A study by the Oregon State University and U.S. Forest Service found that the use of Artificial Intelligence for analysis of the acoustic data recorded by these devices is a prospective approach for monitoring the marbled murrelet and other similar species.
Innovative approach unveiled
The endangered marbled murrelet is a distinctively marked murrelet that lives on the west coast of British Columbia and Washington, their closeness to puffins and murres. While both these species provide for their young on the coast side, they differ in the fact that the murrelets make it as far as 60 miles deep into the forests of mature and old-growth areas.
According to Matt Betts, co-author and faculty member at OSU’s College of Forestry, studying such a reclusive species is challenging, and conspicuously comparable cases are rare. Betts stated, “Next, we’re testing whether murrelet sounds can actually predict reproduction and occupancy in the species, but that is still a few steps off.”
A research team consisting of Adam Duarte of the US Forest Service’s Pacific Northwest Research Station used data from audio recordings that were initially installed to assist in monitoring northern spotted owl populations in hundreds of locations managed by federal agencies in the Oregon Coast Range and Olympic Peninsula in Washington.
Researchers created a machine learning system, which is a convolutional neural network, in order to detect the murrelet calls in the recordings. Scientists from Ecological Indicators published their findings that were tested against known murrelets population data and it was determined that the recorders and AI provided an excellent accuracy of identifying at least 90% of murrelets in a given area.
In addition, Betts suggests that Betts indicates that the study will lead to examining the sounds of murrelets in order to predict reproductive success and occupancy of habitats as this will be the key avenue for the forthcoming research.
Accurate population insights
The small marbled murrelet spends most of its time in coastal waters where it feeds on krill, other invertebrates and forage fish like herring, anchovies, smelt, and capelin. If the nests are successful, birds can still only produce a single young each year, and the nutrient-rich forage fish are essential for the baby murrelets’ proper growth and development.
The marbled murrelets are usually discovered along the coastline of West, from Santa Cruz, California to the Aleutian Islands. The species is designated as threatened under U.S. Endangered Species Act in state of Washington, Oregon and California for the time being.
However, most of the detections in this research usually take place where forest maturity dominates as well as near the coastal habitats,” Duarte reported. Nevertheless, with its nesting and resting spots constantly being lost to predators like Steller’s jays and human-induced habitat degradation, that is why the monitoring strategies have to be really effective to save the Black-footed Albatross.
Towards sustainable conservation
Duarte highlights how this AI-driven surveillance system can contribute to species distribution models and act as an underlying basis of long-term population monitoring—the two essential elements in the conservation of rare and threatened species. Through the elimination of labor-intensive approaches of the past, such as telemetry and nest ground search, their method will enable conservation efforts to be carried out in a much more streamlined fashion while conservation is still the main focus.
