Mapping the global prey-field: Combining acoustics, optics and net samples to reduce uncertainty in estimates of mesopelagic biomass

Abstract

Mesopelagic (200–1000 m depth zone) biomass largely comprises zooplankton, jellyfish, squid and fish. Part of this biomass actively migrates between the surface and the mesopelagic zone on a daily basis, playing an important role in global biogeochemical cycling. When observed using echosounders, aggregated components of this community (largely made up of fish and siphonophores) are known as Deep Scattering Layers (DSLs). DSLs have been used to partition the Global Ocean into biogeographically distinct regions, based on DSL vertical distribution and echo energy, and estimates of mesopelagic biomass have stemmed from this work. These estimates are typically derived using a combination of biological sampling, optics, and active-acoustics, but are often biased (e.g., through trawl-based escapement and avoidance) and uncertain. We have developed a new method that combines multiple strands of observations, both from vessels and ocean probes, to minimise sampling bias and produce accurate estimates of mesopelagic biomass. We apply the method to the specific case of estimating mesopelagic fish biomass in the Tasman Sea, combining observations made from a vessel and a Profiling Lagrangian Acoustic Optical System. We compare our new biomass estimates to previous estimates and ecosystem model predictions. Our new method provides a robust framework for future ocean biomass estimates.