Integrating Fish-Friendly Hydropower Solutions with the Nature Restoration Policy Through River Barrier Modification

Abstract

The recently adopted EU Nature Restoration law emphasises the urgent need to address the ecological impacts of river barriers, which fragment habitats and disrupt natural flows. However, efforts to remove barriers are often constrained by prohibitive costs, regulatory hurdles, and public opposition. In Ireland, barrier removal costs range between EUR 200,000 and EUR 500,000 per structure, representing a substantial financial burden given that more than 73,000 barriers are identified nationwide. Although removal would restore ecological function, it would also eliminate the potential to repurpose these structures for hydropower, thereby reducing opportunities to contribute to the national target of 80% renewable electricity generation by 2030. This study outlines the development of a river barrier modification system to serve the dual purposes of upstream and downstream fish lift over barriers and generation of electricity for local consumption using a fish-friendly pump-as-turbine unit. Under normal flows, the unit generates electricity while during low flows it operates in pumping mode to enable fish passage. A prototype was fabricated and tested at a fish farm using both artificial and live fish. An assessment of the regional potential was also extrapolated from preliminary results suggesting that the BMS offers a cost-effective alternative to full barrier removal, potentially offsetting costs by 50–85% while contributing to both EU restoration targets and national renewable energy goals.

1. Introduction

The recent enactment of the European Union’s (EU) Nature Restoration Regulation (NRR) proposes to restore at least 20% of the EU’s land and sea areas by 2030 and repair all ecosystems that are in need of restoration by 2050. This highlights the EU’s determination to restore ecosystems for improved habitats and mitigation of the impacts of climate change. The NRR ensures the implementation of the EU 2030 biodiversity strategy by setting legally binding targets to restore nature.

The EU biodiversity strategy calls for greater efforts to restore freshwater ecosystems and the natural functions of rivers [1]. In addition to calling for better implementation of existing legislation on freshwater, the biodiversity strategy sets the target to make at least 25,000 km of rivers free flowing again by 2030, by removing primarily obsolete barriers and restoring floodplains and wetlands. In Ireland, despite its relatively small size, there are over 73,000 identified barriers nationally [2] with 2000–7000 estimated to require mitigation/modification [3].

Complete removal of existing barriers is the preferred approach within the NRR; however, in most cases modification of existing barriers will be implemented to achieve nature restoration. This is due to several challenges facing complete removal of barriers such as ownership of the barriers, the need for access through adjacent land, protecting architectural heritage, and public opposition. Complete removal of a barrier in its entirety is a simpler process in terms of demolition practices than barrier modification; however, it is a complex undertaking to obtain the required regulatory permissions to do so. There are over eleven different environmental and heritage focused surveys that are required to be undertaken with satisfactory outcomes, prior to barrier removal or modification. These include land, archaeological, architectural, hydro-morphological, hydraulic, pearl mussel studies, contaminated sediments assessment, etc. This whole process of obtaining permission for barrier removal can take up to two years to complete at a significant cost. Should the outcome of any of the eleven required surveys prove negative then barrier removal cannot proceed, and instead barrier modification may be adopted. Furthermore, the aforementioned issue of ownership and consent for land access on both sides of the river are separate issues which are equally challenging. In addition, opposition to barrier removal from the general public or from interest groups may also arise, posing further challenges.

While the EU is aiming to restore nature, it is also aiming to reach 45% penetration of renewables and 55% greenhouse gas emissions reduction by 2030, including through the generation of hydropower. The Irish government has set a target to increase its electricity generation capacity from renewable sources to 80% by 2030. The removal of barriers will be detrimental to the attainment of such targets, since the adoption of sustainable hydropower, as one of the oldest and largest sources of renewable energy, will be highly affected. Prior to the enactment of the NRR, hydropower could be a viable investment in run-of-river schemes by utilising existing barriers which did not contain hydropower turbines. Quaranta et al. [3,4] estimated 1.6 TWh/y in energy potential was available using hydropower from water wheels in old mills and 5.2 TWh/y from other historic weirs across the EU.

A major reason why hydropower projects fail is the often high cost of consenting (obtaining licences, permissions, etc.), which can be of the order of 50% of the total project cost [5]. This can also take a significant length of time to complete. Where a barrier already exists, these costs can be reduced, making hydropower projects more viable. However, the enactment of the NRR could lead to removal of such barriers in the future, greatly increasing the risk to such energy investments. This may reduce future growth of micro-hydropower renewable energy and might also lead to the demolition of existing installations. This poses an important challenge as hydropower is essential for the provision of grid flexibility in the face of increasing solar and wind production, and the policy drive for electrification of transport and heat.

In order to create a balance between compliance with the NRR and meeting the targets of Directive (EU) 2023/2413, a multifaceted approach to barrier removal/modification is required. This approach should ensure that the benefits of river connectivity to upstream and downstream reaches are achieved without eliminating the benefits of potential hydropower installations in the localities of these barriers, which in most cases are rural areas which can be isolated from the grid.

This paper presents one such approach where the two significant and connected challenges relating to the modification of river barriers and renewable energy generation, especially in rural areas, are combined into an opportunity to support the implementation of the NRR while generating green energy through hydropower. A river barrier modification system is outlined where existing river barriers can be modified to ensure their effects on fish migration and sediment transport are mitigated to benefit both upstream and downstream reaches, while retaining some of the benefits of the existence of the barrier to local communities via green energy production.

In Section 2, the proposed approach is described in detail, followed by the estimation of its renewable energy potential in Section 3. Section 4 describes the details of the conducted fish impact assessment in ascertaining the fish-friendly nature of the proposed barrier modification system. This is followed by the concluding remarks in Section 5.

2. Barrier Modification System Design

Central to the proposed barrier modification system (BMS), is a fish-friendly centrifugal screw pump capable of generating power in its reverse operation as a turbine (Screw-PAT) and of pumping fish upstream in a safe manner. The Screw-PAT relies on the pump-as-turbine (PAT) concept whereby conventional pumps are operated in reverse as turbines [6]. PATs are well known in the field of micro-hydropower energy generation and are associated with several advantages over conventional hydropower machines at this micro power output scale. In particular their low-cost nature is a key advantage due to mass manufacturing [7], while their lower overall efficiency and lack of flow regulation capabilities are known disadvantages of these machines. Extensive research has addressed the conversion of centrifugal pumps to turbines and long-term operation of PAT-based micro-hydropower systems [8]. Most studies, however, have concentrated on single-stage centrifugal pumps, given their broad coverage of flow and head capacities, while the application of fish-friendly centrifugal screw pumps has received little detailed investigation to date.

Centrifugal screw pumps, shown in Figure 1, were developed in the 1950s for the safe transport of live fish and have been in used extensively in the fisheries and aquaculture sectors. It has been shown to be fish-friendly in testing with a minimum ratio of injuries lower than 3% during pumping of eels [9] where it also resulted in no latent mortality of fish. In a separate study, Rodgers and Patrick [10] used a centrifugal screw pump in conjunction with filtered mercury lighting to attract, concentrate, and transport fish. They found that both attraction and transport mortality varied significantly among species with rainbow trout (Salmo gairdneri) significantly hardier than yellow perch (Perca flavescens) and alewife (Alosa pseudoharengus). Higher pumping speeds were also found to generally increase mortality with speeds lower than 600 rpm generally yielding mortality rates less than 10%. However, Patrick and Sim [11] figured out that in the live transfer of American eels (Anquilla americana) under laboratory conditions, there was no significant increase in mortality with an increase in pump impeller speed (890–1204 rpm) or eel density (low–high). It is hypothesised here that given the fish-friendly nature of this device in pump mode, that in reverse as a turbine it will also be fish-friendly. However, this remains to be proven.

Figure 1. Centrifugal screw pump impeller design [12].

The performance curves of 42 fish-friendly centrifugal Screw-PATs in pump mode were collected and their turbine performance was predicted [13] using one-dimensional techniques for the conversion of pump performance into turbine performance [6,7,14,15]. Figure 2 shows the obtained performance curves and the comparison of their operation region with other types of PATs. The results indicate that the Screw-PAT models can be operated in reverse as turbines under head and flow ranges of 10–100 m and 2–1650 L/s, respectively. The results also show a potential power generation of 2–750 kW per Screw-PAT at efficiencies up to 82%.

Figure 2. (a) Characteristic Q-H curve predictions for the 42 Screw-PAT models; (b) Operating region in comparison with other PATs. Each colored line refers to a single pump model used in the analysis.

However, the one-dimensional PAT conversion techniques developed by several authors [7,14,15] and those used in the field of PATs in general, are overwhelmingly based on the measured performance of single-stage centrifugal pumps. The accuracy of the conversion techniques from pump to turbine mode have been shown to lie within 8–10% for single-stage centrifugal pumps, and this gives sufficient accuracy for hydropower plant design. However, a large uncertainty exists over the level of accuracy that these prediction methods can offer for centrifugal screw pumps considering their significantly distinctive impeller geometry, flow characteristics, and hydraulic behaviour. Further experimental and/or numerical simulations are required to validate the predictions shown in Figure 2. These are not conducted in the current study as its objective was primarily the design and assessment of the BMS, while the wider study of Screw-PATs remains for future work.

The design of the BMS integrates the Screw-PAT within an existing barrier installed in reverse and combines its pumping and power generation capabilities to allow upstream and downstream fish movement and sediment transport. The pump outlet faces the upstream direction such that water will naturally flow backwards through it, allowing it to behave as a turbine without electrical energy inputs. In PAT mode the pump motor acts as a generator to produce electricity, while in pump mode the motor will be driving the pump to transport fish, thus facilitating the upstream fish movement while consuming electricity. The motor and Screw-PAT are contained within a stainless-steel frame to support the elongation of the shaft connecting the motor and turbine. This is conducted to ensure the motor (and control system) are raised above the high-water level within the river.

Figure 3 shows the conceptual model of the BMS integrated within an existing low-head barrier and consisting of the Screw-PAT, fish intake system, control system, and an energy storage system. The purpose of the fish intake system is to attract downstream fish that congregate at the foot of the barrier so that they can be pumped upstream in sufficient numbers. The control system incorporates algorithms to manage alternation between pumping and turbine mode, control of the pump/turbine during operation (RPM, flow), and data acquisition for monitoring the efficacy of the system. Since pumping consumes energy, the energy storage system is required to act as a power bank for the times when the BMS will be required to pump, lowering or eliminating the need to tap into the grid for such where a grid connection is available.

Figure 3. Barrier modification system—conceptual design. The arrow shown in the Screw PAT indicates a flow streamline in pump mode.

The control function starts with the detection of fish at the fish intake system through an installed sensor suite comprising optical and infrared cameras. When an adequate number of fish has been identified in the fish trap system, an activation signal will be initiated that starts by decelerating the turbine rotor, reversing the torque, and accelerating the pumping mode. When all the fish that were in the fish intake have been moved upstream, the restore-turbine-mode signal is initiated. This ensures that pumping is stopped and water is allowed to freely flow back through the turbine, rotating the rotor and restoring power generation.

3. Renewable Energy Potential

The use of existing barriers in rivers is central to the feasibility of new run-of-river micro-hydropower projects amid changing energy dynamics as wind and solar gain lots of attention in the renewable energy space. With hydropower being the oldest and largest source of renewable energy, one of the hurdles to the advancement of hydropower remains its impacts on the environment, including potential harm to fish populations [16]. However, micro scale hydropower is considered a more environmentally friendly option due to the use of existing barriers and utilisation of micro-hydropower turbines [4].

While over 73,000 barriers have been identified in Ireland, data on the height and available flow conditions at each of these sites is relatively limited. The Reconnect Project [2] created and collated a database of 373 barriers to fish migration in Irish rivers in various sub-catchment areas. A total of 141 barriers in this database were weirs as per Table 1. This database provides information on barrier heights to aid in the estimation of hydropower potential.

Table 1. Number of investigated weirs per sub-catchment area [2].

Sub-Catchment Area	Number of Weirs
Moynalty	5
Annalee	5
Erriff	2
Glashaboy	5
Dodder	110
Nanny	7
Owenboliska	1
Bann	2
Clodiagh (upper)	4
Total	141

Another important parameter required for the estimation of the hydropower potential is the streamflow rate at the specific site, which was not available for all the 141 studied sites. This therefore had to be estimated and was carried out by extrapolation from related sites with known flow rates. The spatially lumped version of the EXP Hydro model [17,18,19] was used to predict daily streamflow at the 141 weir sites using daily precipitation, air temperature, and potential evapotranspiration data from the Europe-wide 0.1° gridded Copernicus E-OBS dataset [20]. The EXP Hydro model is a simple daily time-step rainfall-runoff model that conceptualises the catchment as a bucket store whose water balance equation is as follows [17]:

$${\displaystyle \frac{dS}{dt}}=P_r+M-ET-Q_{bucket}-Q_{spill}$$

(1)

The variable S is the water stored in the catchment bucket, P_r is the precipitation that falls as liquid rainfall, and M is the snowmelt that occurs from the snow accumulation store. ET is the evapotranspiration, Q_bucket is the runoff generated based on the available stored water in the bucket, and Q_spill is the capacity excess precipitation and/or snowmelt that is available to infiltrate into the catchment bucket when the storage S has reached full capacity S_max.

Historical streamflow data for the 10-year duration ending on 31 December 2022 for sites with gauging stations were obtained from either the Environmental Protection Agency or Office of the Public Works (in Ireland). Using the area–discharge method the daily streamflow was deduced from stream gauge locations to the individual weir sites [21,22]. Figure 4 shows the location of the investigated weirs and streamflow gauging sites for the Dodder sub-catchment area. This contained most of the investigated sites outlined in Table 1. The resultant available power time series data for each weir site is provided as Supplementary Material S1.

Figure 4. Map of some of the studied barrier sites in the Dodder sub-catchment area.

In Figure 5 the breakdown of the hydropower potential by output power at the investigated weir sites is shown, assuming mean flow conditions. It is apparent that about half the existing weirs investigated are not suitable for hydropower generation as part of the BMS due to their lower height and streamflow combination resulting in potential power output of less than 3 kW.

Figure 5. Breakdown of hydropower potential by output power for the investigated weir sites.

Even though this analysis led to elimination of many of these sites as potential micro-hydropower installations, most of the sites making up the remaining 50%, with power above 3 kW, present a promising regional result. These are potential contenders for modification as per the NRR due to various reasons that could make complete removal impossible such as aesthetic appearance, historical significance, ownership rights, public opposition, etc.

An analysis of the available potential hydropower energy across these sites revealed an annual energy generation potential of up to 13.07 GWh with an average barrier generating over 92,000 kWh annually. An 80% operating time in turbine mode was initially assumed, allowing for the dual use nature of the system, i.e., pumping during 20% of operation time to allow for fish movements. Assuming savings generated from the mean domestic energy price of 35.83 EUR cent/kWh, the average modified barrier would produce in excess of 19,925 EUR/year. With an expected average lifetime of the BMS of over 15 years, this would thus reduce the cost of barrier modification by EUR 298,881 on average, which is 50–85% of the typical barrier modification cost.

Table 2 presents a sensitivity analysis of these results whereby the split between turbine and pump mode operation was varied from 85:15 to 55:45. This was carried out as the typical split between operating modes is unknown and would depend on the number and species of fish present. Upstream and downstream fish migration periods can vary. Some examples include the following: Atlantic salmon (Salmo salar) typically migrate upstream from January to September, but there is an overlap with downstream movements of both juvenile and smolts from March to October. Sea trout (Salmo trutta) migrate upstream from July to January and downstream from March to June. European eels (Anguilla, anguilla) and lamprey species (Petromyzon marinus, Lampetra planeri, and Lampetra fluviatilis) migrate throughout the year at various stages of their lifecycles [23].

Table 2. Sensitivity analysis of the barrier modification cost (€) offset based on various streamflow conditions and proportion of time operated as a turbine.

		Generation Time Proportion
		0.85	0.80	0.75	0.70	0.65	0.60	0.55
Streamflow variation (EUR)	Q30	448,863	384,740	320,617	256,493	192,370	128,247	64,123
	$\overline{\mathit{Q}}$	348,695	298,881	249,068	199,254	149,441	99,627	49,814
	Q50	256,660	219,994	183,328	146,663	109,997	73,331	36,666
	Q75	50,984	43,701	36,417	29,134	21,850	14,567	7283
	Q95	8532	7313	6094	4876	3657	2438	1219

$\overline{\mathit{Q}}$ Moreover, the sensitivity analysis also shows the potential barrier modification cost offset under varying streamflow conditions. $\overline{Q}$ represents mean streamflow conditions, whereas Q_n refers to the streamflow value that is exceeded n percentage of times, with n taking the values of 30, 50, 75, and 95. For the mean streamflow case, reducing the BMS turbine mode operation from 80% to 75% yields a 16.67% reduction in the potential barrier modification cost offsets. Further 5% reductions in operating time results in 20%, 25%, 33.33%, and 50% reduction in potential modification cost offsets. This demonstrates that when the time the BMS operates in turbine mode is reduced the potential cost offsets is eroded, thus becoming expensive to operate.

Case Studies

To demonstrate the available renewable energy generation potential from a typical BMS installation, three weir sites from the Dodder sub-catchment area are considered, namely Weir1, Weir2, and Weir3 in Figure 4. Weir1 is located on the Owendoher river just upstream of its confluence with the Glendoo Brook river. This weir presents a head of 1.4 m and a mean streamflow value of 0.37 m³/s. The mean available power generation potential stands at 5.1 kW. Annually, this would equate to an average energy generation capacity of 35,741 kWh from the BMS and a power consumption of 8935 kWh assuming 80% turbine mode operation. This equates to an annual net income of EUR 9605 which yields a cost saving of EUR 144,075 over the entirety of its 15-year life span.

Weir2 is located on the upstream end of the river Dodder just downstream of the Glenasmole reservoir. It is characterised by a head of 3.65 m and an average streamflow value of 0.41 m³/s. The estimate of the average available hydropower potential at this site is 14.7 kW. Installation of the BMS at this site with 80% turbine mode operation time in a year will generate 103,018 kWh of renewable energy while consuming 25,754 kWh. Ultimately, this results in an annual net income of EUR 27,684 and cost saving of EUR 415,260 over 15 years.

Lastly, Weir3 is the Orwell waterfall that is in the river Dodder downstream of the Waldron’s Bridge. The weir is characterised by a head of 3 m and an average streamflow value of 1.98 m³/s. Estimating the average available hydropower potential yields a value of 58.3 kW. This translates to an annual generation output of 408,566 kWh of clean energy from the BMS while consuming 102,142 kWh of energy in the same duration assuming the system is operating 80% of the time as a turbine. This equates to an annual net income of EUR 36,597 yielding a cost saving of EUR 548,955 over its life.

The analysis of these three weir sites in the Dodder sub-catchment demonstrate that the BMS has the potential of generating meaningful renewable electricity while maintaining ecological function. Even at modest sites (e.g., Weir1 with 5.1 kW potential), the system achieves a positive net income, while larger sites such as Weir2 and Weir3 offer substantially higher outputs and long-term savings. Across all cases, the BMS shows the capacity to offset modification costs, generate consistent revenue and contribute to renewable energy and nature restoration targets, confirming its technical and financial feasibility at varying scales of hydropower potential.

4. Fish Impact Assessment

A key feature to the wide acceptance of the BMS is confirmation of the fish-friendly nature of the Screw-PAT. The centrifugal screw pump has been in use in the fisheries and aquaculture sectors since the 1950s for the safe transport of live fish. The device has been shown to be fish-friendly in testing with a minimal ratio of injuries lower than 3% during pumping of 2300 eels [9]. Significantly, it has also been shown to result in no latent mortality of fish [9]. In reverse mode as a turbine the Screw-PAT requires to be assessed for its fish impact. In the current study, preliminary assessments of the fish-friendliness of the Screw-PAT were undertaken.

Field testing was conducted at the Inland Fisheries Ireland (IFI) fish farm in Roscrea, Ireland in a concrete rearing pond shown in Figure 6a. Water enters the pond at one end (upstream) and flows through the pond, exiting at the other end (downstream). The internal dimensions of the pond measured 32.4 m in length, 4.6 m in width, with a maximum dept of 1 m. There was a dividing concrete wall in the centre of the pond at 16 m, separating the upstream and downstream parts of the pond. Within the dividing wall there were two 0.9 m wide openings. One of the openings was blocked and the Screw-PAT was installed in the other. This allowed the levels of the upstream and downstream sides to be managed during the tests.

Figure 6. (a) Concrete fish rearing pond layout view (dimensions in mm); (b) Field test setup of the BMS for fish behaviour assessment.

The smallest available centrifugal screw pump model was adapted for reverse operation as shown in Figure 6b. Table 3 outlines the key characteristics of the chosen model. The smallest available model was chosen such that it could be accommodated at the test site, incorporating live fish to prove the concept of a Screw-PAT, but the site had limited flow and head conditions in comparison to the predicted performance which resulted in the Screw-PAT not rotating, partly confirming the predictions from the one-dimensional techniques. The envisaged turbine performance characteristics are also presented to provide a full description of the model.

Table 3. Main parameters of the test Screw-PAT.

Parameter	Pump Mode	Turbine Mode †
Nominal discharge (L/s)	12.36	20.74
Nominal head (m)	4.55	10.68
Efficiency (%)	54.35	54.35
Power (kW)	1.01	1.18
Free passage (mm)	75	75
Suction nozzle (mm)	100	80
Discharge nozzle (mm)	80	100

^† Estimated values.

Preliminary fish-friendliness assessments were conducted to investigate the following: (i) possible fish injuries through the Screw-PAT; and (ii) the behaviour/reaction of fish to the on/off operation of the system. For part (i) a synthetic sensor fish was used instead of live fish for ethical considerations, since the reverse operation of the Screw-PAT had never been tested for fish-friendliness to date. The ARC 800 sensor fish (Advanced Telemetry Systems) is a small neutrally buoyant autonomous device that analyses the physical conditions fish experience as they pass through dams and other hydro structures. It measures 90 mm × 25 mm and weighs 43 g and, during its passage through a hydro unit, measures acceleration up to 200 times the force of Earth’s gravity (200 g), rotational velocity on three axes, temperature, and pressure. A smaller sensor fish was not available. The sensor fish was passed through the Screw-PAT in both pumping and turbine mode ten times and its sensor data was analysed.

Figure 7 shows the typical results of the ARC 800 deployment. A number of possible strike events are indicated where acceleration peaked to above 95 g, which is an experimentally obtained threshold using both sensor fish and juvenile Chinook salmon [24]. Strike events are classified as collisions when 70% of the absolute value of the maximum amplitude lasts less than 0.0075 s. When 70% of the absolute value of the maximum amplitude of the event is longer than 0.0075 s, the event is classified as a shear event. Changes in pressure and rotational velocity are then used for validation of the classification as pressure and rotational velocity increase more dramatically during a collision event than during a shear event [24]. These may be indicative of potential injury to fish but may also be due to the small size of the Screw-PAT used relative to the size of the sensor fish (90 mm × 25 mm sensor fish vs. 100 mm inlet diameter Screw-PAT).

Figure 7. Results of the deployment of the ARC 800 sensor fish in pump mode, showing (a) ARC 800 fish sensor; (b) acceleration, pressure, and possible strike events—possible strike event E₁ approximately occurs at the PAT inlet, E₂ and E₃ inside the screw impeller, while E₄ at the 90 degrees pipe bend towards the outlet; and (c) rotation in degrees per second along the X and Y axis.

Further investigations are required of the fish-friendliness of the device on this basis using a large Screw-PAT unit. This requires testing in a larger test or field site facility with greater available flow/head. The results of this assessment also indicated that the 1.18 kW Screw-PAT is not fish-friendly for fish > 90 mm in length in reverse mode. However, fish-friendliness may be retained for smaller species. The results here also indicate that the design of the BMS may be limited by the upper limit of fish sizes at a particular site, in addition to available flow and head, when sizing the Screw-PAT to be used 4.49/8.91.

Fish species migrating through Irish rivers vary greatly in size. Notably, the Atlantic salmon (Salmo solar), European eel (Anguilla anguilla), and river lamprey (Lampetra fluvialtilis) are among the most significant species, all of which are strictly protected under the EU Habitats Directive. While these species differ in body form, both adult Atlantic salmon and European eels can reach lengths of up to ~900 mm, highlighting the need for passage solutions that accommodate larger individuals. Other species such as the pike (Esox lucius) can grow to very large sizes making it challenging for a Screw-PAT to accommodate various fish sizes. Table 4 presents the typical lengths that can be reached by adult fish for various species found in Irish rivers. Safe fish passage can be correlated to the ball passage diameter of a pump. Ball passage diameter is a practical indicator of a pump’s solids-handling and clogging resistance capability, balancing hydraulic efficiency with reliability in real-world fluid applications [25]. It refers to the maximum size of spherical solids the pump can handle without being blocked.

Table 4. Key Irish River fish species.

Species	Length (mm)	Reference
Atlantic Salmon (Salmo salar)	500–900	[26]
Brown Trout (Salmo trutta)	200–400	[27]
European Eel (Anguilla anguilla)	130–800	[28]
River Lamprey (Lampetra fluvialtilis)	300–400	[29]
Pike (Esox lucius)	300–1100	[30]

Table 5 shows the distribution of Screw-PAT models analysed in Section 2, according to ball passage diameter. The majority fall within the 50–150 mm, with the largest model accommodating up to 230 mm. Therefore, these PATs are unlikely to be suitable for fully grown individuals of key migratory species such as perch and three spined sticklebacks, as well as for juvenile stages.

Table 5. Distribution of the ball passage diameter for the 42 Screw-PAT models investigated.

Ball Passage Diameter (mm)	Number of Screw-PAT Models
0–50	2
51–100	17
101–150	11
151–200	6
200–250	6

For part (ii) to assess the fish reactions to the on/off operation of the system, thirty brown trout were placed downstream of the BMS and left undisturbed for 20 h to acclimatise. They were observed using two underwater cameras (GoPro Hero 11 V3 27 MP/30 fps, GoPro, Inc. San Mateo, CA, USA) to be swimming in a shoal. Figure 6 shows the fish behaviour testing setup with the inlet screened to avoid fish swimming through the Screw-PAT unit. The operation of the BMS required switching between turbine and pump mode. Determining fish behaviour near the inlet is important in ascertaining whether the on/off operation will result in a fright reaction and inform the design of the intake system. Fish migration has also been shown to be hindered by the start-up operations of fish-friendly turbines causing fish to be startled and disperse [31].

The flight reactions of fish to switching the BMS on/off were observed on ten separate occasions and timed to coincide with the fish being within 5 m of the intake. In 80% of the activations there was no observed deviation from their previously observed swimming patterns and the fish approached within 30 cm of the intake. These results provide positive indications that fish will swim through the BMS and will also inform the design of the fish intake system and associated components. Further testing is required to fully establish the fish-friendliness of the system and to obtain formal certification prior to implementation at weir sites across the country and within the EU.

5. Limitations of the Proposed Barrier Modification System

While the presented conceptual design of the BMS offers a promising multifunctional approach to low-head river barrier modification by combining fish passage, sediment transport, and micro-hydropower generation, several technical and operational limitations constrain its development and ultimate widespread implementation. The primary limitation lies in the absence of experimental validation of centrifugal screw pumps operated in reverse as turbines. Existing one-dimensional conversion techniques used to predict turbine performance from pump characteristics were developed predominantly for conventional single-stage centrifugal pumps. Although such methods provide acceptable accuracy (±8–10%) for those machines, their applicability to the markedly different geometry and flow characteristics of screw impellers remains uncertain. Consequently, the predicted efficiency and operating head–flow ranges should be regarded as indicative rather than definitive until validated through laboratory tests.

The complexity of the control system also presents some limitations that should be overcome prior to the wide deployment of the system. The system must alternate between generation and pumping modes based on real-time fish detection signals and system energy status. Achieving smooth transitions between modes is technically demanding. The Screw-PAT rotor must be decelerated, the torque reversed, and then accelerated in the opposite direction without imposing hydraulic shocks or sudden velocity changes that could endanger fish or destabilise the system. Furthermore, the reliability of the control algorithms depends on accurate fish detection via optical and infrared imaging, which may be affected by turbidity, biofouling, or poor lighting conditions. Any delay or false trigger could lead to missed passage opportunities or inefficient operation. Thus, a need for robust fail-safe logic and adaptive control strategies remains a critical development requirement.

The other limitations stem from the unverified fish-friendliness in turbine mode. Although the centrifugal screw pump has demonstrated excellent fish survival rates in pumping mode, with injury rates below 3% and negligible latent mortality, the same cannot be assumed for reverse operation. In turbine mode, the flow direction, velocity gradients and pressure fluctuations differ substantially from those in pumping mode. Fish may encounter shear stresses, pressure drops, or strike risks depending on the rotational speed and hydraulic loading during generation. More in-depth tests in reverse mode for fish-friendliness assessment are necessary before wide adoption. Moreover, the interspecies variability may deem the use of a centrifugal Screw-PAT unsuitable for sites where fish of various types and sizes are expected. Therefore, the use of a more open reversible turbine option with large free flows and lower water velocities such as the Archimedes screw pump/turbine is recommended.

Beyond fish passage, the environmental performance of the BMS, including sediment transport dynamics, attraction flow behaviour, and potential modification of local hydraulics remain unquantified. Regulatory acceptance will depend on demonstrating that the system does not adversely affect the upstream water levels, downstream habitats, or flood conveyance capacity. Therefore, further environmental impact assessments and engagements with fisheries and water authorities will be required prior to field deployment.

Implementation of the proposed BMS concept in real river environments will inevitably require navigation of complex authorisation and permitting processes. These include environmental impact assessments, planning and constructions permits, and energy generation licences. Such procedures can be time consuming and financially demanding, particularly for heritage structures or ecologically sensitive sites. While these regulatory requirements fall beyond the scope of the current conceptual study, they constitute an essential phase in any future demonstration or commercial deployment of the BMS. Early stakeholder engagement, integration with existing river-basin management plans, and alignment with EU Water Framework and Habitats Directive will therefore be critical to ensure environmental compliance and project acceptance.

6. Conclusions

The Nature Restoration Regulation represents a pivotal step toward revitalising ecosystems across the EU, with a clear objective of restoring all ecosystems by 2050. Meanwhile in Ireland, the Climate Action Plan 2023 (CAP23) presents a potentially conflicting target of increasing electricity generation capacity from renewable sources to 80% by 2030, which requires hydropower production to ensure grid flexibility is maintained under a large reliance on intermittent sources. The significant number of river barriers existing in Irish rivers pose as a significant challenge to river connectivity and an opportunity for new micro-hydropower projects.

In an attempt to strike a balance between the needs of the two polices, we proposed a barrier modification system presenting a potential solution that harmonises the goals of the NRR and CAP23. The system features a fish-friendly centrifugal screw pump-as-turbine (Screw-PAT) that offers a dual benefit of enhancing river connectivity and generating renewable energy. The system not only facilitates river connectivity and restoration of ecosystems but also harness green energy, particularly benefiting rural areas with limited grid access. Through detailed design, performance analysis, and preliminary fish impact assessments, we have demonstrated the feasibility and potential of the BMS to generate renewable energy at regional and case-study levels, while providing preliminary assessment of any harm caused to aquatic life.

With further research, integration of the Screw-PAT within existing barriers, the BMS, may enable transformation of these structures into dual function systems that support both ecological restoration and renewable energy production. This approach underscores the importance of sustainable hydropower in the broader context of renewable energy strategies, providing essential grid flexibility amidst growing solar and wind contributions.

Moving forward, more detailed analyses of renewable energy potential and fish impact assessments will be conducted to affirm the viability of the BMS. With further refinements and formal certification, this system can play a pivotal role in achieving the dual objectives of the NRR and CAP23, thereby contributing to a more sustainable and ecologically balanced future for the EU.