# Impact of Sparse Benthic Life on Seafloor Roughness and High-Frequency Acoustic Scatter

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Regional Setting

#### 2.2. Field Measurements

^{2}was recorded during the HE486 survey utilizing a roll-stabilized NORBIT iWBMSe multibeam echo sounder operating at a frequency of 200kHz. Multibeam echo sounder raw data is available for download in the Pangea database (refer to the section Supplementary Materials). Hypack 2016 (Hypack / Xylem Inc., Middeltown, CT, USA) was used as a recording software. Data processing was involved the application of sound velocity corrections, the interpolation of asynchronous navigation records, and tidal effects. Backscatter mosaics were subsequently created using Hypack Geocoder with a resolution of 0.5 m, correcting for the insonified area and the effects beam pattern by applying an angle varied gain.

^{2}at a resolution of 0.001 m. The scan resolution is controlled by the attachment height of 0.5 m above the seafloor, the rotation range of 60°, the rotation step size of 0.09°, and the fixed opening swath width of 50° with 480 equidistant capture points. At each deployment location, four surface scans were performed, with each scan lasting, on average, 110 seconds. To mitigate the impact of particle flow within the water column and rapidly changing light conditions in a shallow water environment, two digital filters were applied during data recording, provided by 2G Robotics. The ambient light filter performs multiple line scans to measure and subtract background noise. A defined recording window of ± 0.1 m above and below the seafloor reduces reflection from particles close to the laser head.

^{2}. For this study, we focus on the transducer mounted at a grazing angle of 50° because backscatter intensities recorded at grazing angles between 30° to 60° are less responsive to slight changes in grazing angle [21]. The chosen footprint size of 0.16 m

^{2}at an angle of 50 degrees corresponds approximately to the footprint size of a single beam of a multibeam system. The acoustic footprint is located within the area captured by the laser scanner. Recording windows of the transducers were adjusted to avoid multiple reflections, system reverberation, and nearfield effects. To improve the signal-to-noise ratio at each shot, 20 time series were logged and stacked during post-processing. For transmitting, a driving voltage of 80 V generated a 102 µs linear chirp pulse signal with a bandwidth range of 80 to 310 kHz. For receiving, the sampling rate was set to 10 MHz. The time-varying response function (TVR) and the oscillation circuit voltage (OCV) were provided by the manufacturer to convert the recorded voltage into an acoustic signal. The TVR describes the frequency-dependent relationship between the transmitted signal amplitude at a 1 V driving voltage and the pressure change at a one-meter distance in µPa. The OCV describes the frequency-dependent amplitude conversion of a 1 µPa pressure field into a voltage response. TVR and OCV are defined for the transducer center frequency at 194 kHz (wavelength 7.73 mm), with amplitudes strongly decreasing for both higher and lower frequencies. Because the transducer-specific response function could not be accurately measured, a constant transducer configuration ensures a comparable signal between the stations for each grazing angle. To ensure no reflection or scattering effects caused by the lander frame or optic unit occurred, the system was tested while elevated in the water column.

#### 2.3. Data Processing

#### 2.3.1. Optical Data

_{x,y}) is zero. An example of a gridded surface is demonstrated in Figure 3a. Each surface contains a spatial wavelength range between 0.002 to 0.17 m. The gridded surfaces were used to derive interface roughness parameters in the form of the root-mean-square roughness (RMS roughness). The RMS roughness was computed according to Equation (1) and describes the mean height variation of the zero mean surface grid.

_{kx,ky}) in Equation (2) is obtained by applying the FFT routine onto the windowed surface S

_{x,y}and shifting the zero-frequency component to the center [36]. By radially averaging the amplitudes of a given spatial wavelength of the 2D power spectrum following the method described in reference [36], an average radial power spectrum (PSD) was computed using Equation (3) and is displayed in Figure 3c.

- PSD: 2D power spectral density [m
^{4}] - W: Fourier transformed windowed surface, given by fft2(winS
_{x,y}) [m] - A: area of the surface grid, given by MΔxNΔy [m
^{2}] - k
_{x}: the wave number in the x direction [m^{−1}], ${k}_{x}=\frac{fs}{M}\left(\left(0,\dots ,m-1\right)-\frac{M}{2}\right),m=1,2,\dots ,M,$ - k
_{y}: the wave number in the y direction [m^{−1}], ${k}_{y}=\frac{fs}{N}\left(\left(0,\dots ,n-1\right)-\frac{N}{2}\right),n=1,2,\dots ,N,$ - f
_{s}: sampling frequency = $\frac{1}{\mathsf{\Delta}\mathrm{x}}$ = $\frac{1}{\mathsf{\Delta}\mathrm{y}}$

- PSD: 2D radial averaged power spectral density [m
^{4}] - N
_{r}: total number of points, which lie upon a circle with radius K - K: 2D wave vector length [m
^{−1}], given by

_{K}is used to derive interface roughness parameters in the form of the spectral slope (γ

_{2}) and spectral intercept (ω

_{2}). The parameters “spectral slope,” which describes the slope of a linear fit, and “spectral strength,” which describes the intercept of the linear fit at a certain spatial wavelength, were calculated as described by references [24,37] and are outlined in Figure 3c. For acoustic modeling, the spectral strength is extrapolated to a spatial wavenumber of K = 1 m

^{−1}[5], which corresponds to a wavelength of 1 m. Details on the used acoustic model are given in Appendix A. Additionally, for comparison with past studies of seafloor roughness [6], the spectral strength was computed at a spatial wavenumber of K = 100 m

^{−1}, which corresponds to a wavelength of 0.01 m.

_{kx,ky}and PSD

_{K}were divided into two intervals (Figure 3c–g). The wavelength ranges for the subdivision were set according to the information gathered from the surface scans and video footage. The D

_{1}roughness domain encompasses the spatial wavelength range 0.03–0.17 m (Figure 3c,e) corresponding to hydrodynamic bedforms (Figure 3d). The D

_{2}domain contains the spatial wavelength range 0.005–0.03 m (Figure 3c,g) corresponding to the size of observed biological features (Figure 3f). An inverse FFT algorithm was applied onto the two PSD

_{kx,ky}intervals D

_{1}(Figure 3e) and D

_{2}(Figure 3g), to visualize the corresponding seafloor structures (Figure 3d,f) and to derive the RMS roughness values for the D

_{1}and D

_{2}intervals by utilizing Equation (1). Considering the prefactor “A” and the squared amplitude of “W” in Equation (2), the RMS roughness values computed from S

_{x,y}and PSD

_{kx,ky}by Equation (1) are equivalent within each interval [37]. Further, the RMS roughness is also directly related to the psd

_{K}by a summation of the radial averaged spatial height variations (blue line in Figure 3c) [37].

_{2}domain could be utilized to obtain the small-scale structures (0.005–0.03 m) of the seafloor surface. A threshold level of 0.001 m was found to best isolate benthic structures, and a particle counter was then used to cluster and count the biotic objects. The benthic coverage (BC) was computed by dividing the number of grid cells covered by the detected objects by the total number of grid cells (Figure 3h).

#### 2.3.2. Acoustic Data

^{2}) is the insonified area in m

^{2}, and TL the transmission loss defined by

#### 2.3.3. Acoustic Scatter Model

## 3. Results

#### 3.1. Ship-Based Acoustic Survey and Ground Truthing

^{2}to account for position inaccuracies and the resulting intensities are displayed in Figure 4. Between all lander stations, the difference in the ship-based backscatter strength is 6.6 dB.

#### 3.2. Lander Experiment

#### 3.2.1. Seafloor Roughness

_{2}(Figure 3). Here, elevated magnitudes in the roughness spectra are observed with increasing biological presence (Figure 6). The curves of roughness magnitudes converge at the larger boundary of this interval, while convergence at the smaller boundary is incomplete. The maximum difference in roughness magnitude is observed at spatial wavelengths of 0.01 m. Correspondingly, there is no correlation found between the BC and the RMS roughness of the D and D

_{1}interval (both R = 0.06), but a strong correlation for the smaller-scale roughness features in the domain D

_{2}(R = 0.73) exists (Table 1). The mean spectral intercept at K = 100 m

^{−1}(equal to a wavelength of 0.01 m) for the complete roughness spectrum is 0.00012 ± 0.00012 over all stations, while the spectral exponent is −4.1 ± 0.27 (Figure 6). Generally, a strong correlation is observed between the benthic abundance and the spectral intercept @ K = 100 m

^{−1}(R = 0.91), while the relationship to the spectral slope is moderate (R = 0.64). No significant correlation is found between the spectral slope and the RMS roughness of any interval. The spectral intercept at K = 1 m

^{−1}contains the extrapolated spatial roughness parameters of seafloor features at 1 m wavelength and is primarily affected by the trend of the roughness spectrum over all wavelengths. Correspondingly, the spectral intercept shows a moderate correlation with the RMS roughness. The spectral intercept at K = 100 m

^{−1}is affected by the roughness spectrum with wavelengths <0.01 m and only correlates with the RMS roughness of the D

_{2}interval.

#### 3.2.2. Acoustic Scatter

## 4. Discussion

_{1}interval, which are dominated by larger hydrodynamic bedforms and the BC (Table 1) in the available laser scanner–derived data. While high abundances of deposit feeders may increase large-scale seafloor roughness due to the local presence of larger mounds [6,7,40], the impact of benthic life on seafloor roughness is usually focused on higher spatial frequencies [19]. A further possibility for benthic life to impact large-scale roughness is the deconstruction of hydrodynamic bedforms [41], which decreases large-scale roughness. This is rarely observed in the video data (an eventual example is shown in Figure 5a) and depends both on the abundance of benthic life and the present hydrodynamic conditions. However, a clear positive correlation exists between BC and roughness derived from higher spatial frequencies (Figure 6), despite the lower limit of the BC being much less than in previously reported studies [7,41,42]. Because the substrate type does not differ between the stations and the present hydrodynamic bedforms are not expressed at high spatial frequencies (Figure 3), the strong correlation is caused by biological presence with a peak impact at spatial wavelengths of 0.01 m. This can also be observed in older 1D power spectra derived from stereo-photogrammetry [43]. Due to this peak, the presence of the local benthic life has a marked control on both spectral slope and intercept (Figure 7a). While both spectral slope and intercept increase, the spectral intercept at K = 100 m

^{−1}was more clearly impacted (Figure 7). The reason for the reduced correlation with spectral slope is the remaining scatter in large-scale roughness caused by variable ripple amplitudes and wavelengths, which changes the slope of the linear regression of the psd

_{K}. At higher benthic abundances above 2%, the two available data points may indicate the presence of a threshold effect limiting spectral slope and intercept. To constrain this effect, additional surface models are required.

_{1}and D

_{2}RMS roughness), which differ over a limited spatial frequency domain (D

_{1}or D

_{2}interval) without being related by a continuous alteration process, is referred to follow a scale-free, power-law behavior. A scale-free, power-law behavior is not applicable to derive input parameters for high-frequency acoustic modeling [5]. Therefore, slope and intercept values derived by the Heincke data with benthic coverage <2% must be considered to introduce error associated with the backscatter prediction. Unfortunately, no lander-based acoustic data was available for the Mya data [7] with benthic coverage >2.5%, where an interaction between epibenthos and hydrodynamic bedforms was demonstrated.

## 5. Conclusions

## Supplementary Materials

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## Appendix A

## References

- Lyons, B.P.; Thain, J.E.; Stentiford, G.D.; Hylland, K.; Davies, I.M.; Vethaak, A.D. Using biological effects tools to define Good Environmental Status under the European Union Marine Strategy Framework Directive. Mar. Pollut. Bull.
**2010**, 60, 1647–1651. [Google Scholar] [CrossRef] [PubMed] - Report of the Working Group on Acoustic and Egg Surveys for Sardine and Anchovy in ICES Areas VIII and IX (WGACEGG); ICES: Palma de Mallorca, Spain, 2007; p. 163.
- Bouwma, I.; Schleyer, C.; Primmer, E.; Winkler, K.J.; Berry, P.; Young, J.; Carmen, E.; Jana, Š.; Bezák, P.; Preda, E. Adoption of the ecosystem services concept in EU policies. Ecosyst. Serv.
**2018**, 29, 213–222. [Google Scholar] [CrossRef] - Roadmap for Maritime Spatial Planning: Achieving Common Principles in the EU; European Commission: Brussels, Belgium, 2008; pp. 1–11.
- Jackson, D.R.; Richardson, M. High-Frequency Seafloor Acoustics; Springer Science & Business Media: Berlin, Germany, 2007; ISBN 9780387369457. [Google Scholar]
- Self, R.F.L.; A’Hearn, P.; Jumars, P.A.; Jackson, D.R.; Richardson, M.D.; Briggs, K.B. Effects of macrofauna on acoustic backscatter from the seabed: Field manipulations in West Sound, Orcas Island, Washington, U.S.A. J. Mar. Res.
**2001**, 59, 991–1020. [Google Scholar] [CrossRef] - Schönke, M.; Feldens, P.; Wilken, D.; Papenmeier, S.; Heinrich, C.; von Deimling, J.S.; Held, P.; Krastel, S. Impact of Lanice conchilega on seafloor microtopography off the island of Sylt (German Bight, SE North Sea). Geo-Marine Lett.
**2016**, 37, 305–318. [Google Scholar] [CrossRef] - Wever, T.; Jenkins, C. Physik biologisch besiedelter Meeresböden. J. Appl. Hydrogr.
**2017**, 106, 8–13. [Google Scholar] - Briggs, K.B.; Williams, K.L.; Richardson, M.D.; Jackson, D.R. Effects of Changing Roughness on Acoustic Scattering: (1) Natural Changes. Proc. Inst. Acoust.
**2001**, 23, 343–390. [Google Scholar] - Richardson, M.D.; Briggs, K.B.; Bentley, S.J.; Walter, D.J.; Orsi, T.H. The effects of biological and hydrodynamic processes on physical and acoustic properties of sediments off the Eel River, California. Mar. Geol.
**2002**, 182, 121–139. [Google Scholar] [CrossRef] - Ballard, M.S.; Lee, K.M. The Acoustics of Marine Sediments. Acoust. Today
**2017**, 13, 11–18. [Google Scholar] - Moore, K.D.; Jaffe, J.S. Time-evolution of high-resolution topographic measurements of the sea floor using a 3-D laser line scan mapping system. IEEE J. Ocean. Eng.
**2002**, 27, 525–545. [Google Scholar] [CrossRef] - Richardson, M.D.; Briggs, K.B.; Bibee, L.D.; Jumars, P.A.; Sawyer, W.B.; Albert, D.B.; Bennett, R.H.; Berger, T.K.; Buckingham, M.J.; Chotiros, N.P.; et al. Overview of SAX99: Environmental considerations. IEEE J. Ocean. Eng.
**2001**, 26, 26–53. [Google Scholar] [CrossRef] - Heinrich, C.; Feldens, P.; Schwarzer, K. Highly dynamic biological seabed alterations revealed by side scan sonar tracking of Lanice conchilega beds offshore the island of Sylt (German Bight). Geo-Marine Lett.
**2017**, 37, 289–303. [Google Scholar] [CrossRef] - McGonigle, C.; Grabowski, J.H.; Brown, C.J.; Weber, T.C.; Quinn, R. Detection of deep water benthic macroalgae using image-based classification techniques on multibeam backscatter at Cashes Ledge, Gulf of Maine, USA. Estuar. Coast. Shelf Sci.
**2011**, 91, 87–101. [Google Scholar] [CrossRef] - McGonigle, C.; Brown, C.; Quinn, R.; Grabowski, J. Evaluation of image-based multibeam sonar backscatter classification for benthic habitat discrimination and mapping at Stanton Banks, UK. Estuar. Coast. Shelf Sci.
**2009**, 81, 423–437. [Google Scholar] [CrossRef] - Che Hasan, R.; Ierodiaconou, D.; Laurenson, L. Combining angular response classification and backscatter imagery segmentation for benthic biological habitat mapping. Estuar. Coast. Shelf Sci.
**2012**, 97, 1–9. [Google Scholar] [CrossRef][Green Version] - Le Bas, T.P.P.; Huvenne, V.A.I. Acquisition and processing of backscatter data for habitat mapping—Comparison of multibeam and sidescan systems. Appl. Acoust.
**2009**, 70, 1248–1257. [Google Scholar] [CrossRef] - Feldens, P.; Schulze, I.; Papenmeier, S.; Schönke, M.; Schneider von Deimling, J. Improved Interpretation of Marine Sedimentary Environments Using Multi-Frequency Multibeam Backscatter Data. Geosciences
**2018**, 8, 214. [Google Scholar] [CrossRef] - Roche, M.; Degrendele, K.; Vrignaud, C.; Loyer, S.; Le Bas, T.; Augustin, J.M.; Lurton, X. Control of the repeatability of high frequency multibeam echosounder backscatter by using natural reference areas. Mar. Geophys. Res.
**2018**, 39, 89–104. [Google Scholar] [CrossRef][Green Version] - Lurton, X.; Lamarche, G.; Brown, C.; Lucieer, V.; Rice, G.; Schimel, A.; Weber, T. Backscatter Measurements by Seafloor—Mapping Sonars. In Guidelines and Recommendations; Lurton, X., Lamarche, G., Eds.; 2015; p. 200. Available online: http://geohab.org/wp-content/uploads/2013/02/BWSG-REPORT-MAY2015.pdf (accessed on 19 October 2019).
- Wendelboe, G. Backscattering from a sandy seabed measured by a calibrated multibeam echosounder in the 190–400 kHz frequency range. Mar. Geophys. Res.
**2018**, 39, 105–120. [Google Scholar] [CrossRef] - Richardson, M.; Briggs, K.; Williams, K.; Tang, D.; Jackson, D.; Thorsos, E. The effects of seafloor roughness on acoustic scattering: Manipulative experiments. In Boundary Influences in High Frequency Shallow Water Acoustics; Pace, N., Blondel, P., Eds.; University of Bath: Bath, UK, 2005; Volume 298, pp. 109–116. [Google Scholar]
- Buscombe, D.; Grams, P.E.; Kaplinski, M.A. Characterizing riverbed sediment using high-frequency acoustics: 1. Spectral properties of scattering. J. Geophys. Res. F Earth Surf.
**2014**, 119, 2674–2691. [Google Scholar] [CrossRef] - Huff, L. Acoustic Remote Sensing as a Tool for Habitat Mapping in Alaska Waters. Mar. Habitat Mapp. Technol. Alaska
**2008**, 10, 29–46. [Google Scholar] - Tang, D. Fine-scale measurements of sediment roughness and subbottom variability. IEEE J. Ocean. Eng.
**2004**, 29, 929–939. [Google Scholar] [CrossRef] - Chotiros, N.P.; Isakson, M.J. The evolution of sediment acoustic models. AIP Conf. Proc.
**2012**, 1495, 193–201. [Google Scholar] - Jackson, D.R.; Briggs, K.B. High-frequency bottom backscattering: Roughness versus sediment volume scattering. J. Acoust. Soc. Am.
**2005**, 92, 962–977. [Google Scholar] [CrossRef] - Ivakin, A.N. High frequency scattering from sandy sediments: Roughness vs. discrete inclusions. In Proceedings of the Boundary Influences in High Frequency Shallow Water Acoustics, Bath, UK, 5–9 September 2005. [Google Scholar]
- Ivakin, A.N. Scattering from discrete inclusions in marine sediments. In Proceedings of the Seventh European Conference on Underwater Acoustics, ECUA 2004, Delft, The Netherlands, 5–8 July 2004. [Google Scholar]
- Valerius, J.; Kösters, F.; Zeiler, M. Erfassung von Sandverteilungsmustern zur großräumigen Analyse der Sedimentdynamik auf dem Schelf der Deutschen Bucht. Die Küste
**2015**, 83, 39–63. [Google Scholar] - Zeiler, M.; Schwarzer, K.; Ricklefs, K. Seabed morphology and sediment dynamics. Kuste
**2008**, 74, 31–44. [Google Scholar] - Salzwedel, H.; Eike, R.; Dieter, G. Benthic macrofauna communities in the German Bight. In Veröffentlichungen des Instituts für Meeresforschung in Bremerhaven; Salzwedel, H., Eike, R., Dieter, G., Eds.; Institute for Marine Research in Bremerhaven: Bremerhaven, Germany, 1985. [Google Scholar]
- Benthosuntersuchungen im Rahmen des Beweissicherungsverfahrens für den Ausbau des Hafenbecken IV in Büsum. Available online: https://epic.awi.de/id/eprint/44126/ (accessed on 22 October 2019).
- Panda, S.; Panzade, A.; Sarangi, M.; Chowdhury, S.K.R. Spectral Approach on Multiscale Roughness Characterization of Nominally Rough Surfaces. J. Tribol.
**2017**, 139, 1–10. [Google Scholar] [CrossRef] - Sidick, E. Power spectral density specification and analysis of large optical surfaces. Model. Asp. Opt. Metrol. II
**2009**, 7390, 73900L. [Google Scholar] - Briggs, K.B.; Lyons, A.P.; Pouliquen, E.; Mayer, L.A.; Richardson, M.D. Seafloor roughness, sediment grain size, and temporal stability. Nav. Res. LAB STENNIS Sp. Cent. MS SEAFLOOR Sci. Dir.
**2005**, 298, 9. [Google Scholar] - Bjørnø, L. Applied Underwater Acoustics; Neighbors, T., Bradley, D., Eds.; Elsevier: Amsterdam, The Netherlands, 2017; ISBN 978-0-12-811240-3. [Google Scholar]
- Degraer, S.; Moerkerke, G.; Rabaut, M.; Van Hoey, G.; Du Four, I.; Vincx, M.; Henriet, J.P.; Van Lancker, V. Very-high resolution side-scan sonar mapping of biogenic reefs of the tube-worm Lanice conchilega. Remote Sens. Environ.
**2008**, 112, 3323–3328. [Google Scholar] [CrossRef] - Peine, F.; Friedrichs, M.; Graf, G. Potential influence of tubicolous worms on the bottom roughness length z 0 in the south-western Baltic Sea. J. Exp. Mar. Bio. Ecol.
**2009**, 374, 1–11. [Google Scholar] [CrossRef] - Jackson, D.R.; Richardson, M.D.; William, K.L.; Lyons, A.P.; Jones, C.D.; Briggs, K.B.; Tang, D. Acoustic observation of the time dependence of the roughness of sandy seafloors. IEEE J. Ocean. Eng.
**2009**, 34, 407–422. [Google Scholar] [CrossRef] - Wang, L.; Davies, G.; Bellettini, A.; Pinto, M. Multipath Effect on DPCA Micronavigation of a Synthetic Aperture Sonar. In Impact of Littoral Environmental Variability of Acoustic Predictions and Sonar Performance; Pace, N.G., Jensen, F.B., Eds.; Springer: Dordrecht, The Netherlands, 2002; pp. 465–472. [Google Scholar]
- Williams, K.L.; Jackson, D.R.; Thorsos, E.I.; Tang, D.; Briggs, K.B. Acoustic backscattering experiments in a well characterized sand sediment: Data/model comparisons using sediment fluid and Biot models. IEEE J. Ocean. Eng.
**2002**, 27, 376–387. [Google Scholar] [CrossRef] - Lyons, A.P.; Fox, W.L.J.; Hasiotis, T.; Pouliquen, E. Characterization of the two-dimensional surface roughness of a wave-rippled sea floor using digital photography. IEEE J. Ocean. Eng.
**2002**, 27, 515–524. [Google Scholar] [CrossRef] - Montereale-Gavazzi, G.; Roche, M.; Lurton, X.; Degrendele, K.; Terseleer, N.; Van Lancker, V. Seafloor change detection using multibeam echosounder backscatter: Case study on the Belgian part of the North Sea. Mar. Geophys. Res.
**2018**, 39, 229–247. [Google Scholar] [CrossRef] - Montereale-Gavazzi, G.; Roche, M.; Degrendele, K.; Lurton, X.; Terseleer, N.; Baeye, M.; Francken, F.; Van Lancker, V. Insights into the Short-Term Tidal Variability of Multibeam Backscatter from Field Experiments on Different Seafloor Types. Geosciences
**2019**, 9, 34. [Google Scholar] [CrossRef] - Briggs, K.B.; Tang, D.; Williams, K.L. Characterization of interface roughness of rippled sand off fort Walton Beach, Florida. IEEE J. Ocean. Eng.
**2002**, 27, 505–514. [Google Scholar] [CrossRef]

**Figure 1.**Left hand side: Overview and bathymetry map offshore Sylt. Middle: A ship-based backscatter intensity map of the investigation area. Right hand side: Close up of the study area and the lander deployment stations (red cross), with Norbit multibeam echosounder backscatter data recorded at a frequency of 200 kHz. The dynamic range of the shown uncalibrated backscatter values is approx. 10 dB.

**Figure 2.**Setup of the lander frame. The size of the lander frame is 120 cm × 80 cm × 100 cm. Displayed are the slant range and the grazing angle θ of the selected profiler, which is defined as the angle between the wavefield motion and the zero mean seafloor surface. The red circles mark the areas of the acoustic footprint. In blue the laser swath is highlighted. The ULS200 mid-range laser scan system is manufactured by 2G Robotics, and the acoustic transducer units by Benthowave Instruments Inc.

**Figure 3.**An example of a measured surface by the laser line scanner is displayed in (

**a**), while (

**b**) shows the corresponding power spectral density (PSD) containing the wavelength interval D (0.002–0.17 m). The blue line in (

**c**) shows the radial averaging PSD of (

**b**). The black line in (

**c**)represents the linear fit of the radial averaged PSD required to derive the spectral slope $\gamma $ and the spectral intercepts at the wavelengths 0.01 m and 1 m. The vertical dashed line in (

**c**) indicates the separation of the spectrum D into the intervals D

_{1}and D

_{2}, which is equivalent to separation of the PSD demonstrated in (

**d**)–(

**g**). The inverse transform of the low-pass filtered PSD D

_{1}interval is shown in (

**d**) with the corresponding spectrum in (

**e**). The inverse transform of the PSD D

_{2}covering wavelength <0.03 m is shown in (

**f**), with the corresponding high-pass filtered spectrum in (

**g**). The red circles in (

**e**) and (

**g**) indicate the cutoff wavelength. The red markers in (

**h**) highlight the detected objects from (

**f**), which were used to compute the benthic coverage (BC).

**Figure 4.**Predicted backscatter strength using the small-roughness perturbation approximation (blue), the measured backscatter strength (BS) from the lander (cyan), the measured backscatter strength from the ship-based system (black), root-mean-square roughness (RMS roughness) for the D

_{1}interval (red), RMS roughness for the D

_{2}interval (yellow), and benthic coverage (BC) at the seafloor for the lander stations (green). The last two station digits correspond to the position of the lander stations displayed in Figure 1.

**Figure 5.**Video footage from a survey near station 63_1 (

**a**), (

**b**), (

**d**), (

**e**) (red dots have a 10 cm distance), from the lander station 67 (

**c**) and the corresponding position in the ship-based backscatter mosaic (

**f**). (

**a**) reveals asymmetrical straight to sinusoidal ripples with sparse benthic coverage and organic fluff located inside the ripple troughs. (

**b**) indicates the transition is between the ripple-dominated and the benthic-dominated seafloor type. In (

**c**), degenerate ripple structures were observed with few tubeworm structures, shell fragments, and brittlestars (Ophiuroidea). (

**d**) reveals degraded ripple residuals and a higher coverage of tubeworm structures, few starfish, and larger shell fragments. The morphology in (

**e**) shows no ripple structures and very few visible epibenthic features besides some organic fluff. The ship-based backscatter mosaic in (

**f**) shows a corresponding change from low to high backscatter along the video track. However, the positioning is not sufficiently accurate to relate video snapshots directly to the backscatter mosaic.

**Figure 6.**Left hand side psd

_{K}over all lander stations. The vertical dashed line indicates the cutoff frequency (K = 33 m

^{−1}) between the D

_{1}and D

_{2}domain. The color shows the corresponding benthic coverage. High RMS roughness values in the D

_{2}domain correlate with high benthic coverage. Right hand side, backscatter strength predicted by the small-roughness perturbation approximation given by reference [6] for the lander sites with acoustic data.

**Figure 7.**(

**a**) Spectral slopes and intercepts displayed against the benthic coverage. Grain size is constant at 2.6 phi. (

**b**) A slope vs. intercept diagram for the study site. Symbol color denotes RMS roughness, symbol size denotes the benthic coverage in percent. Arrow 1: Effect of increasing high-frequency roughness due to benthic life (this study). Arrow 2: Effect of relative decrease of low-frequency roughness [9]. Arrow 3: Effect of the relative increase of low-frequency roughness [9].

**Table 1.**Correlation matrix of the studied parameters. Entries highlighted show a probability value << 0.05.

BS_{lander} | BS_{model} | RMS _{D} Roughness | RMS_{D1} Roughness | RMS_{D2} Roughness | Spectral Slope | Spectral Intercept K = 1 m^{−1} | Spectral Intercept K = 100 m^{−1} | BC | |
---|---|---|---|---|---|---|---|---|---|

BS_{ship} | 0.03 | −0.30 | −0.25 | −0.25 | −0.08 | −0.28 | −0.16 | −0.19 | −0.32 |

BS_{lander} | 1 | 0.27 | 0.18 | 0.18 | 0.32 | 0.17 | 0.06 | 0.46 | 0.57 |

BS_{model} | 1 | −0.26 | −0.25 | 0.21 | 0.96 | −0.71 | 0.88 | 0.8 | |

RMS_{D} roughness | 1 | 1 | 0.56 | −0.4 | 0.59 | −0.02 | 0.04 | ||

RMS_{D1} roughness | 1 | 0.55 | −0.38 | 0.57 | −0.02 | 0.04 | |||

RMS_{D2} roughness | 1 | −0.06 | 0.48 | 0.56 | 0.56 | ||||

spectral slope | 1 | −0.86 | 0.74 | 0.64 | |||||

spectral intercept K = 1 m^{−1} | 1 | −0.39 | −0.33 | ||||||

spectral intercept K = 100 m^{−1} | 1 | 0.91 |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Schönke, M.; Wiesenberg, L.; Schulze, I.; Wilken, D.; Darr, A.; Papenmeier, S.; Feldens, P.
Impact of Sparse Benthic Life on Seafloor Roughness and High-Frequency Acoustic Scatter. *Geosciences* **2019**, *9*, 454.
https://doi.org/10.3390/geosciences9100454

**AMA Style**

Schönke M, Wiesenberg L, Schulze I, Wilken D, Darr A, Papenmeier S, Feldens P.
Impact of Sparse Benthic Life on Seafloor Roughness and High-Frequency Acoustic Scatter. *Geosciences*. 2019; 9(10):454.
https://doi.org/10.3390/geosciences9100454

**Chicago/Turabian Style**

Schönke, Mischa, Lars Wiesenberg, Inken Schulze, Dennis Wilken, Alexander Darr, Svenja Papenmeier, and Peter Feldens.
2019. "Impact of Sparse Benthic Life on Seafloor Roughness and High-Frequency Acoustic Scatter" *Geosciences* 9, no. 10: 454.
https://doi.org/10.3390/geosciences9100454