acoustic models is a great challenge. Many different models, which vary in accuracy and generality, have been developed within the past decades. First scattering
models treated zooplankton as a homogeneous fluid sphere (Rayleigh 1945;
Anderson 1950; Johnson 1977). More sophisticated models take shape and material
properties of animals (Stanton 1989; Chu et al. 1992; Stanton and Chu 2000) into
account. In this study we used so called “high-pass” model for a fluid sphere
introduced by Stanton (1989), which proved to work well in the Arctic environment
(Trudnowska et al. 2012).
Zooplankton size spectra obtained by LOPC were divided into 49 size classes
(from 100 μm to 35 mm diameter), which were used as input parameters for
theoretical acoustic backscattering model of zooplankton. To validate implemented
model, the sum of the modeled values of backscattering cross-sections of all
scatterers were transformed into the total S V and compared with the real S V values
measured by the echosounder in the water layer of the LOPC towing depth. For
comparison, the acoustic data were averaged over 1 m depth layers and 10 transmissions (5 s). The same applied to the optical data and the environmental variables, which were averaged over the same time and depth intervals. The
zooplankton sound speed contrast was set at h = 1.027 and the density contrast at
g = 1.0, as proposed for the Arctic copepod species Calanus finmarchicus (Kogeler
et al. 1987). The collected data consist of, in total, 13 h of concurrent acoustical and
optical measurements (5 h in Hornsund and 8 h in Kongsfjorden), which were
organized into 30 min long data series.
4 Results
Two exemplary data series were selected for this report, to illustrate the results of
our study. The first is the data collected in Kongsfjorden on 06 August 2013,
starting at 18:49 UTC. The distribution of backscattering strength in the 0–50 m
layer during the 30 min long survey along the designated line is shown in Fig. 2,
with the LOPC towing route indicated. Distribution of backscatter presumably
reflects the distribution of biomass, most likely of zooplankton, with incidental
strong echoes from other objects like fish. For each time interval the backscattering
strength measured by the 420 kHz echosounder (red curve) was compared with the
backscattering strength calculated from the model (blue curve), for the zooplankton
size and abundance estimations based on the optical measurements at the exact
towing depth (Fig. 3). The repetitive maxima and minima reflect changes in S V
values, caused by higher or lower zooplankton biomass at different water depths.
The linear correlation coefficient between the measured and modeled values of S V
for the whole 30 min long data set in the entire studied water column equaled to
R = 0.72 (Fig. 4).
The data series taken in Hornsund on 26 July 2013, starting at 21:26 UTC
illustrates the technical capability of measuring several marine environment properties simultaneously with optical data. Data on distribution of various
54
L. Hoppe and J. Szczucka
models treated zooplankton as a homogeneous fluid sphere (Rayleigh 1945;
Anderson 1950; Johnson 1977). More sophisticated models take shape and material
properties of animals (Stanton 1989; Chu et al. 1992; Stanton and Chu 2000) into
account. In this study we used so called “high-pass” model for a fluid sphere
introduced by Stanton (1989), which proved to work well in the Arctic environment
(Trudnowska et al. 2012).
Zooplankton size spectra obtained by LOPC were divided into 49 size classes
(from 100 μm to 35 mm diameter), which were used as input parameters for
theoretical acoustic backscattering model of zooplankton. To validate implemented
model, the sum of the modeled values of backscattering cross-sections of all
scatterers were transformed into the total S V and compared with the real S V values
measured by the echosounder in the water layer of the LOPC towing depth. For
comparison, the acoustic data were averaged over 1 m depth layers and 10 transmissions (5 s). The same applied to the optical data and the environmental variables, which were averaged over the same time and depth intervals. The
zooplankton sound speed contrast was set at h = 1.027 and the density contrast at
g = 1.0, as proposed for the Arctic copepod species Calanus finmarchicus (Kogeler
et al. 1987). The collected data consist of, in total, 13 h of concurrent acoustical and
optical measurements (5 h in Hornsund and 8 h in Kongsfjorden), which were
organized into 30 min long data series.
4 Results
Two exemplary data series were selected for this report, to illustrate the results of
our study. The first is the data collected in Kongsfjorden on 06 August 2013,
starting at 18:49 UTC. The distribution of backscattering strength in the 0–50 m
layer during the 30 min long survey along the designated line is shown in Fig. 2,
with the LOPC towing route indicated. Distribution of backscatter presumably
reflects the distribution of biomass, most likely of zooplankton, with incidental
strong echoes from other objects like fish. For each time interval the backscattering
strength measured by the 420 kHz echosounder (red curve) was compared with the
backscattering strength calculated from the model (blue curve), for the zooplankton
size and abundance estimations based on the optical measurements at the exact
towing depth (Fig. 3). The repetitive maxima and minima reflect changes in S V
values, caused by higher or lower zooplankton biomass at different water depths.
The linear correlation coefficient between the measured and modeled values of S V
for the whole 30 min long data set in the entire studied water column equaled to
R = 0.72 (Fig. 4).
The data series taken in Hornsund on 26 July 2013, starting at 21:26 UTC
illustrates the technical capability of measuring several marine environment properties simultaneously with optical data. Data on distribution of various
54
L. Hoppe and J. Szczucka