USF Libraries
USF Digital Collections

Improving ground penetrating radar resolution of features of active sinkholes

MISSING IMAGE

Material Information

Title:
Improving ground penetrating radar resolution of features of active sinkholes
Physical Description:
Book
Language:
English
Creator:
Gooch, Bradley
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
3D migration
FDTD modeling
Covered-karst terrain
Subvertical reflectors
GPRMAX
Dissertations, Academic -- Geology -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Ground penetrating radar (GPR) is widely used to identify locations of sinkholes in covered karst terrain in Florida. Some sinkholes serve as hydraulic conduits between the surficial and underlying aquifers. Their role is critical in determining the surficial aquifer response to pumping in deeper aquifers. Improved methods for discriminating between hydraulically active sinkholes and plugged sinkholes could help regional water management. In the covered karst of west-central Florida a clay-rich weathering horizon forms over the limestone. The clay-rich layer is in turn overlain by surficial sands. Ground penetrating radar profiles typically show a strong reflector from the top of clay-rich horizon as well as internal layering within sands. Active sinkholes are expected to have sandy conduits that broach the clay layer, and perhaps layering in the overlying sand indicative of ongoing subsidence. Three dimensional simulations of GPR profiles over sinkhole with and without conduits were run with the finite-difference time-domain (FDTD) program GPRMAX. Results from the synthetic surveys were then processed with standard techniques, including migration. The modeling confirms that conduits appear in GPR records primarily as gaps in the return from the clay layer. The modeling also shows that non-traditional survey geometries (varying antenna spacing and orientation) are unlikely to recover more information than traditional proximal transmitter-receiver separation. Also examined are GPR profiles and 3D grids over a set of active and inactive sinkholes in Tampa, Florida. Results from these surveys showed decent structural recovery of a small sinkhole similar in structure to that of the modeled ones. Indications of active subsidence and possible conduit structure were apparent from this data. Finally, the dense surveys served as a benchmark to compare interpretations taken with the same surveys at lower spatial resolutions and profiles with 2D-only processing methods in order to understand errors in analysis and interpretation that are possible from 2D surveys. Two-dimensional surveys, 2D processed and migrated, showed some similarity to the 3D results previously mentioned but contained more complexities and artifacts, which led to poorer interpretation ability.
Thesis:
Thesis (M.S.)--University of South Florida, 2010.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Bradley Gooch.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X 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:
usfldc doi - E14-SFE0003428
usfldc handle - e14.3428
System ID:
SFS0027743:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Improving Ground Penetrating Radar Resolution of Features of Active Sinkholes by Bradley Tyler Gooch A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Geology College of Arts and Sci ences University of South Florida Major Professor: Sarah Kruse, Ph.D. Mark Stewart, Ph.D. Mark Rains, Ph.D. Date of Approval: March 12 2010 Keywords: 3D migration FDTD modeling, covered karst terrain, subvertical reflectors GPRMAX Copyrig ht 2010 Brad ley Tyler Gooch

PAGE 2

Dedication I dedicate this work to one who inspired my love for science: Stephen Lee Hopkins (1946 2010)

PAGE 3

Acknowledgements I acknowledge all the help my adviser, Dr. Sarah Kruse, has given me throughout the thesis process. I also thank my committee members, Drs. Mar k Stewart and Mark R ains, for their guidance and review of my thesis. Final thanks go to all my family, friends, and fellow graduate students at USF for all their help and support during my time as a graduate stu dent. Without you, none of this would have been possible. Thank you all!

PAGE 4

i Table of Contents List of Figure s iii Abstract x I. Introduction 1 Active v. Inactive Sinkholes as a Problem 1 Standard Geological Inv estigation Method s for Sinkholes 6 Ground Penetrating Radar as a Modern Method to Delineate Problem 7 Proposed Strategies f or Improved Sinkhole Resolution 10 II. Modeling: Sinkhole Conduits as Difficult G PR Targets 12 Background Information 12 Theory 12 G PRMAX : Electromagnetic Reflection Software 14 GeoPark Sinkhole Model Descriptions 15 Standard 2D Geometry Survey 22 2D Model vs 3D Model (with 2D Migration) 22 Variable Common Offset Analysis 35 3D Snapshots of Electromagnetic Field s in Model Space 41 III. Modeling: Non S t andard Acquisition Geometries for Improved Conduit Detection 6 5 Differences in Receiver Field in XY Plane at Surface (z=0) 65 Model Design Information 65 Parallel Tx Rx Orientation (Ex C ompone nt of EM Field) 68 Perpendicular Tx Rx Orientation (Ey C omponent of EM Field ) 86 IV. Field Data from GeoPark, Tampa, FL 100 High Resolution 3D GPR Grids 100 GeoPark Geology 100 3D GPR Survey and P rocessing 100 Results and Discussion 1 07 2 50 MHz 107 500 MHz 113 Effects on Interpretation by Decreased Resolution 143 Decreasing 3D Survey Grid Resolution 143 Analysis in 2D only 147 V. Conclusions 155

PAGE 5

ii References 157 Appendices 160 Appendix A: G PRMAX Input Files 161 App endix B. MATLAB Codes 188 Appendix C: Shot Gather Analysis 227

PAGE 6

iii List of Figures Figure 1.1 Dissolution process of carbonate rock 2 Figure 1.2 Basic process of sinkhole formation 3 Figure 1.3 C over subsidence sinkhole formation 3 Figure 1.4 C over collapse sinkhole formation 4 Figure 1.5 Locations of reported sinkholes in Florida between 1960 and 1991 5 Figure 2.1 3D FDTD Yee Cell 13 Figure 2.2a Model A: v ertical slice through three dimensional model space at midpoint (y=4.0m) 17 Figur e 2.2b Model B: v ertical slice through three dimensional model space at midpoint (y=4.0 m) 18 Figure 2.3a U nprocessed results from 3D model A Ex field data 23 Figure 2.3b U nprocessed results from 3D model A Ey field data 23 Figure 2.3c U nprocessed re sults from 3D model A Ez field data 24 Figure 2.4a Model A 2D results processed and di splayed as profile; unmigrated 27 Figure 2.4b Model A 2D results processed and displayed as profile; migrated 28 Figure 2.5a Model A 3D results processed and di s played as profile; unmigrated 29 Figure 2.5b Model A 3D results processed and displayed as profile; migrated 30 Figure 2.6a Model B 2D results processed and d isplayed as profile; unmigrated 31 Figure 2.6b Model B 2D results processed and displayed as profile; migrated 32 Figure 2.7a Model B 3D results processed and displayed as profile; unmigrat ed 33

PAGE 7

iv Figure 2.7b Model B 3D results processed and displayed as profile; migrated 34 F igure 2.8a Model B 3D unprocessed re sults with 0.3 m common off set 37 Figure 2.8b Model B 3D unprocessed results with 1.0 m comm on offset 37 Figure 2.8c Model B 3D unprocessed results with 1.8 m common offset 38 Figure 2.9a Model B 3D results processed and di splayed as profile; unmigrated 39 Figure 2.9b Model B 3D results processed and displayed as profile; migrated 40 Figure 2.10a Sna pshot at 20 ns in x plane 43 Figure 2.10b Snapsho t at 20 ns in y plane 43 Figure 2.10c Snapshot at 20 ns in z plane 44 Figure 2.11a Sna pshot at 25 ns in x plane 45 Fig ure 2.11b Snapsho t at 25 ns in y plane 45 Figure 2.11c Snapshot at 25 ns in z plane 46 Figure 2.12a Sna pshot at 30 ns in x plane 47 Figure 2.12b Snapsho t at 30 ns in y plane 47 Figure 2.12c Snapshot at 3 0 ns in z plane 48 Figure 2.13a Sna pshot a t 35 ns in x plane 49 Figure 2.13b Snapsho t at 35 ns in y plane 49 Figure 2.13c Snapshot at 35 ns in z plane 50 Figure 2.14a Sna pshot at 40 ns in x plane 51 Figure 2.14b Snapsho t at 40 ns in y plane 51 Figure 2.14c Snapshot at 4 0 ns in z plane 52 Figure 2.15a Sna pshot at 45 ns in x plane 53 Figure 2.15b Snapsho t at 45 ns in y plane 53

PAGE 8

v Figure 2.15c Snapshot at 45 ns in z plane 54 Figure 2.16 a Sna pshot at 5 0 ns in x plane 55 Figure 2.16 b Snapsho t at 5 0 ns in y plane 55 Figure 2.16 c S napshot at 5 0 ns in z plane 56 Figure 2.17a Sna pshot at 55 ns in x plane 57 Figure 2.17b Snapsho t at 55 ns in y plane 57 Figure 2.17c Snapshot at 55 ns in z plane 58 Figure 2.18a Sna pshot at 60 ns in x plane 59 Figure 2.18b Snapsho t at 60 ns in y plane 59 Figure 2.18c Snapshot at 6 0 ns in z plane 60 Figure 2.19a Sna pshot at 65 ns in x plane 61 Figure 2.19b Snapsho t at 65 ns in y plane 61 Figure 2.19c Snapshot at 65 ns in z plane 62 Figure 2.20a Sna pshot at 70 ns in x plane 63 Figur e 2.20b Snapsho t at 70 ns in y plane 63 Figure 2.20c Snapshot at 70 ns in z plane 64 Figure 3.1a Model A: Vertical slice through three dimensional model space at midpoint (y=3.5m) 66 Figure 3.1b Model B: Vertical slice through three dimensional model space at midpoint (y=3.5m) 66 Fig ure 3.2 Receive r positions used in both models 67 Figure 3.3 Ex d ifference plot at transmitter location 1 70 Figure 3.4 Ex d ifference plot at transmitter location 15 71

PAGE 9

vi Figure 3.5 Ex d ifference plot at transmitter location 25 72 Figure 3.6 Ex d ifference plot at transmitter location 30 73 Figure 3.7 Ex d ifference plot at transmitter location 40 74 Figure 3.8 Ex d ifference plot at transmitter location 45 75 Figure 3. 9 Ex d ifference plot at transmitter location 50 76 Figure 3.10 Ex d ifference plot at transmitter location 60 77 Figure 3.11 Ex d ifference plot at transmitter location 70 78 Figure 3.12 Ex d ifference plot at transmitter location 75 79 Figure 3.13 Ex d ifference plot at transmitter location 95 80 Figure 3.14 Ex d ifference plot at transmitter location 100 81 Figure 3.15 Ex d ifference plot at transmitter location 105 82 Figure 3.16 Ex d ifference plot at transmitter location 110 83 Figure 3.17 Ex d ifference plot at transmitter location 115 84 Figure 3.18 Ex d ifference plot at transmitter location 125 85 Figure 3.19 Ey d ifference plot at transmitter location 10 87 Figure 3.20 Ey d ifference plot at transmitter location 15 88 Figure 3.21 Ey d ifference plot at transmitter location 20 8 9 Figure 3.22 Ey d ifference plot at transmitter location 30 90 Figure 3.23 Ey d ifference plot at transmitter location 40 91 Figure 3.24 Ey d ifference plot at transmitter location 55 92 Figure 3.25 Ey d ifference plot at transmitter location 60 93 Fi gure 3.26 Ey d ifference plot at transmitter location 65 94 Figure 3.27 Ey d ifference plot at transmitter location 70 95

PAGE 10

vii Figure 3.28 Ey d ifference plot at transmitter location 85 96 Figure 3.29 Ey d ifference plot at transmitter location 90 97 Figure 3.30 Ey d ifference plot at transmitter location 105 98 Figure 3.31 Ey d ifference plot at transmitter location 110 99 Figure 4.1 Figure of approximate locations of previous studies conducted in the USF GeoPark 101 Figure 4.2 Figure showing near sur face geology of USF GeoPark 102 Figure 4.3 Migrated time slice of 3D GPR (250 MHz) data at 18.98 ns 103 Figure 4.4 Migrated time slice of 3D GPR (250 MHz) data at 23.87 ns 104 Figure 4.5 Migrated time slice of 3D GPR (250 MHz) data at 29.69 ns 105 Figure 4.6 Migrated time slice of 3D GPR (250 MHz) data at 33.97 ns 106 Figure 4.7 Migrated time slice of 3D GPR (250 MHz) data at 37.95 ns 108 Figure 4.8 Migrated time slice of 3D GPR (250 MHz) data at 43.95 ns 109 Figure 4.9 Migrated time slice of 3D GPR (250 MHz) data at 45.91 ns 110 Figure 4.10 Migrated time slice of 3D GPR (250 MHz) data at 49.89 ns 111 Figure 4.11 Migrated time slice of 3D GPR (250 MHz) data at 52.03 ns 112 Figure 4.12 Migrated 3D GPR (250 MHz) cross line data at 7.042 m 114 Figure 4. 13 Migrated 3D GPR (250 MHz) cross line data at 9.035 m 115 Figure 4. 14 Migrated 3D GPR (250 MHz) cross line data at 11.54 m 116 Figure 4. 15 Migrated 3D GPR (250 MHz) in line data at 9.000 m 118 Figure 4. 16 Migrated 3D GPR (250 MHz) in line data at 9.600 m 119 Figure 4. 17 Migrated 3D GPR (250 MHz) in line data at 10.000 m 120 Figure 4. 18 Migrated 3D GPR (250 MHz) in line data at 10.400 m 121

PAGE 11

viii Figure 4. 19 Migrated 3D GPR (250 MHz) in line data at 11.200 m 122 Figure 4. 20 Migrated 3 D GPR (250 MHz) in line data at 12.000 m 123 Figure 4. 21 Migrated 3D GPR (250 MHz) in line data at 12.600 m 124 Figure 4.22 Migrated time slice of 3D GPR (50 0 MHz) data at 18.95 ns 126 Figure 4.23 Migrated time slice of 3D GPR (50 0 MHz) data at 29.75 ns 127 Figure 4.24 Migrated time slice of 3D GPR (50 0 MHz) data at 42.05 ns 128 Figure 4.25 Migrated time slice of 3D GPR (50 0 MHz) data at 52.02 ns 129 Figure 4.26 Migrated time slice of 3D GPR (50 0 MHz) data at 58.01 ns 130 Figure 4.27 Migrated time slice of 3D GPR (50 0 MHz) data at 65.99 ns 131 Figure 4.28 Migrated 3D GPR (50 0 MHz) cross line data at 3.544 m 1 33 Figure 4.29 Migrated 3D GPR (50 0 MHz) cross line data at 5.236 m 134 Figure 4.30 Migrated 3D GPR (50 0 MHz) cross line data at 6.4 17 m 135 Figure 4.31 Migrated 3D GPR (50 0 MHz) cross line data at 8.039 m 136 Figure 4.32 Migrated 3D GPR (500 MHz) in line data at 6.000 m 138 Figure 4.33 Migrated 3D GPR (500 MHz) in line data at 6.800 m 139 Figure 4.34 Migrated 3D GPR (500 MHz) in line data at 7.700 m 140 Figure 4.35 Migrated 3D GPR (500 MHz) in line data at 8.900 m 141 Figure 4.36 Migrated time slice of low resolution 3D GPR (50 0 MHz) data at 42.05 ns 144 Figure 4.37 Migrated low resolution 3D GPR (50 0 MHz) cross line da ta at 6.510 m 145 Figure 4.38 Migrated low resolution 3D GPR (500 MHz) in line data at 6.00 m 146 Figure 4.39 Unmigrated 3D GPR (500 MHz) in line data at 6.800 m 148

PAGE 12

ix Figure 4.40 Unmigrated 3D GPR (500 MHz) in line data at 7.700 m 149 Figure 4.41 2 D migrated 3D GPR (500 MHz) in line data at 6.800 m 152 Figure 4.42 2D migrated 3D GPR (500 MHz) in line data at 7.700 m 153

PAGE 13

x Improving Ground Penetrating Radar Resolution of Features of Active Sinkholes Bradley Tyler Gooch A BSTRACT Ground penetra ting radar (GPR) is widely used to identify locations of sinkholes in covered karst terrain in Florida. Some sinkholes serve as hydraulic conduits between the surficial and underlying aquifers. Their role is critical in determining the surficial aquifer re sponse to pumping in deeper aquifers. Improved methods for discriminating between hydraulically active sinkholes and plugged sinkholes could help regional water management. In the covered karst of west central Florida a clay rich weathering horizon forms o ver the limestone. The clay rich layer is in turn overlain by surficial sands. Ground penetrating radar profiles typically show a strong reflector from the top of clay rich horizon as well as internal layering within sands. Active sinkholes are expected to have sandy conduits that broach the clay layer, and perhaps layering in the overlying sand indicative of ongoing subsidence. Three dimensional simulations of GPR profiles over sinkhole with and without conduits were run with the finite difference time dom ain (FDTD) program G PRMAX Results from the synthetic surveys were then processed with standard techniques, including migration. The modeling confirms that conduits appear in GPR records primarily as gaps in the return from the clay layer. The modeling als o shows that non traditional survey geometries (varying antenna spacing and orientation) are unlikely to recover more information than traditional proximal

PAGE 14

xi tra nsmitter receiver separation. A lso examine d are GPR profiles and 3D grids over a set of active an d inacti ve sinkholes in Tampa, Florida. Results from the s e surveys show ed decent structural recovery of a small sinkhole similar in structure to that of the modeled ones. Indications of active subsidence and possible conduit structure we re apparent from th is data. Finally, the dense surveys serve d as a benchmark to compare interpretations taken with the same surveys at lower spatial resolutions and profiles with 2D only processing methods in order to understand errors in analysis and interpretation that are possible from 2D surveys. Two dimensional surveys, 2D processed and migrated, show ed some similarity to the 3D results previously mentioned but contain ed more compl exities and artifacts, which le d to poorer interpretation ability.

PAGE 15

1 Introduction Active v. Inactive Sinkholes as a Problem Sinkholes are naturally occurring geologic features caused by the dissolution of underlying bedrock. The mixing of carbon dioxide (CO2) and water (H2O) from the air and soil leads to the creation of carbonic acid, a we ak acid. This acid, formed above the bedrock in the overlying sediment, readily dissolves carbonate (limestone and dolostone) bedrock (Figure 1.1). When this process occurs over a widespread area, it can cause exaggerated topographic irregularity known as karst terrain. This terrain may either be exposed or have a sediment cover over it. The latter is referred to as covered karst terrain. These sediment covers can be excessively thick, sometimes more than 60 meters (m) (Sinclair and Stewart, 1985). The way the sediment cover and the dissolving bedrock interact determines the type of sinkhole that forms. When the overburden sediments (generally thin) that cover the dissolving bedrock fill in the depression at the same rate of dissolution, it is known as a sol ution sinkhole (Tihanksy, 1999; see Figure 1.2). When dissolution of the bedrock takes place primarily inside the bedrock via subsurface process can occur with little se diment displacement, and hence little change to the surface above. Eventually, the void space will fill in with weathered bedrock and overlying

PAGE 16

2 sinkholes are known a s cover subsidence (Tihanksy, 1999). A third way sinkholes form occurs where the overlying sediments are so cohesive (clay rich) that they will not move into the void spaces created by the dissolving bedrock below. Instead, the void migrates into the overb urden. Eventually the surface will collapse and form a depression (Figure 1.4). This is known as a cover collapse sinkhole (Tihanksy, 1999). All three sinkhole types are common to Florida (Sinclair and Stewart, 1985) and can pose significant hazards to t heir environment and people living in their vicinity. Florida is prone to abundant sinkhole development because the entire state is underlain by a thick carbonate platform, with Figure 1.1. Complete process of the dissolution of carbonate rock (from Tihansky, 1999).

PAGE 17

3 Figure 1.2. Basic process of sinkhole formation, where unconsolidated sediments above dissolved bedrock fill into void spaces created by dissolution process (from Tihansky, 1999). Figure 1.3. Process of cover subsidence si nkhole formation where unconsolidated sediments gradually funnel down conduits to larger void spaces creating a general lag in the topography (from Tihansky, 1999).

PAGE 18

4 Figure 1.4. Process of cover collapse sinkhole formation where clay rich overburden hold s surface topography while sediments move into void spaces. Eventually, this process will lead to abrupt surface failure and create the sinkhole (from Tihansky, 1999). mostly mantled, insoluble siliciclastics over it. Occurrence of sinkholes in Florida is hastened by the growing population whose equally growing need of water resources depletes aquifer volumes, which leads to an increase in cavity roof instability due to lack of ground water to support the suspended material (Ford and William, 1989). Most o f the sinkholes in Florida can be found in the mantled karst of West central area of the state. The mantled karst of west central Florida area contains a higher density of sinkholes than other areas of the state (Figure 5). This high level of sinkhole occu rrence is because the karst in this area is heavily dissolved and is overlain by approximately 10 60 m of sand to clayey sand, common to cover subsidence and cover collapse sinkhole development.

PAGE 19

5 Figure 1.5. Locations of reported sinkholes in Florida between 1960 and 1991. Some of the data may be biased towards areas of high population density due to better reporting (Wilson and Shock, 1996). One of the challenging issues currently in sinkhole investigations is determining whether or not the depressi ons are actively subsiding. Once materials fill in the void spaces, it is not well understood how the dissolution process continues (Tihansky, 1999). Does the newly deposited material change the dissolution rate? A second challenge is to assess whether or not a sinkhole is hydraulically active. By hydraulically active it is mean t that the sinkhole contains low permeability zones that function as preferential flow paths for groundwater. Active sinkholes can be an important site source for water recharge in to the underlying aquifers (Stewart, 1998; Stewart and

PAGE 20

6 Parker, 1992). Improved means of imaging subsurface features and structure of sinkholes are desirable to better address these questions. Standard Geological Investigation Methods for Sinkholes Sinkh oles are typically initially identified on the basis of their associated often roughly conical topographical depressions. Growing topographic expressions are often indications of sinkhole activity. Where buildings or foundation work are present, then mea surements of floor elevation change or wall movement (such as cracking) can be indicators of the activity of nearby sinkholes. However these phenomena cannot be unambiguously associated with sinkhole formation, as many other factors can cause ground deform ation. Other factors include tree root activity, expansive clays, and buried decomposing materials. To further study suspected sinkholes, either to understand their precise subsurface structure or their hydrological role in recharge, more detailed geolog ical investigation is required. Relevant information on the regional geology is helpful in setting the field site in context. Locations and morphology of other known sinkholes in the immediate area are also useful when combined with the regional geologic m ap. Direct geologic sampling through techniques such as drilling and boring can resolve zones of thickened overburden. With these data a local site geologic map of the subsurface can be created (e.g. Zisman et al 2005). It is important to note that such ma ps must with care in future investigations because of the potentially rapid (days to years) dissolution and raveling processes associated with sinkhole formation.

PAGE 21

7 Hydrological data from groundwater wells can also contribute to sinkhole investigations. The information can help define the movement of local ground water, which may be influenced by sinkholes. However, direct sampling and hydrologic data are usually costly and thus seldom available until later in the investigation process. Surface geophysical methods are commonly used for target identification and structural imaging after preliminary inspections and before any serious direct sampling methods. There are many different geophysical methods currently used to investigate various anomalous subsurfac e properties that are indicative of sinkholes. The majority of these methods use subsurface electrical and density properties for evaluation (Dobecki and Upchurch, 2006) such as microgravity, electrical resistivity tomography, coupled capacitance resistivi ty, and seismic methods (i.e. reflection, refraction, and surface wave analysis). The Florida Geological Survey has recently detailed the use of geological, geophysical, and geotechnical procedures to be used in sinkhole investigations in the state of Flor ida (Schmidt, 2005). The special publication focuses on identification of sinkholes rather than assessing the levels of either subsidence or hydraulic activity. Zisman (2001) attempted to better define subsidence potential by offering a standardized invest igation process based on site specific characteristics such as soil density variance with depth, drilling properties, stratigraphic conditions, and ground water levels. Zisman (2001) emphasizes the uncertainty of the current, standard investigation procedu res described briefly above. There is still a need to identify improved methods for evaluating both subsidence potential and hydraulic activity of sinkholes Ground Penetrating Radar as a Modern Method to Delineate Problem

PAGE 22

8 Of all the geophysical methods used in sinkhole investigations, ground penetrating radar (GPR) seems to offer the highest spatial resolution (Dobecki and Upchurch, 2006). GPR is a very robust method for investigating the near surface subsurface and has been used for many different appl ications (Mellett, 1995). GPR uses high frequency electromagnetic radio waves to image subsurface structures that reflect and diffract the energy back to the surface where it is recorded. The instrumentation incorporates tra nsmitting and receiving antennas a recording device, power accessories, and a computer interface (Reynolds, 1997). The method is very fast and efficient when compared to most other geophysical methods, enabling individuals to rapidly collect and view field data on site at high resolutio n (Burger et al, 2006). Although radar wave and seismic waves respond to fundamentally different physical properties, in practice the GPR reflection method is very similar in nature to that of common offset seismic reflection methodology and undergoes esse ntially the same data processing as single receiver, common offset seismic reflection data. One fundamental difference is that radar waves are polarized, while seismic waves are not. GPR waves travel at five orders of magnitude faster, with much shorter wa velengths (typically tens of centimeters rather tens of meters), than seismic waves. Traditionally, GPR data are collected with one transmitter and receiver antenna with a constant offset. Antennas may be unshielded or shielded to protect against electroma gnetic interference. Recently new advances in resolution and positioning systems have allowed GPR surveys to become truly three dimensional, enhancing subsurface visualization (Grasmueck et al, 2004; Grasmueck et al, 2005; Kadioglu, 2008).

PAGE 23

9 GPR has been us ed in the karst of west central Florida to identify sinkhole features with much success within the last two decades (Dobecki and Upchurch, 2006; Kruse et al, 2000; Kruse et al, 2006; Zisman et al, 2005). This methodology has also been used in other karst e nvironments of the USA (Carpenter and Ekberg, 2006) as well as used in the karst of Jordan to identify sinkhole features (Batayneh et al, 2002). Most GPR profiles from regions of covered kart terrain resolve the clay rich material mantled on top of the kar stified limestone bedrock. The conductive, clay rich sediments attenuate the signal making it difficult to resolve structure beneath this mantle (Kruse et al, 2006). Even with this hindrance, GPR images of the topography of this mantled clay which can allo w for a somewhat detailed analysis of the structure of the sinkhole depression. Another feature found in GPR profiles over some sinkholes are strong reflections from within a material filling depression, most likely caused by the gradual infilling associat ed with sinkhole generation (Carpenter and Ekberg, 2006; Dobecki and Upchurch, 2006). On many occasions these infilling reflectors are present even when the depression itself is not, causing concern for interpretation of sinkhole structure based solely on infilling patterns that may not actually be related to sinkholes (Batayneh et al, 2002). Truss et al (2005) imaged small dissolution sink features in Miami oolitic limestone with 3D GPR at various times during rainfall input to view their relevance as hyd raulic conduits from the vadose zone to the lower surficial aquifer. The GPR profiles themselves show little in the way of conduit structure below the small depression. visual ization greatly. Where water content changed the most (in the conduit) there were the greatest differences in wet and dry surfaces. Truss et al (2005) observed differences in

PAGE 24

10 both the velocities and amplitudes. Their strategy is clearly successful in resol ving flow concentrated in a conduit. However, it requires multiple surveys in wet and dry scenarios. ng and imaging these structures could lead to improvements in interpretations of the current activity status of a sinkhole. Proposed Strategies for Improved Sinkhole Resolution This study begins with a simple attempt to understand the limitation tradit ional 2D surveys have for resolving sinkhole structure. Forward modeling is used to make simple model geometries to highlight the inaccuracies of 2D GPR profiles over three dimensional features. These profiles are processed like typical 2D processing seque nces, with and without migration. The results suggest that the 2D migration is unable to accurately reconstruct the structure of the three dimensional model. Also, reflected energy from the conduit is mostly lost. This thesis then addresses whether standar d source and receiver survey geometries are optimal to resolve the difference in various structures of active sinkholes. A suite of non traditional source and receiver survey geometries are in the modeling in order to understand if there is added value in using these types of surveys. The simple adjustment of traditional offset distance does not produce better results. Non traditional survey geometries are proven to be as ineffective at accurately recovering reflective energy fro m the conduit and traditiona l spacing appear to be best suited even though they are limited.

PAGE 25

11 Case studies of very high resolution 3D data of sinkholes and sinkhole related features are shown. The data are processed using standard 3D processing methods, including 3D migration of the data. Sinkhole features from these profiles serve as examples to be compared to those of the simp lified models. Results from these surveys show decent structural recovery of a small sinkhole similar in structure to that of the modeled ones. Indications of active subsidence and possible conduit structure are apparent from this data. Finally, dense surveys serve as a benchmark to compare interpretations taken with the same surveys at lower spatial resolutions and profiles with 2D only processing methods in or der to understand errors in analysis and interpretation that are possible from 2D surveys. Two dimensional surveys, 2D processed and migrated, show some similarity to the 3D results previously mentioned but contain more complexities and artifacts, which le ad to poorer interpretation ability.

PAGE 26

12 Modeling: Sinkhole Conduits as Difficult GPR Targets Background Information Numerical modeling of radar wave response to conduit structure is used to investigate optimal survey geometries and to understand arti facts of 2D profiling across 3D sinkholes. To apply this, thorough experimentation of the circumstance needs to be conducted. Numerical simulatio ns should be conducted in three dimensional model space and occupy volumes analogous to those of sinkhole with conduits, but simplified in order to capture first order effects. Radar waves are governed by Maxwell's equations. These four partial differential equations describe properties of electric and magnetic fields relative to their sources, specifically, curr ent and charge densities (Fleisch, 2008). To model radar wave propagation these equations must be solved with respect to time and space. Simulating spatial volumes of geologic GPR surveys, including moving source and receiver, would take hundreds of hours of calculation time even with multi core workstations. However, computational speeds permit simulations of limited numbers of transmissions. One of the most popular computational solutions to Maxwell's equations is the finite difference time domain (FDTD) method (Yee, 1966; Taflove, 1980). FDTD modeling allows for conditionally stable, grid based, and differential time domain

PAGE 27

13 numerical modeling. Given a specific dimension of unit space and a particular instance of time, Maxwell's equations may be solved wit h specified initial conditions and grid boundary conditions. The key to the model's ability to find realistic solutions lie in the discretization of time and space domains over which models are run. In Yee's (1966) seminal work on FDTD methodology, he fo rmulated what is now known as the Yee cell, which is the most basic space unit making up a multidimensional grid based model space (see below). Key to the numerical effectiveness is t he staggered positions at which electric (E) and magnetic (H) fields are calculated. Figure 2.1. 3D FDTD Yee Cell showing electric and magnetic field directions (Yee, 1966). In the standard Cartesian Yee cell, Maxwell's Equations may be solved in three or two dimensions at each time discretization step. In practice, almost a ny model geometry

PAGE 28

14 edging at some level. Absorbing boundary conditions (ABCs) are applied to the outermost cells to ensure that the model space allows electromagnetic energ y to The freely available computer modeling program GPRMAX is a recently developed FDTD program specifically designed for GPR ( Giannopoulos, 2005) The program allows users to produce realistic two or three dimen sional models. The source code is written in C; the executable form of both of these programs may be used in Windows, Mac, or Linux operating systems installed on any standard personal computer. GPRMAX utilizes a command line interface with ASCII text file s as a means for model parameter and geometry input. The program allows users to introduce layers and shapes of almost any geologic material to be placed inside the parallelepiped model space of specified dimension. Radar transmitters and receivers can als o be positioned anywhere in the model space. Repeated runs using the same model space can be performed to allow for the simulation of the typical moving scans of the tra nsmitting and receiving antennas along a line, known as a common offset gather. GPRMAX has been used to simulate shallow ice covered features, rebar location in concrete structures, and dense non aqueous phase liquids (DNAPLs) in groundwater (Brandt, 2006; Giannopoulos, 2005; Gerhard et al, 2008). Versions of the program (available from its author) allow it to access multiple processors (via OpenMP parallelization) for improved performance (Brandt, 2006). Small conical sinkholes in covered karst terrain have diameters a few meters across at depths a few meters below the ground surface. They typically have

PAGE 29

15 complementary depressions of sediments within them which possibly deepen as they approach the central conduit of the sinkhole. There may also be weathering layers of clay that line the top of the limestone (Parker, 1992). Two basic models were designed to simulate simplified versions of such sinkholes. One model (Model A, seen in Figure 2.2a) contains a simple conical depression theoretically representative of a small inactive sinkhole. The other (Model B, seen in Figure 2.2b) is identical to the first model, but containing an added conduit structure that could exist in a more hydraulically active sinkhole. Models A and B have total space dimensions (xyz format, z being the depth axis) of 8.0 x 8.0 x 4.0 meters and 8.0 x 8.0 x 4.5 meters, re spectively. The discretization of space (Yee cell size) in all three dimensions is set to 0.02 meter. This choice is appropriate because according to GPRMAX's rule of thumb (based on stability criterion proposed by Taflove [1980]), cell size should be at l east ten times smaller than the smallest wavelength of the propagating electromagnetic field. A 250 MHz frequency antenna (common to geologic investigations of this scale) generates a wavelength of 0.12 m for velocities in saturated sediments. Models A and B are composed of more than 30 million Yee cells. The basic geologic structure of both models, from the base up, is a block layer of clay 2.5 to 3.0 m in thickness (Figure 2.2). Above the clay is a 1.0 m thick sand surface horizon. Above the sand layer ( ground level) is a 0.5 m thick layer of air. The sinkhole is simulated by a series of progressively smaller thin cylinders to make a conical section. The cylinders are one Yee cell (0.02 m) in thickness. The cylinders are made of sand and start at the base of the sand layer and protrude into the clay layer below producing a three dimensional conical breach of sand 2.0 meters deep and 4.0 meters across at the

PAGE 30

16 top. The added conduit in Model B is introduced inside the original sand cone where its width is 1.0 m across and steeply descends another 1.0 m down into the clay. All GPRMAX model input text files are shown in Appendix A.

PAGE 31

17 Figure 2.2a

PAGE 32

18 Figure 2.2b

PAGE 33

19 Figure 2.2a. Model A: Vertical slice through three dimensional model space at midpoint (y=4.0m). Orange material represents clay; green, sand; and blue, free space. Values for distance given as model coordinate values where (0.0, 0.0, 0.0) represents lower left forward most point and (8.0, 8.0, 4.0) is the upper right back most point in the model space. Figure 2.2b. Model B: V ertical slice through three dimensional model space at midpoint (y=4.0 m). Orange material represents clay; green, sand; and blue, free space. Values for distance given as model coordinate values where (0.0, 0.0, 0.0) represents lower left forward most poi nt and (8.0, 8.0, 4.5) is the upper right back most point in the model space. The MATLAB codes used to create these figures and all others in this work, unless otherwise speci fied, may be found in Appendix B The physical parameters used for sand are 9.0 for the DC (static) relative permittivity, 0.005 Siemens/m for the DC (static) conductivity, and 1.0 for the relative magnetic permeability. The physical parameters used for clay are 25.0 for the DC (static) relative magnetic permittivity, 0.05 Siemens/m for the DC (static) conductivity, and 1.0 for the relative magnetic permeability. The values for the air layer are those of free space. Limestone, although the basis for sinkhole genesis, is not used in the modeling because the GPR waves rarely penetrate p ast the clay layer as they attenuate almost completely in the clay before reaching the limestone reflecting surface. The antenna specifications chosen for the models are a reasonable approximation of a 250 MHz shielded antenna. In the GPRMAX program thi s is chosen by selecting

PAGE 34

20 model parameters of a Hertzian dipole with unit amplitude (1.0) and a center frequency of a Ricker wavelet of 250 MHz. This command technically simulates an infinitesimal dipole (although it does have a very small, but not nearly i nfinitesimal in length) which is not exactly used in shielded systems but allows a decent approximation nonetheless. Polarization directions may either be in x or y orientations. In this particular model the polarization is in the x direction. The total tr ansmitting time was set so that the two way travel time for radar waves propagating from one end of the surface to the bottom of the conical depression of sand could be recorded. The total recording time was 100 nanoseconds (ns) for Model A and 120 ns for Model B. When working in the time domain, the time iteration of the model is governed by the discretization of space so it cannot be independently assigned. Rather, for the FDTD method to be conditionally stable numerically, it must satisfy the Courant, Fr edrichs, and Lewy (CFL) condition (see below). Equation 2.1. The CFL condition, where c is the speed of light and t is the time discretization based on the values of the space discretization in the x y and z directions. In this particular model case, the time discretization is approximately 38.5 picoseconds (ps) given a space discretization of 0.02 meters. Using this time step, 2597 iterations for Model A and 3116 iterations for Model B are required for each transmitter

PAGE 35

21 location. In order to acquire G PR trace imaging along the lines of realistic surveying, model runs must be repeated, moving transmitting and receiving antennas (spaced 0.3 m apart) pair almost 150 times at 0.04 m increments to simulate a transect crossing the sinkhole. To simulate a si ngle profile, this required solving Maxwell's Equations in three dimensions for each of the approximately 30 million cells for each iteration at each transmitting location. A single processor (~2.5 GHz) on a computer requires over a month of total run time Also, the amount of memory (Random Access Memory, RAM) required for the model to run was about 700 1500 Megabytes (Mb) in size. Total binary output files sizes for models could be anywhere in size from a few hundred Mb to tens of Gigabytes (Gb), dependin g on the number of receivers included in the model. Fortunately, by utilizing the Open MP commands built into the source code of GPRMAX and recompiling the program, GPRMAX could be used with many more processors to speed up run times for the models. Using a desktop workstation containing dual CPU sockets with Quad core processors (for a total of 8 processors) and 16 Gb of RAM, run times for acquired profiles were shortened from over a month to about four to five days.

PAGE 36

22 Standard 2D Geometry Survey Here we examine what users may miss or misinterpret with 2D profiling and data processing over conical sinkholes. Models were simulated in both 2D and 3D versions of models A and B (see Figure 2.2), utilizing the same parameters as those described i n the previous section. All of the model results were processed as typical field GPR data would be treated using the software package ReflexW (Sandmeier). This package allows for standard 2D and 3D data processing including filtering, statics, gaining, and migration. Although GPRMAX outputs results of the six components of the electromagnetic fields ( E x E y E z H x H y, and Hz) as well as the currents ( I x I y, and I z) calculated as the curl of the magnetic fields at each electric field location, only the Ex values are used for processing. These values are the same as those collected in the field with both standard shielded antennas oriented in the x direction while recording along the y direction. Examples of unprocessed Ex, Ey, and Ez field data for 3D mode l A can be compared in Figures 2.3a, 2.3b, and 2.3c.

PAGE 37

23 Figure 2.3a Figure 2.3b

PAGE 38

24 Figure 2.3c

PAGE 39

25 Figure 2.3 a This is an example of unprocessed results from 3D model A Ex field data. Time in ns is given on the y axis (z axis in model coordinates) and distance along the simulated survey lin e is given in m along the x axis (y axis in model coordinates) for all three figures. Figure 2.3 b. This is an example of unprocessed results from 3D model A Ey field data. Figure 2.3 c. This is an example of unproces sed results from 3D model A Ez field dat a. Notice that the Ex field having a particularly strong and symmetric signal over the other fields, making it the choice field of study. It is clear that 2D profiles will record out of plane features from 3D structures. We also seek to understand artifac ts associated with 2D migrations of out of plane features. Here, the raw profiles over 2D and 3D models are shown (Figures 2.4 2.7), along with the results of 2D migrations of each. All of the 2D and 3D models generally produce similar synthetic profiles. When the 2D model A data are migrated (diffraction stack, time to depth conversion of 0.1 m/ns), it produces an image that perfectly mimics the model geometry it is given (Figure 2.4b). This confirms that the 2D migration process is robust for this 2D geom etry. However, it is clear that there are artifacts associated with the 2D migration of the 3D structure (Figure 2.5b). These artifacts are associated with features outside the viewing plane and the 3D nature of the wave, in this case, the side of the coni cal sink. The 3D results of model A give an unexpected curvature through an area at about the middle of the conical area of the modeled sinkhole (Figure 2.5b) which are the implied out of plane effects of the conical sinkhole. The bottom of the sinkhole

PAGE 40

26 f eature is still visible and able to be located as in the 2D migrated results. However, there are also some unexplained strong reflecting artifacts below the sinkhole feature, which may be complex wave reflections. Both the 2D and 3D pre migrated results of model B look very similar to those of model A's (Figures 2.4a & 2.5a) except that they are slightly more 'noisy' in appearance, most likely from numerical artifacts associated with the GPRMAX simulation (see Figures 2.6a & 2.7a). The reason for the increa se in noise level in model B simulations is not clear. The migrated results of the 2D model B resemble those of model A's in that there is the same basic larger depression, but the added conduit in model B (Figure 2.6b) appears as a loss of reflected energ y and imperfectly migrated weak returns. The migrated results of the 3D model B data look of plane artifact and basic depression expected (Figure 2.7b). There is also a loss in the refl ective energy at the mouths of the conduit. It is clear from these simple models that out of plane returns can dominate 2D profile and introduce significant artifacts during migration. Acquiring data in three dimensions is a necessary step in improved sink hole structure imaging.

PAGE 41

27 Figure 2.4a

PAGE 42

28 Figure 2.4b

PAGE 43

29 Figure 2.5a

PAGE 44

30 Figure 2.5b

PAGE 45

31 Figure 2.6a

PAGE 46

32 Figure 2.6b

PAGE 47

33 Figure 2.7a

PAGE 48

34 Figure 2.7b

PAGE 49

35 Note on previous figures: All model data was processed with custom static and gain filters. The diffraction stack migration (2D) used a constant time to depth conversion of 0.1 m/ns. A ll these figures have bo th a time axis (y axis) in ns and a secondary y axis with depth (conversion 0.1 m/ns). Also, all common offset spacings are 0.3 meters. Figure 2.4 a. Model A 2D results processed and displayed as profile; unmigrated. Figure 2.4 b. Model A 2D results proc essed and displayed as profile; migrated. Figure 2.5 a. Model A 3D results processed and displayed as profile; unmigrated. Figure 2.5 b. Model A 3D results processed and displayed as profile; migrated. Figure 2.6 a. Model B 2D results processed and disp layed as profile; unmigrated. Figure 2.6 b. Model B 2D results processed and displayed as profile; migrated. Figure 2.7 a. Model B 3D results processed and displayed as profile; unmigrated. Figure 2.7 b. Model B 3D results processed and displayed as pro file; migrated. To see whether non traditional transmitter receiver offsets might better resolve conduit geometry, model B was run with one transmitter that moved along the center line of the simulated sinkhole. As discussed previously, there was also a receiver at a fixed distance (0.3 m) from the transmitter, also along the same survey line. In addition to this initial receiver, 15 others were placed in line spaced at 0.1 m intervals. This spacing can simulate various offsets for common offset s urveys. Specific intervals used for comparison were the 0.3, 1.0, and 1.8 m offset spacings. Comparison of the unprocessed results of each of these offset lengths may be seen in Figures 2.8a c. It can be seen that there is not much initial difference amon g these three unprocessed profiles. In order to

PAGE 50

36 understand the differences that could occur, the modeling results must be processed, including migration. Only the maximum and minimum offset spacing were used to observe the greatest difference in varying th e offset. The processed results (both migrated and unmigrated) of the minimum 0.3 m offset spacing were previously displayed in Figures 2.7a and 2.7b.

PAGE 51

37 Figure 2.8a Figure 2.8b

PAGE 52

38 Figure 2.8a. Top, previous page. Model B 3D unprocessed results with 0.3 m common offset. Figu re 2.8b. Bottom, previous page. Model B 3D unprocessed results with 1.0 m common offset. Figure 2.8c. Top, current page. Model B 3D unprocessed results with 1.8 m common offset. Note: in constructing the profiles the length of the obtainable p rofile decrea ses as the offset spacing increases. Also, increasing offset distance increases travel time for reflections. These two things account for the greatest discrepancies in similarity of the figures. Figure 2.8c

PAGE 53

39 Figure 2.9a

PAGE 54

40 Fi gure 2.9b

PAGE 55

41 Figure 2.9 a. Model B 3D results processed and displayed as profile; unmigrated. Figure 2.9 b. Model B 3D results processed and displayed as profile; migrated. Both figures have a common offset spacing of 1.8 m and Figure 2.9b was diffraction stack migrated with a c onstant time to depth conversion of 0.1 m/ns The processed results (both migrated and unmigrated) of the maximum 1.8 m offset spacing are displayed in Figures 2.9a&b. In comparison, the two different offsets give similar profiles but are definitely not the same. It is clear from the proc essed results that the shorter offset has a better signal strength than the wider offset. This is most likely due to there being a more direct return of reflected energy to the receiver than if traveling on a longer path. Also, the reflection coefficient d epends on the angle of incidence. The migrated results suggest that the shorter offset will allow for better reconstruction of a subsurface depression. The shorter offset also gives a stronger return of artifacts below the sand clay interface. The wider of fset has an overall weaker return on reflected energy, which is an expected result of the longer travel path through matter. What was not expected was the anomalous pattern of reflected energy where the conduit should exist. The larger offsets were no bett er than traditional smaller ones at reconstructing the steep, narrow conduit geometry of the model. In order to gain better insight into how radar energy propagates and reflects inside the model space, a 3D method similar to 2D ray tracing was emp loyed. GPRMAX allows its user the ability to save model volume information at specified time instants, known as snapshots. Specifying many snapshots over time allows for detailed analysis of the radar

PAGE 56

42 wave as it propagates through the media similar to ray tracing methods. This type of analysis is useful for understanding how the radar energy interacts with the narrow conduit geometry at the base of the modeled sinkhole. This process could be useful for understanding which receiver positions would record the strongest reflected energy from the conduit and interpreting the profiles. One arbitrary transmission location (specifically at [4.5,4.0,4.0]) was chosen from the simulated survey line (center line) from model B. Snapshots were collected from 5 ns to 12 0 ns at 5 ns intervals and can be viewed in three orthogonal planes. These snapshots may be compared directly to Figure 2.9a&b's profile at 4.8 meters (4.5 m plus 0.3 m spacing). Shown following this section are snapshots at 20 to 70 ns (Figures 2.10abc 2.20abc). Following the initial contacts of the first reflections on the larger depression, the radar waves begin to enter the conduit structure after 30 nanoseconds (Figure 2.12ab). The energy as it enters the conduit, is sharply diminished (Figure 2.13ab ). Most of the returned energy that is received is from the slope break at the neck of the conduit (Figure 2.14ab), which is why the record indicates no conduit but a weak bottomed basic form depression (Figure 2.7ab). For the remaining time record, reflec tion multiples can be clearly seen but none of which seem to originate from the conduit structure. These later time intervals are useful for identifying strong reflections after first arrivals seen in Figure 2.7ab. The snapshot function demonstrates that n o direct or indirect returns from the conduit boundary arrive at the surface.

PAGE 57

43 Figure 2.10a Figure 2.10b

PAGE 58

44 Figure 2.10a. Top, previous page. Snapshot at 20 ns in x plane. Media boundaries drawn in for visual assistance. Distances are in meters for all snapshots. Figure 2.10b. Bottom, previous page. Snapshot at 20 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.10c. Top, current page. Snapshot at 20 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimposed on image. Figure 2.10c

PAGE 59

45 Figure 2.11a Figure 2.11b

PAGE 60

46 Figure 2.11a. Top, previous page. Snapshot at 25 ns in x plane. Media boundaries drawn in for visual assistance. Distances are in meters for all snapshots. Figure 2.11b. Bottom, previous page. Snapshot at 25 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.11c. Top, current page. Snapshot at 25 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimpose d on image. Figure 2.11c

PAGE 61

47 Figure 2.12a Figure 2.12b

PAGE 62

48 Figure 2.12a. Top, previous page. Snapshot at 30 ns in x plane. Media boundaries drawn in for visual assistance. Distances are in meters for all snapshots. Figure 2.12b. Bottom, previous page. Snapshot at 30 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.12c. Top, current page. Snapshot at 30 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimposed on image. Figure 2.12c

PAGE 63

49 Figure 2.13a Figure 2.13b

PAGE 64

50 Figure 2.13a Top, previous page. Snapshot at 35 ns in x plane. Media boundaries drawn in f or visual assistance. Distances are in meters for all snapshots. Figure 2.13b. Bottom, previous page. Snapshot at 35 ns in y plane. Transmitter location visible at surface, (4.5 ,4.0,4.0). Figure 2.13c. Top, current page. Snapshot at 35 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimposed on image. Figure 2.13c

PAGE 65

51 Figure 2.14a Figure 2.14b

PAGE 66

52 Figure 2.14a. Top, previous page. Snapshot at 40 ns in x plane. Media boundaries drawn in for visual assistance. Distances are in meters for all snapshots. Figure 2.14b. Bottom, previous page. Snapshot at 40 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.14c. Top, cur rent page. Snapshot at 40 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimposed on image. Figure 2.14c

PAGE 67

53 Figure 2.15a Figure 2.15b

PAGE 68

54 Figure 2.15a. Top, previous page. Snapshot at 45 ns in x plane. Media boundari es drawn in for visual assistance. Distances are in meters for all snapshots. Figure 2.15b. Bottom, previous page. Snapshot at 45 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.15c. Top, current page. Snapshot at 45 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimposed on image. Figure 2.15c

PAGE 69

55 Figure 2.16a Figure 2.16b

PAGE 70

56 Figure 2.16a. Top, previous page. Snapshot at 50 ns in x plane. Media boundaries drawn in for visual assistance Distances are in meters for all snapshots. Figure 2.16b. Bottom, previous page. Snapshot at 50 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.16c. Top, current page. Snapshot at 50 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimposed on image. Figure 2.16c

PAGE 71

57 Figure 2.17a Figure 2.17b

PAGE 72

58 Figure 2.17 a. Top, previous page. Snapshot at 55 ns in x plane. Media boundaries drawn in for visual assistance. Distances are in meters for all snapshots. Figure 2.17 b. Bottom, previous page. Snapshot at 55 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.17 c. Top, current page. Snapshot at 55 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimposed on image. Figure 2.17c

PAGE 73

59 Figure 2.18a Figure 2.18b

PAGE 74

60 Figure 2.18a. Top, previous page. Snapshot at 60 ns in x plane. Media boundaries drawn in for visual assistance. Distances are in meters for all snapshots. Figure 2.18b. Bottom, previous page. Snapshot at 60 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.18c. Top, current page. Snapshot at 60 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larg er depression superimposed on image. Figure 2.18c

PAGE 75

61 Figure 2.19a Figure 2.19b

PAGE 76

62 Figure 2.19a. Top, previous page. Snapshot at 65 ns in x plane. Media boundaries drawn in for visual assistance. Distances are in meters for all snapshots. Figure 2.19b. Bottom, previous page. Snapshot at 65 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.19c. Top, current page. Snapshot at 65 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimposed on ima ge. Figure 2.19c

PAGE 77

63 Figure 2.20a Figure 2.20b

PAGE 78

64 Figure 2.20a. Top, previous page. Snapshot at 70 ns in x plane. Media boundaries drawn in for visual assistance. Distances are in meters for all snapshots. Figure 2.20b. Bottom, previous page. Snapshot at 70 ns in y plane. Transmitter location visible at surface, (4.5,4.0,4.0). Figure 2.20c. Top, current page. Snapshot at 70 ns in z plane, specifically at surface. Center point of model, outer rim of conduit, and outer rim of larger depression superimposed on i mage. Figure 2.20c

PAGE 79

65 Modeling: Non S tandar d Acquisition Geometries for Improved Conduit Detection Differences in Receiver Field in XY Plane at Surface ( z =0) One advantage of the modeling software GPRMAX is that it efficiently simulate s nontraditional data collection methods. The key fea ture is the ability to place any number of receivers virtually any place in the model space. The rx_box command enables the user to define a 2D box of evenly spaced receivers over specified dimensions. Each of the receivers in this grid will record all the of the available EM information over time for each transmission. This process can lead to very large storage space requirements when large numbers of receivers are positioned in larger size models. Two different rx_box models (based on the previously de scribed models A and B) were created to identify transmitter receiver locations most sensitive to the presence/absence of the conduit. Rx_box model A and B are identical to model A and B except that they are slightly trimmed around the edges to save proces sing time and storage space. The new models have dimensions of 5.0 x 7.0 x 4.0 meters and 5.0 x 7.0 x 4.5 meters, respectively (Figures 3.1a and 3.1b). The receiver field used was an evenly spaced grid only on the modeled ground surface with a receiver eve ry 0.1 meter in both the x and y directions (Figure 3.2).

PAGE 80

66 Figure 3.1a Figure 3.1b

PAGE 81

67 Figure 3.1a. Top of previous page, Model A: Vertical slice through three dimensional model space at midpoint (y=3.5m). Orange material represents clay; green, sand ; and blue, free space. Valu es for distance given as model coordinate values where (0.0, 0.0, 0.0) represents lower left forward most point and (5.0, 7.0, 4.0) is the upper right back most point in the model space. Figure 3.1b. Bottom of previous page, Model B: Vertical slice t hrou gh three dimensional model space at midpoint (y=3.5m). Orange material represents clay; gree n, sand; and blue, free space. Values for distance given as model coordinate values where (0.0, 0.0, 0.0 ) represents lower left forward most point and (5.0, 7.0, 4. 5) is the upper right back most point in the model space. Figure 3.2

PAGE 82

68 Figure 3.2. Bottom, previous page. Receiver positions used in b oth models. Receiver spread on modeled ground surface only. Outer rims of both conical features superimposed on the image. Th e first test examined receiver field results for receivers oriented in the same direction as the trans mitter (Ex) in the model (note: this is the same as used in previous sections). The data for the first 100 ns from one model is subtracted from the other. The absolute values of this difference trace are summed to capture a single value represent ative of the total difference b etween models A and B at that receiver location. These difference values are then displayed as a contour map over receiver locations (e.g. Figure 3.3). Transmitter locations along the same survey line as used in the previous models are used for the rx_box models. This entire process is then repeated for different transmitter positions (Figures 3.3 3.18). Ideally, this process could yiel d areas on the surface where difference in the two model's subsurface geometry is the greatest. Finding these locations could help plan where to place receivers when trying to maximize sensitivity to conduit presence/absence. Select contour images from a total of 145 transmitter locations can be seen in Figures 3.3 3.18. Although the output images should be symmetric from symmetric model geometry, computational instabilities produced some distortion in the images. The basic patterns are still visible and u seful in determining areas of greatest difference in the receiver field between conduit and no conduit models. There is a pattern that suggests the curvature of depression and conduit cause some of the symmetric distortion (Figures 3.3,

PAGE 83

69 3.14 3.18). Also, s ome of the difference appears on the opposite side of the conduit when the transmitter approaches the center but this pattern is not stable (Figures 3.7 3.9). In most cases, the strongest difference occurs for receiver locations closest to the transmitter. Thus, the principal conclusion of this modeling is that traditional acquisition geometries, in which the receiver antenna is close to the transmitting antenna, maximize sensitivity to the presence or absence of the modeled conduit.

PAGE 84

70 Figure 3.3. Ex Di fference plot at transmitter location 1 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewise, the y axis refers to the y direction at the surface. Color bar units are relative differences with red be ing more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 85

71 Figure 3.4. Ex Difference plot at transmitter location 15 (displayed as a star on plot). The x axis refers to the x direction at the su rface and is in meters. Likewise, the y axis refe rs to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 86

72 Figure 3.5. Ex Difference plot at transmitter location 25 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative d ifferences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 87

73 Figure 3.6. Ex Difference plot at transmitter location 30 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are su perimposed on plot.

PAGE 88

74 Figure 3.7. Ex Difference plot at transmitter location 40 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color ba r units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 89

75 Figure 3.8. Ex Difference plot at transmitter location 45 (displayed as a star on plot). Th e x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both coni cal depressions are superimposed on plot.

PAGE 90

76 Figure 3.9. Ex Difference plot at transmitter location 50 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 91

77 Figure 3.10. Ex Difference plot at transmitter location 60 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 92

78 Figure 3.11. Ex Difference plot at transmitter location 70 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewise, the y axis refe rs to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 93

79 Figure 3.12. Ex Difference plot at transmitte r location 75 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blu e being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 94

80 Figure 3.13. Ex Difference plot at transmitter location 95 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. L ikewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 95

81 Figure 3.14. Ex Diffe rence plot at transmitter location 100 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red bei ng more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 96

82 Figure 3.15. Ex Difference plot at transmitter location 105 (displayed as a star on plot). The x axis refers to the x direction at the s urface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 97

83 Figure 3.16. Ex Difference plot at transmitter location 110 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relativ e differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 98

84 Figure 3.17. Ex Difference plot at transmitter location 115 (displayed as a star on plot). The x axis refers t o the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions a re superimposed on plot.

PAGE 99

85 Figure 3.18. Ex Difference plot at transmitter location 125 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. C olor bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 100

86 An additional receiver geometry examined here is the case in which the receivin g antenna is oriented perpendicular to the transmitting antenna. This methodology has been described in previous studies (e.g. Lutz et al, 2003; Orlando and Slob, 2009). This receiver geometry response can be extracted from the GPRMAX models from the comp uted co mponent of the E field in the y direction at the surface. Once again, select figures from the total available 145 transmitting locations are displayed in Figures 3.19 3.31. The results show that there is almost no difference between the two models records near the transmitter at any location. Also, the greatest differences appear strongest as the transmitter enters the area above the outer rim of the sinkhole depression. Once inside this area, the locations of greatest differences seem to be isolate d in quadrants at 45 degree angles to the lines parallel to the x and y axes resembling somewhat of a 'butterfly' pattern (see Figure 3.25). Although it is possible to record a stronger Ey signal from the conduit, it would not be feasible to implement it because the difference quadrants wander as the transmitter is moved along its survey line (x = 2.5 m). This means that having perpendicular receivers set at specified distances 45 degrees off axis would not have a constant record of the conduit's signal du e to the wander of the difference magnitudes. Not to mention, that trying to employ this method when subsurface conditions are unknown would be entirely guesswork and prove to be ineffective. Simply put, there is not a simple, different transmitter receive r arrangement that will improve resolution over the standard one.

PAGE 101

87 Figure 3.19. Ey Difference plot at transmitter location 10 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewis e, the y axis refe rs to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 102

88 Figure 3.20. Ey Difference plot at transmitte r location 15 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blu e being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 103

89 Figure 3.21. Ey Difference plot at transmitter location 20 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. L ikewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 104

90 Figure 3.22. Ey Diffe rence plot at transmitter location 30 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red bein g more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 105

91 Figure 3.23. Ey Difference plot at transmitter location 40 (displayed as a star on plot). The x axis refers to the x direction at the surf ace and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 106

92 Figure 3.24. Ey Difference plot at transmitter location 55 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative di fferences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 107

93 Figure 3.25. Ey Difference plot at transmitter location 60 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are su perimposed on plot.

PAGE 108

94 Figure 3.26. Ey Difference plot at transmitter location 65 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewise, the y axis re fers to the y direction at the surface. Color b ar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 109

95 Figure 3.27. Ey Difference plot at transmitter location 70 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both co nical depressions are superimposed on plot.

PAGE 110

96 Figure 3.28. Ey Difference plot at transmitter location 85 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 111

97 Figure 3.29. Ey Difference plot at transmitter location 90 (displa yed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less differen t. Outer rims of both conical depressions are superimposed on plot.

PAGE 112

98 Figure 3.30. Ey Difference plot at transmitter location 105 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different and blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 113

99 Figure 3.31. Ey Difference plot at transm itter location 110 (displayed as a star on plot). The x axis refers to the x direction at the surface and is in meters. Likewi se, the y axis refers to the y direction at the surface. Color bar units are relative differences with red being more different an d blue being less different. Outer rims of both conical depressions are superimposed on plot.

PAGE 114

100 Field Data from GeoPark, Tampa, FL High Resolution 3D GPR Grids In order to better understand the connection between the modeled results and real world structure, high resolution 3D GPR data was collecte d from an area with sinkholes and karstic features of dimensions comparable to the sinkhole models The field site is located in the Geopark on the campus of the University of South Florida in T ampa, Florida (see Figure 4.1). The site is a covered karst terrain with high sinkhole density (Stewart and Parker, 1992). Figure 4.2 shows a generalized geology of the GeoPark with 1 2 m of unconsolidated sands underlain by a 2 3 m sequence of fining sand s to clays. Below that, limestone that varies from very weathered on top to a less weathered limestone underneath. Mapped sinkholes in the area range in diameter from a few meters across to approximately 15 meters (see Figure 4.2). In October, 2008 a 3D GPR survey (250 MHz) was collected in order to obtain a 3D data set in an area of the GeoPark that had not been previously studied at high resolution A year later, in October, 2009, another 3D GPR survey (500 MHz ) was acquired to look at a specific section of that area in more detail. The locations of these survey areas are in Figure 4.1. The 3D surveys were conducted by collecting GPR data

PAGE 115

101 an area of approximate ly 35 m by 13 m and used 0.10 m in line spacing (trace interval, 0.05 m) The second survey (500 MHz) covered an area of approximately 16 m by 9 m (inside the previous survey area) and used 0.05 m in line spacing (trace interval, 0.025 m) The specific rad ar system used for both surveys was manufactured by MAL Geoscience. Figure 4.1. Figure modified from Kruse et al (2006) showing approximate locations of previous studies conducted in the USF GeoPark. In addition, the bl ack shaded area represents the ap proximate location of the 250 MHz 3D GPR grid and the red shaded, the 500 MHz grid.

PAGE 116

102 Figure 4.2. Figure showing near surface geology of USF Ge oPark from Kruse et al (2006), originally from Stewart and Parker (1992). Data compiled from CPT and well data a t sites along profile A A' shown in Figure 4.1. Processing the GPR data first required the flipping and fitting of each line of data to minimize offset jitters due to the errors of switching direction at the end of each line. Data were then processed with the program ReflexW (Sandmeier). Both data sets were processed by dewowing, statics, background removal, custom gain functions, and fk filtering. The data were also 3D migrated after this using ReflexW's constant velocity diffraction stack migrations. The 250 MHz data were processed with a velocity of 0.10 m/ns and the 500 MHz data with 0.12 m/ns; found by applying a best fitting velocity

PAGE 117

103 F igure 4.3, 250 MHz, depth = 0.94 m

PAGE 118

104 Figure 4.4, 250 MHz, depth = 1.19 m

PAGE 119

105 Figure 4.5, 250 MHz, depth = 1.48 m

PAGE 120

106 Figure 4.6, 250 MHz, depth = 1.70 m

PAGE 121

107 Figures 4.3 4.6. Migrated time slice 3D GPR (250 MHz) data or iented in the same position as that of the bla ck shaded area it is represented by in Figure 4.1 Figure numbers also increase with time. analysis to various diffraction hyperbolas in the profiles. The differing velocities are presumably due to varying water content in the ground on different dates. P rofiles were acquired along the long axis of the survey grid and those produced from the cross lines are interpolated when making the 3D cube. The cross line profiles are less sharp in appearance than the in line pr ofiles acquired for this reason as well as the fact the spatial sampling is half as dense in the cross line direction. The 250 MHz GPR shows the prominent features (Figures 4.3 4.6). First, there are two general areas of higher hill like reflection surfaces that gradually deepen with t ime. These are most likely the tops of the silty clayey sand horizon which follows the topography of the deeper geologic structure. The structure appears to trend North South with a sag in the middle and steeper drop offs to the east and west. The second f eature of interest is a small sinkhole in the mid to lower SE section of the grid. Only certain selected lines from the entire 3D data cube have been displayed to emphasis certain features and to save space in this publication. The entire area of the sink hole was not surveyed (the sinkhole lies partially off the grid) but enough of it existed on record due to its symmetry to analyze and interpret it as a sinkhole. The small sinkhole, as it appears in the 250 MHz data (see Figure 4.7 4.21), is roughly a cir cular

PAGE 122

108 Figure 4.7, 250 MHz, depth = 1.90 m

PAGE 123

109 Figure 4.8, 250 MHz, depth = 2.19 m

PAGE 124

110 Figure 4.9, 250 MHz, depth = 2.30 m

PAGE 125

111 Figure 4.10, 250 MHz, depth = 2.50 m

PAGE 126

112 Figure 4.11, 250 MHz, depth = 2.60 m

PAGE 127

113 Figures 4.7 4.11. Migrated time slice 3D GPR (250 MHz) data or iented in the same position as that of the black shaded area it is represented by in Figure 4.1 Figure numbers also increase with time. feature about 4 6 m across (Figur e 4.8) and about 2 m deep (Figure 4.14 & 4.20). In all of the GPR figures, the structure can be seen to form from a general sloping of the topography. These two objects are the most prominent identifiable features acquir ed with the 250 MHz GPR antenna freq uency. Anything else that may appear on the profiles is most likely highly complex, near surface features or data artifacts, possibly from migration. The 500 MHz data grid was collected in a smaller area within the previous survey mapped with th e 250 MHz antenna system. The grid location of the high frequency data can be seen in Figure 4.1. As with the previous survey, the in line data were collected along the long axis of the grid (~N S). The selected figures can be seen in Figures 4.22 4.35. Ov erall, the data from the 500 MHz system are generally of a higher resolution than those of the 250 MHz system so the data appear sharper. The prominent features in this 3D survey are essentially the same as those in the other data set. The mound like struc ture of the first major reflection surface between the surface sands and underlying clay rich sands are visible in the near surface (Figures 4.22 & 4.23). They are similar in form when compared to that of the previous survey (Figures 4.3 & 4.4). The sinkho le is once again visible in the mid to lower right hand (E) side of the survey grid (Figure 4.24) and becomes visible at about 40 ns in time slice view. The sinkhole is about

PAGE 128

114 Figure 4.12, 250 MHz

PAGE 129

115 Figure 4.13, 250 MHz

PAGE 130

116 Figure 4.14, 250 MHz

PAGE 131

117 Figures 4.12 4.14. Migrated (time to depth correction of 0. 1 m/ns) 3D G PR (250 MHz) cross line data. Data interpolated across in line profiles and extends acr oss the y distance on the time slice plots. 5 m across (Figure 4.24) and about 2 m deep (Figures 4.30 & 4.35). Even with an imprecise migration time to depth (as seen i n diffraction artifacts in the migrated image), the conical sinkhole collapses to a point and completely disappears in the time record appropriately for typical sinkholes (Carpenter and Ekberg, 2006; Truss et al, 2005). This particular sinkhole geometry is comparable to the model geometry seen in the previous chapters. Results from the models show that steep conduit walls will not be recoverable in the GPR record but shallower slopes will show up. It is also possible that if the base of the sinkhole gradual ly steepens the last part of the recoverable slope will appear as the bottom of the sinkhole in the data profile. This could be the case for an example like that in Figure 4.35 where a slope break exists near the bottom of the sink at about 6.5 m along th e profile. Comparing that figure to Figure 4.34 shows that there might be conduit structure below the bottom due to the bright reflections beneath the inferred bottom reflections. However it is not possible to definitively determine whether or not this si nkhole has active sediment piping or not in the sense that it has well defined conduit structure. Although some studies (Carpenter and Ekberg, 2006; Dobecki and Upchurch, 2006; Batayneh et al, 2002) suggest that more active sinkholes could contain a bright spot of reflections from sediment infilling, this particular sinkhole does not appear to have any sign of it. Any dense collection of reflectors in the top layer of this data are

PAGE 132

118 Figure 4.15, 250 MHz

PAGE 133

119 Figure 4.16, 250 MHz

PAGE 134

120 Figure 4.17, 250 MHz

PAGE 135

121 Figure 4.18, 250 MHz

PAGE 136

122 Figure 4.19, 250 MHz

PAGE 137

1 23 Figure 4.20, 250 MHz

PAGE 138

124 Figure 4.21, 250 MHz

PAGE 139

125 Figures 4.15 4.21. Migrated (time to depth correction o f 0.1 m/ns) 3D GPR (250 MHz) in line data. Data extends across the x distance on the time slice plots. mostly small buried objects, roots, and small slump features that can be collapsed to points with varying migration parameters. On the whole this sinkho le may be considered more of a plugged type. The 500 MHz data show many more small anomalous features than the 250 MHz data The first is a noticeable concentric ring like structure that can be seen in the time slice images (Figures 4.25 4.27). This is an artifact from a buried bundle of wires currently being used for a self potential experiment taking place in the same study area. Many small circular rings abound in the near surface, but are actually small diffractions from buried objects that have been mi grated with the wrong time to depth conversion (in this case, too high for being shallow, dry sands). As they are mainly very shallow features, they were not the focus of this study and not given a best fitting value for migration. It is good practice to identify and analyze features that exist in both data sets and not give much attention to those that exist only in one or the other. This consistency of appearance with varying frequencies is definitely a good method for determining subsurface structures. The feature's structure appears very similar to others seen in GPR profiles collected elsewhere (Carpenter and Ekberg, 2006; Truss et al, 2005). These

PAGE 140

126 Figure 4.22, 500 MHz, depth = 1.14 m

PAGE 141

127 Figure 4.23, 500 MHz, depth = 1.79 m

PAGE 142

128 Figure 4.24, 500 MHz, depth = 2.52 m

PAGE 143

129 Figure 4.25, 500 MHz, depth = 3.12 m

PAGE 144

130 Figure 4.26, 500 MHz, depth = 3.48 m

PAGE 145

131 Figure 4.27 500 MHz, depth = 3.96 m

PAGE 146

132 Figures 4.22 4.27. Migrated time slice 3D GPR (500 MHz) data or iented in the same posi tion as that of the red shaded area it is represented by in Figure 4.1 Figure numbers also increase with time sinkhole images may even contain evidence for conduit structure recovery (Figure 4.20 21) with stronger amplitude reflections below the center of the sinkhole. Although the results cannot be definitive, this particular sinkhole does not seem to be aggressively forming but may be technically active regardless of showing no signs of surface activity (e.g. depression).

PAGE 147

133 Figure 4.28, 500 MHz

PAGE 148

134 Figure 4.29, 500 MHz

PAGE 149

135 Figure 4.30, 500 MHz

PAGE 150

136 Figure 4.31, 500 MHz

PAGE 151

137 Figures 4.28 4.31 Migrated (time to depth correction of 0.1 2 m/ns) 3D GPR (500 MHz) cross line data. Data interpolated across in line profiles and extends acr oss the y distance on the time slice plots.

PAGE 152

138 Figure 4.32, 500 MHz

PAGE 153

139 Figure 4.33, 500 MHz

PAGE 154

140 Figure 4.34, 500 MHz

PAGE 155

141 Figure 4.35, 500 MHz

PAGE 156

142 Figures 4.32 4.35. Migrated (time to depth correction o f 0.12 m/ns) 3D GPR (500 MHz) in line data. Data extends across the x di stance on the time slice plots.

PAGE 157

143 Effects on Interpretation by Decreased Resolution Typically, spatial GPR sinkhole investigations in the Tampa, FL area are not conducted at the high resolution of the surveys shown here (grid spacings may be on the order of meters). In order to understand what this might do to survey interpretations, the 500 MHz 3D GPR data to spacing between lines of 0.50 m instead of 0.05 meter. Migration w as not performed on the down sampled cube as it typically would not be performed on such a sparse data set. A few specific profiles were selected to compare to those displayed in the previous section. The first is a time slice at 42.05 ns (in Figures 4.24 & 4.36) which shows the maximum depression of the sinkhole previously discussed in detail earlier. Figure 4.36 displays the grid with lower resolution. Even with the bright spot in the same position as the sinkhole, it is no longer possible to claim any so rt of sinkhole or depression like feature in the same area with this time slice. Diffraction hyperbolas are also not visible with this spacing interval. The lower resolution cross line profile at 6.510 m (Figure 4.37) cannot provide any useful information when compared to its higher resolution equivalent at 6.417 m (Figure 4.30). The in line profiles should provide a little more useful information since data was collected at this interval and because it has the greatest resolution compared to the other view ing planes. The in line profile at 6.00 m (Figure 4.38) is in fact clearer than the other viewing planes but is still very hard to interpret because of its noisy appearance, due to a high density of shallow diffractions. In some spots the lower reflection horizon clearly visible in Figure 4.32 can be seen in the lower resolution profile but they are not as noticeable or well connected when viewing the higher resolution profile. Collecting, processing, and interpreting 3D

PAGE 158

144 Figure 4.36, 500 MHz, depth = 2.52 m

PAGE 159

145 Figure 4.37, 500 MHz

PAGE 160

146 Figure 4.38, 500 MHz

PAGE 161

147 Figures 4.36 4.38. Selecte d viewing planes from a lower r esolution migrated 3D GPR data cube. [Figure 4.36 Figure 4.37 (y distance p rofile on Figure 4.36) Figure 4.38 (x distance p rofile on Figure 4.36) ]. GPR data volumes at this low resolution is not appropriate for sinkholes a nalysis at this scale. This experiment proves how necessary it is to use such a small interval spacing (0.05 m for 500 MHz) to process 3D data volumes with 3D processing techniques. Here we compare the effective level of interpretation of a 3D GPR data set to conventional 2D processing of a sparse grid. Usually, sparsely spaced GPR profiles are interpreted without any sort of data migration. To see the effect on interpretation, sample profiles from the 500 MHz data cube were taken and left unmigrate d, but processed in every other way exactly the same, in order to compare them to ones with a 3D migration applied. Figure 4.39 displays in line data at the 6.80 m interval which has not been migrated but shows a time to depth conversion of 0.12 m/ns just as the 3D migrated profile seen in Figure 4.33. The lower bright reflector is still visible without migration. Above the lower reflection, there are a great deal of shallow diffractions that are very densely packed together and even give the impression of some sort of shallow structure towards the right edge of the profile. They also blur what could be inferred as a water table at 20 nanoseconds. Figure 4.40 displays a similar result except that the continuous

PAGE 162

148 Figure 4.39, 500 MHz

PAGE 163

149 Figure 4.40, 500 MHz

PAGE 164

150 Fig ures 4.39 4.40. Figures 4.39 and 4.40 represent unmigrated versions of Figures 4.33 and 4.34, respectively. lower reflection seen in Figure 4.34 is not as well observed. In fact, without the 3D migration, the lower reflection may not even be interpreted as much more than some buried objects with larger diffractions. Clearly without the 3D migration, the semi continuity of this reflecting horizon is lost. Another approach to working with sparse GPR data sets is to migrate the data in only two dimensions. Although the validity of this 2D meth od was deemed inappropriate for recovering 3D structures in Chapter 2, the method may still offer some improvement in interpretation of non 3D grid data sets. Figure 4.41 once again looks at the in line data at the 6.80 m interval but has been 2D migrated with a 0.12 m/ns time to depth conversion. The 2D migration does a decent job at clearing the densely packed diffractions above the lower reflector but actually produces false structure in the upper right hand corner which is not produced in the 3D migrati on in Figure 4.33. The lower reflector is recovered similar to that in the 3D migration but is less connected overall. Figure 4.42 displays the in line profile at the 7.70 m interval and has been 2D migrated the same as the previous figure. The 2D migratio n in this profile did a good job at reducing the dense diffractions above the lower reflector. It did not do as well at recovering the continuous structure of the lower reflector as the 3D migration did.

PAGE 165

151 It seems that 2D migrations on data over 3D structur es can recovery some real subsurface structure but can also just as easily introduce new false structures, as the models in Chapter 2 document. Nevertheless it is still probably best to 2D migrate data in hopes of understanding basic subsurface structure w hen 3D migrations are not possible due to insufficient data volumes

PAGE 166

152 Figure 4.41, 500 MHz

PAGE 167

153 Figure 4.42, 500 MHz

PAGE 168

154 Figures 4.41 4.42. Figures 4.41 and 4.42 represent 2D migrated versions of Figures 4.33 and 4.34, respectively.

PAGE 169

155 Conclusions This study started with a simple attempt to understand the limitation traditional 2D surveys for resolving sinkho le structure. Forward modeling wa s used to make simple model geometries to highlight the inaccuracies of 2D GPR profiles over three dimensio nal features. These profiles were processed li ke typical 2D processing sequences, with and without migration. The results suggest that the 2D migration wa s unable to accurately reconstruct the structure of the three dimensional model. Also, ref lected energy from the conduit wa s mostly lost. This thes is then addressed whether standard source a nd receiver survey geometries wer e optimal to resolve the difference in various structures of active sinkholes. A suite of non traditional source and receiver survey geometries were used in the modeling in order t o understand if there wa s added value in using these types of surveys. The simple adjustment of traditional offset distance did not produce better re sults. Non traditional survey geometries wer e proven to be as ineffective at accurately recovering ref lecti ve energy fro m the conduit and traditional spacing appear ed to be best suited even though they are limited. Case studies of very high resolution 3D data of sinkholes and sinkhole related features we re then shown. The data wer e processed using standard 3D processing methods, including 3D migration of the data. Sinkhole features from these profiles serve d

PAGE 170

156 as examples compared to those of the sim plified models. Results from the s e surveys show ed decent structural recovery of a small sinkhole similar in structu re to that of the modeled ones. Indications of active subsidence and possible conduit structure we re apparent from this data. Finally, the dense surveys serve d as a benchmark to compare interpretations taken with the same surveys at lower spatial resoluti ons and profiles with 2D only processing methods in order to understand errors in analysis and interpretation that are possible from 2D surveys. Two dimensional surveys, 2D processed and migrated, show ed some similarity to the 3D results previously mention ed but contain ed more compl exities and artifacts, which le d to poorer interpretation ability. Future work in this specific field of research should include a wider range of model geometries in order to obtain a better understanding of the robustness of tr aditional which the radar energy is still returned from the bottom of th e conduit. Also, producing more model geometries that incorporate simplified versions of complex infilling patterns seen in real world examples of sinkholes would be helpful in the understanding of what those patterns may suggest about the activity of the sinkholes below them.

PAGE 171

157 References Batayneh A Abueladas A & Moumani K A (2002). Use of ground penetrating radar for assessment of potential sinkhole conditions: an example from Ghor al Haditha area, Jordan Environ mental Geol ogy 41:977 983. B randt O., Langley K., Giannopoulos A. & Kohler J. (2006). 3D Modelling of EM Waves in High Arctic Glaciers (II) HPC Europa report. Burger, H. R. Sheehan, A., & Jones, C. (2006). Introduction to applied geophysics: exploring the shallow subsurface W.W. Norton. Carpenter, P. J. & Ekberg, D. W. (2006). Identification of buried sinkholes, fractures and soil pipes using ground penetrating radar and 2D elec trical resistivity tomography. I n Anderson, N.L. (Editor) Proceedings of the 2006 Highway Geophys ics NDE Conference, 437 449. Dobecki, T. L. & Upchurch S B (2006) Geophysical applications to detect sinkholes and ground subsidence Lead ing Edge 25:336 341 Ford, D. C. & Williams P. W ( 1989 ) Karst Geomorphology and Hydrology Unwin Hyman, L ondon, UK. Gerhard, J I Power, C .,Wilson, V Giannopoulos, A & Grant, G (2008 ). DNAPL Mapping by Ground Penetrating Radar Investigated via Numerical Simulation Eos Trans. AGU, 89(53), Fall Meet. Suppl., Abstract H51G 0923 Grasmueck, M., Weger R., & Horstmeyer H ( 2004 ) Three dimensional ground penetrating radar imaging of sedimentary structures, fractures, and archaeological features at submeter resolution Geology, 32(11), 933 936. Grasmueck, M., Weger R., & Horstmeyer H (2005) F ull r esolution 3 D GPR imaging Geophysics, vol. 70, no. 1, pp. K 12 K 19 Giannopoulos, A. (2005a). Modelling of ground penetrating radar using GprMax Construction and Building Materials, 19 pp. 755 762.

PAGE 172

158 Giannopoulos, A. ( 2005b ). GPRMax User Guide V2 Avai lable f rom http:// www.gprmax.org / Kadioglu S ( 2008 ) Photographing layer thicknesses and discontinuities in a marble quarry with 3D GPR visualization J ournal of Applied Geophysics, vol. 64 109 14 Kruse, S. E., Sc hneider, J. C., Campagna, D. J., Inman, J. A., & Hickey, T. D. ( 2000 ) Ground penetrating radar imaging of cap rock, caliche and carbonate strata J ournal of Applied Geophysics, vol. 43, pp. 239 249. Kruse S Grasmueck M Weiss M Viggiano D. ( 2006 ) Sinkhole structure imaging in covered Karst terrain Geophysical Research Letters 33: L16405. DOI: 10.1029/2006GL026975. Mellet J. (1995). Ground penetrating radar applications in engineering, environ mental management, and geology J ournal of Applied Geophysics, vol. 33, pp. 157 166 Orlando, L., Slob E. (2009). Using multicomponent GPR to monitor cracks i n a historical building Journal of Applied Geophysics, v ol 67, iss. 4, p. 327 334. Parker, J. W. (1992). Surficial Aquifer Hydrogeology in a C overed Karst Terrane Masters Thesis, University of South Florida, 228pp. Reynolds, J. M. ( 1997 ) An Introduction to Applied and Environmental Geophysics Wiley, Chichester. Schmidt, W. ( 2005 ). Geological and Geotechnical Investigation Procedures for Eva luation of the Causes of Subsidence Damage In Florida Florida Geological Survey, Special Publication No. 57 Sinclair, W. C., & Stewart, J. W. (1985). Sinkhole type, development, and distribution in Florida Florida Burea u of Geology, Map series No. 11 Sandmeier, K.J. (2005) Software Reflexw. Available f rom http://www.sandmeier g eo.de/ Stewart, M. & Parker, J. (1992). Localization and seasonal variation of recharge in a covered karst aquifer system, Florid a, USA International Contributions Hydrogeology, vol. 13, pp. 443 460. Stewart, M. T. (1998). The Florida water wars: A geologic perspective Geotimes, vol. 43, pp. 24 27. Taflove A. (1995). The Finite Difference in Time Domain Method Artech House, Bost on, London.

PAGE 173

159 Tihansky, A. B. (1999). Sinkholes, west central Florida, in Land Subsidence in the United States edited b y D. Galloway et al., U.S. Geologic Sur vey Circular 1182, 121 140. Truss, S., Grasmueck M., Vega S., & Viggiano D. A. (2007). Imagin g rainfall drainage within the Miami oolitic limestone using high resolution time lapse ground penetrating radar Water Resour. Res., 43, W03405 Wilson, W. L. & Shock, E. J. (1996). New sinkhole data spreadsheet manual (v1.1): Winter Springs, Fla., Subsu rface Evaluations, Inc., 31 p. 3, app., 1 disk. Yee K. S. (1966). Numerical solution of initial boundary value problems involving IEEE Trans. Antennas Propagat., v ol. AP 14, pp. 302 307 Zisman E D (2001) A st andard method for sinkhole detection in the Tampa, Florida area. Environ mental Eng ineering Geosci ence, vol. 7(1):31 50 Zisman E. D. Wightman M. J. & Taylor C. (2005). The Effectiveness of GPR in Sinkhole Investigations ASCE Conf. Proc. 177, 65 DO I:10.1061/40796(177)65.

PAGE 174

160 Appendices

PAGE 175

161 Appendix A: GPRMAX input files Geopark 2D Model A GPRMAX input file #medium: 9.0 0.0 0.0 0.005 1.0 0.0 sand_geopark #medium: 25.0 0.0 0.0 0.05 1.0 0.0 clay_geopark -------------------------------------------------------------------#domain: 8.0 4.0 #dx_dy: 0.02 0.02 #time_window: 100e 9 ---------------------------------------------------------------------#box: 0.0 0.0 8.0 2.5 clay_geopark #box: 0.0 2.5 8.0 3.5 sand_geopark #box: 0.0 3.5 8.0 4.0 free_space --------------------------------------------------------------------#triangle: 2.0 2.5 4.0 0.5 6.0 2.5 sand_geopark ------------------------------------------------------------#line_source: 1.0 250e6 ricker MyLineSource -----------------------------------------------------------#analysis: 128 geopark2_2D.out a #tx: 0.0 3.5 MyLineSource 0.0 100e 9 #rx: 0.3 3.5 #tx_steps: 0.05 0.0 #rx_steps: 0.05 0.0 #end_analysis: -----------------------------------------------------------#geometry_file: geopark2_2D.geo #title: geopark 1 sink in 2D #messages: y -----------------------------------------------------------Geopark 2D Model B GPRMAX input file #medium: 9.0 0.0 0.0 0.005 1.0 0.0 sand_geopark #medium: 25.0 0.0 0.0 0.05 1.0 0.0 clay_geopark -------------------------------------------------------------------#domain: 8.0 4.5 #dx_dy: 0.02 0.02 #time_window: 100e 9

PAGE 176

Appendix A (Continued) 162 ---------------------------------------------------------------------#box: 0.0 0.0 8.0 3.0 clay_geopark #box: 0.0 3.0 8.0 4.0 sand_geopark #box: 0 .0 4.0 8.0 4.5 free_space ---------------------------------------------------------------------#triangle: 2.0 3.0 4.0 1.0 6.0 3.0 sand_geopark #triangle: 3.5 1.5 4.0 0.5 4.5 1.5 sand_geopark ------------------------------------------------------------# line_source: 1.0 250e6 ricker MyLineSource ------------------------------------------------------------#analysis: 128 geopark3_2D.out a #tx: 0.0 4.0 MyLineSource 0.0 100e 9 #rx: 0.3 4.0 #tx_steps: 0.05 0.0 #rx_steps: 0.05 0.0 #end_analysis: ----------------------------------------------------------#geometry_file: geopark3_2D.geo #title: geopark 3 sink in 2D #messages: y Geopark 3D Model A GPRMAX input file #medium: 9.0 0.0 0.0 0.005 1.0 0.0 sand_geopark #medium: 25.0 0.0 0.0 0.05 1.0 0.0 clay_geopa rk --------------------------------------------------------------------#domain: 8.0 8.0 4.0 #dx_dy_dz: 0.02 0.02 0.02 #time_window: 100e 9 ---------------------------------------------------------------------#box: 0.0 0.0 0.0 8.0 8.0 2.5 clay_geopark # box: 0.0 0.0 2.5 8.0 8.0 3.5 sand_geopark #box: 0.0 0.0 3.5 8.0 8.0 4.0 free_space ---------------------------------------------------------------------#cylinder: z 2.480 2.500 4.000 4.000 2.000 sand_geopark #cylinder: z 2.460 2.480 4.000 4.000 1.980 san d_geopark #cylinder: z 2.440 2.460 4.000 4.000 1.960 sand_geopark #cylinder: z 2.420 2.440 4.000 4.000 1.940 sand_geopark

PAGE 177

Appendix A (Continued) 163 #cylinder: z 2.400 2.420 4.000 4.000 1.920 sand_geopark #cylinder: z 2.380 2.400 4.000 4.000 1.900 sand_geopark #cylinder: z 2.360 2.3 80 4.000 4.000 1.880 sand_geopark #cylinder: z 2.340 2.360 4.000 4.000 1.860 sand_geopark #cylinder: z 2.320 2.340 4.000 4.000 1.840 sand_geopark #cylinder: z 2.300 2.320 4.000 4.000 1.820 sand_geopark #cylinder: z 2.280 2.300 4.000 4.000 1.800 sand_geopar k #cylinder: z 2.260 2.280 4.000 4.000 1.780 sand_geopark #cylinder: z 2.240 2.260 4.000 4.000 1.760 sand_geopark #cylinder: z 2.220 2.240 4.000 4.000 1.740 sand_geopark #cylinder: z 2.200 2.220 4.000 4.000 1.720 sand_geopark #cylinder: z 2.180 2.200 4.000 4.000 1.700 sand_geopark #cylinder: z 2.160 2.180 4.000 4.000 1.680 sand_geopark #cylinder: z 2.140 2.160 4.000 4.000 1.660 sand_geopark #cylinder: z 2.120 2.140 4.000 4.000 1.640 sand_geopark #cylinder: z 2.100 2.120 4.000 4.000 1.620 sand_geopark #cylin der: z 2.080 2.100 4.000 4.000 1.600 sand_geopark #cylinder: z 2.060 2.080 4.000 4.000 1.580 sand_geopark #cylinder: z 2.040 2.060 4.000 4.000 1.560 sand_geopark #cylinder: z 2.020 2.040 4.000 4.000 1.540 sand_geopark #cylinder: z 2.000 2.020 4.000 4.000 1 .520 sand_geopark #cylinder: z 1.980 2.000 4.000 4.000 1.500 sand_geopark #cylinder: z 1.960 1.980 4.000 4.000 1.480 sand_geopark #cylinder: z 1.940 1.960 4.000 4.000 1.460 sand_geopark #cylinder: z 1.920 1.940 4.000 4.000 1.440 sand_geopark #cylinder: z 1 .900 1.920 4.000 4.000 1.420 sand_geopark #cylinder: z 1.880 1.900 4.000 4.000 1.400 sand_geopark #cylinder: z 1.860 1.880 4.000 4.000 1.380 sand_geopark #cylinder: z 1.840 1.860 4.000 4.000 1.360 sand_geopark #cylinder: z 1.820 1.840 4.000 4.000 1.340 san d_geopark #cylinder: z 1.800 1.820 4.000 4.000 1.320 sand_geopark #cylinder: z 1.780 1.800 4.000 4.000 1.300 sand_geopark #cylinder: z 1.760 1.780 4.000 4.000 1.280 sand_geopark #cylinder: z 1.740 1.760 4.000 4.000 1.260 sand_geopark #cylinder: z 1.720 1.7 40 4.000 4.000 1.240 sand_geopark #cylinder: z 1.700 1.720 4.000 4.000 1.220 sand_geopark #cylinder: z 1.680 1.700 4.000 4.000 1.200 sand_geopark #cylinder: z 1.660 1.680 4.000 4.000 1.180 sand_geopark #cylinder: z 1.640 1.660 4.000 4.000 1.160 sand_geopar k

PAGE 178

Appendix A (Continued) 164 #cylinder: z 1.620 1.640 4.000 4.000 1.140 sand_geopark #cylinder: z 1.600 1.620 4.000 4.000 1.120 sand_geopark #cylinder: z 1.580 1.600 4.000 4.000 1.100 sand_geopark #cylinder: z 1.560 1.580 4.000 4.000 1.080 sand_geopark #cylinder: z 1.540 1.560 4.000 4.000 1.060 sand_geopark #cylinder: z 1.520 1.540 4.000 4.000 1.040 sand_geopark #cylinder: z 1.500 1.520 4.000 4.000 1.020 sand_geopark #cylinder: z 1.480 1.500 4.000 4.000 1.000 sand_geopark #cylinder: z 1.460 1.480 4.000 4.000 0.980 sand_geopark #cylin der: z 1.440 1.460 4.000 4.000 0.960 sand_geopark #cylinder: z 1.420 1.440 4.000 4.000 0.940 sand_geopark #cylinder: z 1.400 1.420 4.000 4.000 0.920 sand_geopark #cylinder: z 1.380 1.400 4.000 4.000 0.900 sand_geopark #cylinder: z 1.360 1.380 4.000 4.000 0 .880 sand_geopark #cylinder: z 1.340 1.360 4.000 4.000 0.860 sand_geopark #cylinder: z 1.320 1.340 4.000 4.000 0.840 sand_geopark #cylinder: z 1.300 1.320 4.000 4.000 0.820 sand_geopark #cylinder: z 1.280 1.300 4.000 4.000 0.800 sand_geopark #cylinder: z 1 .260 1.280 4.000 4.000 0.780 sand_geopark #cylinder: z 1.240 1.260 4.000 4.000 0.760 sand_geopark #cylinder: z 1.220 1.240 4.000 4.000 0.740 sand_geopark #cylinder: z 1.200 1.220 4.000 4.000 0.720 sand_geopark #cylinder: z 1.180 1.200 4.000 4.000 0.700 san d_geopark #cylinder: z 1.160 1.180 4.000 4.000 0.680 sand_geopark #cylinder: z 1.140 1.160 4.000 4.000 0.660 sand_geopark #cylinder: z 1.120 1.140 4.000 4.000 0.640 sand_geopark #cylinder: z 1.100 1.120 4.000 4.000 0.620 sand_geopark #cylinder: z 1.080 1.1 00 4.000 4.000 0.600 sand_geopark #cylinder: z 1.060 1.080 4.000 4.000 0.580 sand_geopark #cylinder: z 1.040 1.060 4.000 4.000 0.560 sand_geopark #cylinder: z 1.020 1.040 4.000 4.000 0.540 sand_geopark #cylinder: z 1.000 1.020 4.000 4.000 0.520 sand_geopar k #cylinder: z 0.980 1.000 4.000 4.000 0.500 sand_geopark #cylinder: z 0.960 0.980 4.000 4.000 0.480 sand_geopark #cylinder: z 0.940 0.960 4.000 4.000 0.460 sand_geopark #cylinder: z 0.920 0.940 4.000 4.000 0.440 sand_geopark #cylinder: z 0.900 0.920 4.000 4.000 0.420 sand_geopark #cylinder: z 0.880 0.900 4.000 4.000 0.400 sand_geopark #cylinder: z 0.860 0.880 4.000 4.000 0.380 sand_geopark

PAGE 179

Appendix A (Continued) 165 #cylinder: z 0.840 0.860 4.000 4.000 0.360 sand_geopark #cylinder: z 0.820 0.840 4.000 4.000 0.340 sand_geopark #cylin der: z 0.800 0.820 4.000 4.000 0.320 sand_geopark #cylinder: z 0.780 0.800 4.000 4.000 0.300 sand_geopark #cylinder: z 0.760 0.780 4.000 4.000 0.280 sand_geopark #cylinder: z 0.740 0.760 4.000 4.000 0.260 sand_geopark #cylinder: z 0.720 0.740 4.000 4.000 0 .240 sand_geopark #cylinder: z 0.700 0.720 4.000 4.000 0.220 sand_geopark #cylinder: z 0.680 0.700 4.000 4.000 0.200 sand_geopark #cylinder: z 0.660 0.680 4.000 4.000 0.180 sand_geopark #cylinder: z 0.640 0.660 4.000 4.000 0.160 sand_geopark #cylinder: z 0 .620 0.640 4.000 4.000 0.140 sand_geopark #cylinder: z 0.600 0.620 4.000 4.000 0.120 sand_geopark #cylinder: z 0.580 0.600 4.000 4.000 0.100 sand_geopark #cylinder: z 0.560 0.580 4.000 4.000 0.080 sand_geopark #cylinder: z 0.540 0.560 4.000 4.000 0.060 san d_geopark #cylinder: z 0.520 0.540 4.000 4.000 0.040 sand_geopark #cylinder: z 0.500 0.520 4.000 4.000 0.020 sand_geopark ----------------------------------------------------------------------#hertzian_dipole: 1.0 250e6 ricker MyDipole #analysis: 145 geo park1.out a #tx: x 4.0 0.0 3.5 MyDipole 0.0 100e 9 #rx: 4.0 0.3 3.5 #tx_steps: 0.0 0.05 0.0 #rx_steps: 0.0 0.05 0.0 #end_analysis: -----------------------------------------------------------------------#messages: y #title: geopark 1 #geometry_file: geopa rk1.geo Geopark 3D Model B GPRMAX input file #medium: 9.0 0.0 0.0 0.005 1.0 0.0 sand_geopark #medium: 25.0 0.0 0.0 0.05 1.0 0.0 clay_geopark --------------------------------------------------------------------#domain: 8.0 8.0 4.5 #dx_dy_dz: 0.02 0.02 0 .02 #time_window: 120e 9

PAGE 180

Appendix A (Continued) 166 ---------------------------------------------------------------------#box: 0.0 0.0 0.0 8.0 8.0 3.0 clay_geopark #box: 0.0 0.0 3.0 8.0 8.0 4.0 sand_geopark #box: 0.0 0.0 4.0 8.0 8.0 4.5 free_space --------------------------------------------------------------------#cylinder: z 2.980 3.000 4.000 4.000 2.000 sand_geopark #cylinder: z 2.960 2.980 4.000 4.000 1.980 sand_geopark #cylinder: z 2.940 2.960 4.000 4.000 1.960 sand_geopark #cylinder: z 2.920 2.940 4.000 4.000 1.940 sand_ge opark #cylinder: z 2.900 2.920 4.000 4.000 1.920 sand_geopark #cylinder: z 2.880 2.900 4.000 4.000 1.900 sand_geopark #cylinder: z 2.860 2.880 4.000 4.000 1.880 sand_geopark #cylinder: z 2.840 2.860 4.000 4.000 1.860 sand_geopark #cylinder: z 2.820 2.840 4 .000 4.000 1.840 sand_geopark #cylinder: z 2.800 2.820 4.000 4.000 1.820 sand_geopark #cylinder: z 2.780 2.800 4.000 4.000 1.800 sand_geopark #cylinder: z 2.760 2.780 4.000 4.000 1.780 sand_geopark #cylinder: z 2.740 2.760 4.000 4.000 1.760 sand_geopark #c ylinder: z 2.720 2.740 4.000 4.000 1.740 sand_geopark #cylinder: z 2.700 2.720 4.000 4.000 1.720 sand_geopark #cylinder: z 2.680 2.700 4.000 4.000 1.700 sand_geopark #cylinder: z 2.660 2.680 4.000 4.000 1.680 sand_geopark #cylinder: z 2.640 2.660 4.000 4.0 00 1.660 sand_geopark #cylinder: z 2.620 2.640 4.000 4.000 1.640 sand_geopark #cylinder: z 2.600 2.620 4.000 4.000 1.620 sand_geopark #cylinder: z 2.580 2.600 4.000 4.000 1.600 sand_geopark #cylinder: z 2.560 2.580 4.000 4.000 1.580 sand_geopark #cylinder: z 2.540 2.560 4.000 4.000 1.560 sand_geopark #cylinder: z 2.520 2.540 4.000 4.000 1.540 sand_geopark #cylinder: z 2.500 2.520 4.000 4.000 1.520 sand_geopark #cylinder: z 2.480 2.500 4.000 4.000 1.500 sand_geopark #cylinder: z 2.460 2.480 4.000 4.000 1.480 sand_geopark #cylinder: z 2.440 2.460 4.000 4.000 1.460 sand_geopark #cylinder: z 2.420 2.440 4.000 4.000 1.440 sand_geopark #cylinder: z 2.400 2.420 4.000 4.000 1.420 sand_geopark #cylinder: z 2.380 2.400 4.000 4.000 1.400 sand_geopark #cylinder: z 2.360 2.380 4.000 4.000 1.380 sand_geopark #cylinder: z 2.340 2.360 4.000 4.000 1.360 sand_geopark #cylinder: z 2.320 2.340 4.000 4.000 1.340 sand_geopark #cylinder: z 2.300 2.320 4.000 4.000 1.320 sand_geopark

PAGE 181

Appendix A (Continued) 167 #cylinder: z 2.280 2.300 4.000 4.000 1.300 sand_ge opark #cylinder: z 2.260 2.280 4.000 4.000 1.280 sand_geopark #cylinder: z 2.240 2.260 4.000 4.000 1.260 sand_geopark #cylinder: z 2.220 2.240 4.000 4.000 1.240 sand_geopark #cylinder: z 2.200 2.220 4.000 4.000 1.220 sand_geopark #cylinder: z 2.180 2.200 4 .000 4.000 1.200 sand_geopark #cylinder: z 2.160 2.180 4.000 4.000 1.180 sand_geopark #cylinder: z 2.140 2.160 4.000 4.000 1.160 sand_geopark #cylinder: z 2.120 2.140 4.000 4.000 1.140 sand_geopark #cylinder: z 2.100 2.120 4.000 4.000 1.120 sand_geopark #c ylinder: z 2.080 2.100 4.000 4.000 1.100 sand_geopark #cylinder: z 2.060 2.080 4.000 4.000 1.080 sand_geopark #cylinder: z 2.040 2.060 4.000 4.000 1.060 sand_geopark #cylinder: z 2.020 2.040 4.000 4.000 1.040 sand_geopark #cylinder: z 2.000 2.020 4.000 4.0 00 1.020 sand_geopark #cylinder: z 1.980 2.000 4.000 4.000 1.000 sand_geopark #cylinder: z 1.960 1.980 4.000 4.000 0.980 sand_geopark #cylinder: z 1.940 1.960 4.000 4.000 0.960 sand_geopark #cylinder: z 1.920 1.940 4.000 4.000 0.940 sand_geopark #cylinder: z 1.900 1.920 4.000 4.000 0.920 sand_geopark #cylinder: z 1.880 1.900 4.000 4.000 0.900 sand_geopark #cylinder: z 1.860 1.880 4.000 4.000 0.880 sand_geopark #cylinder: z 1.840 1.860 4.000 4.000 0.860 sand_geopark #cylinder: z 1.820 1.840 4.000 4.000 0.840 sand_geopark #cylinder: z 1.800 1.820 4.000 4.000 0.820 sand_geopark #cylinder: z 1.780 1.800 4.000 4.000 0.800 sand_geopark #cylinder: z 1.760 1.780 4.000 4.000 0.780 sand_geopark #cylinder: z 1.740 1.760 4.000 4.000 0.760 sand_geopark #cylinder: z 1.720 1.740 4.000 4.000 0.740 sand_geopark #cylinder: z 1.700 1.720 4.000 4.000 0.720 sand_geopark #cylinder: z 1.680 1.700 4.000 4.000 0.700 sand_geopark #cylinder: z 1.660 1.680 4.000 4.000 0.680 sand_geopark #cylinder: z 1.640 1.660 4.000 4.000 0.660 sand_ge opark #cylinder: z 1.620 1.640 4.000 4.000 0.640 sand_geopark #cylinder: z 1.600 1.620 4.000 4.000 0.620 sand_geopark #cylinder: z 1.580 1.600 4.000 4.000 0.600 sand_geopark #cylinder: z 1.560 1.580 4.000 4.000 0.580 sand_geopark #cylinder: z 1.540 1.560 4 .000 4.000 0.560 sand_geopark #cylinder: z 1.520 1.540 4.000 4.000 0.540 sand_geopark #cylinder: z 1.500 1.520 4.000 4.000 0.520 sand_geopark

PAGE 182

Appendix A (Continued) 168 #cylinder: z 1.480 1.500 4.000 4.000 0.500 sand_geopark #cylinder: z 1.460 1.480 4.000 4.000 0.480 sand_geopark #c ylinder: z 1.440 1.460 4.000 4.000 0.460 sand_geopark #cylinder: z 1.420 1.440 4.000 4.000 0.440 sand_geopark #cylinder: z 1.400 1.420 4.000 4.000 0.420 sand_geopark #cylinder: z 1.380 1.400 4.000 4.000 0.400 sand_geopark #cylinder: z 1.360 1.380 4.000 4.0 00 0.380 sand_geopark #cylinder: z 1.340 1.360 4.000 4.000 0.360 sand_geopark #cylinder: z 1.320 1.340 4.000 4.000 0.340 sand_geopark #cylinder: z 1.300 1.320 4.000 4.000 0.320 sand_geopark #cylinder: z 1.280 1.300 4.000 4.000 0.300 sand_geopark #cylinder: z 1.260 1.280 4.000 4.000 0.280 sand_geopark #cylinder: z 1.240 1.260 4.000 4.000 0.260 sand_geopark #cylinder: z 1.220 1.240 4.000 4.000 0.240 sand_geopark #cylinder: z 1.200 1.220 4.000 4.000 0.220 sand_geopark #cylinder: z 1.180 1.200 4.000 4.000 0.200 sand_geopark #cylinder: z 1.160 1.180 4.000 4.000 0.180 sand_geopark #cylinder: z 1.140 1.160 4.000 4.000 0.160 sand_geopark #cylinder: z 1.120 1.140 4.000 4.000 0.140 sand_geopark #cylinder: z 1.100 1.120 4.000 4.000 0.120 sand_geopark #cylinder: z 1.080 1.100 4.000 4.000 0.100 sand_geopark #cylinder: z 1.060 1.080 4.000 4.000 0.080 sand_geopark #cylinder: z 1.040 1.060 4.000 4.000 0.060 sand_geopark #cylinder: z 1.020 1.040 4.000 4.000 0.040 sand_geopark #cylinder: z 1.000 1.020 4.000 4.000 0.020 sand_ge opark ----------------------------------------------------------------------#cylinder: z 1.480 1.500 4.000 4.000 0.500 sand_geopark #cylinder: z 1.460 1.480 4.000 4.000 0.490 sand_geopark #cylinder: z 1.440 1.460 4.000 4.000 0.480 sand_geopark #cylinder: z 1.420 1.440 4.000 4.000 0.470 sand_geopark #cylinder: z 1.400 1.420 4.000 4.000 0.460 sand_geopark #cylinder: z 1.380 1.400 4.000 4.000 0.450 sand_geopark #cylinder: z 1.360 1.380 4.000 4.000 0.440 sand_geopark #cylinder: z 1.340 1.360 4.000 4.000 0.430 sand_geopark #cylinder: z 1.320 1.340 4.000 4.000 0.420 sand_geopark #cylinder: z 1.300 1.320 4.000 4.000 0.410 sand_geopark #cylinder: z 1.280 1.300 4.000 4.000 0.400 sand_geopark #cylinder: z 1.260 1.280 4.000 4.000 0.390 sand_geopark #cylinder: z 1.240 1.260 4.000 4.000 0.380 sand_geopark #cylinder: z 1.220 1.240 4.000 4.000 0.370 sand_geopark

PAGE 183

Appendix A (Continued) 169 #cylinder: z 1.200 1.220 4.000 4.000 0.360 sand_geopark #cylinder: z 1.180 1.200 4.000 4.000 0.350 sand_geopark #cylinder: z 1.160 1.180 4.000 4.000 0.340 sand_ge opark #cylinder: z 1.140 1.160 4.000 4.000 0.330 sand_geopark #cylinder: z 1.120 1.140 4.000 4.000 0.320 sand_geopark #cylinder: z 1.100 1.120 4.000 4.000 0.310 sand_geopark #cylinder: z 1.080 1.100 4.000 4.000 0.300 sand_geopark #cylinder: z 1.060 1.080 4 .000 4.000 0.290 sand_geopark #cylinder: z 1.040 1.060 4.000 4.000 0.280 sand_geopark #cylinder: z 1.020 1.040 4.000 4.000 0.270 sand_geopark #cylinder: z 1.000 1.020 4.000 4.000 0.260 sand_geopark #cylinder: z 0.980 1.000 4.000 4.000 0.250 sand_geopark #c ylinder: z 0.960 0.980 4.000 4.000 0.240 sand_geopark #cylinder: z 0.940 0.960 4.000 4.000 0.230 sand_geopark #cylinder: z 0.920 0.940 4.000 4.000 0.220 sand_geopark #cylinder: z 0.900 0.920 4.000 4.000 0.210 sand_geopark #cylinder: z 0.880 0.900 4.000 4.0 00 0.200 sand_geopark #cylinder: z 0.860 0.880 4.000 4.000 0.190 sand_geopark #cylinder: z 0.840 0.860 4.000 4.000 0.180 sand_geopark #cylinder: z 0.820 0.840 4.000 4.000 0.170 sand_geopark #cylinder: z 0.800 0.820 4.000 4.000 0.160 sand_geopark #cylinder: z 0.780 0.800 4.000 4.000 0.150 sand_geopark #cylinder: z 0.760 0.780 4.000 4.000 0.140 sand_geopark #cylinder: z 0.740 0.760 4.000 4.000 0.130 sand_geopark #cylinder: z 0.720 0.740 4.000 4.000 0.120 sand_geopark #cylinder: z 0.700 0.720 4.000 4.000 0.110 sand_geopark #cylinder: z 0.680 0.700 4.000 4.000 0.100 sand_geopark #cylinder: z 0.660 0.680 4.000 4.000 0.090 sand_geopark #cylinder: z 0.640 0.660 4.000 4.000 0.080 sand_geopark #cylinder: z 0.620 0.640 4.000 4.000 0.070 sand_geopark #cylinder: z 0.600 0.620 4.000 4.000 0.060 sand_geopark #cylinder: z 0.580 0.600 4.000 4.000 0.050 sand_geopark #cylinder: z 0.560 0.580 4.000 4.000 0.040 sand_geopark #cylinder: z 0.540 0.560 4.000 4.000 0.030 sand_geopark #cylinder: z 0.520 0.540 4.000 4.000 0.020 sand_ge opark #cylinder: z 0.500 0.520 4.000 4.000 0.010 sand_geopark ----------------------------------------------------------------------#hertzian_dipole: 1.0 250e6 ricker MyDipole #analysis: 154 geopark3a.out b #tx: x 4.0 0.0 4.0 MyDipole 0.0 120e 9

PAGE 184

Appendix A (Continued) 170 #rx: 4.0 0.3 4.0 #rx: 4.0 0.4 4.0 #rx: 4.0 0.5 4.0 #rx: 4.0 0.6 4.0 #rx: 4.0 0.7 4.0 #rx: 4.0 0.8 4.0 #rx: 4.0 0.9 4.0 #rx: 4.0 1.0 4.0 #rx: 4.0 1.1 4.0 #rx: 4.0 1.2 4.0 #rx: 4.0 1.3 4.0 #rx: 4.0 1.4 4.0 #rx: 4.0 1.5 4.0 #rx: 4.0 1.6 4.0 #rx: 4.0 1.7 4.0 #rx: 4.0 1.8 4.0 #tx_steps: 0.0 0.05 0.0 #rx_steps: 0.0 0.05 0.0 #end_analysis: -----------------------------------------------------------------------#messages: y #title: geopark 3a #geometry_file: geopark3a.geo ---------------------------------------------------------------------------------------------------------------------------------------------Geopark 3D Model A rx_box GPRMAX input file #medium: 9.0 0.0 0.0 0.005 1.0 0.0 sand_geopark #medium: 25.0 0.0 0.0 0.05 1.0 0.0 clay_geopark -------------------------------------------------------------------#domain: 5.0 7.0 4.0 #dx_dy_dz: 0.02 0.02 0.02 #time_window: 100e 9 ---------------------------------------------------------------------#box: 0.0 0.0 0.0 5.0 7.0 2.5 clay_geopark #box: 0.0 0.0 2.5 5.0 7.0 3.5 sand_geopark #box: 0.0 0.0 3.5 5.0 7.0 4.0 free_space ---------------------------------------------------------------------#cylinder: z 2.480 2.500 2.500 3.500 2.000 sand_geopark

PAGE 185

Appendix A (Continued) 171 #cylinder: z 2.460 2.480 2.500 3.500 1.980 sand_geopark #cylinder: z 2.440 2.460 2.500 3.500 1.960 sand_geopark #cylinder: z 2.420 2.440 2.500 3.500 1.940 sand_geopark #cylinder: z 2.400 2.420 2.500 3.500 1.920 sand_geopark #cylinder: z 2.380 2.400 2.500 3.500 1.900 sand_geopark #cylinder: z 2.360 2.380 2.500 3.500 1.880 sand_geopark #cylinder: z 2.340 2.360 2.500 3.500 1.860 sand_geopark #cylinder: z 2.320 2.340 2.500 3.500 1.840 sand_geopark #cylinder: z 2.300 2.320 2.500 3.500 1.820 sand_geopark #cylinder: z 2.280 2.300 2.500 3.500 1.800 sand_geopark #cylinder: z 2.260 2.280 2.500 3.500 1.780 sand_geopark #cylinder: z 2.240 2.260 2.500 3.500 1.760 sand_geopark #cylinder: z 2.220 2.240 2.500 3.500 1.740 sand_geopark #cylinder: z 2.200 2.220 2.500 3.500 1.720 sand_geopark #cylinder: z 2.180 2.200 2.500 3.500 1.700 sand_geo park #cylinder: z 2.160 2.180 2.500 3.500 1.680 sand_geopark #cylinder: z 2.140 2.160 2.500 3.500 1.660 sand_geopark #cylinder: z 2.120 2.140 2.500 3.500 1.640 sand_geopark #cylinder: z 2.100 2.120 2.500 3.500 1.620 sand_geopark #cylinder: z 2.080 2.100 2. 500 3.500 1.600 sand_geopark #cylinder: z 2.060 2.080 2.500 3.500 1.580 sand_geopark #cylinder: z 2.040 2.060 2.500 3.500 1.560 sand_geopark #cylinder: z 2.020 2.040 2.500 3.500 1.540 sand_geopark #cylinder: z 2.000 2.020 2.500 3.500 1.520 sand_geopark #cy linder: z 1.980 2.000 2.500 3.500 1.500 sand_geopark #cylinder: z 1.960 1.980 2.500 3.500 1.480 sand_geopark #cylinder: z 1.940 1.960 2.500 3.500 1.460 sand_geopark #cylinder: z 1.920 1.940 2.500 3.500 1.440 sand_geopark #cylinder: z 1.900 1.920 2.500 3.50 0 1.420 sand_geopark #cylinder: z 1.880 1.900 2.500 3.500 1.400 sand_geopark #cylinder: z 1.860 1.880 2.500 3.500 1.380 sand_geopark #cylinder: z 1.840 1.860 2.500 3.500 1.360 sand_geopark #cylinder: z 1.820 1.840 2.500 3.500 1.340 sand_geopark #cylinder: z 1.800 1.820 2.500 3.500 1.320 sand_geopark #cylinder: z 1.780 1.800 2.500 3.500 1.300 sand_geopark #cylinder: z 1.760 1.780 2.500 3.500 1.280 sand_geopark #cylinder: z 1.740 1.760 2.500 3.500 1.260 sand_geopark #cylinder: z 1.720 1.740 2.500 3.500 1.240 sand_geopark #cylinder: z 1.700 1.720 2.500 3.500 1.220 sand_geopark #cylinder: z 1.680 1.700 2.500 3.500 1.200 sand_geopark

PAGE 186

Appendix A (Continued) 172 #cylinder: z 1.660 1.680 2.500 3.500 1.180 sand_geopark #cylinder: z 1.640 1.660 2.500 3.500 1.160 sand_geopark #cylinder: z 1.620 1.640 2.500 3.500 1.140 sand_geopark #cylinder: z 1.600 1.620 2.500 3.500 1.120 sand_geopark #cylinder: z 1.580 1.600 2.500 3.500 1.100 sand_geopark #cylinder: z 1.560 1.580 2.500 3.500 1.080 sand_geopark #cylinder: z 1.540 1.560 2.500 3.500 1.060 sand_geo park #cylinder: z 1.520 1.540 2.500 3.500 1.040 sand_geopark #cylinder: z 1.500 1.520 2.500 3.500 1.020 sand_geopark #cylinder: z 1.480 1.500 2.500 3.500 1.000 sand_geopark #cylinder: z 1.460 1.480 2.500 3.500 0.980 sand_geopark #cylinder: z 1.440 1.460 2. 500 3.500 0.960 sand_geopark #cylinder: z 1.420 1.440 2.500 3.500 0.940 sand_geopark #cylinder: z 1.400 1.420 2.500 3.500 0.920 sand_geopark #cylinder: z 1.380 1.400 2.500 3.500 0.900 sand_geopark #cylinder: z 1.360 1.380 2.500 3.500 0.880 sand_geopark #cy linder: z 1.340 1.360 2.500 3.500 0.860 sand_geopark #cylinder: z 1.320 1.340 2.500 3.500 0.840 sand_geopark #cylinder: z 1.300 1.320 2.500 3.500 0.820 sand_geopark #cylinder: z 1.280 1.300 2.500 3.500 0.800 sand_geopark #cylinder: z 1.260 1.280 2.500 3.50 0 0.780 sand_geopark #cylinder: z 1.240 1.260 2.500 3.500 0.760 sand_geopark #cylinder: z 1.220 1.240 2.500 3.500 0.740 sand_geopark #cylinder: z 1.200 1.220 2.500 3.500 0.720 sand_geopark #cylinder: z 1.180 1.200 2.500 3.500 0.700 sand_geopark #cylinder: z 1.160 1.180 2.500 3.500 0.680 sand_geopark #cylinder: z 1.140 1.160 2.500 3.500 0.660 sand_geopark #cylinder: z 1.120 1.140 2.500 3.500 0.640 sand_geopark #cylinder: z 1.100 1.120 2.500 3.500 0.620 sand_geopark #cylinder: z 1.080 1.100 2.500 3.500 0.600 sand_geopark #cylinder: z 1.060 1.080 2.500 3.500 0.580 sand_geopark #cylinder: z 1.040 1.060 2.500 3.500 0.560 sand_geopark #cylinder: z 1.020 1.040 2.500 3.500 0.540 sand_geopark #cylinder: z 1.000 1.020 2.500 3.500 0.520 sand_geopark #cylinder: z 0.980 1.000 2.500 3.500 0.500 sand_geopark #cylinder: z 0.960 0.980 2.500 3.500 0.480 sand_geopark #cylinder: z 0.940 0.960 2.500 3.500 0.460 sand_geopark #cylinder: z 0.920 0.940 2.500 3.500 0.440 sand_geopark #cylinder: z 0.900 0.920 2.500 3.500 0.420 sand_geo park #cylinder: z 0.880 0.900 2.500 3.500 0.400 sand_geopark

PAGE 187

Appendix A (Continued) 173 #cylinder: z 0.860 0.880 2.500 3.500 0.380 sand_geopark #cylinder: z 0.840 0.860 2.500 3.500 0.360 sand_geopark #cylinder: z 0.820 0.840 2.500 3.500 0.340 sand_geopark #cylinder: z 0.800 0.820 2. 500 3.500 0.320 sand_geopark #cylinder: z 0.780 0.800 2.500 3.500 0.300 sand_geopark #cylinder: z 0.760 0.780 2.500 3.500 0.280 sand_geopark #cylinder: z 0.740 0.760 2.500 3.500 0.260 sand_geopark #cylinder: z 0.720 0.740 2.500 3.500 0.240 sand_geopark #cy linder: z 0.700 0.720 2.500 3.500 0.220 sand_geopark #cylinder: z 0.680 0.700 2.500 3.500 0.200 sand_geopark #cylinder: z 0.660 0.680 2.500 3.500 0.180 sand_geopark #cylinder: z 0.640 0.660 2.500 3.500 0.160 sand_geopark #cylinder: z 0.620 0.640 2.500 3.50 0 0.140 sand_geopark #cylinder: z 0.600 0.620 2.500 3.500 0.120 sand_geopark #cylinder: z 0.580 0.600 2.500 3.500 0.100 sand_geopark #cylinder: z 0.560 0.580 2.500 3.500 0.080 sand_geopark #cylinder: z 0.540 0.560 2.500 3.500 0.060 sand_geopark #cylinder: z 0.520 0.540 2.500 3.500 0.040 sand_geopark #cylinder: z 0.500 0.520 2.500 3.500 0.020 sand_geopark ----------------------------------------------------------------------#hertzian_dipole: 1.0 250e6 ricker MyDipole #analysis: 145 geopark1_rxbox.out b #tx : x 2.5 1.1 3.5 MyDipole 0.0 100e 9 #rx_box: 0.1 1.1 3.5 4.9 6.9 3.5 0.1 0.1 0.1 #tx_steps: 0.0 0.05 0.0 #end_analysis: -----------------------------------------------------------------------#messages: y #title: geopark 1 rxbox #geometry_file: geopark1_ rxbox.geo -----------------------------------------------------------------------Geopark 3D Model B rx_box GPRMAX input file #medium: 9.0 0.0 0.0 0.005 1.0 0.0 sand_geopark #medium: 25.0 0.0 0.0 0.05 1.0 0.0 clay_geopark -------------------------------------------------------------------#domain: 5.0 7.0 4.5 #dx_dy_dz: 0.02 0.02 0.02

PAGE 188

Appendix A (Continued) 174 #time_window: 120e 9 ---------------------------------------------------------------------#box: 0.0 0.0 0.0 5.0 7.0 3.0 clay_geopark #box: 0.0 0.0 3.0 5.0 7.0 4.0 sand_ geopark #box: 0.0 0.0 4.0 5.0 7.0 4.5 free_space ---------------------------------------------------------------------#cylinder: z 2.980 3.000 2.500 3.500 2.000 sand_geopark #cylinder: z 2.960 2.980 2.500 3.500 1.980 sand_geopark #cylinder: z 2.940 2. 960 2.500 3.500 1.960 sand_geopark #cylinder: z 2.920 2.940 2.500 3.500 1.940 sand_geopark #cylinder: z 2.900 2.920 2.500 3.500 1.920 sand_geopark #cylinder: z 2.880 2.900 2.500 3.500 1.900 sand_geopark #cylinder: z 2.860 2.880 2.500 3.500 1.880 sand_g eopark #cylinder: z 2.840 2.860 2.500 3.500 1.860 sand_geopark #cylinder: z 2.820 2.840 2.500 3.500 1.840 sand_geopark #cylinder: z 2.800 2.820 2.500 3.500 1.820 sand_geopark #cylinder: z 2.780 2.800 2.500 3.500 1.800 sand_geopark #cylinder: z 2.760 2 .780 2.500 3.500 1.780 sand_geopark #cylinder: z 2.740 2.760 2.500 3.500 1.760 sand_geopark #cylinder: z 2.720 2.740 2.500 3.500 1.740 sand_geopark #cylinder: z 2.700 2.720 2.500 3.500 1.720 sand_geopark #cylinder: z 2.680 2.700 2.500 3.500 1.700 sand_ geopark #cylinder: z 2.660 2.680 2.500 3.500 1.680 sand_geopark #cylinder: z 2.640 2.660 2.500 3.500 1.660 sand_geopark #cylinder: z 2.620 2.640 2.500 3.500 1.640 sand_geopark #cylinder: z 2.600 2.620 2.500 3.500 1.620 sand_geopark #cylinder: z 2.580 2.600 2.500 3.500 1.600 sand_geopark #cylinder: z 2.560 2.580 2.500 3.500 1.580 sand_geopark #cylinder: z 2.540 2.560 2.500 3.500 1.560 sand_geopark #cylinder: z 2.520 2.540 2.500 3.500 1.540 sand_geopark #cylinder: z 2.500 2.520 2.500 3.500 1.520 sand _geopark #cylinder: z 2.480 2.500 2.500 3.500 1.500 sand_geopark #cylinder: z 2.460 2.480 2.500 3.500 1.480 sand_geopark #cylinder: z 2.440 2.460 2.500 3.500 1.460 sand_geopark #cylinder: z 2.420 2.440 2.500 3.500 1.440 sand_geopark #cylinder: z 2.400 2.420 2.500 3.500 1.420 sand_geopark #cylinder: z 2.380 2.400 2.500 3.500 1.400 sand_geopark #cylinder: z 2.360 2.380 2.500 3.500 1.380 sand_geopark #cylinder: z 2.340 2.360 2.500 3.500 1.360 sand_geopark #cylinder: z 2.320 2.340 2.500 3.500 1.340 san d_geopark

PAGE 189

Appendix A (Continued) 175 #cylinder: z 2.300 2.320 2.500 3.500 1.320 sand_geopark #cylinder: z 2.280 2.300 2.500 3.500 1.300 sand_geopark #cylinder: z 2.260 2.280 2.500 3.500 1.280 sand_geopark #cylinder: z 2.240 2.260 2.500 3.500 1.260 sand_geopark #cylinder: z 2.22 0 2.240 2.500 3.500 1.240 sand_geopark #cylinder: z 2.200 2.220 2.500 3.500 1.220 sand_geopark #cylinder: z 2.180 2.200 2.500 3.500 1.200 sand_geopark #cylinder: z 2.160 2.180 2.500 3.500 1.180 sand_geopark #cylinder: z 2.140 2.160 2.500 3.500 1.160 sa nd_geopark #cylinder: z 2.120 2.140 2.500 3.500 1.140 sand_geopark #cylinder: z 2.100 2.120 2.500 3.500 1.120 sand_geopark #cylinder: z 2.080 2.100 2.500 3.500 1.100 sand_geopark #cylinder: z 2.060 2.080 2.500 3.500 1.080 sand_geopark #cylinder: z 2.0 40 2.060 2.500 3.500 1.060 sand_geopark #cylinder: z 2.020 2.040 2.500 3.500 1.040 sand_geopark #cylinder: z 2.000 2.020 2.500 3.500 1.020 sand_geopark #cylinder: z 1.980 2.000 2.500 3.500 1.000 sand_geopark #cylinder: z 1.960 1.980 2.500 3.500 0.980 s and_geopark #cylinder: z 1.940 1.960 2.500 3.500 0.960 sand_geopark #cylinder: z 1.920 1.940 2.500 3.500 0.940 sand_geopark #cylinder: z 1.900 1.920 2.500 3.500 0.920 sand_geopark #cylinder: z 1.880 1.900 2.500 3.500 0.900 sand_geopark #cylinder: z 1. 860 1.880 2.500 3.500 0.880 sand_geopark #cylinder: z 1.840 1.860 2.500 3.500 0.860 sand_geopark #cylinder: z 1.820 1.840 2.500 3.500 0.840 sand_geopark #cylinder: z 1.800 1.820 2.500 3.500 0.820 sand_geopark #cylinder: z 1.780 1.800 2.500 3.500 0.800 sand_geopark #cylinder: z 1.760 1.780 2.500 3.500 0.780 sand_geopark #cylinder: z 1.740 1.760 2.500 3.500 0.760 sand_geopark #cylinder: z 1.720 1.740 2.500 3.500 0.740 sand_geopark #cylinder: z 1.700 1.720 2.500 3.500 0.720 sand_geopark #cylinder: z 1 .680 1.700 2.500 3.500 0.700 sand_geopark #cylinder: z 1.660 1.680 2.500 3.500 0.680 sand_geopark #cylinder: z 1.640 1.660 2.500 3.500 0.660 sand_geopark #cylinder: z 1.620 1.640 2.500 3.500 0.640 sand_geopark #cylinder: z 1.600 1.620 2.500 3.500 0.620 sand_geopark #cylinder: z 1.580 1.600 2.500 3.500 0.600 sand_geopark #cylinder: z 1.560 1.580 2.500 3.500 0.580 sand_geopark #cylinder: z 1.540 1.560 2.500 3.500 0.560 sand_geopark #cylinder: z 1.520 1.540 2.500 3.500 0.540 sand_geopark

PAGE 190

Appendix A (Continued) 176 #cylinder: z 1.500 1.520 2.500 3.500 0.520 sand_geopark #cylinder: z 1.480 1.500 2.500 3.500 0.500 sand_geopark #cylinder: z 1.460 1.480 2.500 3.500 0.480 sand_geopark #cylinder: z 1.440 1.460 2.500 3.500 0.460 sand_geopark #cylinder: z 1.420 1.440 2.500 3.500 0.44 0 sand_geopark #cylinder: z 1.400 1.420 2.500 3.500 0.420 sand_geopark #cylinder: z 1.380 1.400 2.500 3.500 0.400 sand_geopark #cylinder: z 1.360 1.380 2.500 3.500 0.380 sand_geopark #cylinder: z 1.340 1.360 2.500 3.500 0.360 sand_geopark #cylinder: z 1.320 1.340 2.500 3.500 0.340 sand_geopark #cylinder: z 1.300 1.320 2.500 3.500 0.320 sand_geopark #cylinder: z 1.280 1.300 2.500 3.500 0.300 sand_geopark #cylinder: z 1.260 1.280 2.500 3.500 0.280 sand_geopark #cylinder: z 1.240 1.260 2.500 3.500 0.2 60 sand_geopark #cylinder: z 1.220 1.240 2.500 3.500 0.240 sand_geopark #cylinder: z 1.200 1.220 2.500 3.500 0.220 sand_geopark #cylinder: z 1.180 1.200 2.500 3.500 0.200 sand_geopark #cylinder: z 1.160 1.180 2.500 3.500 0.180 sand_geopark #cylinder: z 1.140 1.160 2.500 3.500 0.160 sand_geopark #cylinder: z 1.120 1.140 2.500 3.500 0.140 sand_geopark #cylinder: z 1.100 1.120 2.500 3.500 0.120 sand_geopark #cylinder: z 1.080 1.100 2.500 3.500 0.100 sand_geopark #cylinder: z 1.060 1.080 2.500 3.500 0. 080 sand_geopark #cylinder: z 1.040 1.060 2.500 3.500 0.060 sand_geopark #cylinder: z 1.020 1.040 2.500 3.500 0.040 sand_geopark #cylinder: z 1.000 1.020 2.500 3.500 0.020 sand_geopark --------------------------------------------------------------------#cylinder: z 1.480 1.500 2.500 3.500 0.500 sand_geopark #cylinder: z 1.460 1.480 2.500 3.500 0.490 sand_geopark #cylinder: z 1.440 1.460 2.500 3.500 0.480 sand_geopark #cylinder: z 1.420 1.440 2.500 3.500 0.470 sand_geopark #cylinder: z 1.400 1.420 2.500 3.500 0.460 sand_geopark #cylinder: z 1.380 1.400 2.500 3.500 0.450 sand_geopark #cylinder: z 1.360 1.380 2.500 3.500 0.440 sand_geopark #cylinder: z 1.340 1.360 2.500 3.500 0.430 sand_geopark #cylinder: z 1.320 1.340 2.500 3.500 0.420 sand_geop ark #cylinder: z 1.300 1.320 2.500 3.500 0.410 sand_geopark #cylinder: z 1.280 1.300 2.500 3.500 0.400 sand_geopark #cylinder: z 1.260 1.280 2.500 3.500 0.390 sand_geopark #cylinder: z 1.240 1.260 2.500 3.500 0.380 sand_geopark

PAGE 191

Appendix A (Continued) 177 #cylinder: z 1.220 1.24 0 2.500 3.500 0.370 sand_geopark #cylinder: z 1.200 1.220 2.500 3.500 0.360 sand_geopark #cylinder: z 1.180 1.200 2.500 3.500 0.350 sand_geopark #cylinder: z 1.160 1.180 2.500 3.500 0.340 sand_geopark #cylinder: z 1.140 1.160 2.500 3.500 0.330 sand_geo park #cylinder: z 1.120 1.140 2.500 3.500 0.320 sand_geopark #cylinder: z 1.100 1.120 2.500 3.500 0.310 sand_geopark #cylinder: z 1.080 1.100 2.500 3.500 0.300 sand_geopark #cylinder: z 1.060 1.080 2.500 3.500 0.290 sand_geopark #cylinder: z 1.040 1.0 60 2.500 3.500 0.280 sand_geopark #cylinder: z 1.020 1.040 2.500 3.500 0.270 sand_geopark #cylinder: z 1.000 1.020 2.500 3.500 0.260 sand_geopark #cylinder: z 0.980 1.000 2.500 3.500 0.250 sand_geopark #cylinder: z 0.960 0.980 2.500 3.500 0.240 sand_ge opark #cylinder: z 0.940 0.960 2.500 3.500 0.230 sand_geopark #cylinder: z 0.920 0.940 2.500 3.500 0.220 sand_geopark #cylinder: z 0.900 0.920 2.500 3.500 0.210 sand_geopark #cylinder: z 0.880 0.900 2.500 3.500 0.200 sand_geopark #cylinder: z 0.860 0. 880 2.500 3.500 0.190 sand_geopark #cylinder: z 0.840 0.860 2.500 3.500 0.180 sand_geopark #cylinder: z 0.820 0.840 2.500 3.500 0.170 sand_geopark #cylinder: z 0.800 0.820 2.500 3.500 0.160 sand_geopark #cylinder: z 0.780 0.800 2.500 3.500 0.150 sand_g eopark #cylinder: z 0.760 0.780 2.500 3.500 0.140 sand_geopark #cylinder: z 0.740 0.760 2.500 3.500 0.130 sand_geopark #cylinder: z 0.720 0.740 2.500 3.500 0.120 sand_geopark #cylinder: z 0.700 0.720 2.500 3.500 0.110 sand_geopark #cylinder: z 0.680 0 .700 2.500 3.500 0.100 sand_geopark #cylinder: z 0.660 0.680 2.500 3.500 0.090 sand_geopark #cylinder: z 0.640 0.660 2.500 3.500 0.080 sand_geopark #cylinder: z 0.620 0.640 2.500 3.500 0.070 sand_geopark #cylinder: z 0.600 0.620 2.500 3.500 0.060 sand_ geopark #cylinder: z 0.580 0.600 2.500 3.500 0.050 sand_geopark #cylinder: z 0.560 0.580 2.500 3.500 0.040 sand_geopark #cylinder: z 0.540 0.560 2.500 3.500 0.030 sand_geopark #cylinder: z 0.520 0.540 2.500 3.500 0.020 sand_geopark #cylinder: z 0.500 0.520 2.500 3.500 0.010 sand_geopark ----------------------------------------------------------------------#hertzian_dipole: 1.0 250e6 ricker MyDipole #analysis: 145 geopark3_rxbox.out b

PAGE 192

Appendix A (Continued) 178 #tx: x 2.5 1.1 4.0 MyDipole 0.0 120e 9 #rx_box: 0.1 1.1 4.0 4.9 6.9 4.0 0.1 0.1 0.1 #tx_steps: 0.0 0.05 0.0 #end_analysis: -----------------------------------------------------------------------#messages: y #title: geopark 3 rxbox #geometry_file: geopark3_rxbox.geo ----------------------------------------------------------------------Geopark 3D Model B rx_box off axis GPRMAX input file #medium: 9.0 0.0 0.0 0.005 1.0 0.0 sand_geopark #medium: 25.0 0.0 0.0 0.05 1.0 0.0 clay_geopark --------------------------------------------------------------------#domain: 5.0 7.0 4.5 #dx_dy_dz: 0.02 0.02 0.02 #time_window: 120e 9 ---------------------------------------------------------------------#box: 0.0 0.0 0.0 5.0 7.0 3.0 clay_geopark #box: 0.0 0.0 3.0 5.0 7.0 4.0 sand_geopark #box: 0.0 0.0 4.0 5.0 7.0 4.5 free_space --------------------------------------------------------------------#cylinder: z 2.980 3.000 2.500 3.500 2.000 sand_geopark #cylinder: z 2.960 2.980 2.500 3.500 1.980 sand_geopark #cylinder: z 2.940 2.960 2.500 3.500 1.960 sand_geopark #cylinder: z 2.920 2.940 2.500 3.500 1.940 sand_geopark #cylinder: z 2.900 2.920 2.500 3.500 1.920 sand_geopark #cylinder: z 2.880 2.900 2.500 3.500 1.900 sand_geopark #cylinder: z 2.860 2.880 2.500 3.500 1.880 sand_geopark #cylinder: z 2.840 2.860 2.500 3.500 1.860 sand _geopark #cylinder: z 2.820 2.840 2.500 3.500 1.840 sand_geopark #cylinder: z 2.800 2.820 2.500 3.500 1.820 sand_geopark #cylinder: z 2.780 2.800 2.500 3.500 1.800 sand_geopark #cylinder: z 2.760 2.780 2.500 3.500 1.780 sand_geopark #cylinder: z 2.740 2.760 2.500 3.500 1.760 sand_geopark #cylinder: z 2.720 2.740 2.500 3.500 1.740 sand_geopark #cylinder: z 2.700 2.720 2.500 3.500 1.720 sand_geopark #cylinder: z 2.680 2.700 2.500 3.500 1.700 sand_geopark #cylinder: z 2.660 2.680 2.500 3.500 1.680 san d_geopark

PAGE 193

Appendix A (Continued) 179 #cylinder: z 2.640 2.660 2.500 3.500 1.660 sand_geopark #cylinder: z 2.620 2.640 2.500 3.500 1.640 sand_geopark #cylinder: z 2.600 2.620 2.500 3.500 1.620 sand_geopark #cylinder: z 2.580 2.600 2.500 3.500 1.600 sand_geopark #cylinder: z 2.56 0 2.580 2.500 3.500 1.580 sand_geopark #cylinder: z 2.540 2.560 2.500 3.500 1.560 sand_geopark #cylinder: z 2.520 2.540 2.500 3.500 1.540 sand_geopark #cylinder: z 2.500 2.520 2.500 3.500 1.520 sand_geopark #cylinder: z 2.480 2.500 2.500 3.500 1.500 sa nd_geopark #cylinder: z 2.460 2.480 2.500 3.500 1.480 sand_geopark #cylinder: z 2.440 2.460 2.500 3.500 1.460 sand_geopark #cylinder: z 2.420 2.440 2.500 3.500 1.440 sand_geopark #cylinder: z 2.400 2.420 2.500 3.500 1.420 sand_geopark #cylinder: z 2.3 80 2.400 2.500 3.500 1.400 sand_geopark #cylinder: z 2.360 2.380 2.500 3.500 1.380 sand_geopark #cylinder: z 2.340 2.360 2.500 3.500 1.360 sand_geopark #cylinder: z 2.320 2.340 2.500 3.500 1.340 sand_geopark #cylinder: z 2.300 2.320 2.500 3.500 1.320 s and_geopark #cylinder: z 2.280 2.300 2.500 3.500 1.300 sand_geopark #cylinder: z 2.260 2.280 2.500 3.500 1.280 sand_geopark #cylinder: z 2.240 2.260 2.500 3.500 1.260 sand_geopark #cylinder: z 2.220 2.240 2.500 3.500 1.240 sand_geopark #cylinder: z 2. 200 2.220 2.500 3.500 1.220 sand_geopark #cylinder: z 2.180 2.200 2.500 3.500 1.200 sand_geopark #cylinder: z 2.160 2.180 2.500 3.500 1.180 sand_geopark #cylinder: z 2.140 2.160 2.500 3.500 1.160 sand_geopark #cylinder: z 2.120 2.140 2.500 3.500 1.140 sand_geopark #cylinder: z 2.100 2.120 2.500 3.500 1.120 sand_geopark #cylinder: z 2.080 2.100 2.500 3.500 1.100 sand_geopark #cylinder: z 2.060 2.080 2.500 3.500 1.080 sand_geopark #cylinder: z 2.040 2.060 2.500 3.500 1.060 sand_geopark #cylinder: z 2 .020 2.040 2.500 3.500 1.040 sand_geopark #cylinder: z 2.000 2.020 2.500 3.500 1.020 sand_geopark #cylinder: z 1.980 2.000 2.500 3.500 1.000 sand_geopark #cylinder: z 1.960 1.980 2.500 3.500 0.980 sand_geopark #cylinder: z 1.940 1.960 2.500 3.500 0.960 sand_geopark #cylinder: z 1.920 1.940 2.500 3.500 0.940 sand_geopark #cylinder: z 1.900 1.920 2.500 3.500 0.920 sand_geopark #cylinder: z 1.880 1.900 2.500 3.500 0.900 sand_geopark #cylinder: z 1.860 1.880 2.500 3.500 0.880 sand_geopark

PAGE 194

Appendix A (Continued) 180 #cylinder: z 1.840 1.860 2.500 3.500 0.860 sand_geopark #cylinder: z 1.820 1.840 2.500 3.500 0.840 sand_geopark #cylinder: z 1.800 1.820 2.500 3.500 0.820 sand_geopark #cylinder: z 1.780 1.800 2.500 3.500 0.800 sand_geopark #cylinder: z 1.760 1.780 2.500 3.500 0.78 0 sand_geopark #cylinder: z 1.740 1.760 2.500 3.500 0.760 sand_geopark #cylinder: z 1.720 1.740 2.500 3.500 0.740 sand_geopark #cylinder: z 1.700 1.720 2.500 3.500 0.720 sand_geopark #cylinder: z 1.680 1.700 2.500 3.500 0.700 sand_geopark #cylinder: z 1.660 1.680 2.500 3.500 0.680 sand_geopark #cylinder: z 1.640 1.660 2.500 3.500 0.660 sand_geopark #cylinder: z 1.620 1.640 2.500 3.500 0.640 sand_geopark #cylinder: z 1.600 1.620 2.500 3.500 0.620 sand_geopark #cylinder: z 1.580 1.600 2.500 3.500 0.6 00 sand_geopark #cylinder: z 1.560 1.580 2.500 3.500 0.580 sand_geopark #cylinder: z 1.540 1.560 2.500 3.500 0.560 sand_geopark #cylinder: z 1.520 1.540 2.500 3.500 0.540 sand_geopark #cylinder: z 1.500 1.520 2.500 3.500 0.520 sand_geopark #cylinder: z 1.480 1.500 2.500 3.500 0.500 sand_geopark #cylinder: z 1.460 1.480 2.500 3.500 0.480 sand_geopark #cylinder: z 1.440 1.460 2.500 3.500 0.460 sand_geopark #cylinder: z 1.420 1.440 2.500 3.500 0.440 sand_geopark #cylinder: z 1.400 1.420 2.500 3.500 0. 420 sand_geopark #cylinder: z 1.380 1.400 2.500 3.500 0.400 sand_geopark #cylinder: z 1.360 1.380 2.500 3.500 0.380 sand_geopark #cylinder: z 1.340 1.360 2.500 3.500 0.360 sand_geopark #cylinder: z 1.320 1.340 2.500 3.500 0.340 sand_geopark #cylinder: z 1.300 1.320 2.500 3.500 0.320 sand_geopark #cylinder: z 1.280 1.300 2.500 3.500 0.300 sand_geopark #cylinder: z 1.260 1.280 2.500 3.500 0.280 sand_geopark #cylinder: z 1.240 1.260 2.500 3.500 0.260 sand_geopark #cylinder: z 1.220 1.240 2.500 3.500 0 .240 sand_geopark #cylinder: z 1.200 1.220 2.500 3.500 0.220 sand_geopark #cylinder: z 1.180 1.200 2.500 3.500 0.200 sand_geopark #cylinder: z 1.160 1.180 2.500 3.500 0.180 sand_geopark #cylinder: z 1.140 1.160 2.500 3.500 0.160 sand_geopark #cylinder: z 1.120 1.140 2.500 3.500 0.140 sand_geopark #cylinder: z 1.100 1.120 2.500 3.500 0.120 sand_geopark #cylinder: z 1.080 1.100 2.500 3.500 0.100 sand_geopark #cylinder: z 1.060 1.080 2.500 3.500 0.080 sand_geopark

PAGE 195

Appendix A (Continued) 181 #cylinder: z 1.040 1.060 2.500 3.500 0.060 sand_geopark #cylinder: z 1.020 1.040 2.500 3.500 0.040 sand_geopark #cylinder: z 1.000 1.020 2.500 3.500 0.020 sand_geopark ---------------------------------------------------------------------#cylinder: z 1.480 1.500 2.500 3.500 0.500 sand_geo park #cylinder: z 1.460 1.480 2.500 3.500 0.490 sand_geopark #cylinder: z 1.440 1.460 2.500 3.500 0.480 sand_geopark #cylinder: z 1.420 1.440 2.500 3.500 0.470 sand_geopark #cylinder: z 1.400 1.420 2.500 3.500 0.460 sand_geopark #cylinder: z 1.380 1.4 00 2.500 3.500 0.450 sand_geopark #cylinder: z 1.360 1.380 2.500 3.500 0.440 sand_geopark #cylinder: z 1.340 1.360 2.500 3.500 0.430 sand_geopark #cylinder: z 1.320 1.340 2.500 3.500 0.420 sand_geopark #cylinder: z 1.300 1.320 2.500 3.500 0.410 sand_ge opark #cylinder: z 1.280 1.300 2.500 3.500 0.400 sand_geopark #cylinder: z 1.260 1.280 2.500 3.500 0.390 sand_geopark #cylinder: z 1.240 1.260 2.500 3.500 0.380 sand_geopark #cylinder: z 1.220 1.240 2.500 3.500 0.370 sand_geopark #cylinder: z 1.200 1. 220 2.500 3.500 0.360 sand_geopark #cylinder: z 1.180 1.200 2.500 3.500 0.350 sand_geopark #cylinder: z 1.160 1.180 2.500 3.500 0.340 sand_geopark #cylinder: z 1.140 1.160 2.500 3.500 0.330 sand_geopark #cylinder: z 1.120 1.140 2.500 3.500 0.320 sand_g eopark #cylinder: z 1.100 1.120 2.500 3.500 0.310 sand_geopark #cylinder: z 1.080 1.100 2.500 3.500 0.300 sand_geopark #cylinder: z 1.060 1.080 2.500 3.500 0.290 sand_geopark #cylinder: z 1.040 1.060 2.500 3.500 0.280 sand_geopark #cylinder: z 1.020 1 .040 2.500 3.500 0.270 sand_geopark #cylinder: z 1.000 1.020 2.500 3.500 0.260 sand_geopark #cylinder: z 0.980 1.000 2.500 3.500 0.250 sand_geopark #cylinder: z 0.960 0.980 2.500 3.500 0.240 sand_geopark #cylinder: z 0.940 0.960 2.500 3.500 0.230 sand_ geopark #cylinder: z 0.920 0.940 2.500 3.500 0.220 sand_geopark #cylinder: z 0.900 0.920 2.500 3.500 0.210 sand_geopark #cylinder: z 0.880 0.900 2.500 3.500 0.200 sand_geopark #cylinder: z 0.860 0.880 2.500 3.500 0.190 sand_geopark #cylinder: z 0.840 0.860 2.500 3.500 0.180 sand_geopark #cylinder: z 0.820 0.840 2.500 3.500 0.170 sand_geopark #cylinder: z 0.800 0.820 2.500 3.500 0.160 sand_geopark #cylinder: z 0.780 0.800 2.500 3.500 0.150 sand_geopark

PAGE 196

Appendix A (Continued) 182 #cylinder: z 0.760 0.780 2.500 3.500 0.140 sand _geopark #cylinder: z 0.740 0.760 2.500 3.500 0.130 sand_geopark #cylinder: z 0.720 0.740 2.500 3.500 0.120 sand_geopark #cylinder: z 0.700 0.720 2.500 3.500 0.110 sand_geopark #cylinder: z 0.680 0.700 2.500 3.500 0.100 sand_geopark #cylinder: z 0.660 0.680 2.500 3.500 0.090 sand_geopark #cylinder: z 0.640 0.660 2.500 3.500 0.080 sand_geopark #cylinder: z 0.620 0.640 2.500 3.500 0.070 sand_geopark #cylinder: z 0.600 0.620 2.500 3.500 0.060 sand_geopark #cylinder: z 0.580 0.600 2.500 3.500 0.050 san d_geopark #cylinder: z 0.560 0.580 2.500 3.500 0.040 sand_geopark #cylinder: z 0.540 0.560 2.500 3.500 0.030 sand_geopark #cylinder: z 0.520 0.540 2.500 3.500 0.020 sand_geopark #cylinder: z 0.500 0.520 2.500 3.500 0.010 sand_geopark ---------------------------------------------------------------------#hertzian_dipole: 1.0 250e6 ricker MyDipole #analysis: 145 geopark5_rxbox.out b #tx: x 2.8 1.1 4.0 MyDipole 0.0 120e 9 #rx_box: 0.1 1.1 4.0 4.9 6.9 4.0 0.1 0.1 0.1 #tx_steps: 0.0 0.05 0.0 #end_analysis: -----------------------------------------------------------------------#messages: y #title: geopark 5 rxbox #geometry_file: geopark3_rxbox.geo -----------------------------------------------------------------------Geopark 3D Model B snapshot GPRMAX input file #medium: 9.0 0.0 0.0 0.005 1.0 0.0 sand_geopark #medium: 25.0 0.0 0.0 0.05 1.0 0.0 clay_geopark --------------------------------------------------------------------#domain: 8.0 8.0 4.5 #dx_dy_dz: 0.02 0.02 0.02 #time_window: 120e 9 --------------------------------------------------------------------#box: 0.0 0.0 0.0 8.0 8.0 3.0 clay_geopark #box: 0.0 0.0 3.0 8.0 8.0 4.0 sand_geopark #box: 0.0 0.0 4.0 8.0 8.0 4.5 free_space

PAGE 197

Appendix A (Continued) 183 --------------------------------------------------------------------#cylinder: z 2.980 3.000 4.000 4.000 2.000 sand_geopark #cylinder: z 2.960 2.980 4.000 4.000 1.980 sand_geopark #cylinder: z 2.940 2.960 4.000 4.000 1.960 sand_geopark #cylinder: z 2.920 2.940 4.000 4.000 1.940 sand_geopark #cylinder: z 2.900 2.920 4.00 0 4.000 1.920 sand_geopark #cylinder: z 2.880 2.900 4.000 4.000 1.900 sand_geopark #cylinder: z 2.860 2.880 4.000 4.000 1.880 sand_geopark #cylinder: z 2.840 2.860 4.000 4.000 1.860 sand_geopark #cylinder: z 2.820 2.840 4.000 4.000 1.840 sand_geopark #cyli nder: z 2.800 2.820 4.000 4.000 1.820 sand_geopark #cylinder: z 2.780 2.800 4.000 4.000 1.800 sand_geopark #cylinder: z 2.760 2.780 4.000 4.000 1.780 sand_geopark #cylinder: z 2.740 2.760 4.000 4.000 1.760 sand_geopark #cylinder: z 2.720 2.740 4.000 4.000 1.740 sand_geopark #cylinder: z 2.700 2.720 4.000 4.000 1.720 sand_geopark #cylinder: z 2.680 2.700 4.000 4.000 1.700 sand_geopark #cylinder: z 2.660 2.680 4.000 4.000 1.680 sand_geopark #cylinder: z 2.640 2.660 4.000 4.000 1.660 sand_geopark #cylinder: z 2.620 2.640 4.000 4.000 1.640 sand_geopark #cylinder: z 2.600 2.620 4.000 4.000 1.620 sand_geopark #cylinder: z 2.580 2.600 4.000 4.000 1.600 sand_geopark #cylinder: z 2.560 2.580 4.000 4.000 1.580 sand_geopark #cylinder: z 2.540 2.560 4.000 4.000 1.560 sa nd_geopark #cylinder: z 2.520 2.540 4.000 4.000 1.540 sand_geopark #cylinder: z 2.500 2.520 4.000 4.000 1.520 sand_geopark #cylinder: z 2.480 2.500 4.000 4.000 1.500 sand_geopark #cylinder: z 2.460 2.480 4.000 4.000 1.480 sand_geopark #cylinder: z 2.440 2. 460 4.000 4.000 1.460 sand_geopark #cylinder: z 2.420 2.440 4.000 4.000 1.440 sand_geopark #cylinder: z 2.400 2.420 4.000 4.000 1.420 sand_geopark #cylinder: z 2.380 2.400 4.000 4.000 1.400 sand_geopark #cylinder: z 2.360 2.380 4.000 4.000 1.380 sand_geopa rk #cylinder: z 2.340 2.360 4.000 4.000 1.360 sand_geopark #cylinder: z 2.320 2.340 4.000 4.000 1.340 sand_geopark #cylinder: z 2.300 2.320 4.000 4.000 1.320 sand_geopark #cylinder: z 2.280 2.300 4.000 4.000 1.300 sand_geopark #cylinder: z 2.260 2.280 4.00 0 4.000 1.280 sand_geopark #cylinder: z 2.240 2.260 4.000 4.000 1.260 sand_geopark #cylinder: z 2.220 2.240 4.000 4.000 1.240 sand_geopark

PAGE 198

Appendix A (Continued) 184 #cylinder: z 2.200 2.220 4.000 4.000 1.220 sand_geopark #cylinder: z 2.180 2.200 4.000 4.000 1.200 sand_geopark #cyli nder: z 2.160 2.180 4.000 4.000 1.180 sand_geopark #cylinder: z 2.140 2.160 4.000 4.000 1.160 sand_geopark #cylinder: z 2.120 2.140 4.000 4.000 1.140 sand_geopark #cylinder: z 2.100 2.120 4.000 4.000 1.120 sand_geopark #cylinder: z 2.080 2.100 4.000 4.000 1.100 sand_geopark #cylinder: z 2.060 2.080 4.000 4.000 1.080 sand_geopark #cylinder: z 2.040 2.060 4.000 4.000 1.060 sand_geopark #cylinder: z 2.020 2.040 4.000 4.000 1.040 sand_geopark #cylinder: z 2.000 2.020 4.000 4.000 1.020 sand_geopark #cylinder: z 1.980 2.000 4.000 4.000 1.000 sand_geopark #cylinder: z 1.960 1.980 4.000 4.000 0.980 sand_geopark #cylinder: z 1.940 1.960 4.000 4.000 0.960 sand_geopark #cylinder: z 1.920 1.940 4.000 4.000 0.940 sand_geopark #cylinder: z 1.900 1.920 4.000 4.000 0.920 sa nd_geopark #cylin der: z 1.880 1.900 4.000 4.000 0.900 sand_geopark #cylinder: z 1.860 1.880 4.000 4.000 0.880 sand_geopark #cylinder: z 1.840 1.860 4.000 4.000 0.860 sand_geopark #cylinder: z 1.820 1.840 4.000 4.000 0.840 sand_geopark #cylinder: z 1.800 1. 820 4.000 4.000 0.820 sand_geopark #cylinder: z 1.780 1.800 4.000 4.000 0.800 sand_geopark #cylinder: z 1.760 1.780 4.000 4.000 0.780 sand_geopark #cylinder: z 1.740 1.760 4.000 4.000 0.760 sand_geopark #cylinder: z 1.720 1.740 4.000 4.000 0.740 sand_geopa rk #cylinder: z 1.700 1.720 4.000 4.000 0.720 sand_geopark #cylinder: z 1.680 1.700 4.000 4.000 0.700 sand_geopark #cylinder: z 1.660 1.680 4.000 4.000 0.680 sand_geopark #cylinder: z 1.640 1.660 4.000 4.000 0.660 sand_geopark #cylinder: z 1.620 1.640 4.00 0 4.000 0.640 sand_geopark #cylinder: z 1.600 1.620 4.000 4.000 0.620 sand_geopark #cylinder: z 1.580 1.600 4.000 4.000 0.600 sand_geopark #cylinder: z 1.560 1.580 4.000 4.000 0.580 sand_geopark #cylinder: z 1.540 1.560 4.000 4.000 0.560 sand_geopark #cyli nder: z 1.520 1.540 4.000 4.000 0.540 sand_geopark #cylinder: z 1.500 1.520 4.000 4.000 0.520 sand_geopark #cylinder: z 1.480 1.500 4.000 4.000 0.500 sand_geopark #cylinder: z 1.460 1.480 4.000 4.000 0.480 sand_geopark #cylinder: z 1.440 1.460 4.000 4.000 0.460 sand_geopark #cylinder: z 1.420 1.440 4.000 4.000 0.440 sand_geopark

PAGE 199

Appendix A (Continued) 185 #cylinder: z 1.400 1.420 4.000 4.000 0.420 sand_geopark #cylinder: z 1.380 1.400 4.000 4.000 0.400 sand_geopark #cylinder: z 1.360 1.380 4.000 4.000 0.380 sand_geopark #cylinder: z 1.340 1.360 4.000 4.000 0.360 sand_geopark #cylinder: z 1.320 1.340 4.000 4.000 0.340 sand_geopark #cylinder: z 1.300 1.320 4.000 4.000 0.320 sand_geopark #cylinder: z 1.280 1.300 4.000 4.000 0.300 sand_geopark #cylinder: z 1.260 1.280 4.000 4.000 0.280 sa nd_geopark #cylinder: z 1.240 1.260 4.000 4.000 0.260 sand_geopark #cylinder: z 1.220 1.240 4.000 4.000 0.240 sand_geopark #cylinder: z 1.200 1.220 4.000 4.000 0.220 sand_geopark #cylinder: z 1.180 1.200 4.000 4.000 0.200 sand_geopark #cylinder: z 1.160 1. 180 4.000 4.000 0.180 sand_geopark #cylinder: z 1.140 1.160 4.000 4.000 0.160 sand_geopark #cylinder: z 1.120 1.140 4.000 4.000 0.140 sand_geopark #cylinder: z 1.100 1.120 4.000 4.000 0.120 sand_geopark #cylinder: z 1.080 1.100 4.000 4.000 0.100 sand_geopa rk #cylinder: z 1.060 1.080 4.000 4.000 0.080 sand_geopark #cylinder: z 1.040 1.060 4.000 4.000 0.060 sand_geopark #cylinder: z 1.020 1.040 4.000 4.000 0.040 sand_geopark #cylinder: z 1.000 1.020 4.000 4.000 0.020 sand_geopark ---------------------------------------------------------------------#cylinder: z 1.480 1.500 4.000 4.000 0.500 sand_geopark #cylinder: z 1.460 1.480 4.000 4.000 0.490 sand_geopark #cylinder: z 1.440 1.460 4.000 4.000 0.480 sand_geopark #cylinder: z 1.420 1.440 4.000 4.000 0.470 sa nd_geopark #cylinder: z 1.400 1.420 4.000 4.000 0.460 sand_geopark #cylinder: z 1.380 1.400 4.000 4.000 0.450 sand_geopark #cylinder: z 1.360 1.380 4.000 4.000 0.440 sand_geopark #cylinder: z 1.340 1.360 4.000 4.000 0.430 sand_geopark #cylinder: z 1.320 1. 340 4.000 4.000 0.420 sand_geopark #cylinder: z 1.300 1.320 4.000 4.000 0.410 sand_geopark #cylinder: z 1.280 1.300 4.000 4.000 0.400 sand_geopark #cylinder: z 1.260 1.280 4.000 4.000 0.390 sand_geopark #cylinder: z 1.240 1.260 4.000 4.000 0.380 sand_geopa rk #cylinder: z 1.220 1.240 4.000 4.000 0.370 sand_geopark #cylinder: z 1.200 1.220 4.000 4.000 0.360 sand_geopark #cylinder: z 1.180 1.200 4.000 4.000 0.350 sand_geopark #cylinder: z 1.160 1.180 4.000 4.000 0.340 sand_geopark #cylinder: z 1.140 1.160 4.00 0 4.000 0.330 sand_geopark

PAGE 200

Appendix A (Continued) 186 #cylinder: z 1.120 1.140 4.000 4.000 0.320 sand_geopark #cylinder: z 1.100 1.120 4.000 4.000 0.310 sand_geopark #cylinder: z 1.080 1.100 4.000 4.000 0.300 sand_geopark #cylinder: z 1.060 1.080 4.000 4.000 0.290 sand_geopark #cyli nder: z 1.040 1.060 4.000 4.000 0.280 sand_geopark #cylinder: z 1.020 1.040 4.000 4.000 0.270 sand_geopark #cylinder: z 1.000 1.020 4.000 4.000 0.260 sand_geopark #cylinder: z 0.980 1.000 4.000 4.000 0.250 sand_geopark #cylinder: z 0.960 0.980 4.000 4.000 0.240 sand_geopark #cylinder: z 0.940 0.960 4.000 4.000 0.230 sand_geopark #cylinder: z 0.920 0.940 4.000 4.000 0.220 sand_geopark #cylinder: z 0.900 0.920 4.000 4.000 0.210 sand_geopark #cylinder: z 0.880 0.900 4.000 4.000 0.200 sand_geopark #cylinder: z 0.860 0.880 4.000 4.000 0.190 sand_geopark #cylinder: z 0.840 0.860 4.000 4.000 0.180 sand_geopark #cylinder: z 0.820 0.840 4.000 4.000 0.170 sand_geopark #cylinder: z 0.800 0.820 4.000 4.000 0.160 sand_geopark #cylinder: z 0.780 0.800 4.000 4.000 0.150 sa nd_geopark #cylinder: z 0.760 0.780 4.000 4.000 0.140 sand_geopark #cylinder: z 0.740 0.760 4.000 4.000 0.130 sand_geopark #cylinder: z 0.720 0.740 4.000 4.000 0.120 sand_geopark #cylinder: z 0.700 0.720 4.000 4.000 0.110 sand_geopark #cylinder: z 0.680 0. 700 4.000 4.000 0.100 sand_geopark #cylinder: z 0.660 0.680 4.000 4.000 0.090 sand_geopark #cylinder: z 0.640 0.660 4.000 4.000 0.080 sand_geopark #cylinder: z 0.620 0.640 4.000 4.000 0.070 sand_geopark #cylinder: z 0.600 0.620 4.000 4.000 0.060 sand_geopa rk #cylinder: z 0.580 0.600 4.000 4.000 0.050 sand_geopark #cylinder: z 0.560 0.580 4.000 4.000 0.040 sand_geopark #cylinder: z 0.540 0.560 4.000 4.000 0.030 sand_geopark #cylinder: z 0.520 0.540 4.000 4.000 0.020 sand_geopark #cylinder: z 0.500 0.520 4.00 0 4.000 0.010 sand_geopark ----------------------------------------------------------------------#hertzian_dipole: 1.0 250e6 ricker MyDipole #analysis: 1 geopark3a_snap.out b #tx: x 4.5 4.0 4.0 MyDipole 0.0 120e 9 #rx: 3.5 4.0 4.0 -----------------------------------------------------------------------------#snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 5.0e 9 snap_gp3a_5.out b

PAGE 201

Appendix A (Continued) 187 #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 10.0e 9 snap_gp3a_10.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 15.0e 9 snap_gp3a_15.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 20.0e 9 snap_gp3a_20.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 25.0e 9 snap_gp3a_25.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 30.0e 9 s nap_gp3a_30.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 35.0e 9 snap_gp3a_35.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 40.0e 9 snap_gp3a_40.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 45.0e 9 snap_gp3a_45.out b # snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 50.0e 9 snap_gp3a_50.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 55.0e 9 snap_gp3a_55.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 60.0e 9 snap_gp3a_60.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 65.0e 9 snap_gp3a_65.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 70.0e 9 snap_gp3a_70.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 75.0e 9 snap_gp3a_75.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0. 02 0.02 0.02 80.0e 9 snap_gp3a_80.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 85.0e 9 snap_gp3a_85.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 90.0e 9 snap_gp3a_90.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 95.0e 9 snap_gp3a_95.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 100.0e 9 snap_gp3a_100.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 105.0e 9 snap_gp3a_105.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 110.0e 9 snap_gp3a_11 0.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 115.0e 9 snap_gp3a_115.out b #snapshot: 1 0.0 0.0 0.0 8.0 8.0 4.5 0.02 0.02 0.02 120.0e 9 snap_gp3a_120.out b ------------------------------------------------------------------------------#end _analysis: -----------------------------------------------------------------------#messages: y #title: geopark 3a snapshot #geometry_file: geopark3a.geo ----------------------------------------------------------------------

PAGE 202

188 Appendix B: MATLAB codes % pl ots the grid from the geopark3D model geometry file % modified bgooch 3/2009 clear all ; close all ; % note comments, restrictions below % ---------input parameters here ---------------------------------------dyplot = 0.1; %spacing between x z slices to plot % dzplot = 0.1; %spacing between x y slices to plot % ---------end input parameters ----------------------------------------[Mesh,ID,Header,Media]=gprmax3g( 'geopark3_rxbox.geo' ); Header size(Mesh) %z is vertical %Mesh = (x,y,z) in dimension dx = Header(1).dx; dy = Header(1).dy; dz = Header(1).dz; nx = Header(1).nx; ny = Header(1).ny; nz = Header(1).nz; x=0:dx:(nx 1)*dx; y=0:dy:(ny 1)*dy; z=0:dz:(nz 1)*dz; colormap(prism); %!!! this assumes you won't have more than 5 different types of mat erial colormin=1; colormax=5; clims = [colormin colormax]; %plot x z (vertical) slices yplot = 0:dyplot:max(y); nyplot = length(yplot); for jplot = 1:nyplot [temp jgrid]=min(abs(y yplot(jplot))) i=1:nx; k=1:nz

PAGE 203

Appendix B (Continu ed) 189 xzplane(i,k)=Mesh(i,jgrid,k); %need to flip so that z axis runs up image, increases with increasing row xzplane = xzplane'; figure; imagesc(x,z,xzplane,clims); %set imagesc so that vertical axis increases upwards axis xy ; xlabel( 'x distance (m)' ); ylabel( 'z distance (m)' ); title([ 'y = num2str(y(jgrid)) m' ]); daspect([1 1 1]); clear xzplane ; end %plot x y (horizontal) slices zplot = 0:dzplot:max(z); nzplot = length(zplot); for kplot = 1:nzplot [temp kgrid]=min(abs(z zplot(kplot))); i=1:nx; j=1:ny; xyplane(i,j)=Mesh(i,j,kgrid); %need to flip so that y axis runs up image, increases with increasing row xyplane = xyplane'; figure; imagesc(x,y,xyplane,clims); %set imagesc so that vertical axis increases upwards axis xy ; xlabel( 'x distance (m)' ); ylabel( 'y distance (m)' ); title([ 'z = num2str(z(kgrid)) m' ]); daspect([1 1 1]); clear xyplane ; end % plots the grid from the geopark2D model geometry file % modified bgooch 3/2009 % -----------e nter input parameters here filein = 'geopark3_2D.geo' ; yground = 0.5; % ------------end input parameters [Mesh,Header,Media]=gprmax2g(filein);

PAGE 204

Appendix B (Continu ed) 190 Header; size(Mesh) Mesh(:,200) Media; dx = Header(1).dx; dy = Header(1).dy; x=0:dx:(Header(1).nx 1)*dx; y= (Header(1).ny 1)*dy+yground:dy:yground; colormap(jet); imagesc(x,y,Mesh); axis xy xlabel( 'distance(m)' ); ylabel( 'depth (m)' ); daspect([1 1 1]); % plot sum from every Rx of single TX for Rxbox models of GprMax % bgooch 8/2009 % FIRST T IME THROUGH clear all ; close all ; %%%%%%%%%%%% INPUT first file name and select transmitter number here %%%%%%%%%%%%%%% infile = 'geopark5_rxbox.out' ; trans_num = 1 ; %[Header,Fields]=gprmax_Exonly(infile); [Header,Fields]=gprmax_Exonly_1tx_fast(infil e,trans_num); % print Header Header % convert time to ns t=Fields(1).t*1.0e9; % position for only 1 Tx converted from cell units to metrics Tx_xposition = ( Header(1).tx Header(1).dx ) + ( ( trans_num 1 ) ( Header(1).TxStepX Header(1).dx ) ) ; Tx_yposition = ( Header(1).ty Header(1).dy ) + ( ( trans_num 1 ) ( Header(1).TxStepY Header(1).dx ) ) ; Tx_position = [ Tx_xposition Tx_yposition ] ;

PAGE 205

Appendix B (Continu ed) 191 % convert all Rx x and y positions from cell units to metrics Rx_xpositions = Header(1).rx Header(1).dx ; Rx_ypositions = Header(1).ry Header(1).dy ; % % plot the Rx and Tx locations % plot (Tx_xposition, Tx_yposition, 'x', Rx_xpositions, Rx_ypositions, 'o') % axis equal tight % extract Ex values and store in array EX = Fields(1).ex ; % first time through save as #1 save gp5_tx1 % plot sum from every Rx of single TX for Rxbox models of GprMax % bgooch 8/2009 % FIRST TIME THROUGH clear all ; close all ; %%%%%%%%%%%% INPUT first file name and select transmitter number here %%%%%%%% %%%%%%% infile = 'geopark1_rxbox.out' ; trans_num = 1 ; %[Header,Fields]=gprmax_Exonly(infile); [Header,Fields]=gprmax_Ey_only_1tx_fast(infile,trans_num); % print Header Header % convert time to ns t=Fields(1).t*1.0e9; % position for only 1 Tx conv erted from cell units to metrics Tx_xposition = ( Header(1).tx Header(1).dx ) + ( ( trans_num 1 ) ( Header(1).TxStepX Header(1).dx ) ) ;

PAGE 206

Appendix B (Continu ed) 192 Tx_yposition = ( Header(1).ty Header(1).dy ) + ( ( trans_num 1 ) ( Header(1).TxStepY Header(1).dx ) ) ; Tx_position = [ Tx_xposition Tx_yposition ] ; % convert all Rx x and y positions from cell units to metrics Rx_xpositions = Header(1).rx Header(1).dx ; Rx_ypositions = Header(1).ry Header(1).dy ; % % plot the Rx and Tx locations % plot (Tx_xpo sition, Tx_yposition, 'x', Rx_xpositions, Rx_ypositions, 'o') % axis equal tight % extract Ex values and store in array EY = Fields(1).ey ; % first time through save as #1 save ey_gp1_tx1 % plot GPRMax output for constant offset gather % bgooch skr use 3/2009 clear all ; close all ; infile = 'geopark3a.out' ; offset = 1.8; % offset to plot in meters ncrop = 5; % crop this many bad traces at start of record crange = 0.1; % uses maximum range of color scale over this fraction of data [Header,Field s]=gprmax_Exonly(infile); Header %plot trace positions as midpoint between transmitter and receiver %find which receiver corresponds to specified offset noffset = round(offset/Header(1).dy); [xshift nrxplot] = min(abs((Header(1).ry Header(1).ty(1)) noffset)); nrxplot

PAGE 207

Appendix B (Continu ed) 193 offset = (Header(1).ry(nrxplot) Header(1).ty(1))*Header(1).dy trace_pos1 = (Header(1).ty(1)+Header(1).ry(nrxplot))/2*Header(1).dy; trace_step = Header(1).RxStepY*Header(1).dy; trace_position = trace_pos1:trace_step:trace_pos1+(Head er(1).NSteps 1)*trace_step; t=Fields(1).t*1.0e9; [nt nrec nstep]=size(Fields(1).ex) Ex = zeros(nt,nstep); % put into an array Ex that contains only traces for the specified offset it=1:nt; iy=1:nstep; Ex(it,iy)= Fields(1).ex(it,nrxplot,iy); % crop ou t bad positions at beginning for n=1:ncrop Ex(:,1)=[]; trace_position(1) = []; end % plot cmin = crange*min(min(Ex)); cmax = crange*max(max(Ex)); imagesc(trace_position,t,Ex); caxis([cmin cmax]); xlabel( 'distance (m)' ); ylabel( 'time (ns)' ) ; title([ 'offset = num2str(offset) m' ]); % %plot a single trace % figure; % plotx = 4.0; % [junk ix] = min(abs(nx plotx)); % plot(t,Ex(:,ix)); %write out x min and x max for user to input into Reflex trace_position(1) trace_position(length(trac e_position))

PAGE 208

Appendix B (Continu ed) 194 D = Ex'; %take transpose of array so that traces are rows, to match format expected % expected for ASCII_MATRIX by Reflex save geopark3a_ascii_180.txt D ASCII ; % plot GPRMax snapshot output % bgooch skruse 8/2009 clear all ; close all ; % ------INPUT PARAMETERS HERE -----------------------------infile = 'snap_gp3a_5.out' xplot = 3.5; yplot = 3.9; zplot = 3.5; %user must make sure these are inside grid % ------END INPUT PARAMETERS HERE -------------------------[Header,Fiel ds]=gprmax(infile); Header [nx ny nz] = size(Fields(1).ex); %assumes one corner of grid is at 0 y=0:Header(1).dy:(ny 1)*Header(1).dy; x=0:Header(1).dx:(nx 1)*Header(1).dx; z=0:Header(1).dz:(nz 1)*Header(1).dz; [temp ixplot] = min(abs(x xplot)); [tem p iyplot] = min(abs(y yplot)); [temp izplot] = min(abs(z zplot)); %looks like fields are saved through the whole grid at the snaptime % z = up direction, will plot differently %direction of dipole source is not clear to me. %if we want all 3 plots wi th same color scale, need to extract all values, %set axis limits, then plot %get plane at ix = ixplot Exx = zeros(ny,nz); Exx(:,:) = Fields(1).ex(ixplot,:,:); Exx = Exx';

PAGE 209

Appendix B (Continu ed) 195 minx = min(min(Exx)); maxx = max(max(Exx)); %get plane at iy = iyplot Exy = zero s(nx,nz); Exy(:,:) = Fields(1).ex(:,iyplot,:); Exy = Exy'; miny = min(min(Exy)); maxy = max(max(Exy)); %get plane at iy = iyplot Exz = zeros(nx,ny); Exz(:,:) = Fields(1).ex(:,:,izplot); Exz = Exz'; minz = min(min(Exz)); maxz = max(max(Exz)); minplot = min([minx miny minz]); maxplot = max([maxx maxy maxz]); % x plane plot figure; subplot(1,3,1) imagesc(y,z,Exx); axis xy ; title([ 'Ex at x=' num2str(xplot) 'm' ]); xlabel( 'y' ); ylabel( 'z' ); daspect([1 1 1]); %caxis([minplot maxplot]); % print djpeg f1 r3 00 snap_x_15.jpg % y plane plot figure; subplot(1,3,2) imagesc(x,z,Exy); axis xy ; title([ 'Ex at y=' num2str(yplot) 'm' ]); xlabel( 'x' ); ylabel( 'z' ); daspect([1 1 1]); %caxis([minplot maxplot]); % print djpeg f2 r300 snap_y_15.jpg % z plane plot % fig ure; % subplot(1,3,3) hold on imagesc(y,x,Exz); axis xy ; R = 2; R1 = 0.5; Center = [ 4.0 4.0]; circle(Center,R,1000, 'k ); circle(Center,R1,1000, 'k );

PAGE 210

Appendix B (Continu ed) 196 hold on plot(Center(1),Center(2), 'k.' ) title([ 'Ex at z=' num2str(zpl ot) 'm' ]); xlabel( 'x' ); ylabel( 'y' ); axis equal tight hold off % daspect([1 1 1]); %caxis([minplot maxplot]); % print djpeg f3 r300 snap_z_15.jpg print djpeg r300 snap_5.jpg % plot difference between two Ex from Rxbox GprMax % bgooc h 8/2009 clear all ; close all ; load gp3_tx1 ; % name of second file EX2 = EX ; save EX2 EX2 clear all ; close all ; % import saved EX matrices load gp1_tx1 ; % name of first file EX1 = EX ; save EX1 load EX2 % reload % crop larger matrix % for i=1:(length(EX2(:,1)) 2597) % % EX2(length(EX2(:,1)),:) = [] ; % % end EX2 = EX2(1:2597,:) ; % much faster way! % take difference of the two arrays

PAGE 211

Appendix B (Continu ed) 197 Ex_diff = abs( EX1 EX2 ) ; Ex_sum = su m(Ex_diff) ; % plotting the new matrix xlin = linspace(min(Rx_xpositions),max(Rx_xpositions)); ylin = linspace(min(Rx_ypositions),max(Rx_ypositions)); [X,Y] = meshgrid(xlin,ylin); Z = griddata(Rx_xpositions,Rx_ypositions,Ex_sum,X,Y, 'cub ic' ); hold on contourf(X,Y,Z); colorbar % sinkcenterx = 2.5; % sinkcentery = 4; % sinkrad = 2; % condrad = 1; hold on plot(Tx_xposition,Tx_yposition, 'xk' ) % for i=1:360 % alpha = i/360*2*pi; % xci rcle(i) = sinkcenterx+cos(alpha)*sinkrad; % ycircle(i) = sinkcentery+sin(alpha)*sinkrad; % end % plot(xcircle,ycircle,'k'); % adds a plot of a circle of radius 2 and 0.5 to graph R = 2; R1 = 0.5; Center = [ 2.5 3.5]; circle(Center,R,1000, 'k ); circle(Center,R1,1000, 'k ); hold on plot(Center(1),Center(2), 'k.' ) axis equal tight hold off %axis([min(Rx_xpositions) max(Rx_xpositions) min(Rx_ypositions) max(Rx_ypositions) min(Ex_sum) 50 min(Ex_sum) 5 0]) % save jpeg image of graph print djpeg r300 Ex_diff_tx1.jpg

PAGE 212

Appendix B (Continu ed) 198 % plot difference between two Ey fields from Rxbox GprMax % bgooch 8/2009 clear all ; close all ; load ey_gp3_tx1 ; % name of second file EY2 = EY ; save EY 2 EY2 clear all ; close all ; % import saved EY matrices load ey_gp1_tx1 ; % name of first file EY1 = EY ; save EY1 load EY2 % reload % crop larger matrix % for i=1:(length(EY2(:,1)) 2597) % % EY2(length(EY2(:,1)),:) = [] ; % % end EY2 = EY2(1:2597,:) ; % much faster way! % take difference of the two arrays Ey_diff = abs( EY1 EY2 ) ; Ey_sum = sum(Ey_diff) ; % plotting the new matrix xlin = linspace(min(Rx_xpositions),max(Rx_xpositions )); ylin = linspace(min(Rx_ypositions),max(Rx_ypositions)); [X,Y] = meshgrid(xlin,ylin); Z = griddata(Rx_xpositions,Rx_ypositions,Ey_sum,X,Y, 'cubic' ); hold on

PAGE 213

Appendix B (Continu ed) 199 contourf(X,Y,Z); colorbar % sinkcenterx = 2.5; % sinkcentery = 4; % sinkrad = 2; % condrad = 1; hold on plot(Tx_xposition,Tx_yposition, 'xk' ) % for i=1:360 % alpha = i/360*2*pi; % xcircle(i) = sinkcenterx+cos(alpha)*sinkrad; % ycircle(i) = sinkcentery+sin(alpha)*sinkrad; % end % plot(xcircle,ycircle,'k'); % adds a plot of a circle of radius 2 and 0.5 to graph R = 2; R1 = 0.5; Center = [ 2.5 3.5]; circle(Center,R,1000, 'k ); circle(Center,R1,1000, 'k ); hold on plot(Center(1),Center (2), 'k.' ) axis equal tight hold off %axis([min(Rx_xpositions) max(Rx_xpositions) min(Rx_ypositions) max(Rx_ypositions) min(Ex_sum) 50 min(Ex_sum) 50]) % save jpeg image of graph print djpeg r300 Ey_diff_tx1.jpg % plots sho t gather from user specified line in receiver field % bgooch 8/2009 clear all ; close all ; % load in a saved matlab workspace with extracted Ex info load gp1_tx1.mat % select which profile along x axis of rxbox to plot of 1.1 4.9 m

PAGE 214

Appendix B (Continu ed) 200 profile_n um = median(Rx_xpositions) ; % extract selected profile from larger EX data start_col = round(59 10 profile_num + 1) ; end_col = start_col + 58 ; Ex = EX(:,start_col:end_col) ; % plot the selected common shot gather y = Rx_ypositions(:,1:59) ; crange = 0.001 ; % uses maximum range of color scale over this fraction of data cmin = crange min( min(Ex) ) ; cmax = crange max( max(Ex) ) ; imagesc( y, t, Ex ) ; caxis( [cmin cmax] ) ; xlabel( 'distan ce (m)' ); ylabel( 'time (ns)' ); title([ 'x = num2str(profile_num) m' ]); colorbar % save jpeg image of graph print djpeg r300 shot_gather_gp1_tx1.jpg % plots shot gather from user specified line in receiver field % b gooch 8/2009 clear all ; close all ; % load in a saved matlab workspace with extracted Ex info load ey_gp1_tx1.mat % select which profile along x axis of rxbox to plot of 1.1 4.9 m profile_num = median(Rx_xpositions) ; % extract selected profile from larger EX data

PAGE 215

Appendix B (Continu ed) 201 start_col = round(59 10 profile_num + 1) ; end_col = start_col + 58 ; Ey = EY(:,start_col:end_col) ; % plot the selected common shot gather y = Rx_ypositions(:,1:59) ; crange = 0.001 ; % use s maximum range of color scale over this fraction of data cmin = crange min( min(Ey) ) ; cmax = crange max( max(Ey) ) ; imagesc( y, t, Ey ) ; caxis( [cmin cmax] ) ; xlabel( 'distance (m)' ); ylabel( 'time (ns)' ); title([ 'x = num2str(profile_num) m' ]); colorbar % save jpeg image of graph print djpeg r300 EY_shot_gather_gp1_tx1.jpg % plot the difference between two Ex shot gathers % bgooch 8/2009 clear all ; close all ; % ----------------------load data -------------------------------------load gp3_tx1 ; % name of second file EX2 = EX ; save EX2 EX2 clear all ; close all ; % import saved EX matrices load gp1_tx1 ; % name of first file EX1 = EX ; save EX1

PAGE 216

Appendix B (Continu ed) 202 load EX2 % reload % ---------------crop larger matrix ------------------------------------EX2 = EX2(1:2597,:) ; % much faster way! % select which profile along x axis of rxbox to plot of 1.1 4.9 m profile_num = median(Rx_xpositions) ; % ex tract selected profile from larger EX data start_col = round(59 10 profile_num + 1) ; end_col = start_col + 58 ; Ex1 = EX1(:,start_col:end_col) ; Ex2 = EX2(:,start_col:end_col) ; % ------------------------take difference --------------------------------Ex_diff = ( Ex1 Ex2 ) ; % -------------------plot the selected common shot gather ---------------y = Rx_ypositions(:,1:59) ; crange = 0.001 ; % uses maximum range of color scale over this fra ction of data cmin = crange min( min(Ex_diff) ) ; cmax = crange max( max(Ex_diff) ) ; imagesc( y, t, Ex_diff ) ; caxis( [cmin cmax] ) ; xlabel( 'distance (m)' ); ylabel( 'time (ns)' ); title([ 'x = num2str(profile_num) m' ]); colorbar % save jpeg image of graph print djpeg r300 diff_shot_gather_tx1.jpg

PAGE 217

Appendix B (Continu ed) 203 % plot the difference between two Ey shot gathers % bgooch 8/2009 clear all ; close all ; % -----------------------load data ------------------------------------load ey_gp3_tx1 ; % name of second file EY2 = EY ; save EY2 EY2 clear all ; close all ; % import saved EY matrices load ey_gp1_tx1 ; % name of first file EY1 = EY ; save EY1 load EY2 % reload % --------------crop larger matrix ------------------------------------EY2 = EY2(1:2597,:) ; % much faster way! % select which profile along x axis of rxbox to plot of 1.1 4.9 m profile_num = median(Rx_xpositions) ; % extract selected profile f rom larger EX data start_col = round(59 10 profile_num + 1) ; end_col = start_col + 58 ; Ey1 = EY1(:,start_col:end_col) ; Ey2 = EY2(:,start_col:end_col) ; % ------------------------take difference -------------------------------Ey_diff = ( Ey1 Ey2 ) ; % -------------------plot the selected common shot gather ---------------

PAGE 218

Appendix B (Continu ed) 204 y = Rx_ypositions(:,1:59) ; crange = 0.001 ; % uses maximum range of color scale over this fraction of data cmin = crange min( min(Ey_diff) ) ; cmax = crange max( max(Ey_diff) ) ; imagesc( y, t, Ey_diff ) ; caxis( [cmin cmax] ) ; xlabel( 'distance (m)' ); ylabel( 'time (ns)' ); title([ 'x = num2str(profile_num) m' ]); colorbar % save jpeg image of graph print djpeg r300 EY_diff_shot_gather_tx1.jpg function [Header,Fields]=gprmax_Exonly_1tx_fast(name,trans_num) % GPRMAX3D Read binary data generated by 'GprMax3D' and 'GprMax2D' % simulators for gro und probing radar. % % [Header, Fields] = gprmax( 'filename' ) % filename is the name of a binary format file generated either % from 'GprMax3D' or 'GprMax2D' % % Header is a structure containing details of the model % % Fields is a structure containing the electromagnetic fields at % the requested output points for a number of requested % steps % % % (NOTE: The field matrices of snapshots cary 3D data % Copyright: Antonis Giannopoulos, 1997, 2002, 2005 % This file is not part of the 'GprMax3D' programme for ground probing % radar simulation and can be freely distributed.

PAGE 219

Appendix B (Continu ed) 205 % MODIFIED March 2009 Sarah Kruse to only store E x to save memory space % modification only valid for 3D step mode FT_3D=30200; FT_2D=20200; SMALL=0; BIG=0; if (nargin==0) error( 'GprMax3D requires at least one argument' ); end ; if (nargin==1) type= 'native' ; end ; if (isstr(name)~=1) error( 'First argument is not a filename' ); end ; fid=fopen(name, 'rb' ); if (fid== 1) error([ 'Can not open =' ,name]); end ; ECHECK1=fread(fid,1, 'char' ); if (strcmp(setstr(dec2hex(ECHECK1)), '2B' )==1 ) SMALL=0; BIG=1; end ; if (strcmp(setstr(dec2hex(ECHECK1)), '67' )==1 ) SMALL=1; BIG=0; end ; ECHECK2=fread(fid,1, 'char' ); if (BIG==1) if (strcmp(setstr(dec2hex(ECHECK2)), '67' ) == 0) error([ 'This is not a GprMax2D/3D file.' ]); end ; end ; if (SMALL==1) if (strcmp(setstr(dec2hex(ECHECK2)), '2B' ) == 0)

PAGE 220

Appendix B (Continu ed) 206 error([ 'This is not a GprM ax2D/3D file.' ]); end ; end ; % If you are here you have a valid file. Unless someone is playing a but joke !! % Close and open again to make sure you will read it properly. fclose(fid); if (SMALL==1) fid=fopen(name, 'rb' 'ieee le' ); end ; if (BIG==1) fid=fopen (name, 'rb' 'ieee be' ); end ; % Read Endian again but no check temp=fread(fid,1, 'short' ); % Read type of file FileType=fread(fid,1, 'short' ); SWORD=fread(fid,1, 'short' ); SREAL=fread(fid,1, 'short' ); TITLELENGTH=fread(fid,1, 'short' ); SOURCELENGTH=fread(fid,1, 'short' ); MEDIALENGTH=fread(fid,1, 'short' ); RESERVED=fread(fid,2, 'char' ); if (SWORD==2) word= 'short' ; end ; if (SWORD==4) word= 'long' ; end ; if (SREAL==4) real= 'float' ; end ; if (SREAL==8) real= 'double' ; end ; % Set Defaults for all Model=struct([]); Rx=stru ct([]); % Set title to none %Model(1).title='No title'; switch FileType

PAGE 221

Appendix B (Continu ed) 207 case FT_2D+4 % 2D snapshot disp([ 'Reading GprMax2D #snapshot: file ...' ,name]); Model(1).title=fread(fid,TITLELENGTH, 'char' ); Model(1).title=set str(Model(1).title'); Model(1).iterations=fread(fid,1,real); Model(1).dx=fread(fid,1,real); Model(1).dy=fread(fid,1,real); Model(1).dt=fread(fid,1,real); Model(1).globalTx=fread(fid,1,word); Model(1).Snapx1=f read(fid,1,word); Model(1).Snapy1=fread(fid,1,word); Model(1).Snapx2=fread(fid,1,word); Model(1).Snapy2=fread(fid,1,word); Model(1).Snapxs=fread(fid,1,word); Model(1).Snapys=fread(fid,1,word); stime=fread(fid,1, real); Model(1).snaptime=stime*Model(1).dt/1e 9; Model(1).Snapxsam=fread(fid,1,word); Model(1).Snapysam=fread(fid,1,word); Rx(1).ez=zeros(Model(1).Snapxsam,Model(1).Snapysam); Rx(1).hx=zeros(Model(1).Snapxsam,Model (1).Snapysam); Rx(1).hy=zeros(Model(1).Snapxsam,Model(1).Snapysam); Rx(1).ez=fread(fid,[Model(1).Snapxsam Model(1).Snapysam],real); Rx(1).hx=fread(fid,[Model(1).Snapxsam Model(1).Snapysam],real); Rx(1).hy=fread(fid ,[Model(1).Snapxsam Model(1).Snapysam],real); case FT_3D+4 % 3D snapshot disp([ 'Reading GprMax3D #snapshot: file ...' ,name]); Model(1).title=fread(fid,TITLELENGTH, 'char' ); Model(1).title=setstr(Model(1).title'); Model(1).iterations=fread(fid,1,real); Model(1).dx=fread(fid,1,real); Model(1).dy=fread(fid,1,real); Model(1).dz=fread(fid,1,real); Model(1).dt=fread(fid,1,real); Model(1).globalTx=fread(fid,1,word);

PAGE 222

Appendix B (Continu ed) 208 Model (1).Snapx1=fread(fid,1,word); Model(1).Snapy1=fread(fid,1,word); Model(1).Snapz1=fread(fid,1,word); Model(1).Snapx2=fread(fid,1,word); Model(1).Snapy2=fread(fid,1,word); Model(1).Snapz2=fread(fid,1,word); Mod el(1).Snapxs=fread(fid,1,word); Model(1).Snapys=fread(fid,1,word); Model(1).Snapzs=fread(fid,1,word); stime=fread(fid,1,real); Model(1).snaptime=stime*Model(1).dt/1e 9; Model(1).Snapxsam=fread(fid,1,word); Mo del(1).Snapysam=fread(fid,1,word); Model(1).Snapzsam=fread(fid,1,word); Rx(1).ex=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).ey=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).ez =zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).hx=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).hy=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).hz=zeros(Model(1).Sna pxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).ix=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).iy=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).iz=zeros(Model(1).Snapxsam,Model(1).Snap ysam,Model(1).Snapzsam); for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).ex(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Sna pzsam Model(1).Snapysam],real); Rx(1).ey(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).ez(i,:,:)=temp';

PAGE 223

Appendix B (Continu ed) 209 end for i=1:Model( 1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).hx(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1) .hy(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).hz(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsa m Model(1).Snapysam],real); Rx(1).ix(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).iy(i,:,:)=temp'; end for i=1:Model(1).Sna pxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).iz(i,:,:)=temp'; end case FT_2D % 2D step disp([ 'Reading GprMax2D #analysis: file ...' ,name]); Model(1).title=fread(fid,TIT LELENGTH, 'char' ); Model(1).title=setstr(Model(1).title'); Model(1).iterations=fread(fid,1,real); Model(1).dx=fread(fid,1,real); Model(1).dy=fread(fid,1,real); Model(1).dt=fread(fid,1,real); Model(1).NSteps=fr ead(fid,1,word); Model(1).TxStepX=fread(fid,1,word); Model(1).TxStepY=fread(fid,1,word); Model(1).RxStepX=fread(fid,1,word); Model(1).RxStepY=fread(fid,1,word); Model(1).ntx=fread(fid,1,word); Model(1).nrx=fr ead(fid,1,word); Model(1).nrx_box=fread(fid,1,word);

PAGE 224

Appendix B (Continu ed) 210 for i=1:Model(1).ntx Model(1).tx(i)=fread(fid,1,word); Model(1).ty(i)=fread(fid,1,word); Model(1).source(i,1:SOURCELENGTH)=fread(fid,SOURCELENGTH, 'char )'; Model(1).delay(i)=fread(fid,1,real); Model(1).removed(i)=fread(fid,1,real); end Model(1).source=char(Model(1).source); for i=1:Model(1).nrx Model(1).rx(i)=fread(fid,1,word); Model(1).ry (i)=fread(fid,1,word); end TotalOuts=Model(1).nrx; kk=Model(1).nrx; for i=1:Model(1).nrx_box Model(1).rx_box(i).nouts=fread(fid,1,word); TotalOuts=TotalOuts+Model(1).rx_box(i).nouts; for k=kk+1:kk+Model(1).rx_box(i).nouts Model(1).rx(k)=fread(fid,1,word); Model(1).ry(k)=fread(fid,1,word); end kk=kk+Model(1).rx_box(i).nouts; end Rx(1).t=(0:Model(1).iterations 1)'*M odel(1).dt; %Read the data in single vector for speed F=fread(fid,inf,real); % Short out data in (Outputs,Iterations,Steps) ez=reshape(F(1:3:end),TotalOuts,Model(1).iterations,Model(1). NSteps); hx=reshape(F(2:3:end),TotalOuts,Model(1).iterations,Model(1).NSteps); hy=reshape(F(3:3:end),TotalOuts,Model(1).iterations,Model(1).NSteps);

PAGE 225

Appendix B (Continu ed) 211 % Save the data in (Iterations,Outputs,Steps) format if Model(1).NSteps == 1 Rx(1).ez=ez'; Rx(1).hx=hx'; Rx(1).hy=hy'; else for i=1:TotalOuts Rx(1).ez(:,i,:)=ez(i,:,:); Rx(1).hx(:,i,:)=hx(i,:,:); R x(1).hy(:,i,:)=hy(i,:,:); end end case FT_3D % 3D step disp([ 'Reading GprMax3D #analysis: file ...' ,name]); Model(1).title=fread(fid,TITLELENGTH, 'char' ); Model(1).title=setstr(Model(1).title'); Model(1).iterations=fread(fid,1,real); Model(1).dx=fread(fid,1,real); Model(1).dy=fread(fid,1,real); Model(1).dz=fread(fid,1,real); Model(1).dt=fread(fid,1,real); Model(1).NSteps=fread(fid,1,word); Model(1) .TxStepX=fread(fid,1,word); Model(1).TxStepY=fread(fid,1,word); Model(1).TxStepZ=fread(fid,1,word); Model(1).RxStepX=fread(fid,1,word); Model(1).RxStepY=fread(fid,1,word); Model(1).RxStepZ=fread(fid,1,word); Model(1).ntx=fread(fid,1,word); Model(1).nrx=fread(fid,1,word); Model(1).nrx_box=fread(fid,1,word); for i=1:Model(1).ntx Model(1).polarization(i,:)=fread(fid,1, 'char' ); Model(1).polarization(i,:)=char(setstr( Model(1).polarization(i,:)')); Model(1).tx(i)=fread(fid,1,word); Model(1).ty(i)=fread(fid,1,word); Model(1).tz(i)=fread(fid,1,word);

PAGE 226

Appendix B (Continu ed) 212 Model(1).source(i,1:SOURCELENGTH)=fread(fid,SOURCELENGTH, 'char' ); Model(1).delay(i)=fread(fid,1,real); Model(1).removed(i)=fread(fid,1,real); end Model(1).source=char(Model(1).source); for i=1:Model(1).nrx Model(1).rx(i)=fread(fid,1,word); Model(1).ry(i)=frea d(fid,1,word); Model(1).rz(i)=fread(fid,1,word); E nd TotalOuts=Model(1).nrx; kk=Model(1).nrx; for i=1:Model(1).nrx_box Model(1).rx_box(i).nouts=fread(fid,1,word); TotalOuts=TotalOuts+Mode l(1).rx_box(i).nouts; for k=kk+1:kk+Model(1).rx_box(i).nouts Model(1).rx(k)=fread(fid,1,word); Model(1).ry(k)=fread(fid,1,word); Model(1).rz(k)=fread(fid,1,word); end kk=kk +Model(1).rx_box(i).nouts; end Rx(1).t=(0:Model(1).iterations 1)'*Model(1).dt; %Read the data in single vector for speed % Impractical for large files with memory limitations in Windows % F=fread(fid,inf,real); % % % % Short out data in (Outputs,Iterations,Steps) % % ex=reshape(F(1:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % ey=reshape(F(2:9:end),TotalOuts,Model(1).iter ations,Model(1).NSteps); % ez=reshape(F(3:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps);

PAGE 227

Appendix B (Continu ed) 213 % hx=reshape(F(4:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % hy=reshape(F(5:9:end),TotalOuts,Model(1).iterations,Model(1). NSteps); % hz=reshape(F(6:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % ix=reshape(F(7:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % iy=reshape(F(8:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % iz=reshape(F(9:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % % % Save the data in (Iterations,Outputs,Steps) format % if Model(1).NSteps == 1 % Rx(1).ex=ex'; % Rx(1).ey=ey'; % Rx(1) .ez=ez'; % Rx(1).hx=hx'; % Rx(1).hy=hy'; % Rx(1).hz=hz'; % Rx(1).ix=ix'; % Rx(1).iy=iy'; % Rx(1).iz=iz'; % else % for i=1:TotalOuts % Rx(1).ex(:,i,: )=ex(i,:,:); % Rx(1).ey(:,i,:)=ey(i,:,:); % Rx(1).ez(:,i,:)=ez(i,:,:); % Rx(1).hx(:,i,:)=hx(i,:,:); % Rx(1).hy(:,i,:)=hy(i,:,:); % Rx(1).hz(:,i,:)=hz(i,:,:); % Rx(1).ix(:,i,:)=ix(i,:,:); % Rx(1).iy(:,i,:)=iy(i,:,:); % Rx(1).iz(:,i,:)=iz(i,:,:); % end % end % That is a rather slower way to read data !!! tic % --skipping over results f rom all transmitters before ntrans -------------for i = 1:trans_num 1 % number of transmitter for which we are writing out receiver results

PAGE 228

Appendix B (Continu ed) 214 for j=1:Model(1).iterations disp([ 'Reading Iteration # ,num2str(j), of ,num2str(Model( 1).iterations), Iterations, on Transmitter #: num2str(i), of num2str(trans_num)]); for k=1:Model(1).nrx temp=fread(fid,9,real); % skipping over receivers listed in nrx end ; kk=Model(1).nrx; for p=1:Model(1).nrx_box % skipping over receiver listed in nrx_boxes % disp(['Reading rx_box # ',num2str(p),' of ',num2str(Model(1).nrx_box),' rx_box']); for k=kk+1:kk+Model(1).rx_box(p).nouts temp=fread(fid,9,real); end kk=kk+Model(1).rx_box(p).nouts; end end end toc % --------------------------end skip -----------------------------------tic % ---------store data for transm itter ntrans -------------------------for j=1:Model(1).iterations disp([ 'Reading Iteration # ,num2str(j), of ,num2str(Model(1).iterations), Iterations, for Transmitter #: num2str(trans_num)]); for k=1:Model(1).nrx Rx(1).ex(j,k,1)=fread(fid,1,real); temp=fread(fid,8,real); % skip over non Ex information end ; kk=Model(1).nrx; for p=1:Model(1).nrx_box % disp(['Reading rx_box # ',num2str(p),' of ',num2str(Model(1). nrx_box),' rx_box']); for k=kk+1:kk+Model(1).rx_box(p).nouts Rx(1).ex(j,k,1)=fread(fid,1,real); temp=fread(fid,8,real); % skip over non Ex information end kk=kk+Model(1).rx_box(p).nouts; end end toc

PAGE 229

Appendix B (Continu ed) 215 case FT_3D+5 disp([ 'This is not a data file. It is a geometry file' ]); disp([ 'Use GPRMAX3G.' ]); case FT_2D+5 disp([ 'This is not a data file. It is a geometry file' ]); disp( [ 'Use GPRMAX2G.' ]); otherwise disp([ 'This is not a valid GprMax2D/3D Ver 2.0 data file.' ]); disp([ 'It may be an older version data file' ]); end Header=Model; Fields=Rx; %close file fclose(fid); function [Header,Fields]=gp rmax_Ey_only_1tx_fast(name,trans_num) % GPRMAX3D Read binary data generated by 'GprMax3D' and 'GprMax2D' % simulators for ground probing radar. % % [Header, Fields] = gprmax( 'filename' ) % filename is the name of a binary format file generated either % from 'GprMax3D' or 'GprMax2D' % % Header is a structure containing details of the model % % Fields is a structure containing the electromagnetic fields at % the requested output points for a number of requested % steps % % % (NOTE: The field matrices of snapshots cary 3D data % Copyright: Antonis Giannopoulos, 1997, 2002, 2005 % This file is not part of the 'GprMax3D' pro gramme for ground probing

PAGE 230

Appendix B (Continu ed) 216 % radar simulation and can be freely distributed. % MODIFIED August 2009 Brad Gooch to only store Ey to save memory space % modification only valid for 3D step mode FT_3D=30200; FT_2D=20200; SMALL=0; BIG=0; if ( nargin==0) error( 'GprMax3D requires at least one argument' ); end ; if (nargin==1) type= 'native' ; end ; if (isstr(name)~=1) error( 'First argument is not a filename' ); end ; fid=fopen(name, 'rb' ); if (fid== 1) error([ 'Can not open =' ,name]); end ; ECHECK1= fread(fid,1, 'char' ); if (strcmp(setstr(dec2hex(ECHECK1)), '2B' )==1 ) SMALL=0; BIG=1; end ; if (strcmp(setstr(dec2hex(ECHECK1)), '67' )==1 ) SMALL=1; BIG=0; end ; ECHECK2=fread(fid,1, 'char' ); if (BIG==1) if (strcmp(setstr(dec2hex(ECHECK2)), '67' ) == 0) error([ 'This is not a GprMax2D/3D file.' ]); end ; end ; if (SMALL==1)

PAGE 231

Appendix B (Continu ed) 217 if (strcmp(setstr(dec2hex(ECHECK2)), '2B' ) == 0) error([ 'This is not a GprMax2D/3D file.' ]); end ; end ; % If you are here you have a valid file. Unless someone is playing a but joke !! % Close an d open again to make sure you will read it properly. fclose(fid); if (SMALL==1) fid=fopen(name, 'rb' 'ieee le' ); end ; if (BIG==1) fid=fopen(name, 'rb' 'ieee be' ); end ; % Read Endian again but no check temp=fread(fid,1, 'short' ); % Read type of file FileType=f read(fid,1, 'short' ); SWORD=fread(fid,1, 'short' ); SREAL=fread(fid,1, 'short' ); TITLELENGTH=fread(fid,1, 'short' ); SOURCELENGTH=fread(fid,1, 'short' ); MEDIALENGTH=fread(fid,1, 'short' ); RESERVED=fread(fid,2, 'char' ); if (SWORD==2) word= 'short' ; end ; if (SWORD==4) word= 'long' ; end ; if (SREAL==4) real= 'float' ; end ; if (SREAL==8) real= 'double' ; end ; % Set Defaults for all Model=struct([]); Rx=struct([]); % Set title to none %Model(1).title='No title';

PAGE 232

Appendix B (Continu ed) 218 switch FileType case FT_2D+4 % 2D snapshot disp([ 'Reading GprMax2D #snapshot: file ...' ,name]); Model(1).title=fread(fid,TITLELENGTH, 'char' ); Model(1).title=setstr(Model(1).title'); Model(1).iterations=fread(fid,1,real); Model(1).dx=fread(fid,1,real); Model( 1).dy=fread(fid,1,real); Model(1).dt=fread(fid,1,real); Model(1).globalTx=fread(fid,1,word); Model(1).Snapx1=fread(fid,1,word); Model(1).Snapy1=fread(fid,1,word); Model(1).Snapx2=fread(fid,1,word); Model(1).S napy2=fread(fid,1,word); Model(1).Snapxs=fread(fid,1,word); Model(1).Snapys=fread(fid,1,word); stime=fread(fid,1,real); Model(1).snaptime=stime*Model(1).dt/1e 9; Model(1).Snapxsam=fread(fid,1,word); Model(1). Snapysam=fread(fid,1,word); Rx(1).ez=zeros(Model(1).Snapxsam,Model(1).Snapysam); Rx(1).hx=zeros(Model(1).Snapxsam,Model(1).Snapysam); Rx(1).hy=zeros(Model(1).Snapxsam,Model(1).Snapysam); Rx(1).ez=fread(fid,[Model(1). Snapxsam Model(1).Snapysam],real); Rx(1).hx=fread(fid,[Model(1).Snapxsam Model(1).Snapysam],real); Rx(1).hy=fread(fid,[Model(1).Snapxsam Model(1).Snapysam],real); case FT_3D+4 % 3D snapshot disp([ 'Reading GprMax3D #snapshot: file ...' ,name]); Model(1).title=fread(fid,TITLELENGTH, 'char' ); Model(1).title=setstr(Model(1).title'); Model(1).iterations=fread(fid,1,real); Model(1).dx=fread(fid,1,real); Model(1).dy=fread(fid,1,real); Model(1).dz=fread(fid,1,real); Model(1).dt=fread(fid,1,real);

PAGE 233

Appendix B (Continu ed) 219 Model(1).globalTx=fread(fid,1,word); Model(1).Snapx1=fread(fid,1,word); Model(1).Snapy1=fread(fid,1,word); Model(1).Snapz1=fread(fid,1,word); Model(1).Snapx2=fread(fid,1,word); Model(1).Snapy2=fread(fid,1,word); Model(1).Snapz2=fread(fid,1,word); Model(1).Snapxs=fread(fid,1,word); Model(1).Snapys=fread(fid,1,word); Model(1).Snapzs=fread(fid,1,word); stime=fread(fid,1,real); Model(1).snaptime=stime*Model(1).dt/1e 9; Model(1).Snapxsam=fread(fid,1,word); Model(1).Snapysam=fread(fid,1,word); Model(1).Snapzsam=fread(fid,1,word); Rx(1).ex=zeros(Model(1).Snapx sam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).ey=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).ez=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).hx=zeros(Model(1).Snapxsam,Model(1).Snapys am,Model(1).Snapzsam); Rx(1).hy=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).hz=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).ix=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsa m); Rx(1).iy=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); Rx(1).iz=zeros(Model(1).Snapxsam,Model(1).Snapysam,Model(1).Snapzsam); for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Sna pysam],real); Rx(1).ex(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).ey(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).ez(i,:,:)=temp';

PAGE 234

Appendix B (Continu ed) 220 end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).hx(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).hy(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).S napysam],real); Rx(1).hz(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).ix(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).iy(i,:,:)=temp'; end for i=1:Model(1).Snapxsam temp=fread(fid,[Model(1).Snapzsam Model(1).Snapysam],real); Rx(1).iz(i,:,:)=temp'; end case FT_2D % 2D step disp([ 'Reading GprMax2D #analysis: file ...' ,name]); Model(1).title=fread(fid,TITLELENGTH, 'char' ); Model(1).title=setstr(Model(1).title'); Model(1).iterations=fread(fid,1,real); Model(1).dx=fread(fid,1,real); Model(1).dy=fread(fid,1,real); Model(1).dt=fread(fid,1,real); Model(1).NSteps=fread(fid,1,word); Model(1).TxStepX=fread(fid,1,word); Model(1).TxStepY=fread(fid,1,word); Mo del(1).RxStepX=fread(fid,1,word); Model(1).RxStepY=fread(fid,1,word); Model(1).ntx=fread(fid,1,word); Model(1).nrx=fread(fid,1,word); Model(1).nrx_box=fread(fid,1,word); for i=1:Model(1).ntx

PAGE 235

Appendix B (Continu ed) 221 Model(1).tx(i )=fread(fid,1,word); Model(1).ty(i)=fread(fid,1,word); Model(1).source(i,1:SOURCELENGTH)=fread(fid,SOURCELENGTH, 'char' )'; Model(1).delay(i)=fread(fid,1,real); Model(1).removed(i)=fread(fid,1,real); end M odel(1).source=char(Model(1).source); for i=1:Model(1).nrx Model(1).rx(i)=fread(fid,1,word); Model(1).ry(i)=fread(fid,1,word); end TotalOuts=Model(1).nrx; kk=Model(1).nrx; for i=1:Model(1) .nrx_box Model(1).rx_box(i).nouts=fread(fid,1,word); TotalOuts=TotalOuts+Model(1).rx_box(i).nouts; for k=kk+1:kk+Model(1).rx_box(i).nouts Model(1).rx(k)=fread(fid,1,word); Model(1).ry(k)=f read(fid,1,word); end kk=kk+Model(1).rx_box(i).nouts; end Rx(1).t=(0:Model(1).iterations 1)'*Model(1).dt; %Read the data in single vector for speed F=fread(fid,inf,real); % Short out data in (Outputs,Iterations,Steps) ez=reshape(F(1:3:end),TotalOuts,Model(1).iterations,Model(1).NSteps); hx=reshape(F(2:3:end),TotalOuts,Model(1).iterations,Model(1).NSteps); hy=reshape(F(3:3:end),T otalOuts,Model(1).iterations,Model(1).NSteps);

PAGE 236

Appendix B (Continu ed) 222 % Save the data in (Iterations,Outputs,Steps) format if Model(1).NSteps == 1 Rx(1).ez=ez'; Rx(1).hx=hx'; Rx(1).hy=hy'; else for i=1:TotalOuts Rx(1).ez(:,i,:)=ez(i,:,:); Rx(1).hx(:,i,:)=hx(i,:,:); Rx(1).hy(:,i,:)=hy(i,:,:); end end case FT_3D % 3D step disp([ 'Reading GprM ax3D #analysis: file ...' ,name]); Model(1).title=fread(fid,TITLELENGTH, 'char' ); Model(1).title=setstr(Model(1).title'); Model(1).iterations=fread(fid,1,real); Model(1).dx=fread(fid,1,real); Model(1).dy=fread(fid,1,re al); Model(1).dz=fread(fid,1,real); Model(1).dt=fread(fid,1,real); Model(1).NSteps=fread(fid,1,word); Model(1).TxStepX=fread(fid,1,word); Model(1).TxStepY=fread(fid,1,word); Model(1).TxStepZ=fread(fid,1,word) ; Model(1).RxStepX=fread(fid,1,word); Model(1).RxStepY=fread(fid,1,word); Model(1).RxStepZ=fread(fid,1,word); Model(1).ntx=fread(fid,1,word); Model(1).nrx=fread(fid,1,word); Model(1).nrx_box=fread(fid,1,word) ; for i=1:Model(1).ntx Model(1).polarization(i,:)=fread(fid,1, 'char' ); Model(1).polarization(i,:)=char(setstr(Model(1).polarization(i,:)')); Model(1).tx(i)=fread(fid,1,word); Model(1).ty(i)=fread(fid, 1,word); Model(1).tz(i)=fread(fid,1,word); Model(1).source(i,1:SOURCELENGTH)=fread(fid,SOURCELENGTH, 'char' ); Model(1).delay(i)=fread(fid,1,real);

PAGE 237

Appendix B (Continu ed) 223 Model(1).removed(i)=fread(fid,1,real); end Mod el(1).source=char(Model(1).source); for i=1:Model(1).nrx Model(1).rx(i)=fread(fid,1,word); Model(1).ry(i)=fread(fid,1,word); Model(1).rz(i)=fread(fid,1,word); end TotalOuts=Model(1).nrx; k k=Model(1).nrx; for i=1:Model(1).nrx_box Model(1).rx_box(i).nouts=fread(fid,1,word); TotalOuts=TotalOuts+Model(1).rx_box(i).nouts; for k=kk+1:kk+Model(1).rx_box(i).nouts Model(1).rx(k)=fread(fid,1 ,word); Model(1).ry(k)=fread(fid,1,word); Model(1).rz(k)=fread(fid,1,word); end kk=kk+Model(1).rx_box(i).nouts; end Rx(1).t=(0:Model(1).iterations 1)'*Model(1).dt; %Read the data in single vector for speed % Impractical for large files with memory limitations in Windows % F=fread(fid,inf,real); % % % % Short out data in (Outputs,Iterations,Steps) % % ex=reshape(F(1:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % ey=reshape(F(2:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % ez=reshape(F(3:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % hx=reshape(F(4:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % hy=reshape(F(5:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps);

PAGE 238

Appendix B (Continu ed) 224 % hz=reshape(F(6:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % ix=reshape(F(7:9 :end),TotalOuts,Model(1).iterations,Model(1).NSteps); % iy=reshape(F(8:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % iz=reshape(F(9:9:end),TotalOuts,Model(1).iterations,Model(1).NSteps); % % % Save the data in (I terations,Outputs,Steps) format % if Model(1).NSteps == 1 % Rx(1).ex=ex'; % Rx(1).ey=ey'; % Rx(1).ez=ez'; % Rx(1).hx=hx'; % Rx(1).hy=hy'; % Rx(1).hz=hz'; % Rx(1).ix =ix'; % Rx(1).iy=iy'; % Rx(1).iz=iz'; % else % for i=1:TotalOuts % Rx(1).ex(:,i,:)=ex(i,:,:); % Rx(1).ey(:,i,:)=ey(i,:,:); % Rx(1).ez(:,i,:)=ez(i,:,:); % Rx(1).hx(:,i,:)=hx(i,:,:); % Rx(1).hy(:,i,:)=hy(i,:,:); % Rx(1).hz(:,i,:)=hz(i,:,:); % Rx(1).ix(:,i,:)=ix(i,:,:); % Rx(1).iy(:,i,:)=iy(i,:,:); % Rx(1).iz(:,i,:)=iz(i,:,:) ; % end % end % That is a rather slower way to read data !!! tic % --skipping over results from all transmitters before ntrans -------------for i = 1:trans_num 1 % number of transmitter for which we a re writing out receiver results for j=1:Model(1).iterations

PAGE 239

Appendix B (Continu ed) 225 disp([ 'Reading Iteration # ,num2str(j), of ,num2str(Model(1).iterations), Iterations, on Transmitter #: num2str(i), of num2str(trans_num)]); for k=1: Model(1).nrx temp=fread(fid,9,real); % skipping over receivers listed in nrx end ; kk=Model(1).nrx; for p=1:Model(1).nrx_box % skipping over receiver listed in nrx_boxes % disp(['Readi ng rx_box # ',num2str(p),' of ',num2str(Model(1).nrx_box),' rx_box']); for k=kk+1:kk+Model(1).rx_box(p).nouts temp=fread(fid,9,real); end kk=kk+Model(1).rx_box(p).nouts; end end end toc % --------------------------end skip -----------------------------------tic % ---------store data for transmitter ntrans -------------------------for j=1:Model(1).iterations disp([ 'Reading Iteration # ,num2 str(j), of ,num2str(Model(1).iterations), Iterations, for Transmitter #: num2str(trans_num)]); for k=1:Model(1).nrx temp=fread(fid,1,real); % skip over non Ey information Rx(1).ey(j,k,1)=fread(fid,1,real); temp=fread(fid,7,real); % skip over non Ey information end ; kk=Model(1).nrx; for p=1:Model(1).nrx_box % disp(['Reading rx_box # ',num2str(p),' of ',num2str(Model(1).nrx_box),' rx_box']); for k=kk+1:k k+Model(1).rx_box(p).nouts temp=fread(fid,1,real); % skip over non Ey information Rx(1).ey(j,k,1)=fread(fid,1,real); temp=fread(fid,7,real); % skip over non Ey information end kk=kk+Mo del(1).rx_box(p).nouts; end end toc

PAGE 240

Appendix B (Continu ed) 226 case FT_3D+5 disp([ 'This is not a data file. It is a geometry file' ]); disp([ 'Use GPRMAX3G.' ]); case FT_2D+5 disp([ 'This is not a data file. It is a geome try file' ]); disp([ 'Use GPRMAX2G.' ]); otherwise disp([ 'This is not a valid GprMax2D/3D Ver 2.0 data file.' ]); disp([ 'It may be an older version data file' ]); end Header=Model; Fields=Rx; %close file fclose(fid);

PAGE 241

227 Appendix C. Common Shot Gather Analysis Since the rx_box command allows each receiver to record information for each individual transmitter location, it can be used to construct what is known as a common shot gather or wide angle reflection and refr action (WARR) survey (Reynolds, 1997; Burger et al, 2006). In this survey, a single transmitting source at a fixed location broadcasts an energy pulse while another receiver records at a set distance away. The receiver is then moved farther out at a fixed distance along the profile line for each transmission. If many receivers can be used at the same time, it is possible to lay them all out in a line and record with fewer moving transmission pulses (this known as a normal move out survey). The information c ontained in the rx_box model results can be used to configure this survey type. The survey can also be viewed with receiving antennae positions the same as the transmitter (Ex) or perpendicular to it (Ey). These surveys were constructed for both Model A an d B across the center line of the model (same as in previous section). Sample common shot gather records for both Ex and Ey information of the two models may be viewed in Figures C.1 14. These results do not suggest any other detailed structure than the c ommon shot gather survey results did in Chapter 2. A basic depression in the subsurface is visible but not the actual conduit structure itself. The differences between the common shot gather results of the two models appear very similar, which makes compar ison analysis more difficult. A routine (similar to that used in the previous rx_box analysis) to subtract the first 100 ns of both Model A and B's records made it easier to observe the subtle differences. These difference plots may be seen in Figure s C.15 21. The difference plo ts

PAGE 242

Appendix C (Continued) 228 add very little to the analysis, if any at all. In fact, the time records for both models are identical until about 50 70 ns where a wave that is different begins to arrive and after that there is a great deal of noise. Unfortunat ely, the difference shown is not enough to justify locating conduit structure from modeling efforts

PAGE 243

Appendix C (Continued) 229 Figure C.1a Figure C.1b

PAGE 244

Appendix C (Continued) 230 Figure C.1a. Top, previous page. Ex common shot gather reco rd for transmitter location #5 (same identification numbers as those in previous sections) for Model A. The x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C. 1b. Bottom, previous page. Ey common shot gather reco rd for transmitter location #5 for Model A. All values are the same as previous graph. Notic e Ey signal strength is weaker than Ex record.

PAGE 245

Appendix C (Continued) 231 Figure C.2a Figure C.2b

PAGE 246

Appendix C (Continued) 232 Figure C.2a. Top, previous page. Ex common shot gathe r recor d for transmitter location #30 (same identification numbers as those in previous sections) for Model A. The x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.2b. Bottom, previous page. Ey common shot gather recor d for transmitter location #30 for Model A. All values are the same as previous graph. Notice Ey signal s trength is weaker than Ex record.

PAGE 247

Appendix C (Continued) 233 Figure C.3a Figure C.3b

PAGE 248

Appendix C (Continued) 234 Figure C.3a. Top, previous page. Ex common shot gather recor d for transmitter location #50 (same identification numbers as those in previous sections) for Model A. The x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.3b. Bottom, previous page. Ey common shot gather re cor d for transmitter location #50 for Model A. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record.

PAGE 249

Appendix C (Continued) 235 Figure C.4a Figure C.4b

PAGE 250

Appendix C (Continued) 236 Figure C.4a. Top, previous page. Ex common shot gather record for tra nsmitter location #60 (s ame identification numbers as those in previous sections) for Model A. The x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may be directly found by following air and grou nd waves to their source location at time zero. Figure C.4b. Bottom, previous page. Ey common shot gather recor d for transmitter location #60 for Model A. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record. Note : this is at the center of the conduit (x = 2.5 m, y = 3.5 m).

PAGE 251

Appendix C (Continued) 237 Figure C.5a Figure C.5b

PAGE 252

Appendix C (Continued) 238 Figure C.5a. Top, previous page. Ex common shot gather record for transmitter location #70 (same identification numbers as those in previous sections) for Model A. The x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.5b. Bottom, previous pag e. Ey common shot gather recor d for transmitter location #70 for Model A. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record.

PAGE 253

Appendix C (Continued) 239 Figure C.6a Figure C.6b

PAGE 254

Appendix C (Continued) 240 Figure C.6a. Top, previous page. Ex common shot gather recor d for transmitter location #90 (same identification numbers as those in previous sections) for Model A. The x axis represents the distance in the y direction (for x = 2.5 m). The y axis represents the time axis (in ns) Transmitter locations may be directly foun d by following air and ground waves to their source location at time zero. Figure C.6b. Bottom, previous page. Ey common shot gather recor d for transmitter location #90 for Model A. All values are the same as previous graph. Notic e Ey signal strength is we aker than Ex record.

PAGE 255

Appendix C (Continued) 241 Figure C.7a Figure C.7b

PAGE 256

Appendix C (Continued) 242 Figure C.7a. Top, previous page. Ex common shot gather record for transmitter location #115 (same identification numbers as those in previous sections) for Model A. The x axis represents the distance in the y direc tion (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.7b. Bottom, previous page. Ey common shot gather r ecord for tra nsmitter location #115 for Model A. All values are the same as previous graph Notice Ey signal strength is weaker than Ex record.

PAGE 257

Appendix C (Continued) 243 Figure C.8a Figure C.8b

PAGE 258

Appendix C (Continued) 244 Figure C.8a. Top, previous page. Ex common shot gather reco rd for transmitter location #5 (same identific ation numbers as those in previous sec tions) for Model B. The x axis represents the distance in the y direction (for x = 2.5 m). The y axi s represents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to t heir source location at time zero. Figure C.8b. Bottom, previous page. Ey common shot gather reco rd for transmitter location #5 for Model B. All values are the same as previous graph. Notic e Ey signal strength is weaker than Ex record.

PAGE 259

Appendix C (Continued) 245 Figure C.9a Figure C.9b

PAGE 260

Appendix C (Continued) 246 Figure C.9a. Top, previous page. Ex common shot gather recor d for transmitter location #30 (same identification numbers as those in previous sec tions) for Model B. The x axis represents the distance in the y direction (for x = 2.5 m). The y axi s represents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.9b. Bottom, previous page. Ey common shot gather recor d for transmitter location #30 for Model B. All valu es are the same as previous graph. Notic e Ey signal strength is weaker than Ex record.

PAGE 261

Appendix C (Continued) 247 Figure C.10a Figure C.10b

PAGE 262

Appendix C (Continued) 248 Figure C.10a. Top, previous page. Ex common shot gather record for transmitter loc ation #50 (same identification numbers as those in previous sec tions) for Mo del B. The x axis represents the distance in the y direction (for x = 2.5 m). The y axi s represents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.10b. Bo ttom, previous page. Ey common shot gather r ecord for transmitter location #50 for Model B. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record.

PAGE 263

Appendix C (Continued) 249 Figure C.11a Figure C.11b

PAGE 264

Appendix C (Continued) 250 Figure C.11a. Top, previous page. Ex common shot gather record for transmitter location #60 (same identification numbers as those in previous sections) for Model B. The x axis represents the distance in the y direction (for x = 2.5 m). The y axi s represents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.11b. Bottom, previous page. Ey common shot gather record for trans mitter location #60 for Model B. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record. Note: this is at the center of the conduit (x = 2.5 m, y = 3.5 m).

PAGE 265

Appendix C (Continued) 2 51 Figure C.12a Figure C.12b

PAGE 266

Appendix C (Continued) 252 Figure C.12a. Top, previous page. Ex common shot gather recor d for transmitter location #70 (same identification numbers as tho se in previous sec tions) for Model B. The x axis represents the distance in the y direction (for x = 2.5 m). The y axi s represents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.12b. Bottom, previous page. Ey common shot gather r ecord for transmitter location #70 for Model B. All values are the same as previous graph. Notic e Ey signal strength is weaker than Ex record.

PAGE 267

Appendix C (Continued) 253 Figure C.13a Figure C.13b

PAGE 268

Appendix C (Continued) 254 Figure C.13a. Top, previous page. Ex common shot gather recor d for transmitter location #90 (same identification numbers as those in previous sec tions) for Model B. The x axis represents the distance in the y direction (for x = 2.5 m). The y axi s represents the time axis (i n ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.13b. Bottom, previous page. Ey common shot gather r ecord for transmitter location #90 for Model B. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record.

PAGE 269

Appendix C (Continued) 255 Figure C.14a Figure C.14b

PAGE 270

Appendix C (Continued) 256 Figure C.14a. Top, previous page. Ex common shot gather record for transmitter location #115 (same identification numbers as those in previous sec tions) for Model B. The x axis represents the distance in the y direction (for x = 2.5 m). The y axi s represents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.14b. Bottom previous page. Ey common shot gather r ecord for transmitter location #115 for Model B. All values are the same as previous graph. Notice Ey signa l strength is weaker than Ex record.

PAGE 271

Appendix C (Continued) 257 Figure C.15a Figure C.15b

PAGE 272

Appendix C (Continued) 258 Figure C.15a. Top, previous page. Ex difference plo t between Model A and B for transmitter location #5 (same identification numbers as those in previous se ctions). The x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may b e are the same as those in previous figures. Figure C.15b. Bottom, previous page. Ey difference plot between Model A and B for transmitter location #5. All values are the same as previous gr aph. Notice miniscule order of magnitude of relative difference i n each graph.

PAGE 273

Appendix C (Continued) 259 Figure C.16a Figure C.16b

PAGE 274

Appendix C (Continued) 260 Figure C.16a. Top, previous page. Ex difference plot between Model A and B for transmitter location #30 (same identification numbers as those in previous se ctions). The x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may be are the same as those in previous figures. Figure C.16b. Bottom, previous page. Ey difference plot between Model A and B for transmitter location #30. All values are th e same as previous gr aph. Notice miniscule order of magnitude of relative difference in each graph.

PAGE 275

Appendix C (Continued) 261 Figure C.17a Figure C.17b

PAGE 276

Appendix C (Continued) 262 Figure C.17a. Top, previous page. Ex difference plot between Model A and B for transmitter location #50 (same identification numbers as those in previous se ctions). The x axis represents the distance in the y direction (for x = 2.5 m). The y axis represents the time axis (in ns). Transmitter locations may be are the same as those in previous figures. Figure C.17b. Bottom, previous page. E y difference plot between Model A and B for transmitter location #50. All values are the same as previous graph. Notice min iscule order of magnitude of relative difference in each graph.

PAGE 277

Appendix C (Continued) 263 Figure C.18a Figure C.18b

PAGE 278

Appendix C (Continued) 264 Figure C.18a. Top, previous page. Ex difference plot between Model A and B for transmitter location #60 (same identification numbers as those in previous sections). Th e x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations m ay be are the same as those in previous figures. Figure C.18b. Bottom, previous page. Ey difference plot between Model A and B for transmitter location #60. All values are the same as previous gr aph. Notice miniscule order of magnitude of relative differe nce in each graph. Note: this is at the center of the conduit (x = 2.5 m, y = 3.5 m).

PAGE 279

Appendix C (Continued) 265 Figure C.19a Figure C.19b

PAGE 280

Appendix C (Continued) 266 Figure C.19a. Top, previous page. Ex difference plot between Model A and B for transmitter location #70 (same identification numbers as those in previ ous se ctions). The x axis represents the distance in the y direction (for x = 2.5 m). The y axis represents th e time axis (in ns). Transmitter locations may be are the same as those in previous figures. Figure C.19b. Bottom, previous page. Ey difference pl ot between Model A and B for transmitter location #70. All values are the same as previous graph. Notice mi niscule order of magnitude of relative difference in each graph.

PAGE 281

Appendix C (Continued) 267 Figure C.20a Figure C.20b

PAGE 282

Appendix C (Continued) 268 Figure C.20a. Top, previous page. Ex difference plot between Mo del A and B for transmitter location #90 (same identification numbers as those in previous sections). T he x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may be are the s ame as those in previous figures. Figure C.20b. Bottom, previous page. Ey difference plot between Model A and B for transmitter location #90. All values are the same as previous gr aph. Notice miniscule order of magnitude of relative difference in each gra ph.

PAGE 283

Appendix C (Continued) 269 Figure C.21a Figure C.21b

PAGE 284

Appendix C (Continued) 270 Figure C.21a. Top, previous page. Ex difference plot between Model A and B for transmitter location #115 (same identification numbers as those in previous se ctions). The x axis represents the distance in the y direction (for x = 2.5 m). The y axis rep resents the time axis (in ns). Transmitter locations may be are the same as those in previous figures. Figure C.21b. Bottom, previous page. Ey difference plot between Model A and B for transmitter location #115. All values are the same a s previous graph. Notice the miniscule order of magnitude of relative difference in each graph. All previous simulated surveys were run with the transmitting and receiving pair of antenna in a line through the center of the conduit. One final analy sis was made to observe the results of doing an off axis common shot gather where the transmitting line was off by 0.3 m to one side (specifically, x = 2.8 m) of the center line of the model. The rx_box and model geometry was still the same as used in the Model B. By placing the new line only 0.3 m to one side, the survey line was still over the conduit but not exactly in the middle of it. It was hypothesized that there might be a better return of energy by doing this experiment. The results of this experim ent can be seen in Figure C.22 28. The experiment, unfortunately, did not reveal anything worth further investigation. Essentially, the side of the larger conical depression was discernible but nothing else of value could be gleamed from it. All the prev ious results of the performed common shot gathers were negative in extracting information about the presence of the modeled conduit structure. This result is, however, an important one in that it explains how similar conduit structure might actually

PAGE 285

Appendix C (Continued) 271 exist in subsurface geology but is frequently not recognized. More rigorous modeling research may eventually prove useful in extracting similar conduit structure, but for now the results are inconclusive.

PAGE 286

Appendix C (Continued) 272 Figure C.22a Figure C.22b

PAGE 287

Appendix C (Continued) 273 Figure C.22a. Top, previous page. Ex common shot ga ther reco rd for transmitter location #5 (same identification numbers as those in previous sections) fo r off axis Model B. The x axis represents the distance in the y direction (for x = 2.8 m). The y axi s represents the time axis (in ns). Transmitter locati ons may be directly found by following air and ground waves to their source location at time zero. Figure C.22b. Bottom, previous page. Ey common shot gather reco rd for transmitter location #5 for off axis Model B. All values are the same as previous graph Notice Ey signal strength is weaker than Ex record.

PAGE 288

Appendix C (Continued) 274 Figure C.23a Figure C.23b

PAGE 289

Appendix C (Continued) 275 Figure C.23a. Top, previous page. Ex common shot gather record for transmitter location #30 (same identification numbers as those in previous sections) for off axis Model B. The x axis r epresents the distance in the y direction (for x = 2.8 m). The y axis represents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.23b. Bottom, previous page Ey common shot gather record for transmitter location #30 for off axis Model B. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record.

PAGE 290

Appendix C (Continued) 276 Fi gure C.24a Figure C.24b

PAGE 291

Appendix C (Continued) 277 Figure C.24a. Top, previous page. Ex common shot gather recor d for transmitter location #50 (same identification numbers as those in previous sections) for off axis Model B. The x axis represents the distance in the y direction (for x = 2.8 m). The y axis represents the time axis (in ns). Transmitter locations may b e directly found by following air and ground waves to their source location at time zero. Figure C.24b. Bottom, previous page. Ey common shot gather record for transmitter location #50 for off axis Model B. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record.

PAGE 292

Appendix C (Continued) 278 Figure C.25a Figure C.25b

PAGE 293

Appendix C (Continued) 279 Figure C.25a. Top, previous page. Ex common shot gather record for transmitter location #60 (same identification numbers as those in previous sections) fo r off axis Model B. The x axis represen ts the distance in the y direction (for x = 2.8 m). The y axis represents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.25b. Bottom, previous page. Ey co mmon shot gather record for transmitter location #60 for off axis Model B. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record. Note: this is at the center of the conduit (x = 2.5 m, y = 3.5 m).

PAGE 294

Appendix C (Continued) 280 Figure C.26a Figure C.26b

PAGE 295

Appendix C (Continued) 281 Figu re C.26a. Top, previous page. Ex common shot gather record for transmitter location #70 (same identification numbers as those in previous sections) for off axis Model B. The x axis represents the distance in the y direction (for x = 2.8 m). The y axis repr esents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.26b. Bottom, previous page. Ey common shot gather record for transmitter location #70 for off axis M odel B. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record.

PAGE 296

Appendix C (Continued) 282 Figure C.27a Figure C.27b

PAGE 297

Appendix C (Continued) 283 Figure C.27a. Top, previous page. Ex common shot gather record for transmitter location #90 (same identification numbers as those in pr evious sections) for off axis Model B. The x axis represents the distance in the y direction (for x = 2.8 m). The y axis represents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source locatio n at time zero. Figure C.27b. Bottom, previous page. Ey common shot gather record for transmitter location #90 for off axis Model B. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record.

PAGE 298

Appendix C (Continued) 284 Figure C.28a Figure C.28b

PAGE 299

Appendix C (Continued) 285 Figure C.28a. Top, previous page. Ex common shot gather record for transmitter location #115 (same identification numbers as those in previous sections) for off axis Model B. The x axis represents the distance in the y direction (for x = 2.8 m). The y axis repre sents the time axis (in ns). Transmitter locations may be directly found by following air and ground waves to their source location at time zero. Figure C.28b. Bottom, previous page. Ey common shot gather record for transmitter location #115 for off axis M odel B. All values are the same as previous graph. Notice Ey signal strength is weaker than Ex record.


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 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003428
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Gooch, Bradley.
0 245
Improving ground penetrating radar resolution of features of active sinkholes
h [electronic resource] /
by Bradley Gooch.
260
[Tampa, Fla] :
b University of South Florida,
2010.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
502
Thesis (M.S.)--University of South Florida, 2010.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: Ground penetrating radar (GPR) is widely used to identify locations of sinkholes in covered karst terrain in Florida. Some sinkholes serve as hydraulic conduits between the surficial and underlying aquifers. Their role is critical in determining the surficial aquifer response to pumping in deeper aquifers. Improved methods for discriminating between hydraulically active sinkholes and plugged sinkholes could help regional water management. In the covered karst of west-central Florida a clay-rich weathering horizon forms over the limestone. The clay-rich layer is in turn overlain by surficial sands. Ground penetrating radar profiles typically show a strong reflector from the top of clay-rich horizon as well as internal layering within sands. Active sinkholes are expected to have sandy conduits that broach the clay layer, and perhaps layering in the overlying sand indicative of ongoing subsidence. Three dimensional simulations of GPR profiles over sinkhole with and without conduits were run with the finite-difference time-domain (FDTD) program GPRMAX. Results from the synthetic surveys were then processed with standard techniques, including migration. The modeling confirms that conduits appear in GPR records primarily as gaps in the return from the clay layer. The modeling also shows that non-traditional survey geometries (varying antenna spacing and orientation) are unlikely to recover more information than traditional proximal transmitter-receiver separation. Also examined are GPR profiles and 3D grids over a set of active and inactive sinkholes in Tampa, Florida. Results from these surveys showed decent structural recovery of a small sinkhole similar in structure to that of the modeled ones. Indications of active subsidence and possible conduit structure were apparent from this data. Finally, the dense surveys served as a benchmark to compare interpretations taken with the same surveys at lower spatial resolutions and profiles with 2D-only processing methods in order to understand errors in analysis and interpretation that are possible from 2D surveys. Two-dimensional surveys, 2D processed and migrated, showed some similarity to the 3D results previously mentioned but contained more complexities and artifacts, which led to poorer interpretation ability.
590
Advisor: Sarah Kruse, Ph.D.
653
3D migration
FDTD modeling
Covered-karst terrain
Subvertical reflectors
GPRMAX
690
Dissertations, Academic
z USF
x Geology
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3428