Detection of jellyfish swarms from space, a feasibility study
Abstract
Jellyfish swarms are a major nuisance in coastal environments. From human health (e.g. ~14,000 people stung in Spain in August 2006,, Boero, 2013), via clogging of fishing gear and disruption aquaculture operations to disrupting cooling operations associated with p,ower plants bringing power down for 40M people in the Philippines. In South Korea alone, the annual damage from Jellyfish to the fis,heries industry has been estimated to be between $70-205 million. Jellyfish are also a significant fishery in Asia (have been consum,ed in China for at least 1700yrs) with an aquaculture industry established in 2003.Many jellyfish species are cosmopolitan and form,blooms in a wide range of environmental conditions. The increase in the frequency and magnitude of jellyfish bloom events has been a,ttributed to global climate change and reductions of top-down predator controls. The current state-of-the-art of detection is mostly, in its infancy, requiring human observations as input (e.g. https://www.copernicus-user-uptake.eu/user-uptake/details/satellite-too,ls-for-jellyfish-blooms-detection-165) or using a neural network of environmental data or RGB drone images. The problem with these,approaches is that they are not driven by a direct relationship with the optics of the problem and hence it is not obvious how well,they can be extrapolated beyond the time/space of the data they are based on. A European project to detect jellyfish with ocean colo,r has been funded (2009-2014) that mostly used low resolution (1km pixel data). Here we propose to test the hypothesis that adding m,ore wavebands in the visible and NIR, now available from many remote sensing platforms can improve the efficiency of Jellyfish detec,tion and the accurate delineation of swarms and blooms. We also propose to apply this approach with high-resolution space-based sens,ors such as Sentinal 2A&B and Landsat 8 and that such an algorithm can be made explicitly, that is without the need of a training da,taset but based on the optical properties of the jellies changing the color of the ocean. If available, we will also try using highe,r resolution data (e.g. 50cm pixel) such as collected with the Pleiades satellites.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 01, 2022
- Source ID
- N000142212161
Entities
People
- Emmanuel S. Boss
Organizations
- Office of Naval Research
- United States Navy
- University of Maine System