Results and comparison (pseudo-section vs. geological section)

 

The next pseudo-section was obtained after using the right image enhancement and processing parameters for the GPR data.

 

Figure 5.1 GPR Washington Creek pseudo-section

 

The pseudo-section data give us the velocity and the frequency data transmitted by the antenna to the ground in a certain period of time. This data was used to calculate the resultant subsurface depth from the GPR test. For the total 55 m length of our site the total depth using GPR was about 5 m. Table 5.1 gives the exact depth and velocity values with respect to time. Appendix A shows the different reflected waves in ns per location in our study area. From the analysis of these wavelength values it was possible to obtain the thickness of each layer.

 

 

 

Location

time ns

time (sec)

velocity m/s

depth

4

2.99

2.99E-09

169941166

0.254062044

5

7.85

7.85E-09

169941166

0.667019078

6

7.85

7.85E-09

169941166

0.667019078

7

10.68

1.068E-08

169941166

0.907485828

3

10.92

1.092E-08

169941166

0.927878768

9

12.94

1.294E-08

169941166

1.099519346

10

13.27

1.327E-08

169941166

1.127559638

2

13.43

1.343E-08

169941166

1.141154932

12

14.89

1.489E-08

169941166

1.265211983

1

16.26

1.626E-08

169941166

1.381621682

11

34.71

3.471E-08

169941166

2.949328941

8

49.68

4.968E-08

169941166

4.221338571

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 5.1 Total depth of the layers

 

Shallow penetration of 5 m indicates the presence of water in the soil. The water absorbs the most of the energy from the transmitter and that is why it did not penetrate deeper.

The result obtained from the GPR survey agreed with the results obtained in the two previous seismic and resistivity test. According to our resistivity test and it corresponding values there are also 3 layers of unfrozen silty-clay (~0-1m) , frozen soil(~1-3 m) and permafrost(~3-5) in the area. The next pseudo-section obtained from our resistivity survey illustrates the 3 subsurface layers.

 

Prior seismic and resistivity surveys performed by the team indicated that the results of the GPR match the data obtained before.

 

Resistivity survey results

Figure 5.2 Dipole-dipole resistivity survey of the Washington Creek site

 

Next are the results from the resistivity survey obtained by the team in previous studies.

 

Depth (m) Distance (m) Color Resistivity (Ω.m) Soil/rock

0-0.5 85-95 Blue ~ 100-1299 silty-clay

0.5 -5.0 85-95 Green ~ 1776- 2428 frozen soil

0-8.50 100-110 Green ~ 100-2428 silty-clay with ice

0-0.5 110-130 Blue/green 100- 1299 permafrost

0-5.0 130-135 Yellow 3319 permafrost

8.50-26 85-110 Brown-Purple ~3319-6202 Permafrost

 

 The values obtained from the resistivity data are easier to interpret. The signals penetrate deeper allowing us to see the subsurface layers deeper. Comparing this data with our GPR data the permafrost layer is about 5 m deep for the resistivity and about 2 meters deep for the GPR data. Even though the GPR did not penetrate deep into the subsurface the data is more accurate since this value is close t the value obtained by DOT in some of their studies of the area.

 

Seismic survey results

The seismic survey also gave us a close

 

 

 

 

Figure 5.3 Seismic Survey pseudo-sections

 

According to the seismic survey pseudo- section, for a depth of 6 meters there are 3 subsurface layers in our study area. This also agrees with our results obtained from the other two field test performed by the team. So far our interpretations of the subsurface condition of the soil have been well interpreted.

 

Values for the seismic survey are:

0-0.5 m = Organic layer

0.5-2 m = Active layer

2-6 m = Permafrost

Back to front page

Back to index