USF Libraries
USF Digital Collections

Multibeam observations of mine scour and burial near clearwater, florida, including a test of the vims 2d burial model

MISSING IMAGE

Material Information

Title:
Multibeam observations of mine scour and burial near clearwater, florida, including a test of the vims 2d burial model
Physical Description:
Book
Language:
English
Creator:
Wolfson, Monica L
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Multibeam bathymetry
Sea level datum
Dissertations, Academic -- Marine Science -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The ability to detect buried mines on the seafloor remains one of the most important tasks in mine countermeasures. As such, there is a vested interest in the development of predictive models of mine burial. This research was conducted in support of the Office of Naval Research Program in Mine Burial Prediction. Repeat high-resolution multibeam bathymetry data were collected over the Indian Rocks Beach (IRB) mine burial experiment site during January through March of 2003, in order to observe in situ scour and burial of instrumented inert mines and mine-like cylinders. These data were also used to test the validity of the VIMS 2D mine burial model.
Thesis:
Thesis (M.S.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Monica L. Wolfson.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 256 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001670336
oclc - 62278903
usfldc doi - E14-SFE0001197
usfldc handle - e14.1197
System ID:
SFS0025518:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001670336
003 fts
005 20051216093258.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 051115s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001197
035
(OCoLC)62278903
SFE0001197
040
FHM
c FHM
049
FHMM
090
GC11.2 (Online)
1 100
Wolfson, Monica L.
0 245
Multibeam observations of mine scour and burial near clearwater, florida, including a test of the vims 2d burial model
h [electronic resource] /
by Monica L. Wolfson.
260
[Tampa, Fla.] :
b University of South Florida,
2005.
502
Thesis (M.S.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 256 pages.
3 520
ABSTRACT: The ability to detect buried mines on the seafloor remains one of the most important tasks in mine countermeasures. As such, there is a vested interest in the development of predictive models of mine burial. This research was conducted in support of the Office of Naval Research Program in Mine Burial Prediction. Repeat high-resolution multibeam bathymetry data were collected over the Indian Rocks Beach (IRB) mine burial experiment site during January through March of 2003, in order to observe in situ scour and burial of instrumented inert mines and mine-like cylinders. These data were also used to test the validity of the VIMS 2D mine burial model.
590
Adviser: David Naar.
653
Multibeam bathymetry.
Sea level datum.
690
Dissertations, Academic
z USF
x Marine Science
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1197



PAGE 1

Multibeam Observations of Mine Scour and Burial near Clearwater, Florida, Including a Test of the VIMS 2D Mine Burial Model by Monica L. Wolfson A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida Major Professor: David F. Naar, Ph.D. Peter A. Howd, Ph.D. Stanley D. Locker, Ph.D. Carl T. Friedrichs, Ph.D. Date of Approval: July 19, 2005 Keywords: multibeam bathymetry, sea level datum Copyright 2005, Monica L. Wolfson

PAGE 2

Acknowledgements Funding for this research was provided by the Office of Naval Research Mine Burial Program, under the management of Drs. Dawn Lavoie, Jill Karsten, Roy H. Wilkens, and Thomas Drake. I thank my committee: Drs. David Naar, Peter Howd, Stanley Locker, and Carl Friedrichs for their guidance and patience over the last couple of years. A special thanks goes out to Dr. Art Trembanis for all of his help with the model and for providing invaluable advice. Drs. Mike Richardson and Thomas Wever provided data and assistance. I would like to specifically recognize Dr. Gary Mitchum, Kate Ciembronowicz, Michelle McIntyre, Kara Sedwick, and Sage Lichtenwaler, without whom I would still be trying to figure out ArcMap and Matlab. I would also like to thank Brian Donahue for showing me how to collect and process multibeam data and for taking all my questions in stride. Most importantly, I would like to thank my family for their love and support.

PAGE 3

Table of Contents List of Tables.....................................................................................................................iii List of Figures....................................................................................................................iv Abstract..............................................................................................................................vi Chapter 1. Introduction........................................................................................................1 Chapter 2. Multibeam Observations and Model Comparison of Two Mines in Fine and Coarse Sand...................................................................................................................4 Introduction..............................................................................................................4 The Experiment........................................................................................................5 Multibeam Data.......................................................................................................8 Mine Burial and Scour Models................................................................................8 Obtaining a Vertical Reference Frame....................................................................9 Temporal Analysis of Mine Burial .......................................................................17 Temporal Changes in Scour and Burial over the A3 Mine....................................19 Comparison of A3 Multibeam Observations to the VIMS 2D Burial Model....................................................................................................30 Discussion of the A3 Comparisons........................................................................35 Temporal Changes in Scour and Burial over the F8 Mine....................................36 Comparison of F8 Multibeam Observations to the VIMS 2D Burial Model....................................................................................................45 Discussion of the F8 Comparisons........................................................................50 Conclusions............................................................................................................50 Chapter 3. Multibeam Observations and Model Comparison of the Remaining Mines ...52 Introduction............................................................................................................52 The A1 Mine..........................................................................................................52 Temporal Changes in Scour and Burial.....................................................52 Comparison of A1 Multibeam Observations to the VIMS 2D Burial Model........................................................................................63 The A2 Mine..........................................................................................................68 Temporal Changes in Scour and Burial.....................................................68 Comparison of A2 Multibeam Observations to the VIMS 2D Burial Model........................................................................................75 The A4 Mine..........................................................................................................81 Temporal Changes in Scour and Burial.....................................................81 i

PAGE 4

Comparison of A4 Multibeam Observations to the VIMS 2D Burial Model........................................................................................88 The F5 Mine...........................................................................................................93 Temporal Changes in Scour and Burial.....................................................93 Comparison of F5 Multibeam Observations to the VIMS 2D Burial Model........................................................................................99 The F6 Mine.........................................................................................................106 Temporal Changes in Scour and Burial...................................................106 Comparison of F6 Multibeam Observations to the VIMS 2D Burial Model......................................................................................114 The F7 Mine.........................................................................................................118 Temporal Changes in Scour and Burial...................................................118 Comparison of F7 Multibeam Observations to the VIMS 2D Burial Model......................................................................................125 The F9 Mine.........................................................................................................131 Temporal Changes in Scour and Burial...................................................131 Comparison of F9 Multibeam Observations to the VIMS 2D Burial Model......................................................................................140 The F10 Mine.......................................................................................................140 Temporal Changes in Scour and Burial...................................................140 Comparison of F10 Multibeam Observations to the VIMS 2D Burial Model......................................................................................146 Summary of Results.............................................................................................151 Chapter 4. Analysis of Mine Scour..................................................................................153 Introduction..........................................................................................................153 Methods................................................................................................................153 Scour Analysis.....................................................................................................154 Summary of Analysis...........................................................................................171 Chapter 5. Discussion......................................................................................................175 Chapter 6. Summary........................................................................................................188 References........................................................................................................................191 Bibliography....................................................................................................................193 Appendices.......................................................................................................................194 Appendix A: Calculation of Ambient Seafloor Change......................................194 Appendix B: Description of Equations ...............................................................239 ii

PAGE 5

iii List of Tables Table 1. Data table for the A3 mine.........................................................................31 Table 2. Data table for the F8 mine.........................................................................46 Table 3. Data table for the A1 mine.........................................................................65 Table 4. Data table for the A2 mine.........................................................................78 Table 5. Data table for the A4 mine.........................................................................91 Table 6. Data table for the F5 mine.......................................................................103 Table 7. Data table for the F6 mine.......................................................................115 Table 8. Data table for the F7 mine.......................................................................128 Table 9. Data table for the F9 mine.......................................................................138 Table 10. Data table for the F10 mine.....................................................................148 Table 11. Beam footprint and beam sp acing along the seafloor for various depths and beam pointing angles.............................................................184 Table 12. Along track beam spaci ng for various vessel speeds...............................184

PAGE 6

List of Figures Figure 1. Location of the experiment site off Indian Rocks Beach, Florida...............6 Figure 2. Location of deployed equipment for the fine sand (2A) and coarse sand (2B) study sites..................................................................7 Figure 3. Beam distances to the bed for the fine sand site........................................10 Figure 4. Schematic representation of the Sontek data.............................................12 Figure 5. De-meaned pressure sensor data................................................................13 Figure 6. Filtered pressure data.................................................................................15 Figure 7. Seafloor elevation change..........................................................................16 Figure 8. Dimensions of the quadpods, spiders, and mine-like cylinders visible in the multibeam images................................................................20 Figure 9. January 10 th survey over the A3 mine.......................................................21 Figure 10. January 13 th survey over the A3 mine.......................................................23 Figure 11. January 17 th survey over the A3 mine.......................................................24 Figure 12. January 20 th survey over the A3 mine.......................................................26 Figure 13. February 6 th survey over the A3 mine.......................................................27 Figure 14. March 13 th survey over the A3 mine.........................................................28 Figure 15. ROV video still image of the A3 mine on March 13, 2003.......................29 Figure 16. Comparison of multibeam observed (black) and predicted (gray) mine depth for the A3 mine over the course of the experiment.................32 Figure 17. Comparison of the mine (magenta), predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the A3 mine over the course of the experiment.....................................................33 iv

PAGE 7

Figure 18. January 13 th survey over the F8 mine........................................................37 Figure 19. January 17 th survey over the F8 mine........................................................38 Figure 20. January 20 th survey over the F8 mine........................................................40 Figure 21. February 6 th survey over the F8 mine........................................................41 Figure 22. March 13 th survey over the F8 mine..........................................................43 Figure 23. ROV video still image of the F8 mine on March 13, 2003........................44 Figure 24. Comparison of multibeam observed (black) and predicted (gray) mine depth for the F8 mine over the course of the experiment.................47 Figure 25. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the F8 mine over the course of the experiment..............................................................49 Figure 26. Location of the deep fine sand study site...................................................53 Figure 27. Location of deployed equipment in the deep fine site...............................54 Figure 28. January 10 th survey over the A1 mine.......................................................56 Figure 29. January 13 th survey over the A1 mine.......................................................57 Figure 30. January 17 th survey over the A1 mine.......................................................58 Figure 31. January 20 th survey over the A1 mine.......................................................59 Figure 32. February 6 th survey over the A1 mine.......................................................61 Figure 33. March 13 th survey over the A1 mine.........................................................62 Figure 34. ROV video still image of the A1 mine on March 13, 2003.......................64 Figure 35. Comparison of multibeam observed (black) and predicted (gray) mine depth for the A1 mine over the course of the experiment.................66 Figure 36. Comparison of the mine (magenta), predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the A1 mine over the course of the experiment..............................................................67 Figure 37. January 10 th survey over the A2 mine.......................................................69 v

PAGE 8

Figure 38. January 13 th survey over the A2 mine.......................................................71 Figure 39. January 17 th survey over the A2 mine.......................................................72 Figure 40. January 20 th survey over the A2 mine.......................................................73 Figure 41. February 6 th survey over the A2 mine.......................................................74 Figure 42. March 13 th survey over the A2 mine.........................................................76 Figure 43. ROV video still image of the A2 mine on March 13, 2003.......................77 Figure 44. Comparison of multibeam observed (black) and predicted (gray) mine depth for the A2 mine over the course of the experiment.................79 Figure 45. Comparison of the mine (magenta), predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the A2 mine over the course of the experiment...............................................80 Figure 46. January 10 th survey over the A4 mine.......................................................82 Figure 47. January 13 th survey over the A4 mine.......................................................83 Figure 48. January 17 th survey over the A4 mine.......................................................85 Figure 49. January 20 th survey over the A4 mine.......................................................86 Figure 50. February 6 th survey over the A4 mine.......................................................87 Figure 51. March 13 th survey over the A4 mine.........................................................89 Figure 52. ROV video still image of the A4 mine on March 13, 2003.......................90 Figure 53. Comparison of multibeam observed (black) and predicted (gray) mine depth for the A4 mine over the course of the experiment.................92 Figure 54. Comparison of the mine (magenta), predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the A4 mine over the course of the experiment...............................................94 Figure 55. January 13 th survey over the F5 mine........................................................96 Figure 56. January 17 th survey over the F5 mine........................................................97 Figure 57. January 20 th survey over the F5 mine........................................................98 vi

PAGE 9

Figure 58. February 6 th survey over the F5 mine......................................................100 Figure 59. March 13 th survey over the F5 mine........................................................101 Figure 60. ROV video still image of the F5 mine on March 13, 2003......................102 Figure 61. Comparison of multibeam observed (black) and predicted (gray) mine depth for the F5 mine over the course of the experiment...............104 Figure 62. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the F5 mine over the course of the experiment............................................................105 Figure 63. January 13 th survey over the F6 mine......................................................108 Figure 64. January 17 th survey over the F6 mine......................................................109 Figure 65. January 20 th survey over the F6 mine......................................................110 Figure 66. February 6 th survey over the F6 mine......................................................111 Figure 67. March 13 th survey over the F6 mine........................................................112 Figure 68. ROV video still image of the F6 mine on March 13, 2003......................113 Figure 69. Comparison of multibeam observed (black) and predicted (gray) mine depth for the F6 mine over the course of the experiment...............116 Figure 70. Comparison of the mine predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the F6 mine over the course of the experiment............................................................117 Figure 71. January 13 th survey over the F7 mine......................................................120 Figure 72. January 17 th survey over the F7 mine......................................................121 Figure 73. January 20 th survey over the F7 mine......................................................123 Figure 74. February 6 th survey over the F7 mine......................................................124 Figure 75. March 13 th survey over the F7 mine........................................................126 Figure 76. ROV video still image of the F7 mine on March 13, 2003......................127 vii

PAGE 10

Figure 77. Comparison of multibeam observed (black) and predicted (gray) mine depth for the F7 mine over the course of the experiment...............129 Figure 78. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the F7 mine over the course of the experiment............................................................130 Figure 79. January 13 th survey over the F9 mine......................................................132 Figure 80. January 20 th survey over the F9 mine......................................................133 Figure 81. February 6 th survey over the F9 mine......................................................135 Figure 82. March 13 th survey over the F9 mine........................................................136 Figure 83. ROV video still image of the F9 mine on March 13, 2003......................137 Figure 84. Comparison of multibeam observed (black) and predicted (gray) mine depth for the F10 mine over the course of the experiment.............139 Figure 85. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the F9 mine over the course of the experiment............................................................141 Figure 86. January 13 th survey over the F10 mine....................................................143 Figure 87. January 20 th survey over the F10 mine....................................................144 Figure 88. February 6 th survey over the F10 mine....................................................145 Figure 89. ROV video still image of the F10 mine on March 13, 2003....................147 Figure 90. Comparison of multibeam observed (black) and predicted (gray) mine depth for the F10 mine over the course of the experiment.............149 Figure 91. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the F10 mine over the course of the experiment............................................................150 Figure 92. A1 scour pit..............................................................................................155 Figure 93. A1 scour pit hypsometry..........................................................................156 Figure 94. A2 scour pit..............................................................................................158 viii

PAGE 11

Figure 95. A2 scour pit hypsometry..........................................................................159 Figure 96. A3 scour pit..............................................................................................160 Figure 97. A3 scour pit hypsometry..........................................................................161 Figure 98. A4 scour pit..............................................................................................162 Figure 99. A4 scour pit hypsometry..........................................................................163 Figure 100. F5 scour pit..............................................................................................165 Figure 101. F5 scour pit hypsometry...........................................................................166 Figure 102. F6 scour pit..............................................................................................167 Figure 103. F6 scour pit hypsometry...........................................................................168 Figure 104. F9 scour pit..............................................................................................169 Figure 105. F9 scour pit hypsometry...........................................................................170 Figure 106. F10 scour pit............................................................................................172 Figure 107. F10 scour pit hypsometry.........................................................................173 Figure 108. Comparison of burial rates for mines in the shallow fine site.................176 Figure 109. Comparison of burial rates for mines in the deep fine site......................177 Figure 110. Comparison of burial rates for mines in the coarse site...........................178 Figure 111. A1 January 10th histogram......................................................................198 Figure 112. A1 January 13 th histogram.......................................................................199 Figure 113. A1 January 17 th histogram.......................................................................200 Figure 114. A1 January 20 th histogram.......................................................................201 Figure 115. A1 February 6 th histogram.......................................................................202 Figure 116. A1 March 13 th histogram.........................................................................203 Figure 117. A2 January 10 th histogram.......................................................................204 ix

PAGE 12

Figure 118. A2 January 13 th histogram.......................................................................205 Figure 119. A2 January 17 th histogram.......................................................................206 Figure 120. A2 January 20 th histogram.......................................................................207 Figure 121. A2 February 6 th histogram.......................................................................208 Figure 122. A2 March 13 th histogram.........................................................................209 Figure 123. A3 January 10 th histogram.......................................................................210 Figure 124. A3 January 13 th histogram.......................................................................211 Figure 125. A3 January 17 th histogram.......................................................................212 Figure 126. A3 January 20 th histogram.......................................................................213 Figure 127. A3 February 6 th histogram.......................................................................214 Figure 128. A3 March 13 th histogram.........................................................................215 Figure 129. A4 January 10 th histogram.......................................................................216 Figure 130. A4 January 13 th histogram.......................................................................217 Figure 131. A4 January 17 th histogram.......................................................................218 Figure 132. A4 January 20 th histogram.......................................................................219 Figure 133. A4 February 6 th histogram.......................................................................220 Figure 134. A4 March 13 th histogram.........................................................................221 Figure 135. F5 January 13 th histogram........................................................................222 Figure 136. F5 January 17 th histogram........................................................................223 Figure 137. F5 January 20 th histogram........................................................................224 Figure 138. F5 February 6 th histogram........................................................................225 Figure 139. F5 March 13 th histogram..........................................................................226 Figure 140. F6 January 13 th histogram........................................................................227 x

PAGE 13

Figure 141. F6 January 17 th histogram........................................................................228 Figure 142. F6 January 20 th histogram........................................................................229 Figure 143. F6 February 6 th histogram........................................................................230 Figure 144. F6 March 13 th histogram..........................................................................231 Figure 145. F9 January 13 th histogram........................................................................232 Figure 146. F9 January 20 th histogram........................................................................233 Figure 147. F9 February 6 th histogram........................................................................234 Figure 148. F9 March 13 th histogram..........................................................................235 Figure 149. F10 January 13 th histogram......................................................................236 Figure 150. F10 January 20 th histogram......................................................................237 Figure 151. F10 February 6 th histogram......................................................................238 xi

PAGE 14

Multibeam Observations of Mine Scour and Burial near Clearwater, Florida, Including a Test of the VIMS 2D Mine Burial Model Monica L. Wolfson ABSTRACT The ability to detect buried mines on the seafloor remains one of the most important tasks in mine countermeasures. As such, there is a vested interest in the development of predictive models of mine burial. This research was conducted in support of the Office of Naval Research Program in Mine Burial Prediction. Repeat high-resolution multibeam bathymetry data were collected over the Indian Rocks Beach (IRB) mine burial experiment site during January through March of 2003, in order to observe in situ scour and burial of instrumented inert mines and mine-like cylinders. These data were also used to test the validity of the VIMS 2D mine burial model. A set of six high-resolution multibeam surveys were collected over the IRB experiment site. Three study sites within the IRB site were chosen: two fine sand sites, a shallow one located in ~ 13 meters of water depth and a deep site located in ~ 14 meters of water depth; and a coarse sand site in ~ 13 meters. Results from these surveys indicate that mines deployed in fine sand are upwards of 74.5% buried within two months of deployment. Mines deployed in the coarse sand showed a lesser amount of scour, burying until they presented roughly the same hydrodynamic roughness of the surrounding rippled bedforms. In general, scour around the mines formed pits ~ 0.30 meters deep, with the most pronounced scour occurring at the ends of the mine. xii

PAGE 15

The multibeam data were also used to test the VIMS 2D mine burial model, which estimates percent burial of cylindrical mines based on predictions of wave-induced scour. The model proved valid for use in areas of fine sand, sufficiently predicting burial over the course of the experiment. In the area of coarse sand, the model greatly over-predicted the amount of burial. This is believed to be due to the presence of ripples around the mines, which affect local bottom morphodynamics and are not accounted for in the model. This issue is currently being addressed by modelers. xiii

PAGE 16

Chapter 1 Introduction Mine countermeasures are some of the most pressing issues being addressed by the Navy today. Current methods of mine hunting involve the use of side-scan sonars, which are dependent on the mine casting a shadow for detection. If the mine scours into the seabed and/or becomes buried by sediment, mine hunting techniques may be severely compromised. The Office of Naval Research Program in Mine Burial Prediction was established to study the how, when, and why of mine burial and develop mine burial probability models. Three locations were selected as experiments sites for this program: Corpus Christi, Texas (2001 and 2002); Marthas Vineyard, Massachusetts (2003); and Indian Rocks Beach offshore of Clearwater, Florida (2003). As part of the Indian Rocks Beach (IRB) experiment, repeat high-resolution multibeam surveys were made over the study site in order to observe in situ scour and burial of inert mines and mine-like cylinders. These data were used to perform temporal and spatial analyses of mine scour and burial and to test the validity of one of the probability models. This thesis represents the culmination of that research. The second chapter of this thesis is a manuscript submitted April 15 th 2005 to a special issue of the Journal of Ocean Engineering focused on mine burial and scour (Wolfson et al., 2005). This chapter provides a detailed analysis of the surveys over an instrumented mine deployed in a fine sand site and one deployed in a coarse sand site. 1

PAGE 17

Chapter three of this thesis includes the analyses and model comparisons for the remaining mines deployed as part of the IRB experiment. A more detailed analysis of the morphology of the scour formed around the mines is included in chapter four. Chapter five discusses the results and their significance. Chapter six summarizes the principle findings and conclusions of this thesis. Appendix A is a brief discussion on the method of determining changes in ambient seafloor elevation observed around the mines. Appendix B provides descriptions of the equations used to calculate the phase and amplitude lag of the tide record, as well the equations used to calculate beam width and spacing of the multibeam sonar. 2

PAGE 18

Chapter 2 Multibeam Observations and Model Comparison for Two Mines in Fine and Coarse Sand Abstract High-resolution multibeam bathymetry data collected offshore of Clearwater, Florida, are compared to predictions of mine burial by the VIMS 2D model for wave-induced scour. This paper focuses specifically on two instrumented but inert mines: an acoustic mine located in fine sands; and an optical instrumented mine located in coarse sands. Temporal analyses of the observed scour and burial of the mines and a method for obtaining a vertical frame of reference (MLLW) from pressure sensor data are presented. In the fine sand case, the model initially predicts a greater amount of burial than observed in the multibeam data; however, the values show a convergence during the course of the experiment. When the 5-centimeter vertical uncertainty (RMS error) of the multibeam sonar is considered, the predicted estimates of mine burial fall within the observable range. Correcting for the tilt of the mine (using a pitch sensor within the mine) can reduce the discrepancy between the observed and predicted percent burial. In the coarse sand case, the model does not work as well. Initially the predictions are within the range of the multibeam measurement uncertainty but then they overestimate the amount of observed burial over the rest of the experiment. Rippled bedforms appear to be influencing the mine scour and burial and should be included in future modeling efforts. 3

PAGE 19

Introduction The ability to detect buried mines on the seafloor remains one of the most difficult tasks in mine countermeasures. Morphodynamics of the seafloor are often responsible for the burial of heavy objects, including, but not limited to, pipelines, breakwaters, concrete, debris, and mines (Richardson et al., 2001). Mines are readily buried on impact and by secondary processes such as scour and fill, liquefaction, and changes in seafloor morphology. While mine-hunting techniques successfully locate mines resting on the seafloor, a partially buried mine can avoid sonar detection and requires either mine sweeping or complete area avoidance (Richardson and Briggs, 2000). It is therefore necessary to develop methods of predicting mine burial under different environmental conditions and temporal scales. The ability to predict how quickly scour will form around a mine and how quickly the mine will become buried under different energy and geological conditions is important in designing search strategies. High-resolution multibeam bathymetry data can be used to test current mine burial models by providing direct estimates of the scour and burial of a mine. Herein, the term mine actually refers to inert mine-like cylinders. Repeated passes of a multibeam sonar over a mine will document the amount of scour and percent burial over time, which can then be compared to the model predictions. This will test the validity of mine burial models. We define percent mine burial as percent of mine subsidence with respect to the ambient seafloor (Equation 1). % burial = 100xmDmdsdmD (1) = diameter of mine mD = depth of ambient seafloor sd d = depth of top of mine m 4

PAGE 20

The Experiment Mine burial experiments sponsored by the Office of Naval Research (ONR) were conducted off the coast of Clearwater, West-Central Florida between January 8 th and March 12 th 2003 (Fig. 1). The study area was selected using side-scan, seismic, and multibeam data, as well as sediment cores. Two main sites were selected roughly 20 kilometers west of Indian Rocks Beach: a fine sand site and a coarse sand site, both located in water depth ~ 13 meters relative to mean low low water (MLLW) (Fig. 2). Four acoustic and six optical instrumented and inert mine-like cylinders were deployed in early January. In order to monitor current and wave interactions with the mines and the seafloor, and their subsequent effect, three instrumented quadpods and five tripods (spiders) were deployed in the vicinity of the mines. Each quadpod was fitted with a 1.5 MHz pulse coherent boundary layer profiler (SonTek PC-ADP), a 5 MHz acoustic Doppler point current meter (SonTek Hydra), an in situ grain size sensor (LISST-100), a conductivity/temperature sensor (SeaBird Microcat C-T), and an optical backscatter sensor (Downing OBS). Each spider was equipped with a 1.5 MHz bottom mounted acoustic Doppler profiler with wave directional capabilities (SonTek ADP). All multibeam data were collected aboard the R/V Suncoaster on six cruises throughout the experiment: January 8 th 11 th when the mines were deployed; January 12 th 13 th ; January 16 th 17 th when the quadpods and spiders were deployed; January 19 th 20 th ; February 5 th 6 th ; and March 12 th 13 th when all deployed equipment was retrieved. During each cruise, multiple passes with the multibeam system were conducted over the mines. Once the multibeam data were post-processed, direct measurement of mine scour and burial was performed. We focus specifically on the acoustic instrumented mine number 5

PAGE 21

Figure 1. Location of the experiment site off Indian Rocks Beach, Florida. All depths are referenced to mean low low water (MLLW). The area outlined with a thin black line represents the data obtained in April 2002 during the site survey. The black dot in the middle of the black square and the red dot in the middle of the red square indicate the location of the fine sand site and coarse sand site respectively. Detailed images of these areas can be seen in Figure 2. Additional digital data provided by the USGS (Gelfenbaum and Guy, 2000) and NOAA (courtesy of D. Scharff, 2004). Figure modified from Naar and Donahue (2002). 6

PAGE 22

Figure 2. Location of the deployed equipment for the fine sand (2A) and coarse sand (2B) study sites. The fine sand study site also included an inert bomb, which has yet to be located in the multibeam data. 7

PAGE 23

3 (A3) which was located in the fine sand site, and on the optical instrumented mine number 8 (F8) located in the coarse sand site. Multibeam Data The use of multibeam sonars as tools for both bathymetric mapping and backscatter imaging is well-established (Pohner, 1990; Clarke, 1998; Collins and Preston, 2002; Collins and Galloway, 1998; Gardner et al., 1998; and references therein). For our experiment, we used a Kongsberg Simrad EM 3000, a 300 kHz multibeam swath sonar with 127 overlapping 1.5 x 1.5 beams, producing a 130-degree swath transverse to ship heading. Vertical uncertainty (RMS error) of the EM 3000 is 5 to 10 centimeters depending on depth. Given that the sonar is usually mounted to a ship, its positioning accuracy is greater than that of towed side-scan sonars and ROV mounted devices. Therefore, multiple passes over the same stationary object should result in the same georeferenced position. Our tests suggest less than 1-meter accuracy in position of seafloor objects in multibeam compared with 10 meters for side-scan data (Locker et al., 2002). The high frequency of the multibeam soundings allows it to operate at faster boat speeds than side-scan sonars, which due to the towfish hydrodynamics have a wider swath and a slower ping rate. Mine Burial and Scour Models One of the main goals of the ONR Mine Burial Prediction Program is the development of accurate models to estimate the percent scour and burial as a function of energy, geological conditions, and time. The models must have a known and acceptable 8

PAGE 24

degree of accuracy in areas of interest. Friedrichs (2001) conducted a review of five mine burial models, describing the main processes and discussing the validity of each model. Four of the five models (WISSP, NBURY, DRAMBUIE and Vortex Lattice) each model mine burial on the basis of scour. The fifth model, Mulhearn, models mine burial as a consequence of bedform migration. These models are only applicable in non-cohesive sediments and do not allow for a distribution of grain sizes. It is clear from the review of these models that a new two-dimensional mine scour and burial model was needed, which Friedrichs and Trembanis developed at the Virginia Institute of Marine Science, (Trembanis et al., 2005). This model has been used to forecast and hindcast mine burial for the Clearwater, FL and Marthas Vineyard, MA ONR mine burial experiments. Data from instrumented mines measured percent burial of some mines; however, to properly measure the scour development over time around all the mines and their subsequent burial required systematic repeat multibeam mapping over a larger area. Obtaining a Vertical Reference Frame Converting depths from pressure sensor data to a chart datum such as mean low low water (MLLW) is required to make temporal comparisons as well as model versus data comparisons. These pressure sensor depths do not take into account the height of the pressure sensor above the bed. In this study, a Sontek PC-ADP (with internal quartz pressure sensor) was used to measure the height of the pressure sensor above the seabed (Fig. 3). The data show three distinct shifts (near Julian day 19, 25, and 54). It is necessary to distinguish shifts caused by the quadpod (and subsequently the pressure sensor) settling into the seabed versus changes in seabed elevation due to erosion, accretion, or bedform 9

PAGE 25

10 Figure 3. Beam distances to the bed for the fine sand site. Distances were obtained from a SonTek PC-ADP on quadpod 1 in order to measure height of the integrated pressure sensor off the bed. Profile shown is an average of usable data from the sensors on the PC-ADP. Note the three data shifts near day 19, 25, and 54.

PAGE 26

migration while the quadpod remains stationary. Figure 4 shows a simplified cartoon schematic of the Sontek data through time along the horizontal axis and depth on the vertical axis. D T represents the total depth, which is equal to the depth of the sensor below MLLW (D S ) plus the height of the sensor above the bed (H S ) plus the local tides. If the apparent change in water depth is simply a function of sensor settling, then (D S + H S ) will remain constant, as illustrated by the first and second cases. If the apparent change is due to erosion or accretion of sediment, e.g., between the second and third cases, or the third and fourth cases, then (D S + H S ) will not remain constant. In order to determine which the case is, the tide component must be isolated and subtracted from the pressure sensor depths. Hourly tide data referenced to MLLW were obtained from NOAA station 8726724 in Clearwater, located at the seaward end of Big Pier 60, approximately 21 kilometers east of quadpod 1. The water depths obtained from the pressure sensor were shifted to overlay the NOAA tide heights by subtracting mean levels (de-meaned), and the two were directly compared. Figure 5 shows the NOAA tides minus the de-meaned pressure sensor data in the top diagram, and beam distances to the bed in the bottom diagram. If the two tides match then the difference between the tides should be zero. The cyclic pattern of the line indicates the difference in amplitude and phase between the two locations, changes due to seafloor elevation, as well as any noise in the pressure sensor data. The two solid arrows on the top diagram represent significant data shifts in one of the locations. We make the reasonable assumption that the NOAA station did not change height because there are no tears in the NOAA tide record. Thus, we can be confident that these two shifts occurred at the quadpod location. The open arrow on the bottom diagram indicates a significant shift in sensor height from the seabed but does not show up in the tide record. This means that 11

PAGE 27

Figure 4. Schematic representation of the SonTek data. Time during the experiment is along the horizontal axis and depth is along the vertical axis. 12

PAGE 28

Figure 5. De-meaned pressure sensor data. Top graph shows the NOAA hourly tides minus the de-meaned Pressure sensor 1 data (equivalent of Dt shown in Figure 4 with the mean subtracted). The bottom graph shows the data from one of the SonTek PC-ADP measuring distances to the bed. The two solid arrows represent data shifts visible in both graphs. The open arrow represents a data shift in the bottom graph that is absent in the top. 13

PAGE 29

(D S + H S ) remained constant and indicates the quadpod sank into the sediments. The two shifts denoted by a solid arrow, however, show up on both the graph of the beam distances to the bed and the NOAA tides minus the pressure sensor tides. This means that both H S and D S changed and the sum (D S + H S ) did not remain constant, indicating a change in seafloor elevation. In order to extract the changes in seafloor elevation from the data, the time series of the NOAA tides minus the pressure sensor data needs to be filtered. A lowpass Butterworth filter was applied to the data using a 36 hr period. The low frequency signal obtained represents changes in seafloor elevation and can subsequently be removed from the de-meaned pressure sensor data, leaving only the tidal component (Fig. 6). The phase lag between the pressure sensor tide record and the NOAA tide record was calculated to be approximately 4 minutes (see Appendix B for a description of the equations used). The amplitude of the pressure sensor tide record is off by a factor of 1.06 when compared to the NOAA record, corresponding to a maximum offset of 4.5 centimeters. When the seafloor elevation under quadpod 1 is plotted, there are two significant shifts punctuated by smaller changes (Fig. 7). The inflection point of the first shift in seafloor elevation lines up with the first shift in our initial tide record, peak significant wave height, and peak wind speed. Maximum erosion, however, does not occur until 16 hours later. At the second shift, the tide shift and wind speed peak line up with the inflection point of the seabed elevation change; however, the significant wave height does not peak until 18 hours later at the time of maximum accretion. The reason for this discrepancy is not clear. 14

PAGE 30

Figure 6. Filtered pressure sensor data. The top graph shows the NOAA tides minus the de-meaned pressure sensor data with the 36 hr filter. The bottom graph shows the NOAA tide versus the tide extrapolated from the pressure sensor data. 15

PAGE 31

Figure 7. Seafloor elevation change. The top graph shows the NOAA tides minus the de-meaned pressure sensor data with the 36 hr Butterworth filter. The second graph shows significant wave height obtained from the ADCP, and the third shows wind speed in meters per second obtained from the NOAA buoy. The bottom graph shows calculated seafloor depth based off the pressure sensor data and the isolated tide record. The dashed lines cut through the inflection points of the two significant changes in seafloor elevation, while the dotted line cuts through the maximum change. 16

PAGE 32

Temporal Analysis of Mine Burial The rate at which mines subside relative to the ambient seafloor and become buried is extremely important to mine countermeasures. Current methods of mine hunting involve the use of side-scan sonars, which rely on shadow casting for detection. Once scour has formed around a mine and it subsides below ambient seafloor depth, it becomes more difficult for an acoustic shadow to form, thereby making detection with side-scan difficult if not impossible. Multibeam sonars do not have a nadir blind zone, are not towed deeply, and thus are more able to image the object as they pass directly over it. This has made it possible to image mines in different stages of scour and burial and observe the temporal scales of such processes, until they are fully covered by sediments. Six multibeam surveys of the fine study site were used in the analysis of the A3 mine: January 10 th 13 th 17 th and 20 th February 6 th and March 13 th 2003. The same surveys were used in the analysis of the F8 mine, with the exception of the January 10 th survey since the mine was not deployed until January 11 th Each individual pass of the multibeam over the mines can be used to estimate the amount of scour and burial at that time. These passes were then used to monitor discrete changes in scour and burial during the experiment. All multibeam data were cleaned and processed using CARIS HIPS and SIPS 5.3. All speed jumps greater than 1 knot and all time jumps greater than 1 second between consecutive pings were removed using a linear interpolation. Once the data were cleaned, a tide correction was applied. The multibeam data from surveys before the quadpods were deployed on January 16 th were tide-corrected with data from NOAA station 8726724. Two tide records were obtained from pressure sensors mounted on quadpods deployed near the 17

PAGE 33

mines using the previously discussed method, and used to tide-correct survey data subsequent to the 16 th The two tide records, one from quadpod 1 deployed near the A3 mine in the fine sand site and one from quadpod 3 deployed near the F8 mine in the coarse sand site, were found to be nearly identical. A multiplier of 0.94 was applied to the NOAA tide record to account for the difference in amplitude between the NOAA tide and the tide records obtained from the pressure sensors. After applying the tide correction, the multibeam data was gridded in CARIS using a weighted mean gridding algorithm. The weight that any given sounding contributes to the grid varies with range and grazing angle to the seabed. The range weight is inversely proportional to the distance from the grid node (i.e., the closer to the node, the greater the weight). The grazing angle weight is most important in grids containing adjacent or overlapping track lines. Higher weight is given to beam from the inner part of a swath. Beams with a grazing angle between 75 and 90 degrees are given a weight of 1.0. This weight linearly decreases to 0.01 as the grazing angle with the seabed decreases to 15 degrees. For each survey, 18-by-18 meter grids centered on the mines were created; gridded at a 20-centimeter horizontal resolution and referenced to MLLW. In some instances, the 20-centimeter grid resolution was too small to provide full coverage in areas of sparse data (e.g., the outer beam of the swath). In these cases, the grids were interpolated in order to fill these data gaps. Interpolation was based on a 3 x 3 grid node area with a threshold level of 6 neighbors. For example, if a node in the grid does not contain a value, the interpolation is limited to the neighboring 9 nodes. In order for the interpolation to take place, a minimum of 6 of these neighboring nodes must contain a pixel value. This helps limit the amount of 18

PAGE 34

interpolation and prevents it from expanding the gridded surface outward from the actual survey area. Final imaging, including 3D rendering and artificial sun illumination, was completed using IVS Fledermaus 6.0. Depth of the mine was defined as the shallowest point on the mine surface. Ambient seafloor depth was defined as an average of 35 depths taken around the mine outside the influence of any scour (see Appendix A for a more detailed analysis on ambient seafloor depth). Given that our study site is in shallow water (average depth ~ 13 meters) and we use a POS MV system with RTK for vessel positioning, we assume a vertical uncertainty of 5 centimeters. This decision was also made in an effort to avoid masking our signal with uncertainty; however, we realize that 5 centimeters may be optimistic and the actual uncertainty may be closer to 10 centimeters. Temporal Changes in Scour and Burial over the A3 Mine The A3 mine was situated over fine sand (median grain size .180 mm) at a water depth of 12.81 meters, and was closely surrounded by two quadpods and one spider (Fig. 8). The January 10 th survey was the first to image the A3 mine after its January 8 th deployment. The grid shows only the A3 mine, as the quadpods and spider were not deployed until January 16 th (Fig. 9). In this image, as in all subsequent images, artificial sun illumination is from the northeast (045) at an angle of 45 degrees above horizontal. The mine has only been deployed for approximately two days, and no scour is visible. The depth to the top of the mine is 12.32 meters, with the average depth of the seafloor around the mine at 12.81 meters. The difference, 0.49 meters, indicates that the mine is approximately 8% buried after two days. The beam mode of the multibeam during this 19

PAGE 35

Figure 8. Dimensions of the quadpods, spiders, and mine-like cylinders visible in the multibeam images. 20

PAGE 36

Figure 9. January 10th survey over the A3 mine. No scour is visible. The black dashed line indicates the ships track line during the survey. The two white dashed boxes and the white dashed circle represent the future deployed locations of the two quadpods and one spider respectively. The mine itself is outlined with a faint white line. All outlines are scaled to the actual dimensions of the deployed equipment. The mine outline remains at the same scale and orientation throughout the rest of the images as a reference. 21

PAGE 37

survey was set on target detection. It was discovered that this mode causes a widening of the beams (from 1.5 to 4.0) in order to improve target detection capabilities, but it unfortunately blurs the mine and its orientation. Comparison with data from the heading sensor in the A3 mine itself indicates that the orientation should be north south (-5.7), rather than the northeast southwest orientation apparent in the image. Orientation of the mine is in relation to magnetic north (declination: 019). The January 13 th survey is similar to that of January 10 th and there is no apparent scour around the mine (Fig. 10). The depth to the top of the mine is 12.42 meters, indicating a sinking of 0.10 meters since January 10 th The average depth of the seafloor around the mine is 12.88 meters, indicating a 13% burial of the mine. Again, the beam mode on the multibeam was set on target detection, explaining blurriness of the mine itself and the distortion of its orientation. The survey of January 17 th occurred just one day after the spiders and quadpods were deployed. The mine, quadpod 2, and the spider are all clearly visible, yet quadpod 1 does not show up (Fig. 11). It is unclear why the quadpod is not visible, though it is possible a bubble sweep occurred. A spike filter was also set on the multibeam at the time of this survey, though this filter is an unlikely cause of the quadpods disappearance since the other quadpod shows up. There is still no visible scour at this time, although the average depth of the seafloor around the mine is 12.92 meters. The depth to the top of the mine is 12.48 meters, indicating the mine has now sunk 0.16 meters for a total burial of 17%. Target detection was not used during this or any subsequent survey, therefore the mine is less fuzzy and its orientation agrees with the orientation data from the mine itself. 22

PAGE 38

Figure 10. January 13th survey over the A3 mine. No scour is visible. The black dashed line indicates the ships track line during the survey. 23

PAGE 39

Figure 11. January 17th survey over the A3 mine. The quadpods and spiders have now been deployed and can be seen in the image, with the exception of quadpod 1. The red dashed box with the question mark denotes where quadpod 1 should be located. 24

PAGE 40

In the January 20 th survey, scour around the mine becomes evident (Fig. 12). It is also clear that the mine has sunk even further. The spider is not visible, although the scour pit that formed around the spider is. The cause of the spider not being detected is also unknown. The depth to the top of the mine is 12.62 meters, 0.30 meters deeper than that observed in the January 10 th survey. The average depth of the seafloor around the mine is 12.82 meters, and the mine is now 62% buried. A scour pit has formed around the mine, with the deepest point measuring 13.04 meters. The spider is visible in the February 6 th image, and the scour has continued to develop around both the mine and the spider (Fig. 13). The depth to the top of the mine is 12.72 meters, indicating that the mine has sunk 0.40 meters since the initial survey on January 8 th The average depth of the seafloor around the mine is 12.81 meters, and burial of the mine is now up to 83%. The depth in the scour pit around the mine has increased to 13.18 meters. The March 13 th survey shows that the mine has become nearly flush with the ambient seafloor depth (Figs. 14 & 15). The mine is only visible due to the defining ring of scour around its periphery. The depth to the top of the mine is now 12.80 meters, indicating the mine has sunk a total of 0.48 meters since the start of observations. The average depth of the seafloor around the mine is 12.82 meters, and the mine is 96% buried. The spider has also scoured considerably and has sunk into the seafloor. Scour is also visible around the legs of both quadpods, though any sinking of the quadpods appears to be minimal, according to pressure sensor data on the quadpod and multibeam bathymetry data. Overall, the total amount of scour over the course of the experiment formed a pit around the mine 0.40 meters deeper than the ambient seafloor and the mine sank 0.48 25

PAGE 41

Figure 12. January 20th survey over the A3 mine. Note that both quadpods are now visible, although the spider is not. The red dashed circle with the question mark denotes where the spider should be located. A ring of scour can clearly be seen around the mine. 26

PAGE 42

Figure 13. February 6th survey over the mine. The spider is once again visible. Scour pits can clearly be seen around the spider and the mine. 27

PAGE 43

Figure 14. March 13th survey over the mine. Both quadpods and spiders are readily visible. The sharp ring of scour around the spider and the mine help set them off from the rest of the seafloor. Note the scour around the legs of the quadpods. 28

PAGE 44

Figure 15. ROV video still image of the A3 mine on March 13, 2003. Camera is facing east-northeast showing a side view of the mine within the scour pit. 29

PAGE 45

meters between January 10 th and March 13 th (Table 1; Fig. 16). The diameter of the mine is 0.53 meters, so a sinking of 0.48 meters would result in a 91% burial. Slight changes in the ambient seabed elevation over the course of the experiment, however, have resulted in the maximum amount of burial as observed in the multibeam images to be 96%. Comparison of A3 Multibeam Observations to the VIMS 2D Burial Model The model predicts percent burial of the mine given sediment size, bed stress, and mine diameter. NOAA WaveWatch3 monthly hindcast wave data were used to drive the model for wave-induced scour, by using linear wave theory to estimate near bed wave orbital velocity. The percent burial was then predicted by comparing the depth of the scour to the diameter of the mine. The percent burial as observed in the multibeam images was directly compared to the model predictions (Table 1). The model was initialized with a local water depth of 12.81 meters (obtained from the January 10 th survey over the mine) and 0% burial. There is no multibeam survey over the A3 mine on the day of deployment; however, SCUBA divers repositioned the mine shortly after deployment to ensure no impact burial. This makes certain that the model and the observed data are initialized with the same conditions. The model was run from the time of mine reposition, January 8 th 2003 1600 GMT, to the time of the last multibeam survey over the mine, March 13 th 2003 at 0200 GMT. The first direct comparison between the observed and predicted burial occurs for the January 10 th survey (Figs. 16 & 17). Observed data show the mine to be 7.5% buried; however, the model predicts a burial of 14.9%. This difference of 7.4% is the largest discrepancy between the predicted and observed data throughout the experiment. The 30

PAGE 46

Jan. 8 Jan. 10 Jan. 13 Jan. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 12.32 12.42 12.48 12.62 12.72 12.80 Cumulative Amount of Change _____ _____ 0.10 0.16 0.30 0.40 0.48 Average Depth of Seafloor _____ 12.81 12.88 12.92 12.82 12.81 12.82 Cumulative Amount of Change _____ _____ 0.07 0.11 0.01 0.00 0.01 Scour Visible / Depth of Scour _____ no no no yes 13.04 yes 13.18 yes 13.22 % Mine Burial from Multibeam ( 9.4% due to 5 cm uncertainty of sonar) 0 7.5 13.2 17.0 62.3 83.0 96.2 % Mine Burial from Model 0 14.9 17.8 17.8 60.1 81.2 97.7 Mine Heading (degrees) -5.7 -6.9 -6.6 -6.6 -1.0 3.37 7.1 Mine Pitch (degrees) -0.7 -0.3 -0.4 -0.4 -0.8 -0.9 0.6 Mine Roll (degrees) -0.4 7.8 7.4 7.4 14.1 25.6 33.1 Table 1. Data table for the A3 mine. All numbers are in meters except where noted. There is no multibeam survey on January 8 th 31

PAGE 47

Figure 16. Comparison of multibeam observed (black) and predicted (gray) mine depth for the A3 mine over the course of the experiment. Predicted percent burial of the mine was converted to predicted depth of the mine using the 12.81-meter water depth used to initialize the model. Observed depth of seafloor and depth of scour from the multibeam are plotted as well. Significant wave height is plotted on the right y-axis. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. The light gray oval represents the A3 mine, and is scaled to the actual dimensions of the mine (length ~ 0.53 m, the diameter of the mine). 32

PAGE 48

Figure 17. Comparison of the mine (magenta), predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the A3 mine over the course of the experiment. Tilt-corrected values have been horizontally offset for clarity. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. 33

PAGE 49

m ultibeam sonar has an inherent uncertainty of 5 centimeters in its vertical accuracy, eter, data lution. ed certainty, even w which corresponds to a percent burial of 9.4. Therefore, the model prediction falls withinthe range of multibeam values. The observed percent burial in the multibeam data is basedoff the shallowest point on the top surface of the mine, which could underestimate true burial of the mine due to pitch (tilting of the long axis of the cylinder). The predicted percent burial is based on the depth of the predicted scour in relation to the mine diamand therefore assumes a direct sinking of the mine with no concern for pitch. Sensors within the A3 mine measured roll, pitch, and heading throughout the experiment. The degree of pitch can be used to calculate how much deeper the center point on the top surface of the mine is from the shallowest point observed in the multibeam images. Acorrection can then be applied to the observed values for percent burial. Reviewing thefrom the pitch sensor (Table 1) reveals a -0.3 pitch during the time of this survey; however, this only corresponds to ~ 6 millimeters and is beyond the multibeam reso The January 13th comparison shows a discrepancy of 4.6%, with an observed burial of 13.2% compared to a predicted burial of 17.8%.This falls within the accuracy of the multibeam (Figs. 16 & 17). The mine shows a tilt of -0.4 at the time of this survey, which corresponds to a change of ~ 7 millimeters. The differences between the predicted and observed data begin to narrow in margin around the January 17th survey, with an observburial of 17.0% and a predicted burial of 17.8%, a discrepancy of only 0.8%. Applying the tilt correction of 0.4 only alters the amount of burial by ~ 7 millimeters as well. The January 20th and February 6th predictions fall within the multibeam un ithout the tilt co rr ection (Fig. 16 & 17). The January 20th survey has an observed 34

PAGE 50

a nd predicted burial of 62.3% and 60.1% respectively, a discrepancy of 2.2%. The eter of 3th, 2003 shows a small .7% .6, Discussion of the A3 Comparisons The VIMS 2D Buriah-resolution multibeam ited beam measured degree of tilt during this survey is -0.8, which corresponds to ~ 1 centimburial and increases the discrepancy to 4.1%. The February 6th survey shows an observed burial of 83.0% and has a predicted burial of 81.2%, a discrep a ncy of 1.8%. The measuredtilt during the survey of February 6th is -0.9, which also adds a centimeters worth of burialand would increase the discrepancy to 3.7%. The comparison from the final survey on March 1 discrepancy of 1.5% with an observed burial of 96.2% and a predicted burial of 97(Figs. 16 & 17). This discrepancy decreases to a mere 0.4% when the tilt correction of 0corresponding to a change of 1 centimeter, is applied. The predicted value falls within the accuracy of the multibeam, even without the tilt correction. l Model is compared to six repeat hig surveys over the A3 mine. The mine subsides and becomes partially buried throughout theexperiment, but surrounding scour is not observed until the January 20th survey, twelve days after deployment. Direct comparison between these observations a nd the VIMS 2DBurial Model shows a good agreement (Fig. 17). The model was initialized with a 0% burial. Impact burial for the observed data was assumed to be 0% as well, based on limSCUBA observations, thus allowing the initial conditions to be the same between the predicted and observed data. The overall trend throughout the experiment shows the modeled predictions are consistently within the measurement uncertainty of the multidata. A tilt correction can be applied to the observed values of mine burial in order to 35

PAGE 51

calculate a direct sinking of the mine. This correction, however, only adds a centimeteburial at most, corresponding to an increase 1.9% in observed burial and was not necessaryIn field applications, tilt of cylindrical mines will not be available, but fortunately, their potential effect on true depth of burial is minimal. r of Temporal Changes in Scour and Burial over the F8 Mine ThRSD) over coarse-nd (m ts s tting on a The depth to the top of the mine is 12.80 meters, indicating a sinking of 0.08 meters since e F8 mine was situated within a rippled scour depression ( saedian grain size 0.840 mm) at a water depth of 13.20 meters. The January13th survey was the first to image the F8 mine after its January 11th deployment (Fig. 18). The mine has only been in the environment for approximately two days, and no scour is visible. The observed depth to the top of the mine is 12.72 meters, with the average depthof the ambient seafloor at 13.20 meters, resulting in an observed burial of zero percent. Thedifference of 0.48 meters between the top of the mine and the seafloor is actually one centimeter greater than the diameter of the mine itself (0.47 meters). Pitch measuremenrecorded from orientation sensors within the mine show a zero degree tilt at the time of thisurvey, however, the discrepancy of 1-centimeter falls within the 5-centimeter measurement uncertainty of the multibeam. It is also possible that the mine is simound of sand slightly shallower than the surrounding seabed. It is important to keep in mind that the ambient seafloor depth is also an approximate regional estimate. A northsouth trending ripple field can be seen in the lower left of the image. Maximum height ofthe ripples is ~ 20 centimeters with a maximum wavelength of a ~ 1.25 meters. The ripplefield is no longer apparent in the multibeam image from the January 17th survey (Fig. 19). 36

PAGE 52

37 Figure 18. January 13th survey over the F8 mine. No scour is visible. The thick black dashed line indicates the shduring the survey. The mine itself is outlined with a faint black dashed line scaled to the actual dimensions of the F8 mioutline remains at the same scale and orientation throughout the rest of the images as a reference. ips trackne. mi line Thene

PAGE 53

Figure 19. January 17th survey over the F8 mine. No scour is visible. 38

PAGE 54

Ja nuary 13th. The average depth of the seafloor around the mine is 13.17 meters, resulting an observed burial of 21.3%. There is no scour visible around the mine at the time of this ll defined as in the previous survey (Fig. 21). The apparent wavelength e ting dicated re in survey; however, it does become evident in the survey of January 20th (Fig. 20). Maximumdepth in the scour pit that has formed at the southwest end of the mine is 13.27 meters. In the lower left of the image, ripples are again visible, although they are smaller than those observed in the January 13th survey. Ripple height is on the order of 10 centimeters and thewavelength is approximately 50 centimeters. The mine has sunk 6 centimeters more to a depth of 12.86 meters. Depth of the ambient seafloor is 13.15 meters, resulting in a percent burial of 38.3%. The ripples are still visible in the survey of February 6th; however, they do not appear to be as we of the ripples has increased to ~ 75 centimeters, though ripple height has appeared to remain the same. The scour pit at the southwest end of the mine has grown deeper, with a maximum depth of 13.32 meters. Depth to the top of the mine is 12.84 meters, 2 centimeters shallower than in the previous survey. The degree of tilt has not changed between this survey and the last; however, the 2-centimeter difference is within thuncertainty of the multibeam. Depth of the surrounding seafloor is 13.11 meters, resulin a percent burial of 42.6%. The mine appears to the south of its original position inby the black dashed oval in the center of the image. Data from the orientation sensors within the mine do not indicate that the mine has rolled into its new position, as the roll has only changed by one degree since the last survey. The orientation sensors do not measucumulative roll, however, so if the mine makes a full rotation the sensor will record no change. The maximum offset between the mines current and original position is ~ 1.5 39

PAGE 55

Figure 20. January 20th survey over the F8 mine. Scour is visible at the southwest en d of the mine. 40

PAGE 56

Figure 21. February 6th survey over the F8 mine. The scour pit has developed at the southwest end of the mine. The mine now appears slightly south of its original position as indicated by the faint black dashed oval. 41

PAGE 57

m eters; a complete roll of the mine would account for 1.4 meters. Roll measurements ld tion t survey again shows the mine to the south of its original position by the mine y rtical Other likely scenarios for the 12-centimeter discrepancy include error in the sound velocity within the mine were made every 15 minutes, so if the roll were rapid the sensors wounot record it. A storm event moved through the area causing elevated wave heights on January 24th 2003. It is therefore possible that the mine made a rapid and complete rotaat this time The ability to measure cumulative roll is recommended for future inert mine development in order to record true roll. The multibeam system has a horizontal accuracy of 1-meter; however, there are no other offsets observed for the other mines during the same survey, suggesting that a 1-meter offset is likely to be a true southward displacemenby some mechanism. The March 13th 1.6 meters. This offset is within 10 centimeters of the offset observed in the February 6th image, and strongly supports the notion that this change is unlikely the source of system error and most likely a result of actual change. The mine is nearly flush with the surrounding ripples (Figs. 22 & 23). The ripples appear very well defined, with a wavelength of ~ 1.2 meters and a height of 12 centimeters. The depth to the top ofis 12.72 meters with a surrounding seafloor depth of 13.00 meters, resulting in an observed burial of 40.4%. The data seem to suggest an anomalous shallowing of the mine and ambient seafloor depth by 12 centimeters that we do not understand and cannot readilexplain. The degree of tilt of the mine has decreased since the February 6th survey, indicating that tilt can not be used to explain part of this anomaly. The co m bined veuncertainty of the multibeam system for both the February 6th and March 13th surveys canaccount for 10 centimeters of this discrepancy; the remaining 2 centimeters is negligible. 42

PAGE 58

43 Figure 22. March 13th survey over the F8 mine. Scour is visible at both ends of the mine. There is a clear ritrending north-northwest/south-southeast. As in the last image, the mine appears south of its original po pple sition. field

PAGE 59

Figure 23. ROV video still image of the F8 mine on March 13, 2003. Camera is facing south-southeast showing a side view of the mine in the ripple field. 44

PAGE 60

p rofile used by the multibeam system to calculate depth during the survey. The average und velocity during this survey is 1520.90 meters/second; therefore, it would only take at (Table 2; Fig. 24). An anomalous allowth d by Comparison of F8 Multibeam Observations to the VIMS 2D Burial Model The model was initialized with a local water depth of 13.20 meters (obtained from the Januns. so an error of 14.02 meters/second to account for 12 centimeters. It is also possible that an error exists in the tide record used to correct the data during processing, or that there is a greater vertical uncertainty in the multibeam system. Of these possibilities, we suspect thchanges in the sound velocity profile is the most likely reason for error, because the tide record and multibeam system have worked quite well elsewhere, and it is common to have changes in sound velocity in coastal settings. Between the January 13th survey and the survey of February 6th, the mine sank a total of 12 centimeters and became 42.6% buried shing during the March 13 t h survey resulted in the mine having the same depth at bothe beginning and end of the experiment. The average depth of the seafloor decrease0.20 meters over the course of the experiment indicating localized deposition in the area. As a result of this deposition, there was an observed burial for the March 13th survey of 40.4%. ary 13th survey) and a 0% burial for comparison with the F8 mine observatio SCUBA divers sent down shortly after deployment checked the status of the mine, but theydid not reposition it. The model start time was set at January 11th, 2003 at 2300 GMT, thetime of mine deployment, and was run until March 13th, 2003 at 1000 GMT, the time of the last multibeam survey over the mine (Table 2). 45

PAGE 61

Jan. 11 Jan. 13 Ja n. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 12.72 12.80 12.86 12.84 12.72 Cumulative Amount of Change _____ _____ 0.08 0.14 0.12 0.00 verage Depth of Seafloor umulative Change Scour Visible / _____ no no yes 13.27 yes 13.32 yes 13.11 Depth of Scour % Mine ultibeam ( Burial from M 10.6% due to 5 cm uncertainty 0 0 21.3 38.3 42.6 40.4 of sonar) MineBurial from Model (degrees) ata table fo January 11 8 mi mber eters ex her Th mu A_____ 13.20 13.17 13.15 13.11 13.00 C Amount of _____ ______ -0.03 -0.05 -0.09 -0.20 % 0 0 0 47.4 75.5 92.5 Mine Pitch 0 0 0 -2 -2 -1 Mine Roll (degrees) -12 -11 -8 -1 0 0 Table 2. Dr the Fne. All nus are in mcept we noted. ere is noltibeam survey onth. 46

PAGE 62

Figure 16. Comparison of multibeam observed (black) and predicted (gray) mine depth for the F8 mine over the course of the experimenpercent burial of the mine was converted to predicted depth of the mine using the 13.20-meter water depth used to initialize the model. Obof seafloor and depth of scour from the multibeam are plotted as well. Significant wave height is plotted on the right y-axis. Error bars recentimeter uncertainty inherent in the multibeam system. The light gray oval represents the F8 mine, and is scaled to the actual dimensio(length ~ 0.47 m, the diameter of the mine). t. Predicted served depth present the 5-ns of the mine 47

PAGE 63

The first comparison between the model and the observed data occurs on January s. 24 & 25). Both the model and the observed data show a 0% burial of the mine. There is no tilt of the mine at the time of this survey, indicating that the mine is sitting completely flat on the seafloor. The model continues to predict a 0% burial for the January 17th comparison, resulting in a discrepancy of 21.3% with the observed burial. The degree of tilt is still zero at this time, so no correction factor can be applied to the observed values. The 5-centimeter vertical uncertainty of the multibeam equates to 10.6% burial of the F8 mine; however, this still leaves a discrepancy of 10.7%. The January 20th comparison shows an observed burial of 38.3% versus a predicted burial of 47.4% (Figs. 24 & 25). The discrepancy of 9.1% falls within the measurement uncertainty of the multibeam. The mine has a -2 tilt at the time of this survey; applying a correction factor to the observed value of mine burial adds 3 centimeters of burial, resulting in a 43.9% total burial. This reduces the discrepancy between the predicted and observed values to 3.5%. The model predicts a 75.5% burial of the mine for February 6th, but the observed value is only 42.6%, an offset of 32.9% (Figs. 24 & 25). The mine continues to have a 2 tilt at this time; however, this can only account for 3.5% of the difference. The greatest discrepancy between the model and the observations occurs during the March 13th comparison. The observed data show a 40.4% burial of the mine compared with a predicted value of 92.5%, resulting in a 52.1% offset. There is -1 tilt of the mine at this time, which would add 1 centimeter of burial and decrease the offset to 50%. The 5centimeter vertical uncertainty of the multibeam system can further reduce this discrepancy by another 10.4%. 13th (Fig 48

PAGE 64

Figure 25. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observeburial for the F8 mine over the course of the experiment. Tilt-corrected values have been horizontally offset clarity. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. d (red) pe rcent for 49

PAGE 65

T he anomalous 12-centimeter shallowing of the mine apparent at the time of this survey ould account for another 25.5%. Combining these corrections still leaves a discrepancy of The VIMS 2D Burial Model is compared with five repeat high-resolution multibeam observations ove experiment, the mine ecomeIMS 2D he me Conclusions High-resolution multibeam bathymetry data are useful to successfully document burial of inert mines over time at both a a coarse sand site off Clearwater, nd y w 14.5% between the predicted and observed values of mine burial. Discussion of the F8 Comparisons r the F8 mine. Over the course of the bs ~ 40.4% buried. Direct comparison between these observations and the VBurial Model indicates the model does not work well in areas of coarse sand (Fig. 25). Ttrend throughout the experiment shows the modeled predictions are consistently higher than the actual observed values. Applying a tilt correction and taking the uncertainty of the multibeam into consideration cannot account for the discrepancies. This indicates that soother factor must be affecting mine burial that is not accounted for in the model, such as the presence of rippled bedforms near the mine. This is issue is currently being addressed by the modelers (Trembanis et al., 2005). fine sand and Florida. While the amount of observed burial by subsidence was significant in the fine sasite (96.2 %), the mine remained uncovered by sediment. This has been shown to actuallincrease the likelihood of detection using side-scan sonars as a result of the larger scour pit that forms around the mine. The VIMS 2D burial model was compared with in situ 50

PAGE 66

multibeam observations of mine burial at both sites. The model works well in the fine-sand case, staying consistently within the measurement uncertainty of the multibeam systdoes not work so well in the coarse sand analysis, where initial comparisons are good but quickly diverge throughout the rest of the experiment. Possible sources of error are that themodel uses one water depth. This assumes the local water depth does not change over the course of the experiment; however, localized erosion and accretion has been observed at both study sites. The presence of rippled bedforms at the coarse sand site is also observed during the experiment. These ripples directly affect morphodynamics of the seafloor and thus can affect rates of mine burial. Currently, the addition of a bedform correction to the model is being explored by the modelers. em. It 51

PAGE 67

Chapter 3 Multibeam Observations and Model Comparison of the Remaining Mines Introduction The following mine analyses were completed using the same methodology as described in the previous chapter. There were five remaining mines in the shallow fine sand site, one in the coarse sand site, and two mines in a deep fine sand site (~ 14 meters relative to MLLW) (Figs. 26 & 27). As described in the previous chapter, all multibeam data were cleaned and processed using CARIS HIPS and SIPS 5.3 (see chapter 2 for a detailed description on). All images are 18-by-18 meter grids centered on the mine at a horizontal resolution of 0.20 meters and referenced to MLLW. Final imaging, including 3D rendering and artificial sun illumination, was completed using IVS Fledermaus 6.0. Artificial sun illumination is from the northeast (045) and at an angle of 45 degrees above horizontal. Since the analyses of the A3 and F8 mines showed that tilt correction made little difference in mine burial, and given that tilt of the mine will not be available during actual field applications, it was not included in the following analyses. The A1 Mine Temporal Analysis of Scour and Burial The acoustic instrumented mine 1 (A1) was deployed on January 8 th 2003 in the shallow fine sand site at a water depth of 12.77 meters, and was oriented north-south. The 52

PAGE 68

Figure 26. Location of the deep fine sand study site. Study site is represented by the green dot in the middle of the green square. All depths are referenced to MLLW. 53

PAGE 69

Figure 27. Location of deployed equipment in the deep fine sand site. 54

PAGE 70

survey on January 10 th was the first to image the mine after its deployment (Fig. 28). The depth of the mine is 12.20 meters with an ambient seafloor depth of 12.77 meters. This implies that the mine is just resting on the seafloor and scour has not yet begun to form. The mine appears to have a northwest-southeast orientation; however, data from the mine itself indicates more straight north-south orientation (-2.9). The multibeam system was set on target detection during this survey and may explain the discrepancy in orientation. The mine orientation appears more north-south in the January 13 th survey, although the mine itself is somewhat blurred (Fig. 29). Target detection was still on during this survey as well, which may account for this. The mine has sunk 0.19 meters, to a depth of 12.39 meters, since January 10 th Localized erosion around the mine has caused the ambient seafloor depth to drop to 12.91 meters, resulting in a 1.8% burial. There is no scour evident around the mine. The ambient seafloor depth stays relatively constant between the survey of January 13 th and that of January 17 th 12.90 meters and 12.91 meters respectively, and no scour is evident. The mine does not show up well in this survey, and appears as two separate bumps in the image (Fig. 30). It is possible that a bubble sweep occurred causing interference with the beams. The depth to the top of the mine is now 12.52 meters, 0.13 meters deeper than in the previous survey, resulting in an observed burial of 28.3%. During the time between the January 17 th and January 20 th surveys, two distinct scour pits have formed at the north and south ends of the mine, despite an overall localized deposition around the mine of 0.12 meters (Fig. 31). The maximum depth measured in the scour pits is 13.15 meters, and the ambient seafloor depth is now 12.82 55

PAGE 71

Figure 28. January 10th survey over the A1 mine. In all multibeam images, the black dashed line represents the ships track line during the survey. The mine is outlined with a faint white line scaled to the mines actual dimensions. The mine outline remains at the same orientation throughout the A1 multibeam images as a reference. 56

PAGE 72

Figure 29. January 13th survey over the A1 mine. 57

PAGE 73

Figure 30. January 17th survey over the A1 mine. 58

PAGE 74

Figure 31. January 20th survey over the A1 mine. 59

PAGE 75

m eters. A storm event passing through the area on January 17th caused increased wave eights (peaking at ~ 2.34 meters at 4 pm GMT) and can account for the rapid scour es f the mine and the depth of the 2 the n the pth of than in the h development. The mine is at a depth of 12.65 meters, a sinking of 0.13 meters since the 13th, and is 67.9% buried. The observed orientation of the mine is north-south and agre with the data from the sensor within the mine (-0.3). In the image from the February 6th survey, the mine is only visible due to the defining ring of scour (Fig. 32). The depth to the top o surrounding seafloor are 12.88 and 12.90 meters respectively, a difference of only 0.0meters resulting in a 96.2% burial. The greatest amount of scour occurs at the southernend of the mine, where the maximum depth within the pit reaches 13.33 meters. During the March 13th survey, the mine lays just within the inner beams of the multibeam swath (Fig.33). The wavy pattern to the west of the mine is caused by outer beams of the sonar hitt in g the seafloor at greater grazing angles. Data interpolatioin CARIS was used to patch data holes. The mine depth is 12.80 meters, 0.08 meters shallower than the February 6th survey. This discrepancy can be accounted for by the vertical uncertainty of the multibeam system for both the February 6th and March 13th surveys, which combines to 0.10 centimeters. Other possible explanations include possible changes in the sound velocity profile in the water column ve r sus that used b y multibeam system during data collection to calculate depth. For a more detailed discussion of these possibilities, please refer to the discussion of the temporal observations of scour and burial of the F8 mine in chapter 2 of this thesis. The dethe ambient seafloor during this survey is 12.81 meters, 0.09 meters shallower previous survey. It is not clear whether this difference is related to the shallowing of the 60

PAGE 76

Figure 32. February 6th survey over the A1 mine. 61

PAGE 77

Figure 33. March 13th survey over the A1 mine. 62

PAGE 78

m ine, or whether it represents actual localized deposition around the mine. The observed urial of the A1 mine based off a mine depth of 12.80 meters and an ambient seafloor llowing of the mine in the March 13th survey of e Observations to the VIMS 2D Burial Model The VIMS 2D burial model was initialized with a local water depth of 12.77 meters odel the ot m b depth of 12.81 meters is 98.1% (Fig. 34). In total, the mine sank 0.68 meters between the January 10th and February 6th surveys (Table 3; Fig. 35). An apparent sha reduced this total to 0.60 meters. Over the course of the experimen t the average dep t hthe seafloor surrounding the mine increased by a total of 0.04 meters. Scour became evident around the mine during the January 20th survey and developed into a pit 0.59 meters deeper than the ambient seabed by March 13th. The final observed burial for thA1 mine was 98.1%. Comparison of A1 Multibeam (obtained from the January 10th survey over the mine) and 0% burial. The m was run from the time of mine reposition, January 8th 2003 1600 GMT, to the time oflast multibeam survey over the mine, March 13th 2003 at 0300 GMT. The first direct comparison between the observed and predicted burial occurs for the January 10th survey (Figs. 35 & 36). The multibeam data indicate th e mine is resting on the seabed and is nburied at all. The model; however, predicts a burial of 15.3% at this time. The m u ltibeam sonar has an inherent uncertainty of 5 centimeters in its vertical accuracy, which corresponds to a percent burial of 9.4. The discrepancy is 15.3% between the model and the multibeam observations, so the model does not fall within the range of multibeavalues. 63

PAGE 79

Figure 34. ROV video still image of the A1 mine on March 13, 2003. 64

PAGE 80

Jan. 8* Jan. 10 Jan. 13 Jan. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 12.20 12.39 12.52 12.65 12.88 12.80 Cumulative ount of _____ _____ 0.19 0.32 0.45 0.68 0.60 Am Change Depth of Seafloor Cumulati ve f _____ _____ 0.14 0.13 0.05 0.13 0.04 Amount oChange Scour Visible / Depth of Scour _____ no no no yes yes yes 13.15 13.33 13.39 % MineBurial froMultibeam( 9.4% m due to 5 cm nty 0 0 1.8 28.3 67.9 96.2 98.1 uncertaiof sonar) % Mine Burial froModel Heading(degrees) (degre (degrees) Data table fn January 8 A1 m ll nu ers ex here d. The no mu Average _____ 12.77 12.91 12.90 12.82 12.90 12.81 m 0 15.3 18.1 18.1 60.5 81.6 98.0 Mine -2.9 -2.1 -1.9 -1.8 -0.3 0.7 1.1 Mine Pitch es) -0.2 -0.3 1.0 0.7 0.6 0.6 0.8 Mine Roll 4.8 1.3 -1.3 -1.3 -5.8 -23.4 -32.5 Table 3.or the ine. Ambers are in metcept w notere isltibeam survey oth. 65

PAGE 81

Figure 35. Comparison of multibeam observed (black) and predicted (gray) mine depth for the A1 mine over thexperiment. Predicted percent burial of the mine was converted to predicted depth of the mine using the 12.77-meter watto initialize the model. Observed depth of seafloor and depth of scour from the multibeam are plotted as well. Significant is plotted on the right y-axis. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. The ligrepresents the A1 mine, and is scaled to the actual dimensions of the mine (length ~ 0.53 m, the diameter of the mine). e course of the er depth used wave height ht gray oval 66

PAGE 82

Figure 36. Comparison of the mine (magenta), predicted (dashed), observed (blue), and tilt-corrected opercent burial for the A1 mine over the course of the experiment. Tilt-corrected values have been horizonfor clarity. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. bserved (red) tally offset 67

PAGE 83

T he model predictions from January 13th and January 17th do not fall with the nge of the multibeam data either (Figs. 35 & 36). The predicted value of mine burial for both thrial beam values (Figs. 35 & 36). ly 81.6%, e Temporal Analysis of Scour and Burial The acoustic instrumented mineployed in the shallow fine sand site on January 8th, 2003 at a est orientation. The d to est ra e January 13th and January 20th compari so ns is 18.1%, wh e reas the observed bufor both surveys is 1.8% and 28.3%, respectively. The model performs better during the January 20th evaluation, predicting a b urial of 60.5% compared to an observed value of 67.9%, and falls within the range of multibeam values. The model underestimates the amount of burial during the February 6th comparison, and is once again outside the range of multi Observed burial during this survey is 96.2%; however, the predicted burial is o na difference of 14.6%. The March 13th comparison is the final test of the model for the A1 mine. The model prediction and observed value are nearly identical, with a predicted burial of 98.0% and an observed value of 98.1%. The A2 Min e 2 (A2) was d water depth of 12.87 meters in an east-w observed depth of mine during the first survey on January 10th is 12.40 meters (Fig. 37). The depth of the ambient seafloor is 12.87 meters, resulting in an observed burial of 11.3%. The mine appears quite blurred in this image, presum a bly, because the beam mode was set to target detection on the multibeam system (which later was discoverewiden the beams and blur the image). The orientation of the mine appears to be east-win the image, although this is difficult to determine due to the distortion of the mine. The 68

PAGE 84

69 Figure 37. January 10th survey over the A2 mine. In all multibeam images, the black dashed line represents the ships trackthe survey. In this image, the track line is outside of the gridded area. The mine is outlined with a faint white line scaled tactual dimensions. The mine outline remains at the same orientation throughout the A2 multibeam images as a reference. lino the es e during min

PAGE 85

o rientation sensor within the mine indicates that the mine should have a more astnortheast-westsouthwest trend (~ 70). The mine appears less distorted in the January 13th image, though target detection hile the 12.92 meters, reducing the observed burial to ine has tending around the mine. The depth of e mined end where the maximum depth reaches 13.27 meters. he miary meters deeper than in the previous e was still on (Fig. 38). The mine depth has remained constant at 12.40 meters, wsurrounding seafloor depth has decreased to 1.8%. The mine may have provided protection to the underlying sand while the surrounding sand was locally eroded by currents. Target detection mode remained on during the January 17th survey, explaining why the mine still appears quite distorted in the image (Fig. 39). Although the mclearly sunk into the seabed, no scour is evident ex the is 12.53 meters with a surrounding seafloor depth of 12.98 meters. The observburial during this survey is 15.0%. Scour becomes evident around the mine during the January 20th survey (Fig. 40). There is a small pit of scour developing at eastern end of the mine, but the majority of development appears at the western Tne has sunk a further 0.06 meters since the 17th and is now at a depth 12.59 meters. The ambient seafloor is 12.89 meters, indicating a localized deposition of 0.09 meters and resulting in an observed burial of 43.4%. The scour has continued to develop and surrounds the mine during the Febru6th survey, although the depth within the pit remains constant at 13.27 meters (Fig. 41). The depth to the top of the mine is 12.82 meters, 0.23 survey. The ambient seafloor is 12.86 meters and the observed burial is 92.5%. The observed burial increases to 96.2% during the March 13th survey, with a mine depth of 70

PAGE 86

Figure 38. January 13th surve y over the A2 mine. 71

PAGE 87

Figure 39. January 17th surve y over the A2 mine. 72

PAGE 88

Figure 40. January 20 th survey over the A2 mine. 73

PAGE 89

Figure 41. February 6th survey over the A2 mine. 74

PAGE 90

12.88 meters and an ambient seafloor depth of 12.90 meters (Figs. 42 & 43). The scour has continued to expand out from the around the mine, and maximum depth within the pit is 13.25 meters. Overall, the A2 mine sank a total of 0.48 meters and became 96.2% buried (Table 4; Fig. 44). Scour around the mine formed a pit 0.35 meters deeper than the surrounding seafloor. The ambient seafloor became a total of 0.03 meters deeper over the course of the experiment. Comparison of A2 Multibeam Observations to the VIMS 2D Burial Mode The VIMS 2D burial model was initialized with a local water depth of 12.87 meters (obtained from the January 10th survey over the mine) and 0% burial. Thedel was run from the time of mine reposition, January 8th 2003 1600 GMT, to the timf the last multibeam survey over the mine, March 13th 2003 at 0200 GMT. The first direct comparison between the observed and predicted burial occurs for the January 10trvey (Figs. 44 & 45). The observed percent burial at this time is 11.3% compared to a predicted value of 14.7%. The difference is only 3.4%, and the predicted burial falls within the range of multibeam values. The January 13th comparison shows a discrepancy of 15.6% with a predicted ercent burial of 17.4% and an observed value of 1.8% (Figs. 44 & 45). This discrepancy most likely due to the fact that the mine did not sink between the January 10th and nuary 13th surveys. The predicted burial for the January 17th comparison is also 17.4%. here is an observed burial of 15.4% at this time, resulting in a discrepancy of 2% that is well within the 9.4% uncertainty range. l moe oh su p is Ja T 75

PAGE 91

76 Figure 4rc3theyr th 2. Mah 1 surv ovee A2mine.

PAGE 92

Figure 43. ROVvideo still image of the A2 mine on March 13, 2003. 77

PAGE 93

Jan. 8* Jan. 10 Jan. 13 Jan. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 12.40 12.40 12.53 12.59 12.82 12.88 Cumulative Amount of Change _____ _____ 0.00 0.13 0.19 0.42 0.48 Average Depth of Seafloor _____ 12.87 12.92 12.98 12.89 12.86 12.90 Cumulative Amount of Change _____ _____ 0.05 0.11 0.02 -0.01 0.03 Scour Visible / Depth of Scour _____ no no no yes 13.27 yes 13.27 yes 13.25 % Mine Burial from Multibeam ( 9.4% due to 5 cm uncertainty of sonar) 0 11.3 1.8 15.0 43.4 92.5 96.2 % Mine Burial from Model 0 14.7 17.4 17.4 59.5 80.7 97.3 Mine Heading (degrees) 71.2 69.9 69.8 69.0 64.4 62.7 62.9 Mine Pitch (degrees) 0.2 0.1 0.0 0.6 -0.1 -0.9 -0.9 Mine Roll (degrees) 1.4 5.7 5.6 4.6 8.6 21.6 21.5 Table 4. Data table for the A2 mine. All numbers are in meters except where noted. There is no multibeam rvey on January 8th. su 78

PAGE 94

44. Cntighy Figure omparison of multibeam observed (black) and predicted (gray) mine depth foe A2 mine over the course of the experime. Predicted percent burial of the ne was cverted to predd depth of thne usinge 12.87er watdepth used to initialize thel. Obseed depth seaflod depth of scour from the multibare potted Signit wave het is plotted the righaxis. r bars ent the 5-centimeter uncertaiherent the mueam system. Thelight graoval represents the A2 mine, and is scaled to the actual dimensions of the mine (length ~ 0.53 m, the diameter of the mine). r th mi ofErro onor anrepres icte e mieam nty in thl in -metas well. ltib er fican e mod on rvt y79

PAGE 95

Figure 45. Comparison of the mine (magenta), predicted (dashed), observed (blue), and tilt-corrected observed (repercent burial for the A2 mine over the course of the experiment. Tilt-corrected values have been horizontally ofor clarity. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. d) ffset 80

PAGE 96

The model overestimates the amount of burial during the January 20th comparison, with a predicted burial of 59.5% and an observed burial of 43.4% (Figs. 44 & 45). The 16.1% offset is outside the range of the multibeam values. The offset from the February 6th comparison is also outside the acceptable range, with a predictedobserved burial of 80.7% and 92.5%, respectively. The comparison from Mar shows the predicted value falls well within the 9.4% uncertainty range, with a predicted burial of 97.3% and an observed burial of 96.2%. The A4 Mine Temporal Analysis of Scour and Burial The acoustic instrumented mine 4 (A4) was deployed on January 8th, 2003 in the shallow fine sand site. It was positioned in an east-west orientation (79.3) atdepth of 12.77 meters. The January 10th survey over the mine shows a 1.8% observed burial with a mine depth of 12.25 meters and an ambient seafloor depth of 12rs (Fig. 46). There is no scour evident around the mine at this time. The observeorientation appears in agreement with the data from the orientation sensor wimine itself. The slight blurriness of the mine can be attributed the target detecof the multibeam. There is no scour evident in the January 13th survey either, though thes sunk 0.18 meters for a depth of 12.43 meters (Fig. 47). The seafloor depth around the mine is 12.90 meters, giving an observed burial of 11.3%. The ends of the mine appear blurry in this image; this is also likely due to the multibeam beam mode being set to target detection. Interestingly, the mine appears even more blurry and distorted in the and ch 13th a water .77 meted thin the tion mode mine ha 81

PAGE 97

82 Figure 46.arythey r thIn all multibeam images, the black dashed lints thitrack linrvey. Thneou with a faint white line scaled to the mines actual dns. Thmine outline remains at thon ught the A4 multibeam images as a e represene shps imeionse reference. Janu 10 survovee A4 mine. e during the sue mi is tlinede same orientatithroou

PAGE 98

Figure 47. January 13th survey over the A4 mine. 83

PAGE 99

im age from the January 17th survey even though target detection was turned off at this time (Fig.48). The same phenomenon can be see in the A2 multibeam observations. Again, it is unclear what is causing this. The depth to the top of the mine is 12.40 meters, an apparent shallowing of 0.03 meters since January 13th; however, this is within the 5-centimeter uncertainty of the multibeam. The ambient seafloor depth remains essentially the same at a depth of 12.91 meters, resulting in an observed burial of 3.7%. On January 20th, three days later, the seafloor depth around the mine still appears unchanged at a depth of 12.90 meters, despite the fact that the mine has sunk 0.22 meters and now resting at a depth of 12.62 meters (Fig. 49). The observed burial of the A4 mine at this time is 47.2%. The mine does not appear distorted in this image, and, int, does not, itself, actually show up very well. However, it is visible in this image because of the defining pit of scour wrapping around from the south side of the mine around to the east. The maximum depth measured within the scour pit is 13.22 meters. The mine images quite well during the February 6th image and is surrounded by a ring of scour measuring 13.28 meters at its deepest point (Fig. 50). The mine appears to have rolled 0.42 meters northwest from its original position into the scour pit. The maximum amount of recorded roll up to February 6th is -17.9, which only equates to .08 meters. Orientation sensors within the mine recorded data approximately every 38 minutes and do not record cumulative roll, so it is possible that the mine made a complete roll that was not recorded. A complete roll of the mine would shift its position 1.67 meters (the mine perimeter). If the mine rolled into a pit formed by scour; however, it would roll without shifting its actual position the full 1.67 meters. The depth of the mine fac 84

PAGE 100

85 Figure 48. Janu ary 17th eyth mi surv over e A4ne.

PAGE 101

Figure 49. January 20th survey over the A4 mine. 86

PAGE 102

Figure 50. February 6 survey over the A4 mine. In this image, the track line is outside of the gridded area. th 87

PAGE 103

d uring this survey is 12.77 meters and the ambient seafloor depth is 12.86 meters, indicating an observed burial of 83.0%. The mine appears back in its original position during the March 13th survey, indicating that the shift in position in the February 6th image may be due to the positional accuracy of the multibeam ( 1 meter) rather than actual change (Figs. 51 & 52). Once again, there is an apparent shallowing of the mine in this image. The depth to the top of the mine is 12.68 meters, .09 meters shallower than in the previous image. The depth of the ambient seafloor is 12.81 meters, indicating localized deposition around the m. There also appears to be some infilling of the scour pit as the maximum depth hadecreased to 13.11, a change of 0.17 meters. The observed burial for the March 1survey over the A4 mine is 75.5%. Overall, the A4 mine sank a total of 0.43 meters and became 75.5% burieable 5; Fig. 53). Scour around the mine formed a pit 0.30 meters deeper than the surrounding seafloor. The ambient seafloor became a total of 0.04 meters deeper over the course of the experiment. Comparison of A4 Multibeam Observations to the VIMS 2D Burial Mel The comparison of the VIMS 2D burial model with the A4 mine represents the st of the model tests using the acoustic instrumented mines. The model was initialized ith a local water depth of 12.77 meters (obtained from the January 10th survey over the ine) and 0% burial, and was run from the time of mine reposition, January 8th 2003 600 GMT, to the time of the last multibeam survey over the mine, March 13th 2003 at 0300 GMT. The first direct comparison between the observed and predicted burial occurs ines 3th d (Tod la w m 1 88

PAGE 104

89 Figure 51. Marcthyr the A4 m ine. h 13 surve ove

PAGE 105

90 Figure 52. ROV video still image of the A4 mine on March 13, 2003.

PAGE 106

Jan. 8* Jan. 10 Jan. 13 Jan. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 12.25 12.43 12.40 12.62 12.77 12.68 Cumulative Amount of Change _____ _____ 0.18 0.15 0.37 0.52 0.43 Average Depth of Seafloor _____ 12.77 12.90 12.91 12.90 12.86 12.81 Cumulative Amount of Change _____ _____ 0.13 0.14 0.13 0.09 0.04 Scour Visible / Depth of Scour _____ no no no yes 13.22 yes 13.28 yes 13.11 % Mine Burial from Multibeam ( 9.4% due to 5 cm uncertainty of sonar) 0 1.8 11.3 3.7 47.2 83.0 75.5 % Mine Burial from Model 0 15.4 18.1 18.1 60.5 81.6 98.0 Mine Heading (degrees) 79.3 78.4 78.4 77.98 73.2 69.3 69.4 Mine Pitch (degrees) -0.2 -0.2 -0.2 1.1 -0.1 -0.5 -0.6 Mine Roll (degrees) 2.1 4.0 4.0 2.0 -5.1 -9.9 -9.8 Table 5. Data table for the A4 mine. All numbers are in meters except where noted. There is no multibeam rvey on January 8th. su 91

PAGE 107

53ted t y-axnd i Figure Comparison of multibeam observed (black) and predicted (gray) mine depth for the A4 mine over the course of the experiment.Predicpercent buri the miwas corted to dicted depth ohe mine usin.77-r waterpth us inimodel. Observed deptf seaflooand deptf scourm the multibeam are plotted a. Signint waveight is tted onhe righis. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. The light gray oval represents the A4 mine, as scaled to the actual dimensions of the mine (length ~ 0.53 m, the diameter of the mine). e al ofh o ne r nveh o pre fro f t g the 12s well metefica dee h ed toplo tialize th t 92

PAGE 108

for the January 10th survey (Figs. 53 & 54). The observed percent burial at time is 1.8% compared to a predicted value of 15.4%, a discrepancy of 13.6%. Thevalues have a range of 9.4% due to the vertical uncertainty of the multibeahowever, the prediction for this comparison falls outside this range. The January 13th comparison shows a discrepancy of 6.8%, with an oburial of 11.3% and a predicted burial of 18.1%, which falls within the rangmultibeam values (Figs. 53 & 54). The predicted burial for the January 17th 18.1% as well; however, the observed burial is only 3.7% due to the 0.03 mshallowing of the mine. The predicted value, therefore, lies outside the multThis holds true for the comparison for January 20th as well. The model estimmine should be 60.5% buried at this time; however, the multibeam data onlyburial of 47.2%, leaving a discrepancy of 13.3%. On February 6th, the model and the observed values are in agreemenpredicted value for burial of 81.6% and an observed value of 83.0% (Figs. 53 & 54). The discrepancy of 1.4% falls well within the acceptable range. The same is not final comparison on March 13th. The apparent shallowing of the mine has reobserved burial of only 75.5%, while the model predicts a burial of 98.0%. The F5 Mine Temporal Analysis of Scour and Burial The F5 mine was one of two optical instrumented mine located in th sand site during the 2003 IRB mine burial experiment. It was deployed on January 12th, 2003 in 12.96 meters of water and oriented northeast-southwest. The optical mines have his t observed m system; bserved e of comparison iseter ibeam range. ates that the indicate a t, with a true of the sulted in an e shallow fine 93

PAGE 109

94 Figure 54. mp o (magenta), pred (dd), servlu observ(red) percent burir t4ovehe course rimeTies have been horizontally ofbaresent thcertaintyrent in thltibeam sy ed stem. ictedasheobed (be), and tilt-correctedof the expent. lt-corrected value 5-centimeter un inhee mu Coarisonf the mineal fohe A mine r tfset for clarity. Error rs rep

PAGE 110

steel casings and do not include compasses in their suite of instrumentation; therefore, the apparent orientation cannot be checked against the mine itself. The January 13th survey was the first to pass over the F5 mine after deployment (Fig. 55). The depth to the top of the mine is 12.52 meters and the depth of the surrounding seafloor is 12.96 meters. The amount of observed burial at is 6.4%. The mine does not image very well during this survey and the seafloor appears quite mottled. The reason for the poor appearance of the mine is not clear. Tappearance of the seafloor may be in part due to actual bed morphology at tf the seafloor and in part due to noise in the data. The seafloor appears to have smoothed out in the image from the Jath survey (Fig. 56). The mine shows up quite clearly in this image, though it asomewhat blurry and distorted. The depth to the top of the mine is 12.61 me a surrounding seafloor depth of 13.00 meters. The observed burial in this image is 17.0% and there is no evident scour around the mine. The mine does not appear to show up at all in the image from the January 20th survey (Fig. 57). A scour pit can clearly be seen in the image, with a maximum depth of 13.19 meters and a slight rise in the middle. The rise appears as two separate bumps within the scour pit and cannot be attributed to the mine with any certainty. The shallowest depth of this rise is 12.89 meters, which ertically flush with the surrounding seafloor. If this was indeed, the mine, conditions ould indicate a 100% burial. Due to the combined facts that the mine is not fully buried subsequent images, the rise appears as two separate bumps in this image, and that the 6 mine does not show up during the January 20th survey either (to be discussed), it was ecided to not treat the rise as the mine. its this timehe he time onuary 17ppears ters withsets it v w in F d 95

PAGE 111

96 Figure 55. aryth suy over the F miTheack hedrveyhnetlinedwithnite dede mine e remains at thd orienation throuut thfe Janu 13rve5ne. bldas line iheips tack line he su. Te mi itself is ou a fait whash linscaled to thimens minThe outline same scale antghoe rest of th multibrerence. ndicates t shrduring te actual dionsof the e. e F5eam images as a

PAGE 112

Figure 56. January 17th survey over the F5 mine. 97

PAGE 113

Figure 57. January 20 survey over the F5 mine. The mine is not evident in this image, though a well-defined scour pit can be clearly seen. th 98

PAGE 114

The mine is clearly visible in the image from the February 6th survey, resting at a epth of 12.77 meters and is surrounded by a ring of scour that expands out to the utheast of the mine (Fig. 58). The depth of the ambient seafloor is 12.87 meters and the observed amount of burial is 78.7%. Maximum depth within the surrounding scour is 13.22 meters. The mine continues to show up quite well in the March 13th image (Figs. 59 & 60). It has sunk a further .09 meters, for a total depth of 12.86 meters. The depth of the ambient seafloor and depth within the scour pit has remained unchanged, resulting in an observed burial of 97.9%. Over the course of the experiment, the F5 mine sank a total of 0.34 meters and the surrounding seafloor showed a localized deposition of 0.09 meters (Table 6; Figs. 61). Final observed burial of the F5 mine was 97.9%. Scour around the mine formed a pit 0.35 meters deeper than the surrounding seafloor. Comparison of F5 Multibeam Observations to the VIMS 2D Burial Mode For comparison with the F5 mine, the VIMS 2D burial model was initialiwith a local water depth of 12.96 meters (obtained from the January 13th survey over the mine) and 0% burial. It was run from the time of mine deployment (there was no repositioning of the F5 mine by divers), January 12th 2003 0000 GMT, to the time of the last mam survey over the mine, March 13th 2003 at 0300 GMT. The first direct comparisonbetween the observed and predicted burial occurs for the January 13th survey (Fig62). The predicted burial at this time is 3.9% and the observed burial is 6.4% Thobserved values of burial have a range of 10.6% for the optical mines (0.47-meter diameter, see Fig. 8) due to the vertical uncertainty of the multibeam system. d so l zed ultibe s. 61 & e 99

PAGE 115

100 Figure 58. Februa ry 6th survey ovehe r tF5 mine. 100

PAGE 116

101 Figure 59. March 13th survey over the F5 mine.

PAGE 117

Figure 60. ROV video still image of the F5 mine on March 13, 2003. 102

PAGE 118

Jan. 11* Jan. 13 Jan. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 12.52 12.61 ____ 12.77 12.86 Cumulative Amount of Change _____ _____ 0.09 ____ 0.25 0.34 Average Depth of Seafloor _____ 12.96 13.00 12.89 12.87 12.87 Cumulative Amount of Change _____ ______ 0.04 -0.07 -0.09 -0.09 Scour Visible / Depth of Scour _____ no no yes 13.19 yes 13.22 yes 13.22 % Mine Burial from Multibeam ( 10.6% due to 5 cm uncertainty of sonar) 0 6.4 17.0 ____ 78.7 97.9 % Mine Burial from Model 0 3.9 4.2 62.5 85.0 100.9 Mine Pitch (degrees) 2 1 2 2 1 2 Mine Roll (degrees) 20 3 5 8 16 17 Tsu able 6. Data table for the F5 mine. All numbers are in meters except where noted. There is no multibeam rvey on January 11th. 103

PAGE 119

Figure 61.dictservr be act Comparison of multibeam observed (black) and predicted (gray) mine depth for the F5 mine over the coursehe experiment. Preed percenturial of the mine converted to prected depth of tne usihe 12-meterer deh uso initialize the model. Obed depth seaflod depth of scour from theltibeam are plo as well. Signifit wave ight is otten the right y-axis. Erroars represent the 5-centimeter uncertainty inherent in the multibeam system. The light gray oval represents the F5ine, and is scaled to thual dimensions of the mine (length ~ 0.47 m, the diameter of the mine). The mine did not image on Julian day 20. of ted td o m bof was di mu htte e mid ng t .96can wathe ptpl or an 104

PAGE 120

Figure 62. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observeburial for the F5 mine over the course of the experiment. Tilt-corrected values have been horizontally offset for clarity. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. d (red) percent 105

PAGE 121

The January 17th comparison shows a discrepancy of 12.8%, which falls outside e range of multibeam values (Figs. 61 & 62). The observed burial for this comparison is 17.0% while the model predicts a burial of only 4.2%. There is no comparison for January 20th due to the fact that the mine cannot be distinguished in the multibeam image. The model prediction of percent burial at the time of the survey over the miwever, is 62.5%. On February 6th, the discrepancy between the model and the multibe is 6.3%, within the range of observable values (Figs. 61 & 62). The predicted f burial is 85.0%, while 78.7% is actually observed in the multibeam data. Thancy decreases to a mere 3% for the March 13th comparison, with a predicted burial of 100.9% and an observed burial of 97.9%. The F6 Mine Temporal Analysis of Scour and Burial The optical instrumented mine number 6 (F6) was deployed in the shallow fine sand site on January 12th, 2003. It was situated in 13.00 meters of water depnorthwest-southeast orientation. The first survey to image the mine after det was on January 13th (Fig. 63). The target detection mode on the multibeam sonar has caused the mine to appear blurry in the image. The depth to the top of the mine is 1ters and the surrounding seafloor depth is 13.00 meters, giving an observed burial of 8.5%. arget detection was not set during the January 17th survey over the mine, although the ine still appears blurry. The blurriness may explain along with the 5-centimeter ertical uncertainty of the multibeam for this survey the apparent 6-centimeter th ne; hoam dataamount oe discrepth in a ploymen2.57 me T m v 106

PAGE 122

sh allowing of the mine (Fig. 64). Furthermore, the combined vertical uncertainty of the ultibeam for both the January 13th and January 17th surveys can explain the apparent f the for the January 17th survey is zero. The early stages of scour pit development d r this is not clear, yet we have seen this elsewhere .g., Fig. 57), and it is not known if this phenomenon is related to the F5 case or if it is merely coincidence. The maximumdeptht is 13.18 meters and the average depth of the surroundinging m discrepancy in mine depth between the two. Depth to the top of the mine is now 12.51 meters. The depth of the seafloor around the mine is 13.01, which indicates that the mine is resting 3 centimeters above the bed if the depth of the mine is accurate. The tilt omine has not changed since the January 13th survey, and therefore cannot be the reason for the offset between mine depth and seafloor depth. Consequently, the observed burial of the mine can be seen off the northeast and southwest sides of the mine. Maximum depth measurein the scour is 13.17 meters. As in the case of the F5 mine, the F6 mine does not show up in the image from the January 20th survey (Fig. 65). The scour pit can be clearly seen, but there is no evidence of the mine. The reason fo (e of the scour pi seafloor is 12.95 meters. The image from the February 6th survey shows the mine quite clearly resting within a pit of scour at a depth of 12.75 meters (Fig. 66). The depth of the surroundseafloor is 12.93 meters, giving an observed burial of 61.7%. The scour pit itself has remained relatively constant, with a maximum depth of 13.17 meters. The mine appears to have rolled to the northwest in the March 13th survey image (Figs. 67 & 68). Sensors within the mine recorded a 16 change in roll since the previous survey, which can onlyaccount for 0.06 meters of the 0.27-meter offset. T h is offset; however, is well within the 107

PAGE 123

108 Fig 63thveyr the F6 achline indicates the sips tack line during thrvey. The mineed witht ite ded line actudimenionheThe mine outlinhe same scale andorn througt threst eam efee. mine. The blk dased hr a fainwhashe scaled to thal ss of t ientatiohoue of the F6 multib ure. January 13 sur ovee su itself is outlinmine. e remains at timages as a rrenc 108

PAGE 124

Figure 64. January 17th survey over the F6 mine. 109

PAGE 125

Figure 65.t evident in this image, though a well-defined see January 20th survey over the F6 mine. The mine is noscour pit can be clearly n. 110

PAGE 126

Figure 66. February 6th survey over the F6 mine. 111

PAGE 127

Figure 67. March 13th su rvey over the F6 mine. 112

PAGE 128

Figure 68. ROV video still image of the F6 mine on March 13, 2003. 113

PAGE 129

h orizontal accuracy of the multibeam system. It is also possible that the mine has rolled ack and forth within the scour pit, which would account for the sensors not recording the full amount of roll. The depth of the mine is 12.68 meters, indicating a shallowing of 7 centimeters since the survey of February 6th. The combined 5-centimeter vertical uncertainty of the multibeam for both the February 6th and March 13th surveys can be used to explain the apparent 7-centimeter offset of mine depth. For other possible explanations, refer to the discussion of the F8 mine. Maximum depth within the scour pit and depth of the ambient seafloor is 13.26 meters and 12.80 meters respectively, indicating a final observed burial of 74.5% for the F6 mine. Overall, the F6 mine appears to have sunk a total of 0.11 meters (Table 7; Fig. 69). Final observed burial of the F6 mine is 74.5%. Scour around the mine formed a pit 0.46 meters deeper than the surrounding seafloor. The average depth of the seafloor around the mine shows a localized deposition of 0.20 meters over the course of the experiment. Compaal Model In order that the VIMS 2D burial model could be compared with observational ata from the F6 mine, the model was initialized with a local water depth of 13.00 meters btained from the January 13th survey over the mine) and 0% burial. The model was run from the time of mine deployment on January 12th, 2003 at 0000 GMT to the time of the last survey over the mine, March 13th, 2003 at 0900 GMT. The first comparison between the VIMS model and the multibeam observations occurs for the January 13th survey (Figs. 69 & 70). The observed data shows the mine to b rison of F6 Multibeam Observations to the VIMS 2D Buri d (o 114

PAGE 130

Jan. 11* Jan. 13 Jan. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 12.57 12.51 ____ 12.75 12.68 Cumulative Change Depth of _____ 13.00 13.01 12.95 12.93 12.80 Cumulative Amount of _____ ______ 0.01 -0.05 -0.07 -0.20 Change Scour Depth of _____ no 13.17 13.1 Amount of _____ _____ -0.06 ____ 0.18 0.11 Average Seafloor Visible / Scour yes yes 8 yes 13.17 yes 13.26 % Mine 10.6% due to uncertainty 0 8.5 0 ____ 61.7 74.5 Burial from Multibeam ( 5 cm of sonar) % Min e Burial from Model 0 3.8 4.0 62.1 84.6 100.7 M ine Pitch (degrees) 2 2 2 2 2 1 Mine Roll 0 -19 -19 -13 -13 3 (degrees) Table 7. Data table for the F6 mine. All numbers are in meters except where noted. There is no multibeam survey on January 11th. 115

PAGE 131

opercObservedae ac Figure 69. Cmparison of multibeam observed (black) and predicted (gray) mine depth for the F6 mine over the course of the expriment. Predicted ent burial of the mine was coed to predicted deh of the mine he 13-meteer de used t initialize the model. depth oseafloor depth scour from the ltibeam are plot well. Significant wave hht is ptted on the right y-axis. Error brs represent the 5-centimeter uncertainty inherent in the multibeam system. The light gray oval represents the F6 m, and is scaled to thtual dimensions of the mine (length ~ 0.47 m, the diameter of the mine). The mine did not image on Julian day 20. eine nvertof ptmu usited ng t as .00 r wat ptheig olo f and 116

PAGE 132

Figure 70. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observeburial for the F6 mine over the course of the experiment. Tilt-corrected values hafor clarity. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. d (red) percent ve been horizontally offset 117

PAGE 133

b e 8.5% buried; however, the model only predicts a burial of 3.8%. The discrepancy of .7% falls within the 10.6% range of uncertainty of the observed multibeam values. he 6-centimeter shallowing of the mine that occurs in the January 17th survey results in an observed burial of zero percent. The model predicts a 4.0% burial for the 17th, and therefore remains within the acceptable range. At the time of the January 20th survey, the model predicts a 62.1% balthough no comparison can be made since the mine does not appear in the (Figs. 69 & 70). The next comparison occurs for the February 6th survey and showdiscrepancy of 22.9%, which falls outside the acceptable 10.6% range. The observed burial at the time of this survey is 61.7%, while the model predicts a burial of 84.6%. The model prediction from March 13th of 100.7% falls well outside this range as well, when compared to the observed burial of 74.5%. This offset of 26.2% is the largest discrepancy between the model predictions and the observed data for the 6 mines locateshallow fine sand site. The F7 Mine Temporal Analysis of Scour and Burial The optical instrumented mine number 7 (F7) was located in the coarse sand site lying within a rippled scour depression. It was deployed on Januar03, and repositioned by divers on January 12th, 2003 at ~ 2000 GMT. The Janusurvey was the first to pass over F7 after its deployment (Fig. 71). No ripples can be distinguished in the image, despite their presence around the F8 mine deployed in the me location. The observed orientation of the mine appears north northeast by south 4 T urial, images a d in the y 11th 20ary 13th sa 118

PAGE 134

so uthwest, and the depth to the top of the mine is 13.34 meters. The depth of the ambient e or he 7 mine irectly of tilt nuary 13th; however, the combined vertical uncertainty of the multibeam from the January 13th and January 17th surveys can account for this offset. The depth of the seafloor around the mine o 13.87 meters, a difference of 0.60 meters from the observedeters greater than the diamete of the that ly seafloor around the mine is 13.83 meters, a difference of 0.49 meters. The diameter of thF7 mine is only 0.47 meters; however, the 2-centimeter discrepancy can be accounted fby the 5-centimeter uncertainty of the multibeam system. It is also p o ssible that the mine is resting on a mound of sand slightly shallower than the surrounding seaf l oor. Tobserved percent burial for this survey is zero. A quadpod and one spider were deployed in the coarse sand site near the Fon January 16th, 2003 and should be visible in subsequent images. The mine does not image very well during the January 17th survey, despite the multibeam passing doverhead (Fig. 72). The spider does not show up at all in this image, while the quadpod,on the other hand, is quite apparent. The spider has a relatively small profile (top surface area = 0.07 m2), so it is possible that the sonar was unable to get enough hits off the surface in order to adequately image it. The depth to the top of the mine is 13.27 meters, 0.07 meters shallower than on January 13th. There has been no change in the degree for the F7 mine since Ja has increased t top of the mine. This offset is 13 centim r mine. Assuming the sonars vertical uncertainty accounts for 5 centimeters, there is still a discrepancy of 8 centimeters that cannot be explained. Although uncertainty in the sound velocity profile is a possible explanation, the fa c ta shallowing of the F8 mine did not occur during the same survey, makes it an unlikecause. As mentioned in Chapter 1, the 5-centimeter vertical uncertainty of the sonar 119

PAGE 135

120 Figure 71. Janu 13 survovee F7ne. blk dased linsuey. e mi itself is oued a fait bl dash linscaled thotline ne same scale anrienationught threst o the F7 mu arythey r th miTheache indicates the srack line during te rvThnetlinwithnackede toe actual disionof theuremais at thd ot throoue fltibreferen hips thmens e min. The mine eam images as a ce. 120

PAGE 136

121 Figure 72. January 17th survey over the F7 mine. The quadpod images quite clearly, and is outlined withsquare scaled to the dimensions of the quadpod. The red dashed circle with a question marks denotes the pspider that did not image. a white-dd ositio ashen of the

PAGE 137

w as estimated based on the shallow depth in which this study took place, and use of the OS MV positioning system equipped with RTK. This discrepancy may be an indicator that the actual vertical uncertainty of the multibeam system may be closer to 10 centimeters. Given this offset between the mine depth and the depth of the ambseafloor, the amount of burial at the time of this survey is assumed to be zero. Fripples trending approximately north south are apparent in the image to the nore track line. The ripples have an average wavelength of ~ 1.5 meters and a height of ~ 10 centimeters. The mine has sunk 0.10 meters in the January 20th image, resulting in a depth of 13.35 meters (Fig. 73). The depth of the surrounding seafloor is 13.79 meters, giving a percent burial of 10.6. The ripple field is no longer visible at this time, and there is no scour evident. Neither the quadpod nor the spider show up during this survey osubsequent one from February 6th (Fig. 74). The quadpod presents a greater profile than the spider; however, the legs of the quadpod are quite slim and come up over thquapods top platform to form a t-junction (Fig. 8). It is possible that the beams of the sonar hit these legs and were reflected away rather than back to the sonar. It is possible that a bubble sweep occurred causing interference at the time the sonar passed over these instruments. The depth to the top of the mine in the February 6th ima13.35 meters. This apparent 2-centimeter decrease in depth can be accounted for by the vertical uncertainty of the multibeam. The depth of the seafloor is 13.81 mersdecreasing the amount of burial observed to 2.1%. The March 13th survey over the F7 mine is the only one to image both the spider and the quadpod as well as the mine. A ripple field is clearly evident trending north P ient aint th of thr the e also ge is te 122

PAGE 138

123 arythey or th min. Neither thimag Thr deplcationare oued a dshed Figure 73.Janu 20 survvee F7ee quadpodnor thidt is e.eioyed los tlin withared line. e sper are evidenin th 123

PAGE 139

Figure 74. February 6th survey over the F7 mine. Again, neither the quad not the spider appear in this image. 124

PAGE 140

n orthwest by south southeast across the image with an average wavelength of ~ 1.0 eters and a height of 15 centimeters (Figs. 75 & 76). The depth to the top of the mine as remained constant at 13.35 meters since the survey of February 6th. The average afloor depth around the mine is 13.76 meters, giving an observed burial of 12.8%. Between the January 13th and March 13th surveys, the F7 mine sunk a total of 1 centimeter (Table 8; Fig. 77). The average depth of the seafloor around the mine decreased 7 centimeters during this time, mainly due to the formation of rippled bedforms in the area. The final observed burial for the F7 mine was 12.8%. Comparison of F7 Multibeam Observations to the VIMS 2D Burial Model It has been shown that the VIMS 2D burial model does not work well for coarse sand sites where rippled bedforms are prevalent (see Chapter 2; Traykovski et al., 2005; Trembanis et al., 2005). The same holds true for the comparisons with the F7 e, which resides in a rippled scour depression. The model was initialized with a lo water depth of 13.83 meters and 0% burial. It was run from the time the mine was redeployed by divers, January 12th, 2003 at 0000 GMT to the time of the last multibeam survey over the mine on March 13th, 2003 at 0500 GMT. The first two comparisons between the model predictions and the observed data, January 13th and January 17th, are in agreement with a 0% burial in both cases (s. 77 & 78). The January 20th comparison; however, shows the model diverging from thmultibeam data with a predicted burial of 39.0% and an observed burial of only 10.1%. The 28.4% discrepancy is well outside the 10.6% range of the multibeamhodel m h se mincalFige e m T 125

PAGE 141

126 Figu5csur ovhminte that both ther d thadpode evt. e spidane qu arnowiden e. No re 7. Marh 13th veyer te F7 126

PAGE 142

Figure 76. ROV video still image of the F7 mine on March 13, 2003. 127

PAGE 143

Jan. 11* Jan. 13 Jan. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 13.34 13.27 13.37 13.35 13.35 Cumulative Amount of Change _____ _____ -0.07 0.03 0.01 0.01 Average Depth of Seafloor _____ 13.83 13.87 13.79 13.81 13.76 Cumulative Amount of Change _____ ______ 0.04 -0.04 -0.02 -0.07 Scour Visible / Depth of Scour _____ no no yes 13.86 yes 13.99 no % Mine Burial from Multibeam ( 10.6% due to 5 cm uncertainty of sonar) 0 0 0 10.6 2.1 12.8 % Mine Burial from Model 0 0 0 39.0 68.4 85.1 Mine Pitch (degrees) 2 1 1 2 1 1 Mine Roll (degrees) -5 -4 0 2 4 2 Tsu able 8. Data table for the F7 mine. All numbers are in meters except where noted. There is no multibeam rvey on January 11th. 128

PAGE 144

inpl Figure 77. Comparison of multibeam observed (black) and predicted (gra) mine depth for the F7 mine over the course of the experiment. Predicted percent burial of the mine was converted to predicted depth of the mine using the 13.83-metater depth used to itialize thodel. served dth of seafloor andepth of scour frohe mueam are plotted as well. t wave height is otted on the right y-axis. Erroars represent theentimeter uncertaity inhet in thultibeam system. Theht gray oval represents the F7 mine, and is scaled to the actual dimensions of the mine (length ~ 0.47 m, the diameter of the mi y er wifican ligne). e m Ob epr b d 5-c m tn ltibren Sign e m 129

PAGE 145

Figure 78. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observed (reburial for the F7 mine over the course of the experiment. Tilt-corrected values have been horizonfor clarity. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. d) percent tally offset 130

PAGE 146

c ontinues to significantly diverge from field observations for both the February 6th and arch 13th comparisons. On February 6th, the model predicts a 68.4% burial of the F7 ine while the multibeam data only show a 2.1% burial, an offset of 66.3%. This offset creases to 72.3% for March 13th, with a predicted burial of 85.1% and an observed burial of only 12.8%. These discrepancies suggest that rippled bedforms cannot be ignored and should be included in future modeling efforts. The F9 Mine Temporal Analysis of Scour and Burial The optical instrumented mine number 9 (F9) was one of two mines located in the deep fine sand site. F9 was deployed on January 11th, 2003 in a water depth of 13.88 meters and repositioned by divers on January 13th to lay in an east-west orieThe first survey to image the mine after deployment was on January 13th (Fig. 79). The mine appears somewhat distorted in the image as a result of the target detection mode on the multibeam sonar. The depth to the top of the mine is 13.37 meters, and the surrounding seafloor has an average depth of 13.88 meters. The diameter of the mine is a discrepancy of 4 centimeters. This discrepancy could be a result of the vertical accuracy of the multibeam or of mine resting on a mound of sand slightly higher thansurrounding seabed. Pitch sensors in the mine recorded a 2 tilt at the time orvey, which could account for up to 3 centimeters of the offset. The observed amount of burial for F9 during this survey is 0%. The survey from January 17th did not pass over the deep fine sand site; therefore, the next observation occurs during the January 20th survey (Fig. 80). The depth to the top M m in ntation. 0.47 meters the f this su 131

PAGE 147

Figure 79. ary sury ovehe Fmine. is image, thtracke falls oute thddee act dinsioof thine.e e oultibeam images as a ce. Janu 13thver t9 The black dashed line indihe sips tack line he survey. In the linside grid area. The mine itself is outlinwith a fainshe scaled to thualmens e m Thmintline remains at the same scale and orienatio throourest of the F9 mureferen cates thrduring ted t black daed lintnught the 132

PAGE 148

133 Figure 80. January 20th survey over the F9 mine.

PAGE 149

o f the mine is now 13.39 meters, 2 centimeters shallower than on the 13th. The ambient afloor shows a localized deposition of 0.18 meters, depth is 13.70 meters, and results in n observed burial of 34.0%. Scour has started to develop around the mine with a aximum depth of 13.86 meters. By February 6th, the scour has extended to form a ring round the mine (Fig. 81). Maximum depth with the scour pit has increased to 13.94 meters. The seafloor depth around the mine is now 13.67 meters and the depth of mine itself is 13.45 meters, giving an observed burial of 53.2%. The mine appears distorted in this image despite the fact that the target detection mode on the multibeam was turned off. The change in seafloor depth is obvious in the March 13th image; average depth of the seafloor around the mine is now 13.47 meters (Figs. 82 & 83). This sedimentation of 0.20 meters is not just observed around the scour pit, but occurs over the whole grid. The depth of the mine is 13.41 meters, a decrease of 4 centimeters since February 6th. The tilt of the mine has actually decreased since the last survey, indicating this discrepa is most likely a result of the 5-centimeter vertical uncertainty of the multibeam rather than a tilt effect. Maximum depth in the scour pit has decreased to 13.85 meters. Thne appears to be 87.2% buried at this time. The F9 mine has sunk a total of 0.04 meters over the course of the expent, resulting in a final observed burial of 87.2% (Table 9; Fig. 84). The average seafloor depth has steadily shallowed since the first survey, for a total shallowing of 0.41 meters. se a m a ncye mirime 134

PAGE 150

Figure 81.February 6th survey ovehe mi r tF9ne. 135

PAGE 151

136 Figure 82.March 13th survey over th e F9 mine.

PAGE 152

Figure 83. ROV video still image of the F9 mine on March 13, 2003. 137

PAGE 153

Jan. 11* Jan. 13 Jan. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 13.37 No image 13.39 13.45 13.41 Cumulative Amount of Change _____ _____ _____ 0.02 0.08 0.04 Average Depth of Seafloor _____ 13.88 _____ 13.70 13.67 13.47 Cumulative Amount of Change _____ ______ _____ -0.18 -0.21 -0.41 Scour Visible / Depth of Scour _____ no _____ yes 13.86 yes 13.94 yes 13.85 % Mine Burial from Multibeam ( 10.6% due to 5 cm uncertainty of sonar) 0 0 _____ 34.0 53.2 87.2 % Mine Burial from Model 0 0 _____ 53.8 77.1 95.1 Mine Pitch (degrees) 2 2 _____ 1 3 1 Mine Roll (degrees) -4 1 _____ -2 -22 -18 Table 9. Data table for the F9 mine. All numbers are in meters except where noted. There is no multibeam rvey on January 11th. su 138

PAGE 154

Figure 84. Comparisonf multibeam observed (black) and predicted (gray) mine depth for th mine over the course of the experiment. Predictepercent burial of the mine was converted to predicted depth of the mine using the 13.88-meter water depth used to initialize the modeObserved dh of soor andpth of scour froe multibeam tted as well. Significane he is plotted on the righaxis. Errorrs represent the 5ntimeter uncertainty inherent in thltibeamstem. The lightay ovrepresents the F9 s scaled to the actual dimensions of the mine (length ~ 0.47 m, the diameter of the mine). There was no survey on Julian day e F9 ept ba eafl de-ce m th are ploe mu t wav gr ightal sy od l. t y-mine, and i17. 139

PAGE 155

Comparison of F9 Multibeam Observations to the VIMS 2D Burial Model The VIMS 2D burial model was initialized with a local seafloor depth of 13.88 meters and 0% burial for comparison with the F9 multibeam observationwas run from January 13th, 2003 at 1500 GMT, the time thatmine, to March 13th, 2003 at 1200 GMT, the time of the last survey over The first comparison occurs for the January 13th survey over the mdata and the model indicate a percent burial of zero at this time (Figs. 84 January 20th, the data show an observed burial of 34.0%, while the modeburial of 53.8%. The model overestimates the amount of burial by 19.8%outside the 10.6% range of the multibeam data. The model continues togreater amount of burial than what is actually observed for the February 6well. The multibeam data indicate a burial of 53.2% at this time; howevepredicts a burial of 77.1%, a difference of 23.9%. It is not until the Marcthe model predictions fall back within the acceptable range. The observeburial for this survey is 87.2%. The model predicts a burial of 95.1% at tdiscrepancy of only 7.9%. The F10 Mine s. T the divers repositioned the tlrhhat dhTemporal Analysis of Scour and Burial The optical instrumented mine number 10 (F10) was the last instrueployed as part of the mine burial experiments off Indian Rocks Beach. eployed on January 11th, 2003 in the deep fine sand site in a water depth of 13.90 eters. SCUBA divers repositioned the on January 13th to lay in a north-south he model he mine. ine; both the & 85). On predicts a which is predict a th survey as the model 13th survey t amount of is time, a mented mine F10 was d d m 140

PAGE 156

141 Figure 85. Comparison ofedserv(be), andlt-d (repercent burial for the F7 minvee coe expmen. Tilt-corrected valhaenhorizontally offset for clarity.rroese 5-centimeter uncertain inht inmultibeam system. d) the predicted (da sh ), obed lu ticorrected observee or thurse of theritues ve be Er bars reprent thtyeren the 141

PAGE 157

o rientation. The first survey to image the mine after deployment was on January 13th (Fig. 86). The mine appears somewhat distorted in the image as a result of the target detection mode on the multibeam sonar. The top of the mine is at a depth of 13.42 meters, and the surrounding seafloor has an average depth of 13.90 meters. The difference, 0.58 meters, is 11 centimeters greater than the diameter of the F10 mine. The 2 tilt of the mine at this time can account for 3 centimeters, and the vertical uncertainty omultibeam can account for another 5. There is no clear explanation for the rem centimeters; however, it is possible that the target detection mode has caused to appear shallower than it actually is, or that the mine is resting on a mound of sand slightly shallower than the surrounding seabed. The observed amount of burial for F10 for the January 13th survey is 0%. The next observation of the F10 mine does not occur until January 20e the January 17th survey did not pass over the deep fine sand site (Fig. 87). The depth to the top of the mine is now 13.60 meters, indicating a sinking of 0.18 meters s13th. The ambient seafloor depth is 13.73 meters, showing a deposition of 0.1, which agrees with the 0.18-meter deposition seen around the F9 mine. The m.3% buried and is only evident in the image as a result of the ring of scour that hasaround it. The maximum observed depth within the scour is 14.07 meters. On February 6th, the observed depth to the top of the mine is 13.59 me88). The 1-centimeter difference between this observation and that of Janury 20th is well within the vertical accuracy of the multibeam sonar and is essentially negligible. The depth of the seafloor around the mine is 13.69 meters, resulting in a percent burial of 78.7%. The scour has continued to extend around the southern end of the mine, although f the aining 3 the mineth, becausince the 7 metersine is 72 formed ters (Fig. a 142

PAGE 158

143 surv. Inthroou Figure 86.January 13rvovee FThe blk dashed lhrey this image track linegriddd area. Thtlined with a faint backedline scaled to the actual dimensions of thee outline e same scale ant ught the rest of te F10mulamgce. ine indicates the sips tack line during the e mine itself is oul dash remains at thd orienation th suey r th10 mine. ac e, the falls outsid the e mine. Thmine h tibe imaes as a referen

PAGE 159

Figure 87. January 20th survey over the F10 mine. Track line runs just along the top of the grid. 144

PAGE 160

Figure 88. February 6th survey over the F10 mine. 145

PAGE 161

th e maximum depth within the pit has decreased to 14.05 meters. The February 6th survey represents the last multibeam data over the mine. lthough the survey from March 13th (Fig. 89) passed over the deep fine sand site, the F10 mine lay in the outer beams of the sonar swath and was not imaged. Between January 13th and February 6th, 2003, the F10 mine sank a total 0.17 meters and became 78.7% buried (Table 10; Fig. 90). The surrounding seafloor depth showed a localized deposition of 0.21 meters during this time, and a ring of scour with a maximum depth of 14.05 meters developed around the mine. Comparison of F10 Multibeam Observations to the VIMS 2D Burial Model The comparison of the VIMS 2D burial model with the F10 multibeam observations was the last test of the model for this project. The model was initzed with a local seafloor depth of 13.90 meters and 0% burial, and run from January 13003 at 1500 GMT, the time of mine reposition, to February 6th, 2003 at 1000 GMT, the time of the last survey over the mine. The first comparison takes place for the Januaryth survey (Figs. 90 & 91). The model does not predict any burial at this time, and there is 0% observed in the multibeam data. On January 20th, the data show an observed burial of 72.3%, while the model predicts a burial of only 53.6%, a discrepancy of 18.7%, which is outside the uncertainty range of the multibeam system. The final comparison between the model and the F10 mine on February 6th shows that the two are agreement. A burial of 78.7% is observed in the data and the model predicts a burial of 76.9%, a discrepancy of only 1.8%. A ialith, 2 13 146

PAGE 162

Figure 89. ROV video still image of the F10 mine on March 13, 2003. There is no ROV video still image from the Feb. 6, 2003 survey. 147

PAGE 163

Jan. 11* Jan. 13 Jan. 17 Jan. 20 Feb. 6 Mar. 13 Depth of Mine _____ 13.42 No image 13.60 13.59 No image Cumulative Amount of Change _____ _____ _____ 0.18 0.17 _____ Average Depth of Seafloor _____ 13.90 _____ 13.73 13.69 _____ Cumulative Amount of Change _____ ______ _____ -0.17 -0.21 _____ Scour Visible / Depth of Scour _____ no _____ yes 14.07 yes 14.05 _____ % Mine Burial from Multibeam ( 10.6% due to 5 cm uncertainty of sonar) 0 0 _____ 72.3 78.7 _____ % Mine Burial from 0 0 _____ 53.6 76.9 _____ Model Mine Pitch (degrees) 2 -1 _____ 2 2 _____ Mine Roll (degrees) -4 -2 _____ 2 8 _____ T able 10. Data table for the F10 mine. All numbers are in meters except where noted. There is no ultibeam survey on January 11th. m 148

PAGE 164

ofef sc Figure 90. Comparison multibeam observed (black) and predicted (gray) mine depth for the F10 mine over the course of the expet. edicted percent burial of the min was convto picted de of the mine usinthe 13.90-meter water depted to ialize e modeObsered depth of seafloor and depth oour from tmultim are pled as well. Signicant wave height ilotted ohe rig-axis. or bapreset the 5-centimeter uncertainty inherent in thstem. e light gray ovresents the F10ne, and is scaled to the actu dimeions the mine (length ~ 0.47 m, the diameter of the mine). There is no survey data on Julian days 17 and 72. rimenl. rs rens Prvn of erted he e mu redbealtibeam sy pthottTh g fial rep h usn t initht y thErral s p mi 149

PAGE 165

Figure 91. Comparison of the predicted (dashed), observed (blue), and tilt-corrected observed (red) percent burial for the F10 mine over the course of the experiment. Tilt-corrected values have been horizontally offset for clarity. Error bars represent the 5-centimeter uncertainty inherent in the multibeam system. 150

PAGE 166

Summary of Results High-resolution multibeam surveys were performed over the Indian Rocks Beach mine burial experiment site in order to observe in situ scour and burial of th These data were then used to test the validity of the VIMS 2D burial modelestimates the amount of burial for cylindrical mines by predicting scour forased on the Whitehouse-Soulsby equation. The observational data show that for fine sands (mean grain size 0.18 mm), cylindrical mines were at least 74.5% buried within two months of deployment; with four of the eight mines showing a burial of 96% or greater. The two mines deployed in the coarse sand site were 12.8% (F7) and 40.4% (F8) buried within two months of deployment. Although the mines deployed in fine sand showed a significant amount of burial in terms of subsidence below the ambient seabed, there was very little infilling of the scour pits or covering of the mines with sand. As he ability to detect these mines with side-scan sonar was actually enhanced. Despite the lesser degree of burial for the two coarse site mines, it is possible that they would not be detected in side-scan surveys due to the presence of rippled bedforms of nee size commonly found in shallow water coarse sediments. The VIMS 2D burial model developed by Carl Friedrichs and Art Trembanis at the Virginia Institute of Marine Science was tested using the multibeam surveys of the mines. The model performed well for the mines deployed in fine sands with the exception of the A4 and F6 mines. These two mines both showed an anomalous shallowing during the last multibeam survey of the experiment. Despite this, the performance of the model with the other mines illustrates that it sufficiently predicts burial in areas of fine sand. The anomalous shallowing is likely related to some other unknown source error because e mines., which mation ba result, tarly the sam 151

PAGE 167

it is difficult to imagine a process ine to rise in an absolute sense l. of in mind that the observed ambien5 that would cause the m with respect to the MLLW chart datum used. In the case of coarse sands; however, the model did not perform nearly as welFor both of the mines deployed in the coarse sand site, the model significantly over-predicted the amount of burial. An anomalous 12-centimeter shallowing was observed forthe F8 mine during the March 13th survey; however, this did not appear to be the causethe models inability to adequately predict burial in coarse sands. As is seen in the Martas Vineyard mine burial study site (Traykovski et al., 2005), it is believed that the cause is the presence of rippled bedforms around the mines, which are not accounted for in the model. These ripples directly affect morphodynamics of the seafloor and thus can affect rates of mine burial. This issue is being addressed in current and future modeling efforts (Trembanis et al., 2005). Other possible sources of error involve the ambient seafloor depth around the mine. The model assumes that the local seafloor remains constant throughout the model run; however, localized erosion and deposition over the course of the experiment were observed in the multibeam data. It is also important to keep t seafloor depth around the mine was an approximation based on the average of 3measurements from the multibeam data. All references to localized deposition and erosion refer to the area around the mine and just outside the scour pit. A discussion of how changes in seafloor elevation were calculated within the grids is included in Appendix A. 152

PAGE 168

rial in mines deployed in the rippled scour depression showed little to s f 10 e to the combined 5-centimeter vertical accuracy of the multibeam surveys. Although difficult to estimate, the surface area accuracy is assumed to be 2 meters based on the combined 1-meter positional accuracy of the multibeam; however, Chapter 4 Analysis of Mine Scour Introduction Scour formed around and under mines is the driving mechanism for mine bunon-cohesive fine sand. The scour process is the basis of mine burial probability models.Therefore, an understanding of the temporal and spatial scales of mine scour is essential. This chapter is an analysis of the morphology and hypsometry of the scour that formed around the mines deployed in both the deep and shallow fine sand sites during the IRB mine burial experiment. The two no scour due to the coarse-sized grains and rippled morphology, and thus werenot included in this analysis. Methods For the eight mines deployed in fine sands, bathymetric finite difference grids were created by subtracting the first survey over the mine from the final survey. Thiresulted in a difference grid showing areas of deposition (positive values) and erosion (negative values) between the two surveys. These grids have a vertical accuracy ocentimeters du 153

PAGE 169

the number may be overly conservative. The 0-mter contour on each grid represents the level of zero change in seafloor elevation between the first and last surveys over the mine. The scour pit was then contoured in 1er intervals and the area within each contour was calculated. The firwest contour that formed a closed polygon around the pit. Two cross-sections were taken across each scour pit, a long profile passing through the deepest points of the pit and a e acted from the March 13th survey grid. The scour around the 1 mine formed a pit with an approximate surface area of 20.03 meters2 and volume of 3.67 meters3 (Figs. 92 & 93). The pit wathe and 10th e 0-centimet st contour of the scour was defined as the shallo short profile cutting through the shallowest points. All analyses were done using ArcGIS 9. Hypsometry graphs based off the depth and area of each contour were made in EXCELfor each scour pit. Scour Analysis The A1, A2, A3, and A4 mines were deployed in the shallow fine sand site and were 2.03 meters long with a diameter of 0.53 meters. For these mines, the grid from thJanuary 10th survey was subtr A s divided into 8 contour intervals ranging in depth from -0.08 meters to -0.88 meters. The actual maximum depth measured within pit was -0.90 meters; however, the volume of the pit between the -0.88-meter contour the -0.90 meter maximum depth was negligible (3.7 x 10-6), so the maximum depth forpurposes of this analysis was considered -0.88 meters. The shallowest contour of the scour pit was -0.08 meters, indicating an erosion of the seafloor between the Januaryand March 13th surveys that was not contained within the scour pit. The long cross-section (between points C and D on the grid) passes through the deepest point within the 154

PAGE 170

Figure 92. A1 scour pit. Contours are in 10-cm increments. Yellow outline denotes last position th of mine as observed in the March 13 survey. 155

PAGE 171

156 156 A1 Scour Pit Hypsometry20.038.924.701.990.020.650.250.090.005.0010.0015.0020.0025.00-0.08-0.18-0.28-0.38-0.48-0.58-0.68-0.78depthcumulative areaFigure 93. A1 scour pit hypsometry.

PAGE 172

sc our, ~ -0.88 meters, and was approximately 6.5 meters long. The short cross-section (between points A and B) is approximately 4.5 meters long and reached a depth of -0.36 meters. A pit approximately 13.98 meters2 in surface area and 1.91 meters3 in volume was formed by the scour around the A2 mine (Figs. 94 & 95). The pit was divided into 3 contour intervals ranging in depth from -0.06 to -0.26 meters. Maximum depth measured within the pit was -0.35 meters. The shallowest contour of the scour pit has a negative value, -0.06, indicating that there was regional erosion over the grid between thenuary 10th and March 13th surveys that extended beyond the scour pit itself. The long cross-section (profile C-D) was approximately 6.5 meters long and -0.35 meters at its deepest point. The short cross-section (profile A-B) was approximately 2.6 meters long and -0.12 meters at its deepest point. Scour around the A3 mine was complicated by the presence of the two quadpods and one spider deployed in the same area. Scour formed around all the equipment and merged into one. The most pronounced scour was around the A3 mine, and formed a pit with an approximate surface area of 8.60 meters2 and a total volume of 1.64 meters3 (Figs. 96 & 97). The pit was divided into 3 contour intervals, ranging in depth from -0.16 meters to -0.36 meters. The long cross-section (profile C-D) was approximately 3.75 meters long with a maximum depth of 0.46 meters. The short cross-section (profile A-B) was roughly 2.4 meters long and reached a depth of -0.35 meters. The A4 mine formed a scour pit of approximately 11.80 meters2 in surface area nd 1.54 meters3 in total volume (Figs. 98 & 99). Three contours divided the pit, ranging in depth from -0.10 meters to -0.30 meters. The long cross-section (profile C-D) was Ja a 157

PAGE 173

Figure 94. A2 scour pit. Contours are in 10-cm increments. Yellow outline denotes last positionth of mine as observed in the March 13 survey. 158

PAGE 174

159 159 A2 Scour Pit Hypsometry13.983.062.060.002.004.006.008.0010.0012.0014.0016.00-0.06-0.16-0.26depthcumulative areaFigure 95. A2 scour pit hypsometry.

PAGE 175

Figure 96. A3 scour pit. Contours are in 10-cm increments. Yellow outline denotes last position of mine as observed in the March 13th survey. 160

PAGE 176

161 161 A3 Scour Pit Hypsometry8.605.292.540.001.002.003.004.005.006.007.008.009.0010.00-0.16-0.26-0.36depthcumulative areaFigure 97. A3 scour pit hypsometry.

PAGE 177

Figure 98. A4 scour pit. Contours are in 10-cm increments. Yellow outline denotes last positio n of mine as observed in the March 13th survey. 162

PAGE 178

163 163 A4 Scour Pit Histogram11.802.611.000.002.004.006.008.0010.0012.0014.00-0.10-0.20-0.30depthcumulative areaFigure 99. A4 scour pit hypsometry.

PAGE 179

roughly 5.1 meters long with a maximum depth of -0.35 meters. The short cross-section (profile A-B) was approximately 3.0 meters long and reached a maximum depth of -0.17 meters. The F5 and F6 mines were deployed along with the acoustic mines in the shallow fine sand. They had a length of 1.499 meters and a diameter of 0.47 meters. The first survey over these mines was on January 13th, and the final survey was March 13th. The scour around the F5 mine formed a pit with a surface area of approximately 12.53 meters2 and a volume of 2.04 meters3 (Figs. 100 & 101). The pit was divided into 3 contours, ranging in depth from -0.01 meters to -0.21 meters. The long cross-section (profile C-D) was roughly 4.2 meters long and reached a maximum depth of -0.30 meters. The short cross-section (profile A-B) was approximately 3.3 meters long and reached a depth of -0.21 meters. Scour around the F6 mine formed a pit with an approximate surface area of 10.60 meters2 and a volume of 1.84 meters3 (Figs. 102 & 103). Seven counters divided the pit, ranging in depth from 0.10 to -0.50 meters. The first contour was positive, indicating that there was a deposition of sediment between the January 13th and March 13th surveys around the mine. For the F6 mine, the short cross-section (profile A-B) passed through the deepest point in the pit. Profile A-B was approximately 3.20 meters long and reached a depth of -0.58 meters. The long cross-section (profile C-D) was roughly 3.40 meters long and had a maximum depth of -0.35 meters. The F9 and F10 mines were deployed in the deep fine sand site during the IRB experiment (Figs. 104 &105). For both mines, the first survey was on January 13th. The 164

PAGE 180

Figure 100. F5 scour pit. Contours are in 10-cm increments. Yellow outline denotes last position of mine as observed in the March 13th survey. 165

PAGE 181

166 166 F5 Scour Pit Hypsometry12.535.941.970.002.004.006.008.0010.0012.0014.00-0.01-0.11-0.21depthcumulative areaFigure 101. F5 scour pit hypsometry.

PAGE 182

Figure 102. F6 scour pit. Contours are in 10-cm increments. Yellow outline de notes last position of mine as observed in the March 13th survey. 167

PAGE 183

168 168 F6 Scour Pit Hypsometry10.604.981.410.750.100.200.400.002.004.006.008.0010.0012.000.100.00-0.10-0.20-0.30-0.40-0.50depthcumulative areaFigure 103. F6 scour pit hypsometry.

PAGE 184

169 Figure 104. F9 scour pit. Contours are in 10-cm increments. Yellow outline denotes last position of mine as observed in the March 13th survey.

PAGE 185

F9 Scour Pit Hypsometry5.795.764.350.001.002.003.004.005.006.007.000.300.200.10depthcumulative areaFigure 105. F9 scour pit hypsometry. 170

PAGE 186

sc our around the F9 mine formed a pit roughly 5.79 meters2 in surface area and 1.59 eters3 in volume. The pit was divided into 3 contour intervals, ranging in depth from 0.30 meters to 0.10 meters. The contours all had positive values, indicating that deposition occurred over the area before the pit started to form. The maximum depth within the pit should have been zero, since the seafloor cannot accrete underneath the mine. The actual maximum depth measured within the pit was 0.03 meters, well within the 20-centimeter accuracy of the grid. The long cross-section (profile C-D) was roughly 3.8 meters long and reached the maximum depth of 0.03 meters. The short cross-section (profile A-B) was approximately 2.0 meters long and had a depth of 0.22 meters. The survey of March 13th did not capture the F10 mine, so the Februar survey grid was used along with the January 13th grid for the finite difference. The scour around the F10 mine formed a pit with an approximate surface area of 6.53 meters2 and a volume of 1.06 meters3 (Figs. 106 & 107). Three contours divided the pit, ranging in depth from 0.15 meters to -0.05 meters. The long cross-section (profile C-D) was approximeters long and -0.10 meters deep. The short cross-section (profile A-B) was roughly 2.4 meters long and about 0.01 meters deep. Summary of Analysis With the exception of A1 and F6, the scour around the mines formed pits roughly 0.30 meters deep contained around the mine. The A1 pit was approximately .80 meters at its deepest point; however, 99% of the pit was contained within the first 0.40 meters. The 6 scour formed a pit approximately 0.58 meters deep, with ~ 98% of the pit contained ithin the first 0.40 meters. The deepest scour occurred along the flat ends of the mines, m y 6thmately 3.5 F w 171

PAGE 187

Figure 106. F10 scour pit. Contours are in 10-cm increments. Yellow outline denotes last position of mine as observed in the February 6th survey. 172

PAGE 188

173 173 F10 Scour Pit Hypsometry6.533.010.520.001.002.003.004.005.006.007.001.002.003.00cumulative areaFigure 107. F10 scour pit hypsometry. depth

PAGE 189

w hile the shallowest scour tended to occur along the sides. In general, scour around the coustic mine formed the largest pits, with an average length of 5.5 meters and an verage width of 3.1 meters. The scour around the optical mines formed pits with an average length of 3.9 meters and an average width of 2.8 meters. The largest scour pit formed around the A1 mine and had a surface area of 20.03 meters2 and a volume of 3.67 meters3. The smallest scour pit formed around the F9 mine and had a surface area of 5.79 meters2 and a volume of 1.59 meters3. a a 174

PAGE 190

Chapter 5 Discussion Over the course of the two month experiment, the 8 mines deployed in both shallow and deep fine sand showed a substantial observed burial, upwards of 74.5% (Figs. 108 & 109). Four of these min e s had an observed burial of 96.2% or greater. Mines eployed in the coarse sand site showed significantly less burial, and appeared to scour to the bed until they presented approximately the same relief as the surrounding rippled edforms (Fig. 110). These results are similar to those observed at Marthas Vineyard uring the winter 2003 to spring 2004 MVCO mine burial experiment. The final ultibeam survey over the MVCO site occurred approximately 7 months after eployment. The mines deployed in the fine sand sites completely buried with no traces f them were evident in the multibeam data. The mines deployed in the coarse sand site uried until they presented the same hydrodynamic roughness as the wave-orbital ripples ayer et al., 2005; Traykovski et al., 2005). In a laboratory study by Voropayez et al. (2002), scour of cylindrical objects was epressed by the presence of ripples and burial did not occur. Periodic burial of the ylinders was observed when the ripple crest overtook the cylinder; however, this only occurred when the ripple heights were comparable or greater than the cylinder diameter. In the case of the MVCO and IRB experiments, observed ripple heights were significantly less than the mine diameter. d in b d m d o b (M d c 175

PAGE 191

176 Figu10 o burial rates for mins in thalfin re 8. Comparisonfee shlow e site. 176

PAGE 192

Figure 109. Comparison of burial rates for mines in the deep fine site. 177

PAGE 193

Figure 110 Comparison of burial rates for mines in the coarse site. 178

PAGE 194

Despite the significant amount of burial seen at the two fine sand sites during this experiment, there was very little observed infilling of the scour pits and the mines remained relatively uncovered by sediment. This is in contrast to what was seen at the MVCO fine sand sites, where higher energy environments and a greater supply of muds resulted in scour pit infilling within two months, and complete burial and cover of the mines in seven months (Traykovski et al., 2005). Sediment infilling of scour pits is quite important, as it can signify the difference between mines that can and cannot be readily detected by side-scan sonar. The mines deployed in the fine sand sites off Indian Rocks Beach became more visible in the side-scan imagery over time as a result of the surrounding scour pits, which served to form greater targets. In the case of the MVCO fine sand sites, the mines became completely covered with sediment within seven months and no traces of them were evident in rotary side-scan images (Traykovski et al., 2005). It should be noted; however, that while the MVCO experiment lasted seven months, the experiment off Indian Rocks Beach only lasted two. Four of the 10 mines deployed during the IRB experiment (A1, A4, F6, and F8) showed an anomalous shallowing between the February 6th and March 13th surveys. The exact cause of this shallowing is not known; however, there are several possibleexplanations. When comparing the depth of the mine between two surveys, it ist to note that the vertical uncertainty of the multibeam becomes combined. Therefore, the mine depth from one survey can fall within a 10-centimeter range of the mine depth from another survey, even if the mine itself does not move. It is also possible that the shallowing may be related to some unknown source of error related to multibeam system parameters or sound velocity profile used by the multibeam system to calculate depth importan 179

PAGE 195

during the survey. Incorrect heave settings for the POS MV, the positioning and attitude system used during the surveys, may also be responsible for this shallowing. Often, errors in observed depth in multibeam bathymetry are associated with errors in the tide record used to reference the data to a water level datum. A constant erroover the whole tide record would result in errors over the whole multibeam survey. Thiis not seen in the data from Indian Rocks Beach. Although there are 7 instances of mine shallowing seen over the course of the experiment, there is no set pattern between them. Four of these shallowing events (mines F8, A1, A4, and F6) occurred between the February 6th and March 13th surveys. Mines A1, A4, and F6 were deployed in thefine sand site. A1 and A4 were deployed adjacent, approximately 23 meters apart (Fig. 2). For these two mines, the amount of shallowing of the mine was approximately the same amount of shallowing observed in the ambient seafloor depth (See Tables 3 and 5). This may indicate an error in the tide record confined to the period that the survey passover these two mines. The F6 mine was located farthe r s shallow ed r north from A1 and A4, resting 2 ent th approximately 39 meters east of A2 (Fig. 2). There was no observed shallowing of the Amine, nor the A3 or F5 mines, during the March 13th survey; furthermore, the amount of shallowing observed for the F6 mine was roughly half that observed for the ambiseabed around the mine (Table 7). This indicates that the error most likely is not associated with the tide record; however, it is possible that tide errors affecting the ambient seafloor depth are masked by actual localized accretion around the mine. The F8 mine showed the greatest amount of shallowing between the February 6and March 13th surveys, 12 centimeters, that was closely matched by the observed 11-centimeter shallowing of the ambient seafloor depth. The F7 mine; however, also 180

PAGE 196

deployed in the coarse sand site roughly 50 meters away did not show any shallowing of the mine during this survey. This would indicate that if an error in the tide reco rd was ver r m r me mplituat this responsible, it would have to have been confined to the time after the survey passed othe F7 mine, sometime after 0500 on March 13th, 2003 (GMT). A 7-centimeter shallowing of the F7 mine did occur between the January 13th and January 17th surveys; however, during this period there was an observed deepening of the ambient seafloor depth around the mine by 4 centimeters. In short, while an error in the tide record is a possible source of error, the mine shallowing events are more likely a result of othefactors, such as vertical un c ertainty in the multibeam system itself. It should be noted that the 5-centimeter vertical uncertainty of the multibeam system assumed in this study was an estimate. Kongsberg Simrad lists the EM 3000s vertical uncertainty as 5 to 10 centimeters (RMS error) dependant on depth. The average depth of the Indian Rocks Beach study site is ~13 meters, which falls into the shallow water range. Furthermore, vessel positioning was handled by a TSS POS/MV 320 systewith real time kinematics (RTK) using the Clearwater Beach Adams Mark Hotel as a base station. This combined system provides positioning accuracy on the order of 10 centimeters, and roll, pitch, a yaw measurements accurate to 0.02. The positioning accuracy is extended to 1-meter based on other installation parameters and watecolumn properties. The POS/MV system with RTK capabilities also provides real-tiheave correction with a measurement accuracy of 5 centimeters or 5% of the heave ade (whichever is greater) for periods up to 20 seconds. As a result of this, and in an attempt to not mask the entire multibeam signal in noise, a vertical uncertainty of 5 centimeters was used. The anomalous shallowing of some of the mines suggests th 181

PAGE 197

estimate may be overly optimistic and that a more realistic uncertainty is closer to 10 centimeters. In addition, this error estimate does not include error propagation from thepressure sensor used to measure tides, or the NOAA station used in obtaining the tide record. Work is in progress to better determine the vertical uncertainty by completing a full propagation of all system component errors (Wolfson et al., manuscript inpreparation). One other issue with the multibeam data was the blurriness of some of the imagdespite the target detection mode being turned off. Many of the initial surveys over thmine were blurred, due to the beam mode being set to target detectio es, e n. Target detection r for causes a widening of the beams from 1.5 by 1.5 to 4.0 by 4.0, allowing for greater detection capabilities; however, it can also cause a distortion of the target itself, a factonot discovered until the data was processed. The target detection mode was turned off for the survey of January 17th 2003 (with the exception of the A2 mine), and remained off all subsequent surveys. Despite this, 3 of the 8 mines imaged on January 17th (the survey did not cover the coarse sand site) appeared blurry and distorted. The reason for this isnot clear and may be related to bubble sweeps under the sonar or other material in the water distorting the acoustic beam. Additional potential causes for distortion follow. It became apparent during processing that gridding the multibeam data at a resolution of 20 centimeters was pushing the capabilities of sonar. The EM 3000 multibeam has a beam width of 1.5 at nadir, giving it an effective footprint of ~28 centimeters in 13 meters of water depth (the sonar is mounted ~2 meters below the watersurface). The across track beam spacing is 0.9 at nadir, giving an overlap of 0.6, which equates to ~11 centimeters in ~13 meters of water depth. Therefore, a horizontal gridding 182

PAGE 198

resolution of 20 centimeters should be reasonable for this study. Beam width and spacinhowever, increase as the beam pointing angle (angle of the beam with respect to the sonahead) increases (Table 11). Therefore, there is a wider separation of the beams in the outer part of the swath and a gridding resolution of 20 centimeters may be too tight to properly image objects that fall within this area. In som g; r e cases, the multibeam images depicted fairly accurate dimensions for the ines. e n ery .5 t mIn others; however, the mines did not image clearly at all. This is in contrast to the results from the MVCO experiment, where multibeam data not only showed correct dimensions of the mines but could also depict the tapered end of the FWG optical mines.Multibeam surveys for the MVCO study were completed using a Reson 8125 sonar. Thsonar operates at a frequency of 455 kHz with a sub-decimeter resolution. In his article submitted for publication to the Journal of Ocean Engineering, Mayer et el. (2005) states that distortion of true mi n e diameter by the multibeam sonar may be due to the influence of neighboring cells on small targets during the gridding process. It is possi bl e, therefore, that this is the case with the IRB data as well, and may explain some of distortion seen ithe images, especially considering the lower 300 kHz frequency used at the IRB site. The Simrad EM 3000 multibeam sonar has a maximum ping rate of 20 Hz in v shallow water. Average vessel speeds during the IRB surveys ranged from ~ 2.5 9knots. It is possible that at a depth of ~ 13 meters, the observed ping rate of 10 Hz(limited by the two-way travel time from the sonar to the furthest point imaged) is nosufficient to detect the mines at boat speeds of up to 9.5 knots. Indeed, the surveys that imaged the mines most clearly were conducted at vessel speeds of about 6 knots or less. Vessel speed affects the along track distance between consecutive pings (Table 12). At an 183

PAGE 199

water depth 1.5 beam width 2.1 beam width 3.0 beam width vertical (0 degrees) 0.9 beam spacing 45degrees 1.3 beam spacing 60 degrees 1.8 beam spacing 10 meters 0.21 0.13 0.83 0.36 1.68 1.0 13 meters 0.28 1.14 2.31 0.17 0.50 1.38 20 meters 0.47 1.87 3.77 0.28 0.82 2.26 vessel speed (knots) along track beam spacing (at 10 Hz) 2.5 0.13 3.5 0.18 ** 5.5 0.28 6.5 0.33 pointing angles. Beam pointing angle is with respect to the sonar head. Top number is the beamfootprint (width) in meters, bottom number is the beam spacing in meters. These numbers assumthe fact that the sonar was pole mounted to the vessel during surveys, and thus was approximately 2 4.5 0.23 7.5 0.38 8.5 0.43 9.5 0.48 Table 12. Along track beam spacing for various vessel speeds. All numbers are in meters. Alltrack beam spacing becomes greater than the beam footprint in 13 meters of water depth. Refer to Table 11. Beam footprint and beam spacing along the seafloor for various depths and beam e a flat seabed. Water depth is depth of the seafloor below the water surface. Calculations are based on meters below the actual water surface. Refer to Appendix B for a description of the equations used. calculations are based on a ping rate of 10 Hz. At vessel speeds of 5.5 knots or greater, the along Appendix B for a description of the equations used. 184

PAGE 200

average vessel speed of ate s andistance of approximateweeented parallel toy track, thi in a maings osurface. Iters of wateam footprint is ~ 28 centiming a flat bottom). In a water depth of 13 meters at vessels speeds of 5.5 knots and greater, the along track beam spacing is greater than the beam footprint (0.28 meters), indicating that ground coverage is not 100 percent. This may help explain why the mine did not image durin durinmeters at 10 Hz). Furthermore, the EM3000 beam spacing is controlled by fast Fourier transform (FFT) beam forming, causing the angular spacing of the beams to increase with distance from nadir. At naa9 apart; however, this grows to 1.8 at 60 from nadir. As a result, target detection capabilities of the sonar degrade with as the angle of incidence increases, which may help to explain the distortion of some of the images (Table 11). The multibeam data froe IRB experiment were used to test the VIMS 2D mine burial model. The resultsseen in the comparison of the MVCO data with the model (Trembanis et al., 2005; Mayer et al., 2005). In the case of mines located in finexpeover-predicts the amount of burial. In coarse sands, it has been shown that the mines bury until they present approximately the same hydrodynamic roughness as the surrounding orbital ripples (Mayer et al., 2005; Traykovski et al., 2005). The current model does not 9.5 knots and a ping rly 0.48 meters bet of 10 Hz, there in pings. For an FW along track G optical mine ori the surve s would result ximum of 3 p n the mine n ~ 13 me r depth, the be eters (assum g the January 20th 2003 survey over the F5 and F6 mine, where average vessel speedg the survey was approximately 6 knots (corresponding to a ping spacing of 0.3 dir the beam sp cing is approximately 0. m th mirrored those e sand, the model sufficiently predicts percent burial over the course of the riment. In the case of mines deployed in coarse sand; however, the model greatly 185

PAGE 201

address bedform evolution and migration, which appears to be the cause of the mopoor performance in coarse sand. Another possible source of error involves how the model handles ambient seafloor depth. The model assumes that the local seafloor dearound the mine remains constant throughout the model run; however, multibeam data show localized erosion and deposition over the course of the experiment. Scour analyses were performed for each of the mines deployed in fine sands, in order to better understand how the scour formed. The mine was carefully edited out of thdata, and a difference grid was created using the first and last survey. This allowed for a better understanding of how scour formed around the mines over the course of the experiment. The greatest amount of scour occurred along the ends of the mines, while thshallowest scour tended to occur along the sides. On average, scour around the mines formed pits ~ 0.30 meters deep. Little to no infilling of the scour pits was observed over the course of the experiment. This is in contrast to the results seen at MVCO, where infilling was observed to occur in response to increased wave events (Traykovski et al., 2005). A recommendation for future dels pth e e work would be to create difference grids for each hod e inert ip e survey, which would allow for a more detailed examination of the scour. Another metwould be to use IVS Fledermaus to explore 3D images of the scour, as described in Mayer et al. (2005). This study shows that a Kongsberg Simrad EM 3000 can adequately imagand instrumented mine-like cylinders near NADIR at a depth of ~ 13 meters at slow shspeeds. These data provide in situ observations of scour and burial around the mines and are useful in testing the VIMS 2D model. While the model behaves well for mines in finsand, it cannot sufficiently predict burial in coarse sand in the presence of rippled 186

PAGE 202

bedforms. This indicates that wave-orbital bedform evolution cannot be ignored in mine burial models an issue currently being addressed by modelers. Further tests of the Kongsberg Simrad EM 3000 multibeam system for shallow water target detection is recommended in order to attempt to determine the cause for both the blurriness of someof the images and the anomalous shallowing of some of the mines, by usin g independent fixed e levation markers hammered through ~ 3 meters of sediment and into the seafloor limestone to serve as a ground truth elevation datum. 187

PAGE 203

Chapter 6 Summary Sea mines have been used in nearly every conflict since the American Revolutionary War. They are fairly simple to build and deploy, but require more advanced equipment, significant cost, and considerable risk to locate and counter (Griffen t al., 2003). One of the biggest issues in mine countermeasures today is the ability to detect buried mines on the seabed, an issue currently being address by the ONR Program in Mine Burial Prediction. One of the main objectives of this program is to better understand the temporal and spatial scales of mine burial and scour. Another objective is to develop predictive models of mine burial that can be used to determine whether areas should be hunted, swept, or avoided altogether. This study helps to address these objectives by using repeat high-resolution multibeam bathymetry data to monitor in situ scour and burial of inert and instrumented mines deployed off Clearwater, Florida. The multibeam data are used to test the VIMS 2D burial model. In addition, a method for extracting a vertical reference datum from pressure sensor data is presented. The multibeam data show that for cylindrical mines deployed in fine sands (mean grain size 0.18 mm) the amount of burial was at least 74.5% two months after deployment, with half of the mines showing a burial of 96% or greater. For the two mines deployed in coarse sand, the maximum amount of burial reached 40.4% within two months of deployment. In general, it appears that mines in coarse sand scour until they present the same hydrodynamic roughness as surrounding rippled bedforms. Despite the e 188

PAGE 204

significant amount of burial seen at the two fine sand sites during this experiment, there was very little observed infilling of the scour pits and the mines remained relatively uncovered by sediment. The VIMS 2D burial model is tehe multibeam surveys of the mines. he mo exception not be tly e do not ater field sted using t Tdel performs well for the mines deployed in the fine sand sites, with theof the A4 and F6 mines. These two mines show an anomalous shallowing that canaccounted for by the vertical uncertainty of the multibeam. This shallowing is not understood and the cause is not clear. Additional testing of the target detection capabilities of the multibeam sonar are needed to further explore this issue. Despite this, the performance of the mine with the remaining 6 mines illustrates that is sufficienpredicts burial in areas of fine sand. The model did not perform well for the mines deployed in coarse sands, where rippled bedforms complicated the near bottom hydrodynamics. As described in thMarthas Vineyard publications (Mayer et al., 2005; Traykovski et al., 2005), mines in coarse sediment scour until they present roughly the same hydrodynamics as the surrounding rippled bedforms. Ripples directly affect the morphodynamics of the seafloor and can thus affect rates of mine burial. Existing mine burial models account for bedform evolution, an issue currently being addressed by modelers. Scour around mines is the driving force behind mine burial at the Clearwsite and is the basis of mine burial probability models being applied there. Scour analyses of the mines at Clearwater indicate the most prevalent scour occurs at the ends of themines, while the shallowest scour occurred along the sides. In general, scour formed pits roughly 0.30 meters deep around the mines within two months of deployment at the fine 189

PAGE 205

sand locations. Significant scour pits did not form around mines at coarse sand sites. Although infilling of the scour pits was observed during the Marthas Vineyard experiment (Mayer et al., 2005; Tray kovski et al., 2005), there is no evidence of infilling urial redict at the Indian Rocks Beach site. Overall, the results of this study show the mines are clearly distinguishable in the multibeam data, allowing for observed amount of scour and burial to be obtained. Furthermore, these data show the VIMS 2D burial model can sufficiently predict bfor cylindrical mines in fine sand but that additional complexity is required to pburial at the coarse site. 190

PAGE 206

References Collins, W.T., and Galloway J.L (1998), Seabed classification with multibeam bathymetry. Sea Technology 39(9):45 49. Collins, W.T., and Preston, J.M. (2002), Multibeam seabed classification. International Geological Survey Open-File Report 99-417 (CD-ROM). Griffen, S., Bradley, J., and Richardson, M.D. (2003), Improved subsequent burial s. Sea Technology 44(11): 40 -44. ughes Clarke, J.E. (1998), Detecting small seabed targets using high-frequency multibeam sonar. Sea Technology 39(6):87 90. Friedrichs, C. (2001), A review of the present knowledge of mine burial processes. Office of Naval Research, Award No. N00014-01-1-0169. Gardner, J.V., Butman, P.B., Mayer, L.A., and Clarke, J.H. (1998), Mapping U.S. continental shelves. Sea Technology 39(9):10 17. Locker, S.D., and Hine, A.C. (2003) ROV video survey on March 12, 2003 for the Winter 2003 ONR Mine Burial Experiment, Indian Rocks Beach, FL (Also including still images for February and March, 2003). [Digital Versatile Disk] Locker, S.D., Hine, A.C., Wright, A.K., and Duncan, D.S. (2002), Sedimentary framework of an Inner continental shelf sand-ridge system, west-central Florida. AGU 83(47), Fall Meet. Suppl. Abstract: OS61A-0188 Mayer, L.A., Raymond, R., Glang, G., Richardson, M.D., Traykovski, P., and Trembanis, A. (2005), High-resolution mapping of mines and ripples at the Marthas Vineyard Coastal Observatory, submitted Journal of Ocean Engineering. Naar D.F., and Donahue, B.T. (2002) High-resolution multibeam survey of ONR mine burial and scour study area near Clearwater, Florida. EOS, Transactions, AGU Volume 83, Number 47, F692. Pohner, F. (1990), Processing of multibeam echosounder data. The Hydrographic Journal, 50: 10 p. Ocean Systems 6(4):12 15. Gelfenbaum, G., and Guy, K. (2000), Bathymetry of west-central Florida: U.S. instrumented mine H 191

PAGE 207

Richardson, M., Valent, P., Briggs, K., Bradley, J., and Griffin, S. (2001) NRL mine burial experiments. Proceedings of the Second Australian-American Joint Conference on Technology of Mine Countermeasures, Sydney Australia, 27-29 March.. Richardson, M.D., and Briggs, K.B., (2000), Seabed-Structure interactions in coastal y and l, Monterey California, 13-16 March 2000. raykoyer, L., and Irish, J.D. (2005), Mine burial experiments at the Marthas Vineyard Coastal Observatory, submitted to Journal rembanis, A.C., Friedrichs, C.T., Richardson, M.D., Traykovski, P.A., Howd, P.A., cour: assachusetts. submitted toJournal of Ocean Engineering. Voropar, D.L. (2003), Burial and scour around short cylinder under progressive shoaling waves. Ocean Engineering 30: olfson, M.L., Naar, D.F., Howd, P.A., Locker, S.D., Donahue, B.T., Friedrichs, C.T., eam a, to Predictions of Wave-Induced Burial. Submitted to Journal of Ocean Engineering. sediments. Proceedings of the 4th International Symposium on Technolog the Mine Problem, Naval Postgraduate Schoo Tvski, P., Richardson, M.D., Ma of Ocean Engineering. T Elmore, P., and Wever, T. (2005), Predicting seabed burial of cylinders by s Application to the sandy inner shelf off Florida and M yev, S.I., Testik, F.Y., Fernando, H.J.S., and Boye 1647-1667. W Trembanis, A.C., Richardson, M.D., Wever, T. (2005), Comparison of MultibObservations of Mine Scour Near Clearwater, Florid 192

PAGE 208

Bibliography Conference, Mobile, Alabama, April 26 May 1 1999. tliers eng, X., and Zielinski, A. (1999), Precise multibeam acoustic bathymetry. Marine r investigating seafloor processes in the coastal zone and on the continental shelf. Marine Geophysical Researches 18: 607-629. Pratsonn sonar and multibeam bathymetry data. Marine Geophysical Researches 18: 601-605. Brissette, Lt(N) M.B., and Hughes Clarke, J.E. (1999), Side scan versus multibeam echosounder object detection: a comparative analysis. Proceedings of the U.S. Hydrographic Du, Z., Wells, D., and Mayer, L. (1996), An approach to automatic detection of ouin multibeam echo sounding data. The Hydrographic Journal (79): 19-23. G Geodesy 22: 157-167. Hughes Clarke, J.E., Mayer, L.A., and Wells, D.E. (1996), Shallow-water imaging multibeam sonars: a new tool fo L.F., and Edwards, M.H. (1996), Introduction to advances in seafloor mapping using sidesca 193

PAGE 209

Appendix A Calculation of Ambient Seafloor Change In order to determine if these changes in seafloor elevation were either a local or widespread regional change, five by five meter grids were taken from the northeast corner of each of the 18 by 18 meter grid. The seafloor depth was averaged over the 5-meter grid and was then compared to the seafloor depth observed around the mine. This analysis was performed for each mine deployed in both the shallow fine and deep fine sand site. The coarse sand site was not included due to the presence of rippled bedforms, which complicated the seafloor morphology. For the A1 mine, the ambient seafloor changes seen around the mine during the six surveys were also seen in the 5-meter grids, a couple centimeters (Figs. 111 ). The greatest discrepancy between the localized seafloor depth around the mine and the average seafloor depth seen within the 5-meter grid, 6 centimeters, occurred on January 20th. The seafloor around the mine had an average depth of 12.82 meters, while the average depth within the smaller grid was 12.88 meters. The greatest amount of seafloor change within the grid was seen between the January 13th and January 17th surveys, with an observed erosion of 16 centimeters. The seafloor analysis for the A2 mine also showed an agreement between changes seen around the mine and those observed within the 5-meter grid (Figs. 117 122). 194Again, the greatest discrepancy between the average seafloor depth seen around the mine and that seen within the 5-meter grid was 6 centimeters, observed on January 13th. The

PAGE 210

greatest amount of seafloor change within the grid was an erosion of 9 centimeters, which occurred between the surveys of January 10th and January 13th. The greatest discrepancy betwr depth seen around the mine and r both the A2 nuary 13th surveys, which showed an erosion of 9 centimeters. bient seafloor changes seen around the mine during the 29 134). Ton Marunt of seafloor change within the grid was seen between the January 10th and January 13th surveys, with an observed erosion of 12 centimeters. The seafloor analysis for the F5 mine showed a fairly good agreement between changes observed around the mine and those in the 5-meter grid as well (Figs. 135 139). The January 20th, February 6th, and March 13th surveys showed an offset between the average seafloor depth of the 5-meter grid and that around the mine of 4 centimeters each, the greatest offset observed in the F5 analysis. Interestingly, despite the 4-centimeter discrepancy in the actual depth, both the seafloor around the mine and the seafloor in the 5-meter grid remained constant between the February 6th and March 13th surveys. The greatest amount of change within the grid, a sedimentation of 6 centimeters, occurred between the January 17th and January 20th surveys. een the seafloo that seen in the 5-meter grid for the A3 mine was only 3 centimeters, and occurred fo e January 20th and March 13th surveys (Figs. 123 128). As seen with th mine, the greatest amount of seafloor change occurred between the January 10th and Ja For the A4 mine, the am six surveys agreed with those in the 5-meter grids, a couple of centimeters (Figs. 1 he greatest discrepancy between the ambient seafloor depth around the mine and the average seafloor depth seen within the 5-meter grid was 5 centimeters and occurred ch 13th. The greatest amo 195

PAGE 211

The greatest discrepancy between the average seafloor depth seen around the mine and that seen in the 5-meter grids for the F6 mine is only 4 centimeters, and occurred during the March 13th survey (Figs. 140 144). The greatest amount of seafloor change occurred betweewhich showed a depositd 148). The greatest discrepancy between e ambr rveys proximity to the F9 mine, the greatest amount of rge n the February 6th and March 13th surveys, ion of 12 centimeters. The seafloor analyses for the mines in the deep fine site also showed a good correlation between the seabed elevation changes seen around the mines themselves anthose seen in the 5-meter grids. Between February 6th and March 13th, there was an observed deposition of 20 centimeters around the F9 mine. During the same time, a 25-centimeter deposition was seen in the 5-meter grid, indicating that this change was not merely localized around the mine (Figs. 145 thient seafloor depth around the mine and the average seafloor depth of the 5-metegrid was 6 centimeters, and occurred on February 6th. The last of the seafloor analyses were performed on surveys over the F10 mine. Interestingly, the greatest amount of seafloor elevation change within the 5-meter grid, a deposition of 20 centimeters, was seen between the January 13th and January 20th su(Figs. 149 ). This agreed with the greatest amount of elevation change seen in the ambient seafloor around the mine itself, a deposition of 17 centimeters that occurred between the same surveys. Given F10s change would presumably have occurred during the same time. Due to the fact that F10 rested approximately due west of F9, it is possible that this difference represents a labedform moving west to east through the area during that time period. The greatest discrepancy between the average depth of the seafloor within the 5-meter grid and the 196

PAGE 212

ambient seafloor around the mine was 4 centimeters on February 6th, still within the multibeam depth uncertainty of 5 centimeters. For all the mines deployed in both the fine and deep fine sand sites, the changesambient seafloor elevation around the mine were mirrored by the changes seen within t5-meter grids, a few centimeters. in he een f the out half of the is T his indicates that the seafloor elevation changes saround the mines were actual changes occurring across the grid and not just a r e sult omines presence affecting local morphodynamics. Interestingly, ab histograms (20 out of a total of 41) are right-skewed, meaning that the distribution is notsymmetric but leans towards deeper values of seafloor depth. The numbers of histogramswith symmetric and left-skewed (shallow-biased) distributions are about equal, 10 and 11respectively. The reason for this right-skewed trend in the histograms is not clear, but may indicate inappropriate parameter settings in the multibeam sonar causing a bias towards deeper depth values during calculation. Further testing of the multibeam sonar needed to explore this possibility. 197

PAGE 213

198 0510152025-2.71-12.72-12.73-12.74-12Percent Fi re 11 A1 A1nua0tstm1.75-12.7-12.80-12.81-12.82-14 Jary 1h Hiogra6-12.77-12.78-12.79depth 0th hogra. Nothat e distutios ri 2.83-12.8-12.85 AveragStd: 0.02 d. e: -12.77gu1.January 1istmte thribn ight-skewe

PAGE 214

199 A1nua 13Hiam9212.93-12.-9512.96-12.-982.99pth ver -12.St032. A1 aryth hogra. Nothat tdistutios rikewe Jaryth stogr02468101214161820-12.87-12.88-12.89-12.90-12.91-12.-9412.-9712.-1-13.00depercent Aage:93d: 0.Figure 11Janu 13istme the ribn ight-sd.

PAGE 215

A1 January 17th Histogram051015202530-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96depthpercent Average: -12.90Std: 0.02Figure 113. A1 January 17th histogram. Note that the distribution is right-skewed. 200

PAGE 216

A1 January 20th Histogram05101520253035-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94depthpercent Average: -12.88Std: 0.02Figure 114. A1 January 20th histogram. Note that the distribution is right-skewed. 201

PAGE 217

A1 February 6th Histogram02468101214161820-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99depthpercent Average: -12.91Std: 0.02 Figure 115. A1 February 6th histogram. Note that the distribution is normal. 202

PAGE 218

A1 March 13th Histogram0510152025303540-12.75-12.76-12.77-12.78-12.79-12.80-12.81-12.82-12.83depthpercent Average: 12.79Std: 0.01Fig ure 116. A1 March 13th histogram. Note that the distribution is normal. 203

PAGE 219

A2 January 10th Histogram02468101214161820-12.-12.82-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98depthpercent Average: -12.89Std: 0.03Figure 117. A2 January 10th histogram. Note that the distribution is right-skewed. 81 204

PAGE 220

A2 January 13th Histogram02468101214-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00-13.01-13.02-13.03-13.04-13.05-13.06depthpercent Average: -12.98Std: 0.03Figure 118. A2 January 13th histogram. Note that the distribution is left-skewed. 205

PAGE 221

A2 January 17th Histogram0510152025-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00-13.01-13.02depthpercent Average: -12.98Std: 0.02Figure 119. A2 January 17th histogram. Note that the distribution is left-skewed. 206

PAGE 222

A2 January 20th Histogram0246810121416-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98percent depth Average: -12.90Std: 0.03Figure 120. A2 January 20th histogram. Note that the distribution is right-skewed. 207

PAGE 223

A2 February 6th Histogram0510152025-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94depthpercent Average: -12.88Std: 0.02n is right-skewed. Figure 121. A2 February 6th histogram. Note that the distributio 208

PAGE 224

A2 March 13th Histogram0510152025-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94depthpercent Average: -12.87Std: 0.02Figure 122. A2 March 13th histogram. Note that the distribution is right-skewed. 209

PAGE 225

A3 January 10th Histogram02468101214-12.72-12.73-12.74-12.75-12.76-12.77-12.78-12.79-12.80-12.81-12.82-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91depthpercent Average: 12.79Std: 0.03Fig ure 123. A3 January 10th histogram. Note that the distribution is right-skewed. 210

PAGE 226

A3 January 13th Histogram024681012141618-12.80-12.81-12.82-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00depthpercent Average: -12.88Std: 0.03Figure 124. A3 January 13th histogram. Note that the distribution is right-skewed. 211

PAGE 227

A3 January 17th Histogram051015202530-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98depthpercent Average: -12.93Std: 0.02Figure 125. A3 January 17th histogram. Note that the distribution is right-skewed. 212

PAGE 228

A3 January 20th Histogram024681012141618-12.79-12.80-12.81-12.82-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93depthpercent Average: -12.85Std: 0.03d. Figure 126. A3 January 20th histogram. Note that the distribution is right-skewe 213

PAGE 229

A3 Febraury 6th Histogram0510152025-12.74-12.75-12.76-12.77-12.78-12.79-12.80-12.81-12.82-12.83-12.84-12.85depthpercent Average: -12.80Std: 0.02 Figure 127. A3 February 6th histogram. Note that the distribution is left-skewed. 214

PAGE 230

A3 March 13th Histogram0510152025-12.80-12.81-12.82-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92depthpercent Average: -12.85Std: 0.02Figure 128. A3 March 13th histogram. Note that the distribution is right-skewed. 215

PAGE 231

A4 January 10th Histogram02468101214161820-12.74-12.75-12.76-12.77-12.78-12.79-12.80-12.81-12.82-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90depthpercent Average: -12.80Std: 0.02Figure 129. A4 January 10th histogram. Note that the distribution is right-skewed. 216

PAGE 232

217 A4 January 13th Histogram0246810121416-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00depthpercent Average: -12.92Std: 0.03. Figure 130. A4 January 13th histogram. Note that the distribution is normal

PAGE 233

A4 January 17th Histogram0510152025-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00depthpercent Average: -12.93Std: 0.02Figure 131. A4 January 17th histogram. Note that the distribution is left-skewed. 218

PAGE 234

A4 January 20th Histogram024681012141618-12.82-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00depthpercent Average: -12.90Std: 0.03kewed. Figure 132. A4 January 20th histogram. Note that the distribution is right-s 219

PAGE 235

A4 February 6th Histogram05101520253035404550-12.79-12.80-12.81-12.82-12.83-12.84-12.85-12.86depthpercent Average: -12.82Std: 0.01Figure 133. A4 February 6th histogram. Note that the distribution is right-skewed. 220

PAGE 236

A4 March 13th Histogram02468101214161820-12.80-12.81-12.82-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92depthpercent Average: -12.86Std: 0.02Figure 134. A4 March 13th histogram. Note that the distribution is normal. 221

PAGE 237

F5 January 13th Histogram02468101214161820-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00-13.01-13.02-13.03-13.04depthpercent Average: -12.98Std: 0.02Figure 135. F5 January 13th histogram. Note that the distribution is normal. 222

PAGE 238

F5 January 17th Histogram0510152025-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00-13.01-13.02-13.03-13.04depthpercent Average: -12.99Std: 0.02Figure 136. F5 January 17th histogram. Note that the distribution is left-skewed. 223

PAGE 239

F5 January 20th Histogram0510152025-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00-13.01depthpercent Average: -12.93Std: 0.02Figure 137. F5 January 20th histogram. Note that the distribution is right-skewed. 224

PAGE 240

F5 February 6th Histogram0510152025-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95depthpercent Average: -12.91Std: 0.02Figu re 138. F5 February 6th histogram. Note that the distribution is normal. 225

PAGE 241

226 F5 March 13th Histogram02468101214161820-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98depthpercent Average: -12.91Std: 0.02Figure 139. F5 March 13th histogram. Note that the distribution is normal.

PAGE 242

F6 January 13th Histogram0510152025-12.94-12.95-12.96-12.97-12.98-12.99-13.00-13.01-13.02-13.03-13.04-13.05-13.06-13.07-13.08-13.09-13.10-13.11depthpercent Average: -13.03Std: 0.02Figure 14 0. F6 January 13th histogram. Note that the distribution is left-skewed. 227

PAGE 243

F6 January 17th Histogram0510152025-12.98-12.99-13.00-13.01-13.02-13.03-13.04-13.05-13.06-13.07-13.08depth percent Average: -13.03Std: 0.02rmal. Figure 141. F6 January 17th histogram. Note that the distribution is no 228

PAGE 244

F6 January 20th Histogram05101520253035-12.9-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00-13.01-13.02depthpercent Average: -12.97Std: 0.02Figure 142. F6 January 20th histogram. Note that the distribution is left-skewed. 229

PAGE 245

F6 February 6th Histogram02468101214161820-12.83-12.84-12.85-12.86-12.87-12.88-12.89-12.90-12.91-12.92-12.93-12.94-12.95-12.96-12.97-12.98-12.99-13.00-13.01depthpercent Average: -12.91Std: 0.03n is right-skewed. Figure 143. F6 February 6th histogram. Note that the distributio 230

PAGE 246

F6 March 13th Histogram02468101214161820-12.70-12.71-12.72-12.73-12.74-12.75-12.76-12.77-12.78-12.79-12.80-12.81-12.82-12.83-12.84depthpercent Average: -12.76Std: 0.02F igure 144. F6 March 13th histogram. Note that the distribution is right-skewed. 231

PAGE 247

232 F9 January 13th Histogram024681012141618-13.83-13.84-13.85-13.86-13.87-13.88-13.89-13.90-13.91-13.92-13.93-13.94-13.95-13.96-13.97-13.98-13.99depthpercent Average: -13.91Std: 0.02Figure 145. F9 January 13th histogram. Note that the distribution is normal.

PAGE 248

233 F9 January 20th Histogram0510152025-13.68-13.69-13.70-13.71-13.72-13.73-13.74-13.75-13.76-13.77-13.78-13.79-13.80depthpercent Average: -13.73Std: 0.02Figure 146. F9 January 20th histogram. Note that the distribution is right-skewed.

PAGE 249

F9 February 6th Histogram02468101214161820-13.64-13.65-13.66-13.67-13.68-13.69-13.70-13.71-13.72-13.73-13.74-13.75-13.76-13.77-13.78-13.79-13.80depthpercent Average: -13.73Std: 0.03Figure 147. F9 February 6th histogram. Note that the distribution is left-skewed. 234

PAGE 250

F9 March 13th Histogram0246810121416-13.39-13.40-13.41-13.42-13.43-13.44-13.45-13.46-13.47-13.48-13.49-13.50-13.51-13.52-13.53-13.54-13.55depthpercent Average: -13.48Std: 0.03 Figure 148. F9 March 13th histogram. Note that the distribution is left-skewed. 235

PAGE 251

F10 January 13th Histogram0246810121416-13.84-13.85-13.86-13.87-13.88-13.89-13.90-13.91-13.92-13.93-13.94-13.95-13.96-13.97-13.98-13.99-14.00-14.01depthpercent Average: -13.93Std: 0.03Figure 149. F10 January 13th histogram. Note that the distribution is left-skewed. 236

PAGE 252

F10 January 20th Histogram051015202530-13.68-13.69-13.70-13.71-13.72-13.73-13.74-13.75-13.76-13.77-13.78depthpercent Average: -13.73Std: 0.02 is normal. Figure 150. F10 January 20th histogram. Note that the distribution 237

PAGE 253

F10 February 6th Histogram024681012141618-13.64-13.65-13.66-13.67-13.68-13.69-13.70-13.71-13.72-13.73-13.74-13.75-13.76-13.77-13.78-13.79depthpercent Average: -13.73Std: 0.03kewed. Figure 151. F10 February 6th histogram. Note that the distribution is left-s 238

PAGE 254

Appendix B Description of Equations Calculating the phase and amplitude offset between tides The phase difference and amplitude offset between the tide record obtained from the pressure sensor data and the NOAA tide obtained from Clearwater Station 8726724 were calculated using an iterative equation in MATLAB, a powerful mathematics astatistical software package for data analysis (p. 14). The hourly NOAA tide record was interpolated in MATLAB in order to match the time series of the pressure sensor daNext, the following iterative program was run: p = [] STD = [] *standard deviation AMP = [] *amplitude PHASE = [] *phase for A = 1:0.1:2 for T = -1:0.1:1 p = interp1(decimal_date,new_tide,decimal_date+T) m = nanstd(noaa_2hr (A*p)) STD = [STD;m]; AMP = [AMP;A]; PHASE = [PHASE; T]; end end nd ta. 239

PAGE 255

W here: decimal_date = time series for the pressure sensor tide record new_tide = the tide record obtained from the pressure sensor noaa_2hr = the interpolated 2-hour NOAA tide record The code was run several times, changing the range and incremental value for T and A. After each run, the minimum value of STD was found along with its associavalues for AMP and PHASE. The amplitude difference and phase offset between the two tide records were considered found once the values of AMP and PHASE remained unchanged between code runs. The final values of AMP and PHASE were found to1.06 and 0.003 respectively. Calculating beam footprint and beam spacing for the multibeam sonar The across track beam footprint and across track beam spacing (in meters) for the Kongsberg Simrad EM 3000 multibeam sonar were calculated for various water depths (p. 185). These calculations assume a flat seabed and are based on the sonar being pole mounted approximately 2 meters below the water surface. Beam footprint: ted be )tan()cos(2bd where: (degrees) width beamangle pointing beam 2 (m)depth water bd 240

PAGE 256

B eam spacing: ssd21tan(21tan( (degrees) spacing beamangular angle pointing beam 2 (m)depth water sd where: The along track beam spacing (in meters) was also calculated for various vessel speeds for a ping rate of 10 Hertz (p. 185). Along track beam spacing: Vs51.0 (knots) speed vesselVs p where: (Hertz) rate pingsonar P 241