meet increasing demand while progressing towards sustainable energy systems, including the use of new technologies, renewable sources and increased efficiency. Many Indonesian rivers are high gradient with abundant water, so hydropower is becoming an increasingly attractive option as a major future energy source.

Hydropower constitutes ~10.2% of the total energy production of the country (Anonymous 2014). Indonesia’s technical hydropower potential is estimated to be at ~75 000 MW, with untapped resources being concentrated on the islands of Sumatra, Java and Sulawesi (Darmawi and Firdaus 2011). It is estimated that there is currently ~8 GW of undeveloped hydropower potential, which would provide almost 33 TWh of electricity per year. According to the national strategic plan, up to 1 GW of new hydropower capacity has already been developed or is planned in the next decade (Darmawi and Firdaus 2011). Seven hydropower stations totalling 1559 MW are currently operating and a further 10 projects totalling 1819 MW are subject to power purchase-agreement (PPA) negotiations. An additional 19 projects totalling 2131 MW are in the study or design phase (Anonymous 2016).

Globally, hydropower projects have affected fishery production (Williams 2008). In most cases, impacts can be separated into direct and indirect (Bilotta et al. 2016). Direct impacts are where fish are killed, injured or physically blocked by the dam or turbine operation. Fish can experience significant welfare issues during downstream migration, either from blade strike (Deng et al. 2007), shear stress (Cada et al. 1999) or pressure changes (Brown et al. 2014) experienced during turbine or spillway passage. Fish can be physically excluded from upstream habitat simply by the barrier effect of the dam (Dugan et al. 2010) and can often require costly mitigation measures to be implemented (Williams 2008). But hydropower dams also affect downstream flow regimes (through hydropeaking; Gostner et al. 2011) and also alter upstream habitat, changing it from fast-flowing lotic habitat to slow-flowing lentic habitat (Kanehl et al. 1997). So, following hydropower dam construction, fish are faced with two challenges. First, the dam and turbine operation creates physical issues for up- and downstream migrations. Second, the dam significantly alters flow regimes and habitat availability. Both can be somewhat mitigated, but need to be addressed during the early stages of the planning process.

The rapid and widespread expansion of the irrigation sector in Indonesia has placed substantial pressure on fishery resources, on which many people depend for food security and income. Between 1969 and 1989, extensive investment in irrigation and agricultural infrastructure (several thousand dams, weirs and diversion channels) supported a 4.8% increase in rice production per year, which led to a reduction in rural poverty from 40% in 1970 to 16.5% in 1997 (Booth 1977; Vermillion et al. 2000). But despite this increase in rice production, malnutrition (of children 2–5 years old) from a lack of micronutrients normally provided by fish is high. This is the case partly because inland fisheries form the main source of protein and income for many rural families (Melse-Boonstra et al. 2000; Kawarazuka and Béné 2011). But irrigation infrastructure has blocked important migration routes and led to declines in the resource base. For instance, the completion of the irrigation diversion at Perjaya Dam on the Komering River led to the localised extinction of 55 species in the upstream impoundment (Nizar et al. 2014).

Irrigation expansion was rapid in Indonesia and occurred at a time where impacts on inland fisheries were largely unknown. Infrastructure was essential to secure rice production, but has blocked fish seeking to access critical nursery and feeding habitat. In coastal streams, maintaining unimpeded access between freshwater and saltwater is critical for both inland and coastal species to complete their life cycles. Fish cannot access critical upstream habitat unless functional fishways are constructed. Thus, as seen at Perjaya Dam, any barrier that interrupts natural processes can lead to substantial, and rapid, fishery declines, which affects nutrition, jobs and income.

Many tropical rivers in Indonesia are still free-flowing but many are proposed for extensive dam development in the future (Pringle et al. 2000). However, the present understanding of dam-related impacts in Indonesian ecosystems is severely limited (March et al. 2003). Although future tropical dam construction may be limited by social, economic and environmental concerns (McCully 1996), increasing demand for electricity, water and flood control are leading to continued tropical dam construction (Olivas 2004; Sierra et al. 2004), especially given that most suitable sites for dams in temperate regions have already been developed.

Dam and regulator design and operation significantly affect fish history patterns in temperate systems (Poff and Hart 2002). Barriers to fish passage can effectively stop many fish species from breeding and re-populating waterways by restricting their ability to access spawning grounds (Lucas and Baras 2001). Fish attempting to negotiate barriers are also forced to use up precious energy reserves. If this occurs during a breeding event, fish may actually re-absorb reproductive tissue and effectively lose a breeding season. Long-term effects will be supressed population size. For instance, the construction of Tallowa Dam on the Shoalhaven River (Australia) led to the localised extinction of many catadromous species from upstream reaches (Gehrke et al. 2002). Only species with strong climbing abilities were able to maintain populations. In tropical regions, migratory species such as barramundi (Lates calcarifer) are unable to access impounded reaches and experience population declines (Gehrke et al. 2002). Further, migratory catfish has experienced significant declines, following damming, in tropical rivers in Brazil (Agostinho et al. 2008). A common mechanism to mitigate the impact of dam on fish is to provide technical fish-passage solutions (Godinho and Kynard 2009). In developed rivers, provision of bidirectional fish passage will be critical to the long-term persistence of many migratory Indonesian native fish.

Fishways are often used to offset negative effects caused by infrastructure in large tropical rivers (Baumgartner et al. 2012). These structures are effectively channels around or through a migration barrier that promote upstream passage. Commonly used designs include vertical slot, pool and weir and Denil passes (Stuart and Berghuis 1999; Baumgartner et al. 2012). However, despite many designs being available, including fish locks, nature-like bypasses and bypass channels, determining effectiveness, especially in tropical systems, is often overlooked (Schwalme and Mackay 1985; Foulds and Lucas 2013). Much existing knowledge exists mainly for temperate species either from laboratory-based trials (Mallen-Cooper 1992) or in situ field-based experiments (Baumgartner et al. 2012).

Fishways are best addressed during the design phase (Clay 1995). Once a hydropower site has been determined, a fishway design should be progressed alongside plans for civil works. Technical fishways designed for upstream passage must take into consideration local hydrology, local fish ecology, swimming capabilities, hydraulic preferences and physical characteristics of the dam wall (Clay 1995). Further, the entrance location must be situated at the upstream migration limit and internal fishway hydraulics must suit the swimming ability of the target species (Bunt 2001). Developing robust solutions requires that fish ecologists, dam operators and engineers should work collaboratively to design effective solutions (Baumgartner et al. 2014a).

Passing fish downstream is also a challenge at hydropower and irrigation infrastructure (Cada et al. 1999). Indonesia has an abundance of diadromous fish. So any fish that move upstream into the dam reservoir must be provided the opportunity to return to the ocean to complete essential life-history stages. Fish generally move downstream via the spillway or turbine, or specially constructed downstream facilities can be provided for fish (Baumgartner et al. 2014b). Passage via the turbine can be associated with increased injury and mortality arising from blade strike, barotrauma or shear stress (Deng et al. 2005; Brown et al. 2012a, 2012b). In many instances, passing fish upstream can be pointless unless functional downstream facilities are also constructed (Williams 2008; Pelicice et al. 2015).

Establishing that both hydropower and irrigation development can have fishery-related impacts, the following key question remains: when should fish passage be considered as a fishery-mitigation option? Many factors influence the appropriate answer.

First, can an efficient fish pass be constructed? There is mounting global evidence suggesting that, in some instances, providing fish passage may do more harm than good especially if it works suboptimally (Agostinho et al. 2002). Not passing enough fish, or only passing certain size classes may not yield the anticipated results. On the Atlantic coast of the USA, inappropriate application of inadequate technology has not yielded the required fishery outcomes (Brown et al. 2013). In Indonesia, it is extremely important to ensure that appropriate fish-passage designs (both upstream and downstream) are based on criteria suitable for local species. Furthermore, it is very important that the best solution is found for each site. Fish-passage construction is a very precise science. Even a small, seemingly insignificant, design change can lead to significant decreases in efficiency. Compromising too much on design specifications, to save costs or otherwise, can lead to suboptimal fishery outcomes.

The second question is whether the quality of the upstream habitat is suitable? For instance, in Neotropical rivers of South America, it has been suggested that passing fish upstream into unsuitable habitat actually has a negative effect on populations (Pelicice and Agostinho 2008). For instance, many species that have abundant spawning habitat downstream of barriers are disadvantaged when moved upstream into slow-flowing reservoir environments where spawning habitat is limited (Pelicice et al. 2015). In some instances, where fish passes have been constructed, and fish are recorded moving through, there is simply no evidence of overall fishery recovery (Oldani et al. 2007).

The third question is whether the downstream passage is possible? A main argument is that fish migration is often cyclic. An upstream migration may require a subsequent downstream movement (Baumgartner et al. 2014b). Fish may move upstream as adults, spawn and then both adults and larvae may undertake a subsequent return migration (Agostinho et al. 2011). The critical point is that, in progressing a suitable fish-passage solution, the complete life-history strategies must be known and understood for the target species (Kraabøl et al. 2009). A challenge in Indonesia will be sourcing this critical life-history information and then applying it to river-development works through effective designs that are proven in the local context.

The fourth question is whether it is acceptable to decide not to proceed with fish passage? If it is determined that a suitable design is not available, or if upstream habitat is unsuitable or if mechanisms to provide for downstream passage cannot be found, then an appropriate decision may be to not proceed with a fish-passage solution at all. Indeed this occurred in Australia when the government considered the construction of Traveston Dam (Mary River, Queensland; Walker 2008). It was determined that no adequate fish-passage solution could be found for local species and the dam was not approved to proceed. In the USA, areas where inadequate passage has failed to rehabilitate fishery populations has led to significant efforts to decommission and remove dams altogether (Bednarek 2001; Poff and Hart 2002). Such decisions have yielded far better fishery rehabilitation outcomes than have fish-passage programs. However, these are extreme cases, although extremely relevant to Indonesian river development. Hydropower and irrigation construction should be accompanied by environmental-impact considerations and fishery-related impacts should be considered during the design stages of planned projects. Such planning is the only way to ensure that the adequate solutions are found.

Assuming that due diligence has taken place and a need for fish passage is agreed, then it is essential that appropriate mitigation measures are developed and implemented in the local context. Fishways have been constructed worldwide; however, knowledge of the importance and effectiveness of these structures is very limited. It is now accepted that fish-passage solutions should consider local species and conditions (Baumgartner et al. 2012), and not be adopted from studies conducted elsewhere (Mallen-Cooper and Brand 2007).

The Indonesian government has emphasised infrastructure development and modernisation as a major priority. In 1987, the Indonesian government adopted an irrigation operations and maintenance (O&M) policy (Bruns 2004). This included efforts to ensure adequate funding for O&M and the introduction of irrigation service fees. Inadequate uptake of the policy led to unsatisfactory maintenance of existing infrastructure. Although Indonesia achieved self-sufficiency in rice production in 1984, this is presently declining because of inadequate maintenance, rapid deterioration and loss of productivity of irrigation systems (Mears 1984). Thus, irrigation policy is presently being reformed and the majority of irrigation infrastructure is due for urgent upgrade over the next 15 years. These proposed works offer substantial opportunities to provide for fish because improved fishery outcomes (e.g. through the facilitation of fish passage) can be factored into upgrade works.

Because of a scarcity of information about the effectiveness and importance of fishways, only two have ever been built in Indonesia, these being Perjaya Irrigation Dam, on the Komering River, Sumatra Island) and an eel fishway (Poso Hydroelectric Scheme, Sulawesi Island; Fig. 1). However, these were based on solutions developed in North America and were potentially not suitable in the local context. The effectiveness of Perjaya Dam fishway was limited to a single study seeking to quantify species passage (Nizar et al. 2014). Very few species were captured and it was reported that the local fishway, a pool and weir design, was trapped on a daily basis, by local fishermen. No fish are, subsequently, gaining passage to the upstream reservoir. Nevertheless, there are several design flaws in the existing facilities. First, the entrance is located a significant distance downstream from the weir face and would be difficult for fish to locate (Fig. 2, 3). Second, the internal baffle design was based on those for salmonids and no current information is available on fish swimming abilities in the Komering River. Third, the fishway exit is directly adjacent to a large, unscreened irrigation diversion (Fig. 2). It is highly likely that fish exiting the fishway would be diverted into the irrigation canal and be effectively ‘lost’ from the system.

Fig. 2.  Aerial view of Perjaya Dam, showing location of (a) fishway entrance too far down from dam face, (b) fishway exit too close to irrigation scheme intake (image sourced from Google Earth Pro).

Fig. 3.  Close up view of Perjaya fishway, demonstrating that both fishing nets located downstream of the fishway exit and the location of the entrance significantly downstream of the regulator gates.

The fishway at Poso Dam is far more complex (Fig. 4). There was concern that Poso Dam would have significant impacts on local eel populations. Eels fetch a high price for local villagers who export them to a lucrative Japanese market (Honda et al. 2016). Provision was, therefore, made for a short fishway connecting the lower tailrace to the former Poso River, which was diverted to make provision for the Poso hydroelectric project. Conceptually, the fishway is sound. It represents a short section of the channel to allow elvers to pass a major barrier and recolonise upstream habitat. However, from an ecological perspective, the main issue is related to downstream passage. Eels generally migrate long distances to spawn in deep ocean trenches and would return seaward after spending many adult years in the Upper Poso catchment (Honda et al. 2016). However, there are no downstream migration facilities included into the Poso Hydroelectric design. So seaward-migrating broodfish risk injury or mortality when passing through the turbine blades. Therefore, best efforts to provide upstream migration could significantly affect the eel fishery through impacts on breeding individuals.

Fig. 4.  Aerial view of Poso Hydroelectric Dam, demonstrating how the fishway passes fish into the former Poso River. Most flow now passes through the hydropower plant and no downstream passage facilities were included in the scheme construction (image sourced from Google Earth Pro).

In light of design deficiencies at both Poso and Perjaya Dams, the Indonesian government is urgently seeking advice on the importance of fishways to increase the sustainability of fishery resources and guide the irrigation modernisation program. It is expected that each new irrigation structure will include a fishway, but there are presently no formal criteria for fish passage in Indonesia. Thus, there is substantial concern that a one-in-a-generation opportunity will be lost if thousands of irrigation structures are upgraded without considering potential fishery-productivity benefits. The political and industry climate is perfect to boost fishery productivity through the application of fish-passage technology; however, targeted research and development are needed to ensure that sustainable opportunities are developed and implemented within the next 5 years.

A stepwise approach is recommended to further fish-passage advancements in Indonesia. A detailed understanding of the existing two fishways is required to determine applicability to other sites in Indonesia. Specifically, a detailed monitoring plan should be implemented, with the objectives of defining which species are migratory, how well the existing fishways perform, whether any modifications can be made to improve performance, implementing local fishery management plans to prevent exploitation from within the fishways and a concerted effort should be made to identify the ecological life-history traits of target species. Understanding which species have potamodromous or diadromous life-history requirements is important, to determine the impacts of broader river development.

A concerted effort should be made to understand swimming abilities of local fish species. Swimming ability is a critical factor that governs effective fishway design (Castro-Santos 2005). It is specifically important to understand critical swimming speeds and the impact of turbulence on passage success. These two factors can be controlled by effective design specifications and, if known, could be developed into a set of national standards to govern future fishway construction.

Ensuring that fish-passage provision is captured in national policy is essential for effective fishery-recovery programs. There are many global examples where policy changes have been implemented to ensure that riverine development effectively provides for fish passage (Odeh 1999; Calles and Greenberg 2009; Godinho and Kynard 2009; Kemp and O’Hanley 2010; Baumgartner et al. 2014). Specifically, fishery-management policy should recognise obstruction of migration routes as a key threatening process. Further, ensuring that fish passage is mandated in irrigation and hydropower policy is essential to ensure that fish are considered early in the planning stages of infrastructure projects (Larinier 2008). Often, fish are included as an afterthought, usually after a decline has occurred. However, it is always less costly to consider fish during the design phase and ensure that fishery considerations are integrated into scheme operation over the long term.

Finally, fish passage is a two-directional phenomenon. Whereas substantial effort is usually made to ensure that fish are able to migrate upstream at barriers, it is just as important to move fish downstream. This is particularly relevant for diadromous species such as eel, which has an obligate requirement to migrate downstream to spawning grounds. Understanding how the operation of hydropower and irrigation schemes affects downstream migrants is paramount to developing effective solutions. In areas where mortality or injury rates may be high, such as through undershot sluice gates (Baumgartner et al. 2006) or in hydropower turbines (Ferguson et al. 2006), it may be worthwhile considering physical barriers such as screens to divert fish away from areas of welfare concern (Baumgartner and Boys 2012).

Dr Baumgartner is an Associate Editor for Marine and Freshwater Research and is the Handling Editor for this special issue. Despite this relationship, he did not at any stage have Associate Editor-level access to this manuscript while in peer review, as is the standard practice when handling manuscripts submitted by an editor to this journal. Marine and Freshwater Research encourages its editors to publish in the journal and they are kept totally separate from the decision-making processes for their manuscripts. Dr Wibowo declares that he has no conflicts of interest.