Introduction to Ground Penetrating Radar

Presentation on theme: "Introduction to Ground Penetrating Radar"— Presentation transcript:

1 Introduction to Ground Penetrating Radar
Bryan S. Haley

2 Introduction

3 History 1920s: rudimentary GPR, applications such as ice thickness
Air radar used in WWII for aircraft, Radio Detection and Ranging (RADAR) acronym 1950s,60s: ice thickness, geological applications 1972: NASA Apollo 17 on moon, carrying GPR 1980s: engineering applications, concrete assessment, void detection, land mine detection

4 History of GPR in Archaeology
1970s and 1980s Chaco Canyon, Cyprus, Ceren, Japan Analysis of raw profile data from plotters

5 History of GPR in Archaeology
1990s and 2000s Computers more powerful and affordable Onboard storage Time slice maps, 3D modeling, rendering, etc.

6 How Does It Work?

7 How Does It Work? Profile Trace

8 Relative Dielectric Permittivity (RDP)
RDP( ) = (c / V)2 ε c: speed of light in a vacuum (3 X 108 m/s) V: velocity of radar wave through the material Ranges from 1 (air) to 81 (water). Related primarily to water content of materials. Higher ε values mean less radar penetration (more attenuation). Strength of reflection is controlled by RDP contrast between the two materials. A reflection can occur in dielectric contrasts as small as 1.

9 Magnetic Permeability (μ)
Other Properties Conductivity (σ) High σ inhibits radar penetration (more attenuation). Increases with moisture content, and salinity. So highly conductive soils (ie. clays) are not as ideal for GPR investigation as soils with low σ (such as dry sand). Magnetic Permeability (μ) High μ inhibits radar penetration (more attenuation). Most soils have relatively low μ.

10 Conductivity and RDP for Common Materials

11 Strength of Reflection
Reflection Strength = √ε2 - √ε1 / √ε2 + √ε1 ε1: RDP of first material ε1: RDP of second material

12 Strength of Reflection
Reflection coefficient for 2 layer case. From GSSI SIR System-2000 Training Notes 1999.

13 Anomaly Shape Simulations From GPRSIM 2D Forward Modeling Software

14 Antennas Identified by center frequency in MHz
Higher frequency = greater vertical resolution Lower frequency = greater penetration depth Typical penetration depths 100Mhz 4-25m 300Mhz 1-10m 400Mhz .5-4m 500Mhz m 900Mhz 0-1m

15 Antennas Vertical Resolution Tm = c / (4f √ε) ε: RDP.
Tm: minimum thickness resolved. c: speed of light in a vacuum (3 X 108 m/s). f: center frequency of antenna. ε: RDP. Example: For 400 Mhz antenna and RDP of 10, the minimum thickness is about 6 cm.

16 Horizontal Resolution
Antennas Horizontal Resolution A = λ / 4 + D / √ (ε + 1) A = long dimension radius of footprint. λ = center frequency wavelength of antenna. D = depth. ε = RDP. Example: For 400 Mhz antenna, a depth of 50 cm, and a RDP of 10, A is about 21 cm. Therefore the footprint is approximately 42 cm on the front to back axis and 28 cm on the side to side axis.

17 Antennas Simplified antenna patterns.

18 Setup: Gaining No Gain 5 Gain Points

19 Other Setup Parameters
Samples per scan (512) Scans per time (16 to 64 / sec) Bit depth of data (8 bit or 16 bit)

20 Determining Position User Marks
Marks inserted manually with trigger at fixed interval. Survey wheel Calibrated so that distance is determine based on number of revolutions. GPS Location determined by GPS and synched with GPR based on time.

21 Field Notes Must record file name, X value, and Y start and finish
Very important for GPR since software is flexible Basic instrument settings

22 Depth (Velocity) Estimation
Estimate from RDP. Shoot to target of known depth. Hyperbola fitting (geometric scaling). Common Mid Point (CMP) testing.

23 Hardware GSSI SIR 2000 SIR 3000 Sensors and Software Noggin Mala
Others

24 Processing Steps Radargram Processing Background removal
Box car filter Band pass filter Migration Hilbert Transform Topographic Correction Antenna tilt correction

25 Processing Steps Create Info File
Contains file name, X value, and Y start and finish. Reverse Files Align zig-zagged lines. Set Navigation Specify survey wheel, user marks, GPS. Fix marks if there are errors.

26 Processing Steps Slice / Resample Set # of slices, thickness.
Radargrams resampled to constant number per distance unit. Data collected from each radargram. Time slice values computed for each radargram are merged with the navigation. XYZ file created for each slice.

27 Processing Steps Gridding Specify cell size, search radius.
Interpolate the XYZ files already created.

28 Time Slices Processing: low pass, high pass, etc. Set color scheme
Set data range Set transforms

29 3D Data Cubes

30 Isosurface Rendering

31 Animations

32 Support Software Surfer ArcView / ArcGIS

33 Interpretation Anomaly Shape / Size / Orientation Strength / Amplitude
Context Multiple Instrument Response Data from other projects Historic Documents Aerial Photos Lore Etc. Ground Truthing

34 Results

35 For More Reading… Conyers and Goodman 1997 Conyers 2004
Heimmer and Devore 1995 Bevan 1998 Clark 1995 Gaffney and Gater 2003 Johnson 2006

36 Part II: Case Studies Bryan S. Haley

37 Sapelo Island Shell Rings (Georgia)

38 Sapelo Island Reconstruction

39 Sapelo Island Early sketch map. Modern topo map.

40 Sapelo Island

41 Sapelo Island

42 St. Michael’s Cemetery (Pensacola FL)

43 St. Michael’s Cemetery

44 St. Michael’s Cemetery

45 St. Michael’s Cemetery

46 St. Michael’s Cemetery

47 Memorial Cemetery (St. Genevieve Missouri)

48 Memorial Cemetery

49 Memorial Cemetery

50 Memorial Cemetery

51 Belle Alliance (Louisiana)

52 Belle Alliance

53 Belle Alliance

54 Belle Alliance

55 Jackson Barracks (New Orleans)

56 Jackson Barracks Possible Burial

57 Jackson Barracks Possible Burials

58 Jackson Barracks Interpretation

59 Hollywood (NW Mississippi)
1923 Sketch Map of Mounds

60 Hollywood

61 Hollywood Excavated Structures

62 Hollywood

63 Cahal Pech (Belize)

64 Cahal Pech

65 Cahal Pech

66 Cahal Pech Excavated Structure