LICENCIATE THESIS

IN MARINE GEOLOGY

Acoustic characterization of submarine geomorphological features in the Polar Oceans

Francis Freire

2014 Stockholm University

Abstract

Marine glacial environments contain unique seafloor features resulting from the

dynamic glacial processes. Studying these submarine geomorphological features can help

us understand the glacial paleo-environments so that we can predict the likely responses of

present day glaciers and ice sheets to future changes in the climate. This thesis details

different approaches in understanding glacial seafloor features using acoustic systems. It

focuses on the novel technique of automated mapping seafloor properties using the

backscatter intensity collected by acoustic multibeam echosounder systems (MBES). The

aim of this thesis is to assess the potential of this unexploited data source in characterizing

different glacial landforms in the polar oceans. This is done by examining the voluminous

backscatter data collected by Swedish icebreaker Oden from different cruises to the polar

oceans and employing an automated backscatter processing technique, the ARA algorithm,

to extract surficial sediment characteristics. The results from the sediment characterization

are used together with outputs from other marine acoustical systems and sediment core

data to understand formational processes of the glacial submarine features. Operational

issues encountered in using this technology and its viability as a tool in characterization of

glacial seafloor features are discussed and suggestions are given on the improvements

needed to effectively implement the method in future studies. The final part of the

manuscript is a paper, published in Geo-marine Letters, where I and my co-authors show a

practical application of the acoustic systems ability to characterize geomorphological

features of a mass-wasting event in the deepest part of the Arctic, the Molloy Hole.

Table of Contents

Part I - Introduction ....................................................................................................................... 1

Part II. Background ....................................................................................................................... 4

Submarine glaciogenic seafloor features in the Polar Oceans ......................................... 4

Acoustic-based characterization of the seafloor .............................................................. 6

Single beam echo sounder (SBES) ............................................................................ 7

Side-scan sonar ......................................................................................................... 7

Seismic Reflection ..................................................................................................... 8

Multibeam echo-sounder systems (MBES) .............................................................. 9

Backscatter as a tool in studying the ocean seafloor ............................................. 10

Processing of backscatter data for seafloor characterization ................................ 12

Part III. Discussion and Conclusions ............................................................................................ 14

Part IV. Summary of the published manuscript .......................................................................... 19

Part V. Ongoing projects ............................................................................................................. 20

Part VI. Acknowledgments .......................................................................................................... 21

Part VII. References ..................................................................................................................... 25

Part VIII. Paper ............................................................................................................................ 37

1

Part I - Introduction

Sediments and landforms resulting from past glaciations are now preserved on the

present continental shelves and fjords in the Polar oceans and they reflect the dynamic

nature of the advance and retreat of ice-sheets throughout the Quaternary glacial history

[Wellner et al., 2006; Ottesen and Dowdeswell, 2009]. Understanding of the past

glaciological conditions and processes by studying these features can elucidate likely

responses of present day glaciers and ice sheets to future changes in the climate [Bingham

et al., 2010].

Studying submarine features in the deep and cold polar oceans are more complicated

compared to the terrestrial counterpart because of the overlaying water mass that prevents

direct examination of the sea floor. Nevertheless, this also protects the sea floor and its

sediments from weathering and climatic changes, and preserves their past formational

imprints [Jakobsson et al., 2008; Bingham et al., 2010]. Geomorphological information

about the seafloor are collected by actual sediment samples from dredges and core samples

which provide a direct means to observe the undersea sediments (e.g., Kilfeather et al.

2011). However, their results are limited in terms of spatial extent and data collection is

operationally difficult, especially for deep seafloor areas. This limitation is addressed by

using acoustic remote sensing tools that can collect the desired seafloor information over

large areas.

Probing the ocean seafloor through the overlying water mass is best accomplished

using acoustic waves because of their ease in propagating through the water medium

[Lurton, 2010]. The physical characteristics of acoustic waves that are reflected back after

hitting the seafloor and its sub-surface sediments are analyzed and processed to visualize

the seafloor and its underlying strata. These images can accurately capture seafloor and

sub-bottom geological and geomorphological characteristics which are essential sources of

information for reconstructing paleo-environments. However, there are many geophysical

factors in acoustic wave propagation that needs to be considered for its successful

execution. A more detailed discussion on this topic is found in the next chapter.

Technological advancement in the field of marine acoustics has made significant

progress in the last two decades. One of the most important developments in the field is the

emergence of reliable multibeam echosounders (MBES) with accurate positioning through

2

GPS, which allows the collection of high quality bathymetric data over large areas to

visualize and analyze seafloor morphology [Hughes Clarke et al., 1996; Vanneste et al.,

2006; Dowdeswell et al., 2006, 2008, 2010a; Anderson and Fretwell, 2008; Andreassen et

al., 2008; Jakobsson et al., 2008, 2011; Maclean et al., 2010; Dunlop et al., 2011;

Vanneste and L’Heureux, 2012].

Modern MBES systems are capable of producing high quality bathymetric outputs of

down to tens of centimeter in horizontal resolution, which is well sufficient for use in

studying submarine geomorphology. In addition, these new MBES systems can also be

used in seafloor characterization because they also record highly improved, co-registered

acoustic backscatter strength over wide swaths of seafloor – a tool that is used to map

surficial seabed material properties [Dartnell and Gardner, 2004]. During the last decade,

many studies have seen the merits of analyzing the spatial variations in the seabed

backscatter strength as a tool to aid in interpretation of shelf sedimentary processes

[Collier and Brown, 2005; Dunlop et al., 2010; Chiocci et al., 2011; Rzhanov et al., 2011].

However, in spite of its potential to map surficial sediment, very little attention has

been given to the use of automated MBES backscatter classification algorithms to study

seafloor features in glacial environments. This may partly be attributed to the following; a)

very few ships have the financial and operational capability to collect high-quality

backscatter information in the polar regions, b) harsh survey conditions in polar regions

make high-quality data collection difficult, and c) problems in the automated processing,

interpretation, and the sediment characterization potential of MBES backscatter still

remain in spite of the technological advancements [McGonigle et al., 2009; Hughes-

Clarke, 2012].

The primary aim of my thesis is to assess the potential of using automated sediment

characterization approaches in processing and analyzing MBES bathymetry and

backscatter data to study glaciated submarine landscapes. This is done by examining

MBES datasets coming from different acoustic system to determine which among the

various approaches is the most ideal for extracting surficial sediment characteristics. The

results from these studies will be augmented and ground-truthed with evidence collected

from cores or other acoustical systems. My study covers diverse glaciated submarine

landscapes and uses the bathymetry and backscatter information collected from these

landscapes to gain a better understanding of their formation processes (see paper in Part

VI).

3

The next part of this thesis will present a short background of glaciogenic bedforms

in the Polar Oceans and show how they are used to infer paleo-conditions in the glacial

environment. This is followed by a review of the various geophysical mapping systems

used in understanding the glaciogenic bedforms, focusing on the processing of MBES

backscatter for seafloor characterization and a discussion of its potential and challenges for

its adoption into glacial geomorphological research. Part three and four will be a summary

of my paper and a discussion on future work. The final chapter, VI, is a copy of the paper,

”Acoustic evidence of a submarine slide in the deepest part of the Arctic, the Molloy Hole”

by Freire, Gyllencreutz, Jafri, and Jakobsson (2014), published in Geo-Marine Letters.

4

Part II. Background

Submarine glaciogenic seafloor features in the Polar Oceans

Glacial features in the terrestrial environments have been studied since the last

century. Studies of submarine glaciogenic features on the other hand, only escalated in the

last 20 years with the advancement of marine sensor technology in the field of marine

geophysics, and especially, improvements in the positioning accuracy and the introduction

of GPS navigation [Mayer, 2006; Jakobsson et al., 2008; Bingham et al., 2010]. Many

studies have taken advantage of the sonar technology and used it to infer past ice flow

patterns and conditions based on the seafloor features shape and structure information.

Specific questions related to paleo-ice flow patterns that were addressed in these studies

include; the location of the paleo ice-streams [Anderson and Fretwell, 2008; Maclean et

al., 2010], past ice berg size and thickness [Kuijpers et al., 2007; Jakobsson et al., 2010],

extent of glaciation [Evans et al., 2006; Graham et al., 2009; Dunlop et al., 2010],

direction and speed of the ice flow [Dowdeswell et al., 2008; Graham et al., 2010;

Winsborrow et al., 2010], and ice-sheet retreat dynamics [Dowdeswell et al., 2008;

Jakobsson et al., 2012a].

Using relict glaciogenic seafloor features to understand past ice-flow patterns

requires the correct identification and characterization of these landforms. At present,

there are no standard criteria employed in categorization of submarine geomorphological

features. This is probably because of the numerous processes involved, the varying

geologic strata over which the ice flows, and the changing environmental conditions. This

leads to a myriad of features, each having a number of interpretations for their formation

process. The variety of glacial landforms are exhaustively reviewed by Benn and Evans

(2010) wherein they also mention that many of the terms are used interchangeably by

different studies which poses a challenge for new researchers of this field.

Glacial landforms are broadly categorized as either erosional or depositional features.

Studying they shape, size and spatial distribution of depositional features such as

grounding zone wedges, eskers and end moraines help us understand how fast or how

stable was the ice-sheet during its retreat [Wellner et al., 2001; Jakobsson et al., 2012a;

Todd and Shaw, 2012]. Erosional features such as mega-scale glacial lineations (MSGL),

P-forms, and whalebacks on the other hand can provide bed-thermal conditions and how

fast advancing glaciers flowed [Glasser and Bennett, 2004; Ottesen et al., 2007; Graham et

5

al., 2010; Batchelor et al., 2011; Howe et al., 2012], and drift patterns of icebergs can be

reconstructed from erosional scours found on the seafloor [Jakobsson et al., 2008;

Dowdeswell et al., 2010a; Maclean et al., 2010; Todd and Shaw, 2012]. The fact that

glacial morphologic features are preserved at all gives important clues about the rate and

manner of deglaciation [Kuijpers et al., 2007; Graham et al., 2009].

More detailed characterization of seafloor features involves classifying the landforms

according to their shape by differentiating those landforms that are glacially streamlined,

as opposed to transverse ridges or as deep elongated curvilinear-shaped channels. In such

loose categorizations, glacially streamlined landforms such as striae, flutes, drumlins, and

glacial lineations, are used to indicate the direction of the paleo-ice sheet flow [Clark,

1993; Greenwood and Clark, 2009; Stokes et al., 2011] and, depending on their size, the

velocity of ice stream [Dowdeswell et al., 2008; Graham et al., 2010]. Transverse features

such as moraines have a more complex shape and formational process, and can tell us of

the extent of the glaciation or style of ice-retreat [Nygård et al., 2004; Bradwell et al.,

2008; Graham et al., 2009; Maclachlan et al., 2010; Winsborrow et al., 2010]. Channel

features are interpreted to have formed by subglacial and proglacial meltwater flow [Howe

et al., 2012] which delineates it path. These examples are not an exhaustive review off all

glacigenic features, but provide a glimpse of the variety in studied seafloor features.

Aside from inferring past ice conditions, another important topic of glacial

geomorphological studies is the investigation of mass wasting events that occur in the

high-latitude continental margins. Eroded materials from the melting paleo ice-streams

during deglaciation cycles produce massive quantities of sediments that are transported to

the adjacent continental margins [Bugge et al., 1987; Elverhøi et al., 2002]. Eventually, too

much load on the slope can lead to mass-wasting or transfer of these materials from the

margins to the ocean floor, triggered perhaps by a seismic event [Laberg and Vorren,

2000; Lindberg et al., 2004; Freire et al., 2014]. In studies of mass wasting features,

morphological elements commonly identified with slide scars are usually mapped and used

in the interpretation of the slide event [Laberg et al., 2013] and are used to determine not

only its causes but also its consequences. Studies that characterize geomorphological

features of mass wasting events in glacial environment have been able to reveal past glacial

conditions [Vanneste et al., 2006; Hogan et al., 2013], reconstruct and predict future

tsunamigenic conditions [Harbitz et al., 2006; Berndt et al., 2009], determine climate-

changing effects from the release of huge quantities of methane gas [Maslin et al., 2004;

6

Beget and Addison, 2007; Paull et al., 2007] and identify its potential destructive effect to

offshore construction and resource exploitation activities.

Acoustic-based characterization of the seafloor

Acoustic systems have been extensively used in the field of marine science and

technology because of their low absorption in water [Penrose et al., 2005]. There are other

remote sensing tools that are used to study the seafloor that utilize different segments of

the electromagnetic spectrum and even the variations in the gravitational and magnetic

fields. These tools are installed in a variety of platforms (satellite, airborne, ships, remotely

operated or autonomous underwater vehicle or handheld devices) and operate on different

configurations. While each tool has its own capabilities, ocean mapping efforts where

acoustic systems are used provide a cost-effective means to collect high-resolution seafloor

data (i.e., submeter to tens of meter resolution), and the ability to map the deep oceanic

waters over a relatively-wide spatial extent.

The development of advanced acoustic-based technology can be traced to its crucial

role in submarine warfare during the last world wars. These war-borne innovations were

subsequently enhanced and successfully found its use in various commercial applications

such as those in marine resource extraction (oil industry) and in ship navigation [Lurton,

2010]. The importance of this technology for practical and commercial use has led to more

research and development in the last two decades, which have profited the marine

geomorphological research. In fact, acoustic technology has become one of the most

essential tools of marine geomorphologists because direct observation of submarine

landforms is, if not impossible, too costly to undertake [Dunlop et al., 2011; Micallef,

2011].

One of the earliest uses of acoustic systems in marine glacial geomorphology

research was documented by Belderson et al. (1973) who, using a very coarse resolution

side-scan imaging, identified and characterized iceberg plough marks from the Canadian

Arctic which was before only observed on deglaciated terrains. The same technology was

also used by King (1976) to characterize seafloor features that were caused by iceberg

scouring. Although acoustic-based technology already showed its promise during this

period, it was only in the last two decades that it has been routinely used in marine

geomorphological research. This was made possible because of the rapid advancement in

both computer hardware and software for digital data processing and the revolutionary

7

development of global navigation technology from GPS [Lurton, 2010]. Nowadays, marine

geomorphologists have at hand various acoustic-based tools installed on a variety of

platforms to collect data not only from surficial landforms, but also from the sub-surface

strata and even features under thick ice shelves in the Polar oceans [Graham et al., 2013].

Summarized below are the main features of acoustic systems that have contributed in the

field of submarine glacial geomorphology.

Single beam echo sounder (SBES)

Singe beam echo sounders, developed during the Second World War was one of the

first commercial applications of acoustic systems. It was installed in the ship’s hull and

was used to detect shallow shoals and aid in the ships navigation, a function that it

performs even in today’s modern vessels. SBES transmits and measures the time it takes

for a beam of sound to travel from the transducer to the seafloor and back to the receiver,

and a corresponding depth value for every “ping” of sound is calculated. The early

versions of SBES had broad beams that ensonify a very large area of the seafloor (0.5-1

times the water depth) producing a depth value from anywhere within the ensonified area

resulting in great inaccuracies [Mayer, 2006]. And, although newer versions of SBES use

beam forming to produce narrower beams for more accurate soundings, its sparse data

collection capability makes it less useful for geomorphological mapping studies because a

very dense grid of survey lines are needed to get the resolution needed to map glacial

landforms. There are efforts that make use of the SBES data collected by the numerous

commercial ships, e.g., the OLEX data sharing platform [OLEX, 2014], to produce detailed

bathymetric maps. These dense SBES data provide sufficient information to study

submarine glacial morphology in certain areas [Bradwell et al., 2008; Winsborrow et al.,

2010; Dunlop et al., 2011; Jakobsson et al., 2012b], but have limited coverage restricted to

regions commonly trafficked by commercial vessels.

Side-scan sonar

The side-scan sonar is a category of acoustic system that creates an image of

relatively large areas of the seafloor by measuring the amount of returned acoustic energy.

It is one of the earliest acoustic systems used in glacial environments together with the

single beam echosounder [Belderson et al., 1973]. The sonar device is towed from, or

attached to a vessel where it emits fan-shaped sound pulses down towards the seafloor

across a wide angle perpendicular to the path of a sensor. It records the intensity of the

8

reflected energy (backscatter) from the seafloor as a series of cross-track slices and

produces ultra-high resolution images because of its ability to collect data very close to its

target if deep-towed or installed in autonomous underwater vehicle (AUV) or remotely

operated vehicle (ROV) systems. However, deployment is troublesome particularly in

highly rugged underwater terrain which may result in the system getting damaged or even

lost underwater. Another disadvantage of side scan sonars is the difficulty to achieve the

high positional accuracy that the multi and single beam echosounders provides, as GPS

signals cannot penetrate the water column.

Side-scan data have been used to identify deep ice-berg plough marks [Kuijpers et

al., 2007], obtain information about seafloor slide morphology [Laberg et al., 2013] or

sediment characteristics [Hogan et al., 2013]. Newer versions of side-scan sonars have

shown greater potential for use in geomorphological studies with their enhanced capability

to measure high-resolution bathymetry and backscatter at the same time and the capability

to reveal additional information on surface sediment properties such as density,

water/sediment ratios, texture and porosity [Micallef, 2011].

Seismic Reflection

Seismic reflection is a geophysical exploration method to obtain information about

the sub-surface stratigraphy. The general principle involves sending artificially produced

low frequency acoustic energy that travel down the water column and into the seafloor,

where it will interact with the different material properties and structures within the

subsurface. The energy that is reflected back is measured by a hydrophone array in a towed

streamer to record the acoustic impedance [Micallef, 2011]. This information has been

used in depicting the stratigraphic sequences in the seafloor since the early 1960’s to find

oil reserves. In modern systems, extensive seismic reflection surveys employ tens of

parallel lines over three kilometers long of towed seismic streamers to generate data that

reveal geological structures more than 1 km deep and at high spatial resolution producing

3D images of the sub-bottom. These surveys are very expensive to undertake and are

generally only utilized by oil exploration companies. However, exploration seismics has

also found its use in glacial geomorphological studies to reveal the evolution of paleo

seafloor features and the processes that have shaped them [Andreassen et al., 2008, 2013;

Rebesco et al., 2011]. Geophysical surveys that deploy seismic streamers are not much

used in the polar environments because of the difficulty in maneuvering the iceberg littered

seas. On the other hand, the continued warming of the oceans could provide a window of

9

opportunity to use such systems for gathering important information of sub-bottom

characteristics from these glacial environments.

Another less expensive alternative for investigating sub-bottom geology is the use of

high resolution systems such as pingers, chirp sonars or parametric echo sounders,

collectively called sub-bottom profilers. They use the same principle as seismic methods

but utilize frequencies between the seismic and bathymetric sonar ranges, penetrating only

up to100 meters deep into the substratum producing shallow vertical 2D images of the

relatively soft sediment cover at a high resolution. Unlike exploration seismics, sub-bottom

profilers do not require long distances between the sound source and the receivers allowing

them to be easily deployed or hull-mounted resulting in a simple and economical means to

study seafloor sub-bottom characteristics. This allows sub-bottom profilers to be utilized

extensively in many geological surveys where they produce information that is used in

interpreting paleo-environmental development, core correlation or stratigraphic

descriptions. They have also been used in glacial geomorphological studies to show and

analyze glaciosedimentary processes [Maclachlan et al., 2010; García et al., 2012] or to

corroborate the presence of glacial landforms together with MBES bathymetry and

backscatter [Jakobsson et al., 2008; Dowdeswell et al., 2010b; Rebesco et al., 2011].

Multibeam echo-sounder systems (MBES)

MBES systems, unlike its SBES counterpart, transmit and receive sound pulses

consisting of more than 200 fan-shaped beams (depending on the system) at varying angles

to cover a wide swath of the seafloor. They cover an across track distance of 5-7 times that

of the water depth resulting in simultaneous depth measurements across this wide swath.

MBES systems use precise vessel attitude and positioning information coupled with sound

speed measurements of the water column during data processing to compensate for the

complicated beam geometry [Foster et al., 2013]. Improvements in both software (i.e.,

better data processing algorithms) and hardware (i.e., faster computer processing speeds)

together with advancement in acoustic and positioning technology have resulted in faster

data processing and higher quality outputs during the last decades [Mayer, 2006].

However, even with the most advanced MBES systems, the sea conditions during the

survey and the correction of known errors during the data acquisition and processing are

crucial for the successful implementation of the MBES technology.

10

MBES is presently the most preferred means of collecting seafloor morphologic data

because of its ability to collect dense and high accuracy soundings of the seafloor, thereby

producing detailed physiographic maps [Gardner et al., 2003; Dunlop et al., 2011]. In

addition to measuring acoustic pulse travel times, MBES systems are also capable of

recording the intensity of the transmitted acoustic pulse after it has reflected from the

seafloor. This is otherwise known as the Backscattering Strength (BS), which is a record

of the relative amount of energy sent back by the target, the seafloor, to the receiver.

Backscatter as a tool in studying the ocean seafloor

The use of acoustic backscatter to characterize the seafloor was first developed in the

1980’s. However, it was only in the last decade that its potential for large-scale seafloor

characterization was being realized with the development of higher resolution MBES

systems that allow the collection of bathymetry and co-registered backscatter data [Brown

and Blondel, 2009]. The additional surficial geomorphic information that can be collected

at a fraction of the cost makes it a very attractive tool for geomorphological studies. Many

marine glacial studies have benefited from this additional information by using

conventional subjective interpretation in delineating the variations in the backscatter

intensity to infer general sediment characteristics [Noormets et al., 2009; Dunlop et al.,

2010; García et al., 2012; Sacchetti et al., 2012; Todd and Shaw, 2012; Hogan et al.,

2013]. This delineation process takes a lot of time and is subject to human-errors,

especially with the voluminous data produced in a typical MBES survey. This led to the

development of MBES backscatter automated processing and classification methods to

identify surficial sediments characteristics [Chakraborty et al., 2003; Fonseca and Mayer,

2007; Preston, 2009; Simons and Snellen, 2009; Lamarche et al., 2011; Parnum and

Gavrilov, 2011].

However, using acoustic BS for seafloor characterization is not straightforward

because the propagation of the acoustic wave in the water meets a series of obstacles either

in the water column itself, or at the highly complex water-sediment interface, or within the

sediment [Masetti et al., 2011]. The variations in BS measurements depends on the

properties of the sediment-water interface such as surface roughness, impedance contrast,

volume inhomogeneity, and varying grazing angles [Hughes Clarke, 1994; Hughes Clarke

et al., 1996; Siemes et al., 2010; Hughes-Clarke, 2012], and understanding these factors is

critical in predicting surface sediment features from MBES data.

11

For example, the presence or absence of buried shell debris or glacial dropstones can

affect the recorded BS by affecting the sediment’s volume inhomogeneity and cause

changes to the BS signal, but it is difficult to ascertain which factor contributes most to the

variation. In addition, the inherent complexities in the installation and operation of the

MBES and the varying environmental conditions during the data collection can affect the

quality of the recorded data. Below, the factors that need to be considered in analyzing the

backscatter information from MBES for seafloor characterization purposes are discussed.

Acoustic backscatter is best understood by the sonar equation (see Eq. 1) which

states that the sound recorded (echo level = EL) by the transducer system results from the

transmitted acoustic wave intensity (signal level = SL) minus the two way signal

transmission loss (TL) during the water column propagation, plus the target strength (TS)

which is the acoustic intensity reflected from the seafloor.

𝐸𝐿 = 𝑆𝐿 − 2𝑇𝐿 + 𝑇𝑆 (Eq. 1)

TL is attributed to the absorbance of some of the acoustic energy in the water

resulting in the attenuation of the signal and loss through spreading. Accurate sound speed

profile of the water column is critical in the estimation of TL since a change in seawater

properties (temperature, salinity and pressure) significantly affects the sound propagation.

The TS is the sediments backscatter response and can be decomposed into two parts: the

area ensonified and the corresponding backscattering strength for the unit of surface (from

Hughes Clarke 1994; Tang et al. 2005). The ensonified area of the seabed is dependent on

the geometry of the incident beam angle and the bathymetric surface of the ensonified area

which varies depending on the slope of the area (Figure 1). The backscattering strength

represents the sediment’s ability to reflect the sound energy and varies depending on the

sediment’s roughness, impedance and volume heterogeneity. In addition, the backscatter

strength will also vary with grazing angle (no. 1 in Figure 2) with more energy reflected at

nadir beams. This property is usually corrected for a cleaner and more consistent

backscatter mosaics. However, different sediment types have distinctly different grazing

angle dependence (no. 2 in Figure 2). The angular dependence property of the sediment is

an important characteristic in discriminating different seafloor types [Hughes Clarke,

1994] which is why some sediment characterization algorithms preserve this angular

signature and use it for its analysis, rather than remove it to produce visually appealing

mosaics. The interactions between the abovementioned factors is crucial in processing and

analysis of the backscatter data for seafloor characterization, and is the subject of many

12

studies of acoustic seafloor mapping [Tang et al., 2005; Fonseca and Mayer, 2007;

Gavrilov and Parnum, 2010; Siemes et al., 2010].

Processing of backscatter data for seafloor characterization

Backscatter processing first involves the correction of the raw backscatter value for

radiometric and geometric distortions before it can be used for further seafloor

characterization analysis. These distortions are caused by differences in operational

settings of particular systems (e.g., Kongsberg vs Reson MBES) and sensor platform

considerations [Hughes-Clarke, 2012]. Geometric corrections allow the backscatter data to

be registered to the proper bathymetry for the correct estimate of the ensonified area while

radiometric corrections allow for the corrections of system operational settings [Beaudoin

et al., 2002].

Figure 1. Changes in the geometric configuration of the seafloor ensonified area from the inner (A1) to the outer (A2) beams of an MBES system. This variation, together with the slope of the actual bathymetric surface, is taken into consideration by the MBES system to calculate backscatter strength (modified from Tang et al. 2005).

Seafloor characterization analysis of the corrected backscatter is then undertaken

using two main approaches: 1) using a textural analysis or image-based segmentation of

the average backscatter levels [Preston, 2009; McGonigle et al., 2010a; Brown et al., 2011]

or 2) processing using an analysis of the angular dependence of backscattering strength

[Fonseca et al., 2009; Gavrilov and Parnum, 2010; Lamarche et al., 2011]. The first

approach classifies the backscatter mosaic map by looking at the spatial variability of

backscatter strengths. This is done by extracting various statistical measures or “features”

13

(e.g., mean, median, standard deviation and others) from the backscatter strength variations

in the mosaic. The differences between the features are then reduced using single images

by principal component analysis to find the component that causes the most variation.

These are fed into an image processing clustering algorithm to define the different

sediment types.

Figure 2. The effect of grazing angle on multibeam geometry and typical angular response curves (modified from Hughes-Clarke 2012). Backscatter intensity of beams that are closer to nadir (i.e., A in no. 1 and 2) record more of the reflected energy as compare to the outer beams (i.e., B and C in no. 1 and 2). Different sediment types have different angular responses as shown in no. 3, which is used in ARA analysis for sediment characterization.

The signal based classification on the other hand looks at the changes in the

backscatter values as a function of the angle of incidence (angular dependence) to classify

different types of the seafloor (see Figure 3). This can be done by either examining the

curves of the backscatter angular dependence and comparing this with a theoretical model

or empirically by analyzing the relative difference between angular responses to

distinguish the seafloor types [Parnum and Gavrilov, 2011]. One example, the Angular

range analysis (ARA) [Fonseca and Mayer, 2007] is perhaps the most widely used because

it is incorporated in many commercial hydrographic processing software packages. The

14

ARA analysis is bundled in the Geocoder software package, which undertakes both

geometric and radiometric correction and then determines the seafloor sediment type from

the change in backscatter values as a function of its angle of incidence (angular

dependence) using the ARA algorithm.

The ARA algorithm analyzes the returned BS strength across a swath by stacking or

grouping the BS values of one swath in certain ranges. It stacks the BS intensities

collected at 90o to 65o angles into the far range, from 65o to 35o angles as the outer range

and from 35o to 5o angles as the inner range. It then computes the statistical parameters of

slope, intercept and mean from the backscatter values, for each of the three stacks, in order

to create discrete angular regimes. The inversion process is then undertaken by fitting the

computed parameters to a curve of published statistical measures of sediment properties to

produce estimates of the acoustic impedance, roughness and consequently the mean grain

size of the ensonified area of the seafloor [Fonseca et al., 2009].

Part III. Discussion and Conclusions

The use of ARA as a tool to characterize seafloor features from MBES backscatter is

very appealing since it automates the identification and quantitative classification of the

seafloor sediment types. Also, the algorithm is easily accessible as it is available in many

commercial bathymetric processing software packages. At present, most of my research on

the backscatter processing and analysis is focused on the MBES data (EM122), a low-

frequency deep-water echosounder, from the Swedish ice-breaker Oden. The ARA

algorithm incorporated in Geocoder was the main processing tool used because of the

abovementioned reasons. The results from my research have revealed issues with the

outputs of the seafloor characterization some of which are highlighted in the discussion

section of the published manuscript (Part VI of this thesis). This can be attributed to

limitations of both the ARA algorithm and the hardware as well as the operation of the

acoustic system, and are discussed as follows:

a. Limitations with the use of ARA for great depths – The depth range in the

Molloy Hole (>5000 m) is apparently beyond the optimal coverage depth which

results in progressively less angular coverage of the swath which can affect the

calculation in the ARA analysis for sediment characteristics [Freire et al., 2014].

15

b. Problems with the calibration and limitations of the MBES system – artifacts

(i.e., striping) were observed with the Oden backscatter data even with the proper

geometric and radiometric corrections and filtering of the undesired sectors

[Marcussen and Scientific Party, 2012]. This was evident particularly for the

starboard side of the system (Figure 3). Also, the great depth of my survey area

would result in a large beam footprint size that will leading to a very coarse

resolution surficial sediment maps which might not detect slight changes in the

sediment property that I was studying.

c. Lack of ground-truth data - the data that I processed was collected with the main

aim of creating bathymetric maps, and not for seafloor characterization. This

means that critical information that could increase the accuracy and usefulness of

the analysis was lacking, such as ground-truth data and the collection of multiple

survey overlaps that provide wider range of angular coverage [McGonigle et al.,

2010b].

d. The complexity of the seafloor environment – unknown features in the seafloor

such as great changes in slope which can cause a steep grazing angle response or

the effect of the presence of gas bubbles in the thick sediment basins of my study

area [Fonseca et al., 2002; Collier and Brown, 2005] which affects the sediments

heterogeneity and might confuse the interpretation. Such complexities are very

difficult isolate especially in data from challenging survey conditions. Also,

surveys in the ice-littered environment tend to produce echo multiples due to the

blocks of ice flowing under the acoustic path of the transducer that degrades the

backscatter information.

Any or a combination of the abovementioned factors could have contributed to

degrade the results, and pinpointing the exact problem is difficult. Improvements

particularly in the research design, such as better calibration of the system and collection of

ground truth data for validation and as training data for supervised classification [Parnum,

2007], are clearly needed to create better characterization results. These problems and their

solution will be taken into consideration in my present and upcoming research where I

intend to explore other approaches to process backscatter data from glaciated

environments.

16

MBES backscatter and bathymetry characteristics have been used to map the spatial

distribution of physical and geological properties of the glacial sea floor. However, the

automating the direct inversion of acoustic backscatter to actual physical property of the

sediment feature remains constrained. The effects of the complex environmental

conditions, unreliable survey operational conditions and the quality of the MBES system

all contribute to the problem. Studies are still needed to successfully utilize this technology

in the glacial submarine environment.

17

18

Figure 3 (page before). Example of categorization of submarine glaciogenic features. A) The processed bathymetry collected by ice-breaker Oden (in grayscale palette) of Pine Island Bay, Antarctica and locations of gravity cores and their surface sediment composition [Kirshner et al., 2012]. B) The result of classifying the different landforms by Jakobsson et al. (2011) based on multibeam seafloor morphology (MSGL – megascale glacial lineations; GZW – grounding zone wedges). C) The inversion output of sediment characterization analysis using angular response algorithm (Fonseca and Mayer 2007) on the same multibeam data set to classify the seafloor type according to grain size (red = coarser, green = intermediate, blue = finer grain size). Note the different response in the starboard and port sides of the each swath along the survey lines. In this case example, it was not possible to distinguish a good correlation with ground truthing data. D) The processed backscatter signal from the same data set as A with enlarged part (inset) clearly showing striping artifacts, probably caused by a system calibration error. This in turn affects the characterization algorithms (C).

19

Part IV. Summary of the published manuscript

The paper examines submarine landforms partly created by mass-wasting event in

the deepest part of the Arctic Ocean, the Molloy Hole. The study involved the use of

several geophysical mapping systems from different scientific expeditions to identify and

characterize the submarine features. The production of a high resolution bathymetric map

together with other geophysical data from sub-bottom profiles and backscatter analysis

allowed a geomorphologic mapping of the Molloy Hole environment. There, we have

identified prominent features of a deep submarine mass-wasting event in the western

Svalbard continental margin, which we termed the Molloy Slide. The Molloy slide is

different compared to other submarine slides because of its unique, steep topography

resulting in a short slide run-off distance but at a great vertical distance. It is the second

identified submarine slide in very high latitudes of the northeastern Atlantic and the first

for the western Svalbard continental shelf where thick glacial sediment deposits with

considerable gas hydrates deposits are found. The paper discusses possible

processes/mechanisms that might have caused the submarine slide and potential

consequences of any such occurrence in the future.

The paper shows the practical application of using acoustic characterization to study

geomorphological features in the polar oceans. The emphasis of the paper was on the

compilation and morphologic analysis of the high-resolution bathymetry and the sub

bottom profile data. However, it was also, to our knowledge, the first to use automated

sediment characterization algorithms to analyze submarine glacial features in the polar

seas. Although the sediment characterization process produced somewhat useful results,

only crude general patterns could be established. A variety factors were identified to

explain the result and a more thorough discussion is presented in the preceding chapter.

The sediment characterization outputs would have been greatly improved with additional

data sources (i.e., sediment cores and multiple MBES survey overlaps) which have to be

resolved to create useful sediment characterization outputs in such deep polar

environments.

20

Part V. Ongoing projects

My present and future work focuses on processing MBES and backscatter data for

seafloor characterization studies from different MBES systems. I am currently working on

high resolution datasets from both ship-mounted shallow-water MBES and an

interferometric sonar system mounted on an Autonomous Underwater Vehicle (AUV). The

survey was conducted in the offshore area off the western coast of Greenland. The

expedition’s aim was to find Vega, the sunken research vessel of A.E. Nordensköld, and at

the same time, map the sea floor of the area at high resolution. Vega wasn’t found, but very

good quality data was obtained from the MBES and interferometric sonar-AUV survey.

The shallow water MBES EM2040 is a high-frequency system, which was used to

map seafloor areas in 5-m resolution. The map produced by the MBES is used by the AUV

for navigation to map smaller target areas at a higher in resolution. Both systems are

capable of measuring both bathymetry and a co-registered backscatter but differ in terms of

their spatial resolution. Also, the two systems also differ in how they collect and process

the sonar signal for bathymetric mapping. An interferometric sonar measures the two way

travel time from the range and also the angle of the returned acoustic signal, which is

determined by the known spacing between the elements within the transducer, or the wave

phase difference. This offset and the two-way travel time is used to calculate the position

and depth to the seafloor. This type of system can collect higher resolution data than the

MBES, because the AUV is able to survey very close to the seafloor. However, there are

also problems associated with its use, particularly the accuracy of the bathymetric data.

The AUV does not have the navigational accuracy of the ship-mounted MBES, which

constantly receive highly accurate GPS positions and ship motion information from a

motions sensor that is used to correct the swath signals for positional errors. The results of

multibeam and backscattering information from the high-resolution interferometric sonar

shows promise for textural classification, as shown in Figures 4-8 in the succeeding pages.

The data sets are now in the initial stages of interpretation and further processing will be

performed to improve them before publication.

21

Part VI. Acknowledgments

First of all, I would like to thank my main supervisor, Richard Gyllencreutz,

for his encouragement, for his guidance in the preparation of this manuscript and for the

informal scientific discussions which helped me understand more of the cryogenic

environment. He has been so patient and understanding of my shortcomings, and many of

what I have written here will not be possible without his help. I would also like to thank

my co-supervisor, Martin Jakobsson, for sharing his ideas and knowledge in my research.

His passion for his work served as an inspiration in my career.

A huge thank you to all my friends, and fellow-PhD’s at Stockholm University for

the companionships and being there especially during the tough times. Special mention to

my roommates for putting up with me during all this time. A special thanks is given to

the administrative and academic staff of the IGV for the support. Lastly, I would like to

thank my family – elena, tashaa, mom and others – for their love.

22

Some results from present research:

Figure 4. The location and extent of geophysical datasets processed for seafloor characterization of the glacial features off the western coast of Greenland. The datasets comprised of high resolution MBES data and an even higher resolution interferometric sonar.

23

Profile A

Profile B

Profile C

depressions

>75 degree

Figure 5. Geomorphological characterization of the high-resolution MBES data from off the westerncoast of Greenland. A) Bathymetry and, B) Slope map (1st derivative of bathymetry) revealing a complex landform assemblage with steep ledges separating relatively flat areas, and several spurious semi-circular pits. C) Profiles in the bathymetric map shows cliffs with > 75 degrees slope and > 100 m vertical offset over the sharpest ledges, and rounded gently sloping walls of the largest semi-circular pit.

24

Sediment characterization analysis on AUV-Gaviadata

3-m contourVariations in backscatter value

Figure 6. Example of a high resolution side-scan backscatter measurements from the interferometric sonar with resolution of 0.5 m (A) which was subject to automated sediment characterization analysis using the ARA algorithms (B) where correlations with the backscatter intensity values is observed

A

B

25

Part VII. References

Anderson, J. B., and L. O. Fretwell (2008), Geomorphology of the onset area of a paleo-ice stream , Marguerite Bay , Antarctic Peninsula, Earth Surf. Process. Landforms, 512, 503–512, doi:10.1002/esp.

Andreassen, K., J. S. Laberg, and T. O. Vorren (2008), Seafloor geomorphology of the SW Barents Sea and its glaci-dynamic implications, Geomorphology, 97(1-2), 157–177, doi:10.1016/j.geomorph.2007.02.050. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0169555X07002917 (Accessed 17 August 2011)

Andreassen, K., M. C. M. Winsborrow, L. R. Bjarnadóttir, and D. C. Rüther (2013), Ice stream retreat dynamics inferred from an assemblage of landforms in the northern Barents Sea, Quat. Sci. Rev., 1–12, doi:10.1016/j.quascirev.2013.09.015. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S027737911300351X (Accessed 13 April 2014)

Batchelor, C. L., J. A. Dowdeswell, and K. A. Hogan (2011), Late Quaternary ice flow and sediment delivery through Hinlopen Trough , Northern Svalbard margin : Submarine landforms and depositional fan, Mar. Geol., 284(1-4), 13–27, doi:10.1016/j.margeo.2011.03.005. [online] Available from: http://dx.doi.org/10.1016/j.margeo.2011.03.005

Beaudoin, J. D., J. E. Hughes Clarke, E. J. Van Den Ameele, and J. V. Gardner (2002), Geometric and Radiometric Correction of Multibeam Backscatter Derived from Reson 8101 Systems, in Canadian Hydrographic Conference (CHC).

Beget, J. E., and J. A. Addison (2007), Methane gas release from the Storegga submarine landslide linked to early Holocene climate change: a speculative hypothesis, The Holocene, 17(3), 291–295, doi:10.1177/0959683607076435. [online] Available from: http://hol.sagepub.com/cgi/doi/10.1177/0959683607076435 (Accessed 14 March 2013)

Belderson, R. H., N. H. Kenyon, and J. B. Wilson (1973), Iceberg plough marks in the northeast Atlantic, Palaeogeogr. Palaeoclimatol. Palaeoeeology, 13, 215–224.

Benn, D. I., and D. J. A. Evans (2010), Glaciers and Glaciation, Second Edi., Routledge, New York.

Berndt, C., S. Brune, E. Nisbet, J. Zschau, and S. V. Sobolev (2009), Tsunami modeling of a submarine landslide in the Fram Strait, Geochemistry Geophys. Geosystems, 10(4), Q04009, doi:10.1029/2008GC002292. [online] Available from: http://www.agu.org/pubs/crossref/2009/2008GC002292.shtml (Accessed 16 November 2012)

Bingham, R. G., E. C. King, A. M. Smith, and H. D. Pritchard (2010), Glacial geomorphology : Towards a convergence of glaciology and geomorphology, Prog. Phys. Geogr., 34(3), 327–355, doi:10.1177/0309133309360631. [online] Available

26

from: http://ppg.sagepub.com/cgi/doi/10.1177/0309133309360631 (Accessed 3 August 2011)

Bradwell, T. et al. (2008), The northern sector of the last British Ice Sheet: Maximum extent and demise, Earth-Science Rev., 88(3-4), 207–226, doi:10.1016/j.earscirev.2008.01.008. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0012825208000160 (Accessed 19 March 2014)

Brown, C. J., and P. Blondel (2009), Developments in the application of multibeam sonar backscatter for seafloor habitat mapping, Appl. Acoust., 70(10), 1242–1247, doi:10.1016/j.apacoust.2008.08.004. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0003682X0800176X (Accessed 20 July 2011)

Brown, C. J., B. J. Todd, V. E. Kostylev, R. A. Pickrill, and G. Bank (2011), Image-based classification of multibeam sonar backscatter data for objective surficial sediment mapping of Georges Bank, Canada, Cont. Shelf Res., 31(2), S110–S119, doi:10.1016/j.csr.2010.02.009. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0278434310000701 (Accessed 4 August 2011)

Bugge, T., S. Befring, R. H. Belderson, T. Eidvin, E. Jansen, N. H. Kenyon, H. Holtedahi, and P. H. Sejrup (1987), A giant three stage submarine slide off Norway, Geo-Marine Lett., 7, 191–198.

Chakraborty, B., V. Kodagali, and J. Baracho (2003), Sea-Floor Classification Using Multibeam Echo-Sounding Angular Backscatter Data: A Real-Time Approach Employing Hybrid Neural Network Architecture, IEEE J. Ocean. Eng., 28(1), 121–128.

Chiocci, F. L., A. Cattaneo, and R. Urgeles (2011), Seafloor mapping for geohazard assessment: state of the art, Mar. Geophys. Res., 32(1-2), 1–11, doi:10.1007/s11001-011-9139-8. [online] Available from: http://link.springer.com/10.1007/s11001-011-9139-8 (Accessed 29 January 2014)

Clark, C. D. (1993), Mega-scale glacial lineations and cross-cutting ice-flow landforms, Earth Surf. Process. Landforms, 18(1), 1–29, doi:10.1002/esp.3290180102. [online] Available from: http://doi.wiley.com/10.1002/esp.3290180102

Collier, J. S., and C. J. Brown (2005), Correlation of sidescan backscatter with grain size distribution of surficial seabed sediments, Mar. Geol., 214(4), 431–449, doi:10.1016/j.margeo.2004.11.011. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0025322704003196 (Accessed 4 August 2011)

Dartnell, P., and J. V Gardner (2004), Predicting Seafloor Facies from Multibeam Bathymetry and Backscatter Data, Photogramm. Eng. Remote Sens., 70(9), 1081–1091.

27

Dowdeswell, J. A., J. Evans, C. O Cofaigh, and J. B. Anderson (2006), Morphology and sedimentary processes on the continental slope off Pine Island Bay, Amundsen Sea, West Antarctica, Geol. Soc. Am. Bull., 118(5-6), 606–619, doi:10.1130/B25791.1. [online] Available from: http://gsabulletin.gsapubs.org/cgi/doi/10.1130/B25791.1 (Accessed 2 April 2012)

Dowdeswell, J. A., D. Ottesen, J. Evans, C. Ó. Cofaigh, and J. B. Anderson (2008), Submarine glacial landforms and rates of ice-stream collapse, Geology, 36(10), 819–822, doi:10.1130/G24808A.1. [online] Available from: http://geology.gsapubs.org/cgi/doi/10.1130/G24808A.1 (Accessed 21 September 2011)

Dowdeswell, J. A. et al. (2010a), High-resolution geophysical observations of the Yermak Plateau and northern Svalbard margin: implications for ice-sheet grounding and deep-keeled icebergs, Quat. Sci. Rev., 29(25-26), 3518–3531, doi:10.1016/j.quascirev.2010.06.002. [online] Available from: http://dx.doi.org/10.1016/j.quascirev.2010.06.002 (Accessed 21 September 2011)

Dowdeswell, J. A., J. Evans, and C. Ó Cofaigh (2010b), Submarine landforms and shallow acoustic stratigraphy of a 400 km-long fjord-shelf-slope transect, Kangerlussuaq margin, East Greenland, Quat. Sci. Rev., 29(25-26), 3359–3369, doi:10.1016/j.quascirev.2010.06.006. [online] Available from: http://dx.doi.org/10.1016/j.quascirev.2010.06.006 (Accessed 13 January 2012)

Dunlop, P., R. Shannon, M. Mccabe, R. Quinn, and E. Doyle (2010), Marine geophysical evidence for ice sheet extension and recession on the Malin Shelf: New evidence for the western limits of the British Irish Ice Sheet, Mar. Geol., 276(1-4), 86–99, doi:10.1016/j.margeo.2010.07.010. [online] Available from: http://dx.doi.org/10.1016/j.margeo.2010.07.010 (Accessed 14 September 2011)

Dunlop, P., F. Sacchetti, S. Benetti, and C. O. Cofaigh (2011), Mapping Ireland ’ s Glaciated Continental Margin Using Marine Geophysical Data, in Developments in Earth Surface Processes, vol. 15, pp. 339–357, Elsevier B.V. [online] Available from: http://dx.doi.org/10.1016/B978-0-444-53446-0.00011-2

Elverhøi, A., F. De Blasio, F. Butt, D. Issler, C. Harbitz, L. Engvik, A. Solheim, and J. Marr (2002), Submarine mass-wasting on glacially-influeced continental slopes: processes and dynamics, in Glacier-Influenced Sedimentation on High-Latitude Continetal Margins, edited by J. A. Dowdeswell and C. O Cofaigh, pp. 73–78, Geological Society, London, Special Publication.

Evans, J., J. A. Dowdeswell, C. Ó Cofaigh, T. J. Benham, and J. B. Anderson (2006), Extent and dynamics of the West Antarctic Ice Sheet on the outer continental shelf of Pine Island Bay during the last glaciation, Mar. Geol., 230(1-2), 53–72, doi:10.1016/j.margeo.2006.04.001. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0025322706000910 (Accessed 10 March 2012)

Fonseca, L., and L. Mayer (2007), Remote estimation of surficial seafloor properties through the application Angular Range Analysis to multibeam sonar data, Mar.

28

Geophys. Res., 28(2), 119–126, doi:10.1007/s11001-007-9019-4. [online] Available from: http://www.springerlink.com/index/10.1007/s11001-007-9019-4 (Accessed 4 August 2011)

Fonseca, L., L. Mayer, D. Orange, and N. Driscoll (2002), The high-frequency backscattering angular response of gassy sediments: Model/data comparison from the Eel River Margin, California, J. Acoust. Soc. Am., 111(June), 2621–2631, doi:10.1121/1.1471911.

Fonseca, L., C. Brown, B. Calder, L. Mayer, and Y. Rzhanov (2009), Angular range analysis of acoustic themes from Stanton Banks Ireland: A link between visual interpretation and multibeam echosounder angular signatures, Appl. Acoust., 70(10), 1298–1304, doi:10.1016/j.apacoust.2008.09.008. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0003682X08002028 (Accessed 4 August 2011)

Foster, G., A. Gleason, B. Costa, T. Battista, and C. Taylor (2013), Acoustic Applications.

Freire, F., R. Gyllencreutz, R. U. Jafri, and M. Jakobsson (2014), Acoustic evidence of a submarine slide in the deepest part of the Arctic, the Molloy Hole, Geo-Marine Lett., doi:10.1007/s00367-014-0371-5. [online] Available from: http://link.springer.com/10.1007/s00367-014-0371-5 (Accessed 14 April 2014)

García, M., J. A. Dowdeswell, G. Ercilla, and M. Jakobsson (2012), Recent glacially influenced sedimentary processes on the East Greenland continental slope and deep Greenland Basin, Quat. Sci. Rev., 49, 64–81, doi:10.1016/j.quascirev.2012.06.016. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0277379112002557 (Accessed 14 March 2013)

Gardner, J. V, P. Dartnell, L. A. Mayer, and J. E. Hughes Clarke (2003), Geomorphology, acoustic backscatter, and processes in Santa Monica Bay from multibeam mapping., Mar. Environ. Res., 56(1-2), 15–46, doi:10.1016/S0141-1136(02)00323-9. [online] Available from: http://www.ncbi.nlm.nih.gov/pubmed/12648948 (Accessed 4 October 2011)

Gavrilov, A. N., and I. M. Parnum (2010), Fluctuations of Seafloor Backscatter Data From Multibeam Sonar Systems, IEEE J. Ocean. Eng., 35(2), 209–219, doi:10.1109/JOE.2010.2041262. [online] Available from: http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=5439783

Glasser, N. F., and M. R. Bennett (2004), Glacial erosional landforms: origins and significance for palaeoglaciology, Prog. Phys. Geogr., 28(1), 43–75, doi:10.1191/0309133304pp401ra. [online] Available from: http://ppg.sagepub.com/cgi/doi/10.1191/0309133304pp401ra (Accessed 16 April 2014)

Graham, A. G. C., L. Lonergan, and M. S. Stoker (2009), Seafloor glacial features reveal the extent and decay of the last British Ice Sheet , east of Scotland, J. Quat. Sci., 24(September 2008), 117–138, doi:10.1002/jqs.

29

Graham, A. G. C., R. D. Larter, K. Gohl, J. A. Dowdeswell, C.-D. D. Hillenbrand, J. A. Smith, J. Evans, G. Kuhn, and T. Deen (2010), Flow and retreat of the Late Quaternary Pine Island-Thwaites palaeo-ice stream, West Antarctica, J. Geophys. Res., 115(F3), 1–12, doi:10.1029/2009JF001482. [online] Available from: http://www.agu.org/pubs/crossref/2010/2009JF001482.shtml (Accessed 18 October 2011)

Graham, A. G. C., P. Dutrieux, D. G. Vaughan, F. O. Nitsche, R. Gyllencreutz, S. L. Greenwood, R. D. Larter, and A. Jenkins (2013), Seabed corrugations beneath an Antarctic ice shelf revealed by autonomous underwater vehicle survey: Origin and implications for the history of Pine Island Glacier, J. Geophys. Res. Earth Surf., 118(3), 1356–1366, doi:10.1002/jgrf.20087. [online] Available from: http://doi.wiley.com/10.1002/jgrf.20087 (Accessed 13 April 2014)

Greenwood, S. L., and C. D. Clark (2009), Reconstructing the last Irish Ice Sheet 1: changing flow geometries and ice flow dynamics deciphered from the glacial landform record, Quat. Sci. Rev., 28(27-28), 3085–3100, doi:10.1016/j.quascirev.2009.09.008. [online] Available from: http://dx.doi.org/10.1016/j.quascirev.2009.09.008 (Accessed 7 October 2011)

Harbitz, C. B., F. Løvholt, G. Pedersen, and D. G. Masson (2006), Mechanisms of tsunami generation by submarine landslides : a short review, Nor. J. Geol., 86, 255–264.

Hogan, K. A., J. A. Dowdeswell, and J. Mienert (2013), New insights into slide processes and seafloor geology revealed by side-scan imagery of the massive Hinlopen Slide, Arctic Ocean margin, Geo-Marine Lett., 33(5), 325–343, doi:10.1007/s00367-013-0330-6. [online] Available from: http://link.springer.com/10.1007/s00367-013-0330-6 (Accessed 30 September 2013)

Howe, J. A., D. Dove, T. Bradwell, and J. Gafeira (2012), Submarine geomorphology and glacial history of the Sea of the Hebrides, UK, Mar. Geol., 315-318, 64–76, doi:10.1016/j.margeo.2012.06.005. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0025322712001351 (Accessed 3 March 2014)

Hughes Clarke, J. (1994), Toward remote seafloor classification using the angular response of acoustic backscattering: a case study from multiple overlapping GLORIA data, IEEE J. Ocean. Eng., 19(1), 112–127, doi:10.1109/48.289456. [online] Available from: http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=289456

Hughes Clarke, J. E., L. A. Mayer, and D. E. Wells (1996), Shallow-water imaging multibeam sonars: A new tool for investigating seafloor processes in the coastal zone and on the continental shelf, Mar. Geophys. Res., 18(6), 607–629, doi:10.1007/BF00313877. [online] Available from: http://www.springerlink.com/index/10.1007/BF00313877

Hughes-Clarke, J. E. (2012), Optimal use of multibeam technology in the study of shelf morphodynamics, in Sediments, Morphology and Sedimentary Processes on Continental Shelves: Advances in Technologies, Research, and Applications, vol. 44, edited by M. Z. Li, C. R. Sherwood, and P. R. Hill, pp. 1–8, John Wiley & Sons, Ltd.

30

Jakobsson, M., L. Polyak, M. Edwards, J. Kleman, and B. Coakley (2008), Glacial geomorphology of the Central Arctic Ocean : the Chukchi Borderland and the Lomonosov Ridge, Earth Surf. Process. Landforms, 545, 526–545, doi:10.1002/esp.

Jakobsson, M. et al. (2010), An Arctic Ocean ice shelf during MIS 6 constrained by new geophysical and geological data, Quat. Sci. Rev., 29(25-26), 3505–3517, doi:10.1016/j.quascirev.2010.03.015. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0277379110000934 (Accessed 25 May 2012)

Jakobsson, M. et al. (2011), Geological record of ice shelf break-up and grounding line retreat, Pine Island Bay, West Antarctica, Geology, 39(7), 691–694, doi:10.1130/G32153.1. [online] Available from: http://geology.gsapubs.org/cgi/doi/10.1130/G32153.1 (Accessed 13 October 2011)

Jakobsson, M., J. B. Anderson, F. O. Nitsche, R. Gyllencreutz, A. E. Kirshner, N. Kirchner, M. O. Regan, R. Mohammad, and B. Eriksson (2012a), Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay Trough , West Antarctica, Quat. Sci. Rev., 1–10, doi:10.1016/j.quascirev.2011.12.017. [online] Available from: http://dx.doi.org/10.1016/j.quascirev.2011.12.017

Jakobsson, M. et al. (2012b), The International Bathymetric Chart of the Arctic Ocean (IBCAO) Version 3.0, Geophys. Res. Lett., 39(12), 1–6, doi:10.1029/2012GL052219. [online] Available from: http://doi.wiley.com/10.1029/2012GL052219 (Accessed 23 September 2013)

Kilfeather, A. A., C. Ó Cofaigh, J. M. Lloyd, J. A. Dowdeswell, S. Xu, and S. G. Moreton (2011), Ice-stream retreat and ice-shelf history in Marguerite Trough, Antarctic Peninsula: Sedimentological and foraminiferal signatures, Geol. Soc. Am. Bull., 123(5-6), 997–1015, doi:10.1130/B30282.1. [online] Available from: http://gsabulletin.gsapubs.org/cgi/doi/10.1130/B30282.1 (Accessed 30 March 2012)

King, L. H. (1976), Relict iceberg furrows on the Laurentian Channel and western Grand Banks, Can. J. Earth Sci., 13(8), 1082–1092, doi:10.1139/e76-111. [online] Available from: http://www.nrcresearchpress.com/doi/abs/10.1139/e76-111

Kirshner, A. E., J. B. Anderson, M. Jakobsson, M. O. Regan, W. Majewski, F. O. Nitsche, and M. O’Regan (2012), Post-LGM deglaciation in Pine Island Bay, West Antarctica, Quat. Sci. Rev., 38, 11–26, doi:10.1016/j.quascirev.2012.01.017. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0277379112000297 (Accessed 30 April 2012)

Kuijpers, A., F. Dalhoff, M. P. Brandt, P. Hümbs, T. Schott, and A. Zotova (2007), Giant iceberg plow marks at more than 1 km water depth offshore West Greenland, Mar. Geol., 246(1), 60–64, doi:10.1016/j.margeo.2007.05.010. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0025322707001430 (Accessed 28 November 2013)

Laberg, J. S., and T. O. Vorren (2000), The Traenadjupet Slide , offshore Norway -- morphology , evacuation and triggering mechanisms, Mar. Geol., 171, 95–114.

31

Laberg, J. S., K. Kawamura, H. Amundsen, N. Baeten, M. Forwick, T. A. Rydningen, and T. O. Vorren (2013), A submarine landslide complex affecting the Jan Mayen Ridge, Norwegian–Greenland Sea: slide-scar morphology and processes of sediment evacuation, Geo-Marine Lett., 51–58, doi:10.1007/s00367-013-0345-z. [online] Available from: http://link.springer.com/10.1007/s00367-013-0345-z (Accessed 21 January 2014)

Lamarche, G., X. Lurton, A.-L. Verdier, and J.-M. Augustin (2011), Quantitative characterisation of seafloor substrate and bedforms using advanced processing of multibeam backscatter—Application to Cook Strait, New Zealand, Cont. Shelf Res., 31(2), S93–S109, doi:10.1016/j.csr.2010.06.001. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0278434310001925 (Accessed 13 September 2011)

Lindberg, B., J. S. Laberg, and T. O. Vorren (2004), The Nyk Slide—morphology, progression, and age of a partly buried submarine slide offshore northern Norway, Mar. Geol., 213(1-4), 277–289, doi:10.1016/j.margeo.2004.10.010. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0025322704002749 (Accessed 10 October 2012)

Lurton, X. (2010), An introduction to underwater acoustics - principles and applications, Second edi., Springer Netherlands.

Maclachlan, S. E., J. A. Howe, and M. E. Vardy (2010), Morphodynamic evolution of Kongsfjorden-Krossfjorden, Svalbard, during the Late Weichselian and Holocene, Geol. Soc. London, Spec. Publ., 344(1), 195–205, doi:10.1144/SP344.14. [online] Available from: http://sp.lyellcollection.org/cgi/doi/10.1144/SP344.14 (Accessed 3 May 2012)

Maclean, B., S. Blasco, R. Bennett, J. England, W. Rainey, J. Hughes-Clarke, and J. Beaudoin (2010), Ice keel seabed features in marine channels of the central Canadian Arctic Archipelago: evidence for former ice streams and iceberg scouring, Quat. Sci. Rev., 29(17-18), 2280–2301, doi:10.1016/j.quascirev.2010.05.032. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0277379110001770 (Accessed 7 October 2011)

Marcussen, C., and Scientific Party (2012), Lomonosov Ridge off Greenland 2012 ( LOMROG III ) – Cruise Report.

Masetti, G., R. Sacile, A. Trucco, I. Idrografico, P. Osservatorio, and C. Genova (2011), Remote characterization of seafloor adjacent to shipwrecks using mosaicking and analysis of backscatter response, Ital. J. Remote Sens., 43(2), 77–92, doi:10.5721/ItJRS20114326. [online] Available from: http://www.aitjournal.com/articleView.aspx?ID=313 (Accessed 23 February 2012)

Maslin, M., M. Owen, S. Day, and D. Long (2004), Linking continental-slope failures and climate change: Testing the clathrate gun hypothesis, Geology, 32(1), 53, doi:10.1130/G20114.1. [online] Available from: http://geology.gsapubs.org/cgi/doi/10.1130/G20114.1 (Accessed 20 July 2012)

32

Mayer, L. A. (2006), Frontiers in Seafloor Mapping and Visualization, Mar. Geophys. Res., 27(1), 7–17, doi:10.1007/s11001-005-0267-x. [online] Available from: http://link.springer.com/10.1007/s11001-005-0267-x (Accessed 7 March 2014)

McGonigle, C., C. Brown, R. Quinn, and J. Grabowski (2009), Evaluation of image-based multibeam sonar backscatter classification for benthic habitat discrimination and mapping at Stanton Banks, UK, Estuar. Coast. Shelf Sci., 81(3), 423–437, doi:10.1016/j.ecss.2008.11.017. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S027277140800454X (Accessed 14 June 2011)

McGonigle, C., C. J. Brown, and R. Quinn (2010a), Insonification orientation and its relevance for image-based classification of multibeam backscatter, ICES J. Mar. Sci., 1010–1023.

McGonigle, C., C. J. Brown, and R. Quinn (2010b), Operational Parameters , Data Density and Benthic Ecology : Considerations for Image-Based Classification of Multibeam Backscatter, Mar. Geod., 33(1), 16–38, doi:10.1080/01490410903530273. [online] Available from: http://www.tandfonline.com/doi/abs/10.1080/01490410903530273 (Accessed 22 February 2012)

Micallef, A. (2011), Marine Geomorphology : Geomorphological Mapping and the Study of Submarine Landslides, in Developments in Earth Surface Processes, vol. 15, edited by M. J. Smith, P. Paron, and J. S. Griffiths, pp. 377–395, Elsevier B.V. [online] Available from: http://dx.doi.org/10.1016/B978-0-444-53446-0.00013-6

Noormets, R., J. a. A. Dowdeswell, R. D. D. Larter, C. Ó Cofaigh, J. Evans, and C. Ó. Cofaigh (2009), Morphology of the upper continental slope in the Bellingshausen and Amundsen Seas – Implications for sedimentary processes at the shelf edge of West Antarctica, Mar. Geol., 258(1-4), 100–114, doi:10.1016/j.margeo.2008.11.011. [online] Available from: http://dx.doi.org/10.1016/j.margeo.2008.11.011 (Accessed 10 March 2012)

Nygård, A., H. P. Sejrup, H. Haflidason, M. Cecchi, and D. Ottesen (2004), Deglaciation history of the southwestern Fennoscandian Ice Sheet between 15 and 13 14C ka BP, Boreas, 33(1), 1–17, doi:10.1080/03009480310006943. [online] Available from: http://doi.wiley.com/10.1080/03009480310006943 (Accessed 17 April 2014)

OLEX (2014), OLEX platform date accessed: 04-16-2014, [online] Available from: http://www.olex.no

Ottesen, D., and J. A. Dowdeswell (2009), An inter-ice-stream glaciated margin: Submarine landforms and a geomorphic model based on marine-geophysical data from Svalbard, Geol. Soc. Am. Bull., 121(11-12), 1647–1665, doi:10.1130/B26467.1. [online] Available from: http://gsabulletin.gsapubs.org/cgi/doi/10.1130/B26467.1 (Accessed 6 February 2014)

Ottesen, D., J. A. Dowdeswell, J. Y. Landvik, and J. Mienert (2007), Dynamics of the Late Weichselian ice sheet on Svalbard inferred from high-resolution sea-floor morphology, Boreas, 36(3), 286–306, doi:10.1080/03009480701210378. [online]

33

Available from: http://doi.wiley.com/10.1080/03009480701210378 (Accessed 12 April 2012)

Parnum, I. M. (2007), Benthic habitat mapping using multibeam sonar systems, 208 pp., Curtin University, Perth.

Parnum, I. M., and A. N. Gavrilov (2011), High-frequency multibeam echo-sounder measurements of seafloor backscatter in shallow water: Part 2 – Mosaic production, analysis and classification, Underw. Technol. Int. J. Soc. Underw., 30(1), 3–12, doi:10.3723/ut.30.013. [online] Available from: http://openurl.ingenta.com/content/xref?genre=article&issn=1756-0543&volume=30&issue=1&spage=3 (Accessed 28 November 2013)

Paull, C. K., W. Ussler, and W. S. Holbrook (2007), Assessing methane release from the colossal Storegga submarine landslide, Geophys. Res. Lett., 34(4), L04601, doi:10.1029/2006GL028331. [online] Available from: http://www.agu.org/pubs/crossref/2007/2006GL028331.shtml (Accessed 22 November 2012)

Penrose, J. D., P. J. W. Siwabessy, A. Gavrilov, I. Parnum, L. J. Hamilton, A. Bickers, B. Brooke, D. A. Ryan, and P. Kennedy (2005), Acoustic Techniques for Seabed Classification.

Preston, J. (2009), Automated acoustic seabed classification of multibeam images of Stanton Banks, Appl. Acoust., 70(10), 1277–1287, doi:10.1016/j.apacoust.2008.07.011. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0003682X08001801 (Accessed 4 August 2011)

Rebesco, M. et al. (2011), Deglaciation of the western margin of the Barents Sea Ice Sheet — A swath bathymetric and sub-bottom seismic study from the Kveithola Trough, Mar. Geol., 279(1-4), 141–147, doi:10.1016/j.margeo.2010.10.018. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0025322710002896 (Accessed 10 January 2012)

Rzhanov, Y., L. Fonseca, and L. Mayer (2011), Construction of seafloor thematic maps from multibeam acoustic backscatter angular response data, Comput. Geosci., 1–7, doi:10.1016/j.cageo.2011.09.001. [online] Available from: http://dx.doi.org/10.1016/j.cageo.2011.09.001 (Accessed 18 December 2011)

Sacchetti, F., S. Benetti, a. Georgiopoulou, P. M. Shannon, B. M. O’Reilly, P. Dunlop, R. Quinn, and C. Ó Cofaigh (2012), Deep-water geomorphology of the glaciated Irish margin from high-resolution marine geophysical data, Mar. Geol., 291-294, 113–131, doi:10.1016/j.margeo.2011.11.011. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0025322711002684 (Accessed 14 March 2014)

Siemes, K., M. Snellen, A. R. Amiri-simkooei, D. G. Simons, and J. Hermand (2010), Predicting Spatial Variability of Sediment Properties From Hydrographic Data for Geoacoustic Inversion, IEEE J. Ocean. Eng., 35(4), 766–778.

34

Simons, D. G., and M. Snellen (2009), A Bayesian approach to seafloor classification using multi-beam echo-sounder backscatter data, Appl. Acoust., 70(10), 1258–1268, doi:10.1016/j.apacoust.2008.07.013. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0003682X08001813 (Accessed 4 August 2011)

Stokes, C. R., M. Spagnolo, and C. D. Clark (2011), The composition and internal structure of drumlins: Complexity, commonality, and implications for a unifying theory of their formation, Earth-Science Rev., 107(3-4), 398–422, doi:10.1016/j.earscirev.2011.05.001. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0012825211000699 (Accessed 3 October 2011)

Tang, Q., X. Zhou, Z. Liu, and and D. Du (2005), Processing Multibeam Backscatter Data, Mar. Geod., 28(3), 251–258, doi:10.1080/01490410500204595. [online] Available from: http://www.tandfonline.com/doi/abs/10.1080/01490410500204595 (Accessed 4 October 2011)

Todd, B. J., and J. Shaw (2012), Laurentide Ice Sheet dynamics in the Bay of Fundy, Canada, revealed through multibeam sonar mapping of glacial landsystems, Quat. Sci. Rev., 58, 83–103, doi:10.1016/j.quascirev.2012.10.016. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S027737911200412X (Accessed 14 February 2014)

Vanneste, M., and J. L’Heureux (2012), Shallow landslides and their dynamics in coastal and deepwater environments, Norway, in Submarine Mass Movements and Their Consequences 5th International Symposium, edited by Y. Yamada, K. Kawamura, K. Ikehara, Y. Ogawa, R. Urgeles, D. Mosher, J. Chaytor, and M. Strasser. [online] Available from: http://link.springer.com/chapter/10.1007/978-94-007-2162-3_3 (Accessed 3 July 2013)

Vanneste, M., J. Mienert, and S. Bünz (2006), The Hinlopen Slide: A giant, submarine slope failure on the northern Svalbard margin, Arctic Ocean, Earth Planet. Sci. Lett., 245(1-2), 373–388, doi:10.1016/j.epsl.2006.02.045. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0012821X06001865 (Accessed 31 May 2012)

Wellner, J. S., A. L. Lowe, S. S. Shipp, and J. B. Anderson (2001), Distribution of glacial geomorphic features on the Antarctic continental shelf and correlation with substrate : implications for ice behavior, J. Glaciol., 47(158), 397–411.

Wellner, J. S., D. C. Heroy, and J. B. Anderson (2006), The death mask of the antarctic ice sheet: Comparison of glacial geomorphic features across the continental shelf, Geomorphology, 75(1-2), 157–171, doi:10.1016/j.geomorph.2005.05.015. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0169555X05002527 (Accessed 8 February 2012)

Winsborrow, M. C. M., K. Andreassen, G. D. Corner, and J. S. Laberg (2010), Deglaciation of a marine-based ice sheet: Late Weichselian palaeo-ice dynamics and retreat in the southern Barents Sea reconstructed from onshore and offshore glacial

35

geomorphology, Quat. Sci. Rev., 29(3-4), 424–442, doi:10.1016/j.quascirev.2009.10.001. [online] Available from: http://linkinghub.elsevier.com/retrieve/pii/S0277379109003333 (Accessed 22 March 2014)

36

37

Part VIII. Paper

Acoustic evidence of a submarine slide in the deepest part of the Arctic, the Molloy Hole

Francis Freire, Richard Gyllencreutz, Rooh Ullah Jafri and Martin Jakobsson

Department of Geological Sciences, Stockholm University, Sweden

Geo-Marine Letters 2014 - DOI 10.1007/s00367-014-0371-5