Simulated Penetration Tests

To test the penetrometer and study the penetration effect, simulated penetration tests were performed in the laboratory using a calibration chamber and hydraulic loading frame. The chamber was 45 cm in inner diameter and 40 cm in height. A silty sand (SM) was used for the simulated penetration tests. Seven different gravimetric water contents were used to prepare samples in the chamber. The soil and water were mixed thoroughly to obtain the desired water content. The mixed soil was sealed with plastic wrap and allowed to equilibrate for more than 24 h, to yield a uniform soil specimen. The soil was then compacted in the calibration chamber in layers and the total mass of the soil and chamber was measured. The gravimetric water content of the soil specimen ranged from 2% to 10% and the dry density ranged from 1.58 to 1.67 g/cm³.

Two TDR measurements were taken. Simulated penetration test was first conducted by penetrating the TDR penetrometer at the center of the chamber. No surcharge was applied to the soil specimen but a cap was placed on top of the chamber to prevent the soil from heaving during penetration. Hence, the soil around the penetrometer was slightly densified during penetration. After the penetrometer was retracted, another TDR measurement was taken at the location between the penetrated hole and the chamber cylinder using a conventional multi-rod probe (MRP) similar to Fig. 5-11b. The diameter of the multiple rods is 9.5 mm and the spacing between the center conductor and outer conductors is 65 mm. The MRP mimics a coaxial probe in which the electromagnetic field is concentrated around the central rod. The effect of penetration on TDR measurements using the MRP is considered negligible (Siddiqui et al., 2000). Comparing the measurements of TDR penetrometer with that of MRP can reveal the effect of penetration. After all the TDR measurements were taken, samples of the soil were oven-dried to determine the gravimetric water content. The volumetric water content (θ) of each soil sample was determined from the total density and gravimetric water content.

Dielectric properties - θ relationship

TDR measurements show that both dielectric constant and electrical conductivity increase with water content. A good linear relationship between √Ka measured by the TDR penetrometer and the volumetric water content (θ) exists as Fig. 5-18 shows. The

√Ka-θ relationship was shown by Topp et al. (1980) to be relatively independent of soil type and electrical conductivity of pore water. The correlation between √σ and θ also shows great linearity. But the √σ-θ relationship is greatly affected by soil type and pore water electrical conductivity. Therefore, apparent dielectric constant can be used for measuring volumetric water content (or void ratio when the soil is saturated). The electrical conductivity can then provide additional information for determining the characteristic of the pore water. The correlation between √σ and θ also shows great linearity in Fig. 5-19. Unlike the

√Ka-θ relationship, the intercept and slope of the √σ-θ relationship depend on the soil type and electrical conductivity of pore water. Soil water content can be estimated by the Ka

measurement alone, electrical conductivity and future study on dielectric dispersion can provide extra information for characterizing soil type and pore water.

0 5 10 15 20 2.2

2.4 2.6 2.8 3 3.2 3.4 3.6

Volumetric Water Content, θ (%) sq rt (K

)

data point

linear regression

R²=0.9729

Fig. 5-18 Correlation between √Ka and volumetric water content

0 5 10 15 20

0.04 0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12

Volumetric Water Content, θ (%) sq rt ( σ ) ( √ (S m

))

data point

linear regression

Fig. 5-19 Correlation between √σ and volumetric water content

Penetration Effect

In addition to measurements using the TDR penetrometer, TDR measurements were also performed using a 4-rod MRP probe (measurements which were much less disturbed). Fig.

5-20 shows the comparison for Ka measurements. The effect of penetration can be noticed in Ka measurements as expected. Ka increases as soil density increases due to penetration because the volumetric water content increases with density when gravimetric water content remained unchanged. This increase in apparent dielectric constant depends on the degree and extent of the soil densification. As the original soil density increases, the densification due to penetration is less significant but the disturbed zone becomes larger. Therefore, the Ka

measurement is less affected by the penetration effect for dense soil than for loose soil. This can explain the effect of penetration shown in Fig. 5-20. The soil samples prepared for the simulated penetration tests were compacted at dry side of optimum. At dry side, the density increases as gravimetric water content, and hence the apparent dielectric constant, increases.

Therefore, the penetration effect is more pronounced for the soil with lower Ka, as Fig. 11 shows. In summary, the penetration effect induces a coherent Ka error which increases with decreasing soil density. For the soils tested (dry density ranging from 1.58 to 1.67 g/cm3), this error is within the uncertainty associated with the √Ka-θ correlation. But caution should be taken when performing tests in looser soils.

Fig. 5-21 compares electrical conductivity measured by the prototype TDR penetrometer with that by the MRP probe. Unexpectedly, coherent errors due to penetration effect are not noticeable in σ measurements. The radial sampling is less focused on the vicinity of the probe for σ measurements, as shown by comparing Fig. 5-14 to Fig. 5-13. However, this difference can only partly explain the unnoticeable penetration effect. The σ spatial weighting function was experimentally defined by the PVC-tube experiment. This spatial weighting function was valid strictly only for the condition shown in Fig. 5-14. In actual soil measurements, the

conductivity variation due to penetration is smooth. It is believed that the actual radial sampling is much greater than that shown in Fig. 5-14, resulting in an unnoticeable penetration effect

0 5 10 15

0 5 10 15

K

f rom T D R penet reomet er

K

from MRP

data point

linear regression 1:1 line

R²=0.9944

Fig. 5-20 The Ka obtained from TDR penetrometer vs. that from MRP γd=1.67 g/cm³

γd=1.58 g/cm³

0 0.005 0.01 0.015 0.02 0

0.005 0.01 0.015 0.02

σ from MRP (S/m)

σ f rom T D R penetrometer (S /m )

linear regression 1:1 line

R²=0.9781

Fig. 5-21 The electrical conductivity obtained from TDR penetrometer vs. that from MRP γd=1.67 g/cm³

γd=1.58 g/cm³

6 Applications: Characterization of Soil-water Mixture

6.1 Introduction

Shihmen Reservoir is a multi-purpose water resources project, for irrigation, power generation, water supply, flood control and tourism. The Shihmen Dam is an earth-filled dam situated at approximately 50 km south east of Taipei. Since plugging of the diversion tunnel in May, 1963, the hydro-project has made significant contributions to northern Taiwan in agricultural production, industrial and economic developments, as well as alleviating flood or drought losses. The watershed of Shihmen Reservoir has characteristics of being steep in slopes and weak in geologic formations. As a result, during heavy storms, severe surface erosions coupled with land slides often occur. Since its completion in 1963, reservoir siltation has gradually increased, in spite of measures taken on dredging and construction of sediment retention structures. The reservoir was designed to have a total storage of 309 million m³ (volume of water that can be stored in the reservoir) and an effective storage of 252 million m³ (volume of water above the intake level). As of March of 2004, the total storage had been reduced to 253 million m³ and the effective storage was 238 million m³. Aere Typhoon invaded northern Taiwan in August, 2004. The event caused an average rainfall of 973 mm in the watershed which resulted in a total landslide area of 854 hectares, and an estimated inflow of approximately 28 million m³ of sediments into the Reservoir.

Due to aforementioned risks of sedimentation in Shihmen reservoir, there are two problems occur after by. The first one is that the intake valve of the hydro power plant was covered by 10 m of sediment. The other is the requirement of monitoring for suspended sediment concentration during typhoon or deluge. However, the traditional techniques for such purposes are limited in measurement range and durability, such as optical and acoustic sensors, while TDR provides an alternative solution, especially on the soil-water mixture

characterization. Therefore, two applications of soil-water mixture characterization using TDR are presented in this chapter, and the objectives are listed as: (1) Establishing the relationships between electrical properties and sediments concentration with TDR technique, and evaluate the TDR penetrometer performance for basal sediment characterization; (2) Development a high resolution method for monitoring suspended sediment concentration in reservoir area.