Acoustic signatures of gassy sediments in two subtropical lakes – Lake Kinneret (Israel) and Lake Biwa (Japan)

Please download to get full document.

View again

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
"Methane is common in organic-rich marine and freshwater sediments. Escape of this gas via ebullition is the most efficient mechanism of methane transport from the shallow sediments to the atmosphere. Therefore characterization and
  Proceedings of Meetings on Acoustics Volume 17, 2012  ECUA 2012 11th European Conference on Underwater Acoustics  Edinburgh, Scotland2 - 6 July 2012 Session UW: Underwater AcousticsUW100. Acoustic signatures of gassy sediments in two subtropical lakes - Lake Kinneret(Israel) and Lake Biwa (Japan) Jaroslaw Tegowski*, Ilia Ostrovsky, Kumagai Michio, Zamaryka Mateusz and Kanako Ishikawa*Corresponding author’s address: Institute of Oceanography, University of Gdansk, Al. Marszalka Pilsudskiego46, Gdynia, 81-378, pomorskie, Poland, Methane is common in organic-rich marine and freshwater sediments. Escape of this gas via ebullition is the most effi-cient mechanism of methane transport from the shallow sediments to the atmosphere. Therefore characterization andquantification of free gases in surface sediments is an important ecological issue. High spatial heterogeneity of methanedistribution does not allow using conventional sampling methods for representative quantification of sedimentary gasesover large areas, such that development of remote-sensing methodology is required to investigate gassy sediment distribu-tion. In this paper we compare the results of acoustical sampling to study sound scattering in soft gassy sediments in twodeep subtropical lakes - Lake Kinneret and Lake Biwa. Acoustic samplings were carried out using 70 kHz and 120 kHzsingle beam echosounders along standard transects. To determine the typical acoustic features of gassy sediments weconducted echo envelope parameterisation using different spectral, wavelet, fractal, statistical and energy parameters. Anumber of parameters appeared to be sensitive to the presence free gases in sediments. Comparison of the results obtainedon two lakes allows detecting the effect of sediment type, its location, and water level change on acoustic properties of thetop layer of gassy sediments.Published by the Acoustical Society of America through the American Institute of Physics Tegowski et al.© 2012 Acoustical Society of America [DOI: 10.1121/1.4773461]Received 16 Nov 2012; published 13 Dec 2012Proceedings of Meetings on Acoustics, Vol. 17, 070066 (2012) Page 1    1 INTRODUCTION Methane and carbon dioxide are the most important greenhouse gases. They are produced in organic-rich marine and freshwater sediments. However, the fate of these gases in sediments depends onambient conditions, their production, dissolution and consumption rate, formation of bubbles, their growth and migration. The main biogeochemical source of methane in sediments is methanogenesisassociated with anaerobic bacterial decomposition of the deposited organic matter. Other sources of sedimentary gases are submarine geothermal processes, and infiltration of gases (e.g. nitrogen, argon,helium) dissolved in the near-bottom water [1]. Methane is usually the dominant gas in bottomsediments. Because of low dissolution rate of methane comparatively to carbon dioxide, escape of thisgas via ebullition is the most efficient mechanism of methane transport from the shallow sediments tothe atmosphere. Quantification and characterization of free gases in surface sediments is an importantecological issue. High spatial heterogeneity of methane distribution does not allow using conventionalsampling methods for representative quantification of this sedimentary gas over large areas, such thatdevelopment of remote-sensing techniques is required to investigate and map gassy sedimentdistribution. New gas monitoring techniques are also requited for detection and quantification of carbondioxide escaping from storage caverns, which were artificially created under the sea floor for sequestration of this gas. The presence of gas in sediments causes the changes of their elasticproperties, which can be studied by changes of sound attenuation, acoustic wave velocity, andreflective features of the upper sediment layer [1]. Tegowski et al.Proceedings of Meetings on Acoustics, Vol. 17, 070066 (2012) Page 2  Fig.1. Bathymetric maps of Lake Biwa (left) and Lake Kinneret (right). The black lines identifylocations of acoustic transects.In this paper we study the sound scattering in soft gassy sediments in two deep large lakes - Lake Biwa(Japan) and Lake Kinneret (Israel) (Fig. 1) in order to detect specific reflectance properties that canhelp mapping the presence of gas below the bottom surface. 2 MEASUREMENTS ON LAKE BIWA AND LAKE KINNERET Lake Biwa (Fig. 1) is the largest freshwater lake in Japan located in west-central Honsh nj Island. Itssurface area is 672 km 2 , maximal depth is 104 m. Lake Kinneret (Sea of Galilee) is the largestfreshwater lake in Israel. Lake Kinneret surface area is ~165 km 2 , its maximal depth is varied from 39 to44 m depending on water level. It provides about 50% of the country's water demand for drinking water and agricultural needs. Sandy sediments dominate the littoral, while soft mud prevails in deeper area[5]. Large fluctuations in Lake Kinneret water level affect the amount of methane bubbles released frombottom sediments to the overlying water. The most intensive emission of methane from the sedimentswas recorded at the lowest water level and was associated with decreased hydrostatic pressure at thebottom [3]. Methane in Lake Biwa and Lake Kinneret is predominantly biogenic [2,3]. Acoustic samplings of Lake Biwa were carried out using Kaijo KFC-3000 single beam echosounders(Kaijo Corp., Japan) working at 70 kHz (3 dB beam width – 19.8°, pulse width τ =0.48 ms). In LakeKinneret acoustic sampling was done with the Biosonics DE5000 echo sounder operated at 120 kHz(beam width - 6.5°, τ =0.2 ms). In both lakes measurements were conducted along standard transects(Fig.1). To determine typical acoustic features of gassy sediments we carried out an echo envelopeparameterization using different spectral, wavelet, fractal, statistical and energy parameters [5].Examples of 70-kHz (Lake Biwa) and 120-kHz (Lake Kinneret) echograms displaying gas outflow fromthe gas-saturated sediments are shown in Fig. 2. Tegowski et al.Proceedings of Meetings on Acoustics, Vol. 17, 070066 (2012) Page 3  Fig.2. Echograms displaying gas emission in (a) Lake Biwa (18/12/2010) and (b) Lake Kinneret(15/11/2007). Inclined lines represent the tracks of raising individual gas bubbles. The gasseepages are seen as columns of tracks. Long horizontal noisy trails in the interior part of the water column (~30-m depth in Lake Biwa and ~20-m depths in Lake Kinneret) are acoustic scatteringlayers associated with the presence of suspended particles, plankton or gas bubbles in thethermocline. The Lake Biwa echogram shows acoustical traces of fishes below 60-65 m depth. 3 PARAMETRIC ANALYSIS OF ECHOES FROM SINGLE BEAMECHOSOUNDER Parametrical analysis of echo signals has been developed in numerous bottom classification systems[5]. The currently available seabed classification methods either use small set of known echo envelopeparameters (eg. VBT, RoxAnn [4]) or large number of unrevealed envelope descriptors (e.g. QTC [5]).In all systems the parameters are used as input sets to supervised or unsupervised classificationalgorithms (e.g. principal component analysis, fuzzy logic, k-means and neural networks) in order todeliver classified maps of sediments or morphological types of the bottom. In this work we test a set of echo parameters that can be characteristic for sediments containing large amount of gas bubbles.Before parameter computation, the procedures of signals correction were applied. The consecutiveenvelopes were divided for sets of 20 pulses, where only echoes of energy greater than 75% of maximum pulse energy in the set were analysed. The other procedure compensated dependency of echo shape on the bottom depth following the Caughey and Kirlin algorithm [6]  R H t t  0 ' ⋅= , where t  ’ is rescaled time,  H  0 – reference depth (in our case 10-m),  R – distance from transducer to the bottom.The TVG function was established for 30log 10  R . The computations of parameters were conducted for echo envelopes expressed in logarithmic form of  Sv (volume backscattering strength) and in itsrecalculated linear form. The first group of echo features describes the part of energy coming from thesurface scattering – S  att and energy scattered at the sediment volume (echo tail) - S  dec . The set of statistical parameters contained envelope autocorrelation length, statistical moments and moment of inertia. Parameters very sensitive for detection of gas bubbles in the bottom were derived from productsof Fourier transform, as spectral moments – m r  ( r  = 0-12) and their combinations - spectral widths ε 2 ,  ν 2 ,skewness γ  and central frequency ω  0 : Tegowski et al.Proceedings of Meetings on Acoustics, Vol. 17, 070066 (2012) Page 4  ( ) ³ ∞ = 0 ω ω ω  d S m r r  , (1) ,,~~, 010232321202 mmmmmmm === ω γ  ν  (2)where S  ( ω  ) is power spectral density of echo envelope. The slope  β  of spectrum was utilized for estimation of fractal dimension in form of   D = (5-  β  )/2 [6]. The normalized spectrum of echo signal C   f   was the base of classification parameters defined as relationships of integrals of parts power spectraldensities to integral of total spectral density: ³ =  Ny1 f 0f  df  C S   f   , ³ =  Ny1m f m10f f  1 df  C S S   f   (3)were m =2, 4, 8, 16 and  f    Ny is the Nyquist’s frequency. The last large groups of echo envelopeparameters are products of wavelet transformation, which was calculated for Coiflet, Daubechies andMeyer wavelets. Sum of transformation coefficients for chosen dyadic scales a =2  j (where j=1-12) werethe base for computation of the wavelet energies E  wav . Other parameter received from wavelettransform was the Hausdorff exponent  H  and fractal dimension  D wav = 2-  H  calculated for differentwavelets. The listed above echo envelope features were tested for sensitivity to gas in top layer of lakesediments. 4 RESULTS AND DISCUSSION The presence of large amount of gas bubbles in the top layer of bottom sediments can be the mainreason of echo shape fluctuations in otherwise homogenous muddy sediments. Gas bubbles trapped insediments radically change its properties, reflective features [5] and thus affect the echo envelopeparameters. Fig. 4 displays a small section of acoustic transect with two bubble seeps and concomitantvariations of selected echo envelope parameters computed from signal backscattered from the bottom.Rapid changes in parameter values (see the areas confined by dashed lines on Fig.4 were detected onparts of transect from where the gas bubbles were emitted. Such changes are apparently associatedwith locations of high abundance of bubbles in soft sediments. Similar changes in parameter values inother sections of transect probably pinpoint the areas of the enlarged abundance of bubbles insediments without their visible emission to the water column. Similarly, Fig.5 shows good matching of peaks of echo envelope parameters to the location of methane seepage along a transect in LakeKinneret.Fig. 6 presents dependence of (a) spectral width and (b) spectral moment of 7-th order of echoenvelopes on depth of bottom and water level in Lake Kinneret Both parameters achieved their maximain the deepest part of the lake at the lowest water level, when intensity of gas emission was the largest[3, 5]. This is in good agreement with data presented on Figs. 4 and 5. The obtained results suggeststrong relationship between the reflectance properties of sediments and water level (hydrostaticpressure), which controls abundance of gas bubbles in Lake Kinneret soft sediments. Thus, thepresence and abundance of gas bubbles in sediments is one of the key factors affecting its acousticfeatures. For more accurate understanding of the impact of water level fluctuation on acoustic Tegowski et al.Proceedings of Meetings on Acoustics, Vol. 17, 070066 (2012) Page 5
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!