Detection, Classification, and Prediction of Satellite Tagged Marine Mammals for the U.S. Navy
Abstract
As stewards of the sea, the United States Navy is committed to protecting the environment while defending freedom. In recent years, marine mammal researchers have observed and reported an increase in vessel strikes, mortality rates, enigmatic standings, and fishing gear entanglements, particularly among baleen whales. Integrating marine mammal localizations into U.S. Navy Maritime Domain Awareness (MDA) systems will provide the warfighter with the ability to detect, predict, and respond to marine mammals that reside in the maritime battlespace. An MDA system must be able to accurately interpret and predict marine mammals movement in order to provide meaningful alerts to nearby vessels. Satellite-tracking tags have become an increasingly popular tool to track, study migration routes, produce habitat density models, and assess population abundance among marine mammals. Tag data is transmitted to the Argos satellite constellation, reported to the user on the Argos Web Portal, and optionally published on open sourced repositories. This paper explores the positional fidelity of a Wildlife Computers SPOT-303C tag and Wildlife Computers SPOT-258F, two popular tag models used by marine mammal researchers. Tag data was collected from Argos Web Portal, the Collecte Localisation Satellites (CLS) RXG-134/Goniometer receiving system, and Great Scott Gadgets HackRF One Software-Defined Radio (SDR). A Naval Research Lab (NRL) vessel was affixed with the SPOT-258C and SPOT-303C tags and performed a series of runs to simulate marine mammal behavior. Significant positional errors were found in Argos Web and RXG-134/Goniometer data during the vessel test. Argos processed tag data and RXG-134/Goniometer directions should be used as marine mammal presence-only detections. If the U.S. Navy intends to leverages these sources, prior knowledge of the species behavior and the Particle Filter (PF) algorithm provides a general marine mammal movement model.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 12, 2021
- Accession Number
- AD1127649
Entities
People
- Erin C Murnane
Organizations
- United States Naval Research Laboratory