Ecosystem monitoring using high frequency active acoustics

Ce message est une copie de celui mis sur le forum Biodiversite-Ecosystemes, mais il me semble etre interessant pour ce forum.

Ecosystem monitoring using high frequency active acoustics

We aim to address the considerable challenge of monitoring and quantifying the ecologically-key mid-trophic levels (zooplankton) in the rapidly changing Arctic Ocean. We are only just beginning to appreciate the gap in our understanding and expectations of these ecosystems in the Arctic, as exemplified by recent findings (Arrigo et al., 2012) which measured under-ice phytoplankton biomass hugely in excess of expectations and higher than typical open-water concentrations.

While phytoplankton can be relatively easily measured using fluorimetry, estimating the zooplankton population is far more challenging. Traditional, labour-intensive methods such as net sampling have many problems with avoidance by the larger organisms. Traditional techniques are clearly not amenable to future application in autonomous systems, which must be the future of all oceanographic monitoring - especially in the Arctic.

Acoustic methods are “the only means that can efficiently observe the large biomass of the mid-trophic levels at ecologically mportant temporal and spatial scales” (Wiebe et al., 2002). Acoustic data are inherently ambiguous when measured in isolation, however. The amplitude of the back-scattered energy depends on the interacting effects of the number of animals, the species composition and their size distribution and orientation. Inverse modelling techniques are able to untangle these interacting factors, but only with sufficient independent knowledge of the animals present.

This ‘independent knowledge’ (often referred to as ground truthing, though the glaring deficiencies of these complementary techniques are the motivation for using acoustics in the first place) is provided by optical techniques (underwater video profiler, or UVP) and net sampling. Knowledge of the physical environment (CTD, velocity) is also helpful. Scattering/absorption parameters, fluorescence & bioluminescence sensors also aid interpretation. Integrated measurements are the key to understanding.

A key factor in gaining understanding of the acoustic signal is the use of multiple frequencies. Different species have varying responses across the acoustic spectrum and the ratios between backscatter at different frequencies can be used to identify the dominant organisms in a swarm. Comprehensive pan-Arctic observations of the acoustic signatures of key species are sought, in conjunction with a full suite of supporting measurements (optical, physical, nets). The aim is to build a library of acoustic backscatter characteristics, allowing us to move towards a uniquely acoustic characterisation of the functional groups present, ultimately incorporating this knowledge into an autonomous acoustic profiling system, similar in operation to the highly successful Ice Tethered Profilers deployed by LOCEAN and WHOI during DAMOCLES, and under development by the LOCEAN iAOOS project.

Researchers involved: Martin Doble (LOV), Gaby Gorsky (LOV)