Research Engineer U.S. Army Engineer R&D Center, Environmental Laboratory
The decisions of volitional, freely moving fish frequently dictate the success of multi-million dollar engineered structures and management actions. The ability to predict fish response to infrastructure and management actions during the design phase has the potential to save time and money as well as living resources. For the past 25 years, the U.S. Army Corps of Engineers, Research and Development Center (ERDC), has been working on a management tool that can hindcast and predict near-future fish response to infrastructure designs and management actions. Development of the tool – a Eulerian-Lagrangian-agent Method (ELAM) - has leveraged over $65 million dollars’ worth of river and fish movement data. The ELAM model has achieved unique success in predicting near-future 3-D/2-D fish movement, guidance, and entrainment and also has accurately predicted patterns prior to the availability of field data in some cases. Further, the ELAM has performed well on out-of-sample data where the future condition is different from the calibration conditions. The model does not attempt to represent the detailed cognitive architecture of fish; rather, the decision-support tool attempts to leverage researched non-linear relationships between stimuli, perception, and action to make predictions of what fish will do at the scale of river infrastructure and water operations management. Central to model performance is the notion that fish are attuned to more than one environmental signal and more than one timescale. Emerging theoretical developments suggest the potential exists for inverting downstream-moving behavior rules to describe upstream-moving fishes. Fish movement depends on the species, but work unifying past data into a common framework facilitates value-added benefits to existing data, the ability to understand fish behavior more quickly, and the ability to better incorporate animal behavior into the fast-paced nature of engineering design projects.
