Animal Interaction

Project: BioSonics Perimeter Detector 

The BioSonics, Inc. project aims to create a discreet, long-range acoustic monitoring system to assess marine life behavior surrounding MRE systems. This acoustic device uses a perimeter detector that automatically senses targets (marine life) within a 200 m range. The detector provides 360 degrees of coverage from 48 sonar heads that send synchronized signals to detect any organism that enters the detection radius. A secondary sonar, the directed classifier, is then automatically aimed at the detected target to track its position in three dimensions and gather higher-resolution data on the density, size, and rate of movement. This classification will provide researchers with information, including location, depth, behavior, and general type of marine life. In addition to detecting a target from afar, the system can determine whether it is likely to interact with the marine energy device or simply pass by it.

BioSonics, Inc. conducted a series of successful tests in Sequim Bay and the Sequim Bay channel, increasing the system’s scope from 90 degrees to a full 360-degree detection over three testing periods. The system impressed researchers with its ability to recognize a United States Coast Guard navigation day marker 250 m away—50 m longer than its original range of detection. A comparison with a multibeam echosounder, a type of sonar used for imaging and deep-water mapping, demonstrated its improved perimeter coverage and range detection compared to standard practices. The improved technology has even implemented a frequency-modulated (“chirp”) signal transmit transducers that increase signal to noise ratios, which improves target-detection capabilities.

In addition to developing the long-range acoustic monitoring and perimeter detector technologies, Biosonics, Inc. worked with PNNL to reduce the strength of frequency transmissions associated with echosounders. Echosounders often have unwanted frequency transmissions during normal operations. If these unwanted transmissions are in the hearing range of marine mammals, it could affect the behavior of those animals. Reducing the strength of these unwanted transmissions on the system decreases the potential effects on marine mammals.


BioSonics perimeter detector improvements from 90-degrees to 360-degrees.


Divers deploying the BioSonics 360-degree perimeter detector.


Project: Florida Atlantic University UMSLI 


Light Detection and Ranging, or LiDAR, is an advanced remote sensing method that uses light in the form of laser pulses. These data are often used to generate precise, 3-D topographic and bathymetric images to characterize the Earth’s surface or specific features. The Unobtrusive Multi-Static Serial LiDAR Imager (UMSLI) project, led by Florida Atlantic University, is bringing this technology into the water column to improve the detection and classification of marine life. The first-generation UMSLI technology was funded by a DOE FOA in 2014, which developed and tested the optical monitoring system prototype with automated real-time classification of marine animals. The project was awarded a second FOA for further technological development.

The UMSLI system has transmitters that artificially illuminate the water surrounding an MRE device with a series of short laser pulses of light that allow receivers to collect time-resolved returns. Just one year after development, the system covers 360 degrees and classifies targets to 10 m using red lasers. In year two, the project will incorporate green light, which will increase the system’s detection range under water.

The goal of the UMSLI is to monitor and identify marine life surrounding energy devices. A benchmark performance of the baseline prototype occurred in July 2017 in Sequim Bay. To support baseline testing, MSL divers were tasked to emulate the conditions of a representative UMSLI deployment environment by swimming with artificial marine life targets in order to validate the system’s ability to classify species presence or absence. The UMSLI system uses omnidirectional orientation architecture, which is a technology that allows observations to be made at different angles and relative positions. This information was used to improve the UMSLI's classification algorithm. Later this year, the system will be validated against targets at multiple deployment sites, including Ediz Hook, the Dungeness Spit, and the Sequim Bay channel. These deployments will be used to further validate the system performance, allowing comparisons with conventional imagery against the newly developed technology.


Deploying the UMSLI system in Sequim Bay.


The UMSLI system.


Project: University of Washington 3G AMP

Following initial testing of the second-generation Adaptable Monitoring Package (2G-AMP), researchers at MSL are hosting the third-generation version of the system. The latest iteration is designed to enhance software and processing and increase the system’s adaptability to a range of marine renewable energy projects, as well as adding new sensors and the ability to change their tilt.

The AMP already has the capability to trigger multiple sensors based on certain stimuli—evidence they truly work as an integrated package. Its specialized hardware allows its sensors to come in proximity to a marine energy converter and monitor many aspects of the system (check out 2G AMP for more on how this works). The current effort is taking the technology to the next level, with a classifier that can detect, track, and classify marine life in real time.

Ultimately, the focus of the 3G phase of the project is to target observations on the events of greatest importance for understanding environmental risk associated with marine energy systems. When deployed for months at a time, there is an increased opportunity for rare events such as fish strike or interactions with marine mammals to be captured by the AMP. In 2019, the device was improved to be more hydrodynamic, compact, and equipped with tilting instrument heads, as well as to have the ability to operate multiple sonars simultaneously without interference. This upgraded device was successfully deployed for four months in the most energetic location of the Sequim Bay channel, and helped to support a fish mesocosm study being conducted at PNNL to quantify fish behavior around MRE devices. Over that four-month period, the device’s operational uptime was 97 percent. The next step for the project is to integrate the AMP with the Ocean Energy 35 wave energy converter for a long-term deployment at a formal wave energy test site (WETS).


UW AMP improvements from its first iteration in 2016 to the 3G AMP in 2019.


The R/V Jack Robertson retrieving the AMP in Sequim Bay.


Project: Triton Fish Mesocosm Study 


The fish mesocosm study (FMS) is a collaboration between the University of Washington (UW), University of Alaska Fairbanks, University of Maine, and Pacific Northwest National Laboratory (PNNL). The purpose of the study is to quantify fish interactions with tidal turbines and observe their behavior around these devices in a controlled, field environment. Researchers used a UW-developed third-generation adaptable monitoring package (AMP) equipped with sonar, acoustic cameras, hydrophones, and a classification system to record movements of tagged fish in Sequim Bay. In the next phase, fish will be released around a scaled operational tidal turbine to capture fish interactions. Researchers anticipate that study results will verify that there is a low probability of fish interaction with turbines, and the data will allow them to establish avoidance rates for robust fish encounter models.

The FMS began in September 2018 by raising sablefish (Anoplopoma fimbria) and preparing them for acoustic tagging and controlled release. A system for releasing the fish at specific depths and locations was developed.  

In phase I, PNNL researchers tagged 100 fish with JSATS fish tags and released them near a 3G-AMP surrounded by a hydrophone array deployed in Sequim Bay. The advanced PNNL JSATS tagging technology creates fine spatial and temporal resolution fish tracks after release. Combining these tags with the AMP’s multi-sensor technology allowed for detailed observation of the avoidance and evasion behavior of fish. 98% of the tagged fish were detected by the hydrophone array, most of which had at least one 3-D tracked location, which showed that the majority of the fish swam toward the AMP at some point in the tracking period. Data from the AMP are still being processed.

Now that the methodologies have been validated, the study will repeat the experiment on a larger scale in the presence of turbines to observe and quantify fish behavior in the presence of a dynamic MRE device.

Here is an animation of the phase I fish 3-D tracked using the hydrophone array:

Video credit: Jayson Martinez. 


Fish surgeries to tag fish for tracking. Photo credit:  Andrea Starr


Example of a JSATS tag embedded in fish for tracking. Photo credit:  Andrea Starr