Application of geoacoustic inference to assess the diurnal effects of
photosynthetic activity in a seagrass meadow
Abstract
Seagrasses provide a multitude of ecosystem services: they alter water
flow, cycle nutrients, stabilize sediments, support the food web
structure, and provide a critical habitat for many animals. However, due
to threats to seagrass meadows and their associated ecosystems, these
habitats are declining globally. Acoustical methods can be a powerful
remote sensing tool to efficiently monitor seagrass meadows, alleviating
the problem of space and time aliasing associated with traditional spot
measurements. The acoustic field is highly sensitive to the presence of
gas bubbles in liquids, which result in acoustic dispersion, absorption,
and scattering. The biological processes and physical characteristics
associated with seagrass result in trapped gases, which are present in
the aerenchyma of seagrass leaves, roots, and rhizomes as well as in
freely floating bubbles in the water. To investigate the use of
acoustical methods to monitor ecosystem health, an experiment was
conducted in the Lower Laguna Madre where the seabed was covered by a
dense growth of Thalassia testudinum. During the experiment, a
combustive sound source was used to produce broadband signals at ranges
of 25 m to 400 m from an array of receivers. The data were collected
over a period of one day to assess the diurnal effects of photosynthetic
activity on the acoustic measurements. An additional set of measurements
were collected in a nearby sparsely vegetated site to provide a baseline
data set for comparison. Supporting measurements of irradiance,
dissolved oxygen, and water temperature were acquired at the experiment
site. The acoustic data were analyzed for the purposes of inferring
environmental parameters in the seagrass meadow, including the gas
volume present in the environment. An effective medium model for gas
bubbles in liquids was used to characterize the frequency-dependency of
sound speed and attenuation in the waveguide. [Work sponsored by
ARL:UT IR&D and ONR.]