| Modelling
what fish are physically capable of achieving. The third and most novel component is behavioural modelling. Emergent incorporates behavioural rules that describe how fish respond to neighbours and environmental signals, allowing the simulation to reproduce schooling behaviour, alignment and collective movement. “We bring in fish behaviour, which is something new to the mix, and combine these frameworks into a single cohesive model,” Nebiolo says. This integrated approach changes the types of questions engineers can ask when designing fish passage systems. Instead of focusing only on whether a single fish can swim through a certain velocity, the model allows engineers to evaluate whether groups of fish can move efficiently through a system without delays or congestion. “We go from checking ‘can a single fish swim this velocity?’ to asking ‘can a wave of fish go through together without waiting to delay?’” Nebiolo explains. For fishway design, that difference can be critical. Fish
rarely migrate individually; they move in schools whose collective behaviour can influence passage success. By simulating these group dynamics, engineers can better understand how fish interact with hydraulic structures and identify potential issues before construction begins.
Design and validation
Behaviour-based modelling can influence the design and retrofit of fish passage systems in several practical ways. One of the most important applications is evaluating fishway entrance configuration. Simulations can help identify minimum entrance widths required to prevent schools from collapsing into congestion points, ensuring fish can move smoothly through passage structures rather than forming queues at narrow openings. Because each design option influences the surrounding hydraulic conditions, the model can evaluate multiple geometric configurations quickly within the same simulation environment. Another important design consideration involves attraction flows – currents used to guide fish toward fishway entrances. Simulations can test different entrance locations and attraction flow patterns to determine which configurations may reduce search time and guide fish toward passage routes more effectively. According to Nebiolo, even subtle changes in attraction flow placement can influence the behaviour of an entire fish school. “All we need is for that attraction water to hit the leading edge of the school and we’ll see that is enough to pull in fish behind them,” he explains, noting that fish often follow neighbours who have already responded to the hydraulic signal. The modelling framework can also help identify
potential refuge areas, low velocity zones where fish can temporarily rest before attempting another passage effort. These areas can occur naturally within river features or be incorporated deliberately into fishway designs. By simulating these dynamics across different flow conditions, engineers can evaluate a range of design alternatives virtually before any construction takes place. Such tools are particularly valuable in regulatory
environments where hydropower licensing processes require proof that passage systems will perform biologically, not just hydraulically. Validated simulations can demonstrate how fish are expected to behave under different flow conditions and identify areas where delays or congestion may occur. For one project, the model evaluated fish passage efficiency over a natural waterfall under a range of discharge conditions, producing visual
outputs that showed where fish aggregated and where passage routes emerged. The visualisations produced by the model – such as density maps and animated fish movement – are also useful communication tools. They allow engineers, regulators and stakeholders to see how fish interact with hydraulic structures, making complex hydrodynamic processes easier to understand. “You can literally watch a queue form at a constriction or individuals attempt a high velocity corridor and tire and fall back,” Nebiolo says.
Operational insights and potential Beyond design, the Emergent framework can also help
optimise hydro operations. Because the model explicitly represents delay, fatigue and schooling behaviour, it can evaluate operational strategies such as flow releases, spill timing and attraction flow adjustments. This capability allows operators to identify discharge ranges that balance hydro production with biological performance. In one simulation, the model revealed a biological “sweet spot” where discharge levels produced the most efficient fish passage. Too little flow increased delays as fish struggled to locate passage routes, while too much flow created velocities that were difficult to overcome. Within a certain range, however, passage efficiency improved significantly. The simulation also showed how individual fish sometimes discovered favourable passage routes first, with others following through collective navigation behaviour. The framework relies on field data for calibration and validation, particularly telemetry and imagery. Telemetry data such as radio tracking provides information about how fish move through river systems over time, revealing passage success rates and delay patterns. Imagery, including aerial drone footage, helps researchers understand schooling structure and small-scale behaviour. These observations are essential for ensuring that simulated fish behave realistically within the model. Although the initial development focused on sockeye salmon in Alaska, the model is designed to be adaptable to other species and locations. The behavioural and biological parameters can be adjusted to reflect different fish species, meaning the framework could be applied to river systems around the world. The main limitation is the availability of site-specific data needed for calibration and validation. Looking ahead, Nebiolo sees agent-based modelling playing a role in broader digital twin concepts for river management. As computing power increases and more data becomes available, such models have the potential to be integrated with operational systems to support simulatation of broader river systems and evaluate cumulative impacts on migrating fish populations. “Provided the computational capacity and the data are there, I do see agent-based modelling playing a role in broader digital twin concepts for highly regulated river systems,” he says. Ultimately, the goal is to help operators make more informed decisions that support both energy production and ecological outcomes. By better understanding how fish interact with hydraulic structures, engineers can design passage systems that reduce delays, minimise energy expenditure and improve the likelihood that migrating fish reach their spawning grounds. “My hope is that Emergent is used to provide better conditions for migrating fish populations,” Nebiolo says. “The faster we can get a fish above and through a project, the more likely they are to spawn.”
Further information
https://www.kleinschmidtgroup. com/
www.waterpowermagazine.com | April 2026 | 27
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45
