Defense Date

12-9-2021

Document Type

Thesis

Degree Type

Master of Science

Degree Name

Biological Sciences

First Advisor

Mahmood Shivji, Ph.D.

Second Advisor

Jeremy Vaudo, Ph.D.

Third Advisor

Bradley Wetherbee, Ph.D.

Keywords

BRUVS, fishery-independent, forage fish, spatiotemporal

Abstract

Information about pelagic community diversity and ecology generally lags far behind that of coastal communities, and largely derives from fisheries data that do not reflect small and non-target species. We describe spatiotemporal vertebrate species diversity and variability over a 3,486 km2 area of highly productive pelagic marine ecosystem in Pacific Panama using drifting baited remote underwater video stations (BRUVS), a non-invasive fishery-independent sampling technique. We observed 26 taxa from 17 families, including 1 mammal, 3 reptile, 5 elasmobranch, and 17 teleost species. Community assemblages differed on and off the continental shelf and between wet (April – December) and dry (January – March) seasons but did not vary between sampling depths (10 and 40 m). Seasonal differences were largely driven by the relative abundance of three genera of small zooplanktivorous fish: Psenes, Decapterus, and Caranx. Their abundance was greatest in the wet season, lagging peak plankton production in the dry season, and each species was more abundant off the continental shelf. Psenes and juvenile Caranx were rare during the dry season, and while Decapterus abundance decreased in the dry season, they remained along the edge of the shelf year-round. Despite sampling for ~260 h in a highly productive area known for high fish abundance, we did not fully capture the system’s expected vertebrate richness. Differences between expected and observed richness suggest future pelagic BRUVS studies should increase recording time and sample size, especially in less productive systems. Our study demonstrates the practical application of free-drifting BRUVS to detect environmentally driven changes in pelagic communities and fill knowledge gaps in data-limited regions.

Share

COinS