April 26th, 2019
Researchers at the University of Delaware have created a text alert system to help fishermen avoid Atlantic Sturgeon in the Delaware Bay. The system predicts the location and depth of endangered Atlantic Sturgeon with 88% accuracy. The Atlantic Sturgeon has been listed as an endangered species since 2012. It can weigh up to 800 lbs. and reach 14 feet in length and frequently falls victim to bycatch. Bycatch is any marine animal unintentionally captured by commercial fishers targeting a different species of fish.
Matthew Oliver, a professor at UD’s College of Earth, Ocean and the Environment (CEOE), and Matthew Breece, a post-doctoral researcher in CEOE, created the text alert system to protect Atlantic Sturgeon and to preserve the time and energy of fishermen. They partnered with Dewayne Fox, an assistant professor at Delaware State University, and Ed Hale, a fisheries biologist at the Delaware Department of Natural Resources and Environmental Control (DNREC). Without the text alert system, fishermen could potentially encounter Atlantic Sturgeon when searching for other fishes in the same habitat. These species include striped bass, croaker, and monkfish.
The team of researchers used historic Atlantic Sturgeon acoustic telemetry data to develop the statistical model behind the alert system. The acoustic data were retrieved through 88 underwater microphones placed as moorings along the coasts of Maryland, Delaware, and New Jersey between 2009 and 2013. Additional data came from acoustic tags placed on 300 Atlantic Sturgeon.
Gliders are used to validate the data retrieved from the acoustic moorings and tags. Gliders are autonomous underwater vehicles that use multiple sensors to collect oceanic data. To this day, acoustic data continue to be collected so the current model can be improved.
The acoustic data imply that the presence of Atlantic Sturgeon in a given location depends on sea surface temperature, time of year, depth, and ocean color. The researchers observed that Atlantic Sturgeon prefer temperatures around 18° Celsius. They tend to live at shallow depths in the spring time and deeper depths in the fall. Preference for ocean colors is relative to the salinity, amount of nutrients, and suspended particles in the water.
Each prediction entered through the text alert system depends on MARACOOS satellite data. The researchers have entered their second study season following the initial roll out of the product, which initially relied on satellite data from MODIS-Aqua. The second season brings in satellite data from VIIRS, an additional MARACOOS data stream, which passes over the Delaware Bay at different times throughout the day. Gliders are used to verify the predictions derived from satellite data.
“We have a side-by-side comparison of the warning system using two different but similar satellites,” stated Dr. Oliver. Both data streams are currently being tracked to ensure the text alert system is as accurate as possible. “It is important for us to understand the stability of the forecasts we make with two similar but different satellite systems.”
Moving forward, the research team is putting forth effort to improve the text alert system and reach a broader audience, in addition to fishermen. Its initial funding was provided by the Lenfest Ocean Program. Due to the project’s success, the model’s transition into a real-time text alert system was funded by NASA. This original deployment relied on a heavy collaboration with DNREC and received tremendous support from local fishers. Without MARACOOS satellite data, the product’s development would not have been possible.
“The power of MARACOOS is that it is always on and always responding to the Mid-Atlantic environment and needs,” according to Dr. Oliver. “Because that collaboration, expertise, and digital data infrastructure exists through MARACOOS, we were able to quickly deploy the research products developed with Lenfest and NASA to the benefit of our Mid-Atlantic constituents and stakeholders.”
Story By: Jessica Ganim