New advances in hydrocarbon detection using ground penetration radar

Citation

(1999), "New advances in hydrocarbon detection using ground penetration radar", Industrial Lubrication and Tribology, Vol. 51 No. 4. https://doi.org/10.1108/ilt.1999.01851dad.001

New advances in hydrocarbon detection using ground penetration radar

New products and materials

New advances in hydrocarbon detection using ground penetration radar

Keywords Radar, Hydrocarbon

Decision making, whether to remediate a site which is contaminated with hydrocarbon, or perhaps continue to monitor, is made using as much relevant information as possible to determine the correct and most cost-effective solution. Such decisions are invariably based on environmental risk, which in turn is based on a data set that includes; site observation, site history, non-intrusive and intrusive investigations.

The advantage of non-intrusive investigations is the relative higher number of data points that can be cost-effectively produced to assess variability of a particular parameter across a site. Its disadvantage is its two-dimensional interpretation. Boreholes, on the other hand, provide a three-dimensional picture, but fewer data points are generally produced and it is generally assumed (in the absence of other indicators) that the variability between boreholes is linear. There is always the probability that a contaminant hot spot may go undetected between boreholes.

During the 1990s new emerging technologies have been developed and commercially introduced. These can provide an alternative means of identifying ground contamination, plume boundaries and quantification of contaminant variability, but with the need for minimal intrusive investigation. Some of these new technologies have been the re-evaluation of standard geophysical techniques, such as ground penetrating radar (GPR) and resistivity, that are routinely used for shallow investigation (Benson, 1992; Daniels et al., 1995; Maxwell and Schmok, 1995; Monier-Williams, 1995; Bermejo et al., 1997; Atekwana et al., 1998).

The non-intrusive methodology of GPR now has some 200 case histories of experience to demonstrate its ability to map the variability of hydrocarbon contamination in three dimensions (typically to 10m in depth below ground surface). The equipment used during a site survey is a standard set of GPR instrumentation employing 100MHz and 200MHz antennae typically. It is at the data processing stage that new developments in computer software, produced by a Dutch company MAP, enable the signatures of hydrocarbon contamination to be identified and analysed within individual radar scans. When several scans are conducted in close proximity to each other, it is possible to produce three-dimensional data sets of hydrocarbon variability. This environmental application of a GPR survey has been called an Enviscan^TM survey.

GPR surveys collect data by transmitting electromagnetic pulses of a particular radio frequency from an antenna, into the subsoils. When an electromagnetic pulse reaches an electrical interface beneath the antenna, some of the energy is reflected back. As the antenna is moved across the ground a series of electromagnetic pulses are transmitted and recorded, resulting in a continuous radar scan or profile displaying these interfaces. These are typically a function of local changes in geology or structure.

The ability of GPR to detect hydrocarbons relates to further changes in the electrical properties within the subsurface. On a microscopic scale, conductivity variations occur locally via changes in the soil-air-water-contaminant relationship. Mechanisms that could result in either an increase or decrease in apparent conductivity have been previously reported (Monier-Williams, 1995; Grumman and Daniels, 1995).

This can be illustrated by considering the migration of hydrocarbons spilled on the ground surface and migrating through the vadose and saturated zones. Changes in the subsoil electrical properties occur because some of the hydrocarbons can replace both the air and water, which surround the individual grains of subsoil. The hydrocarbons can also alter the local environment by partially dissolving in the water present between soil grains and by partially replacing air by vapour. The ratio of water to air will increase towards the capillary zone, influencing the concentration of hydrocarbons within this horizon and possibly creating an LNAPL within the zone of influence of a fluctuating water table. Bacteria present will also biodegrade hydrocarbons present, to varying degrees by either aerobic or anaerobic processes. Most hydrocarbons are apolar, indicating little change in conductivity from the original product. However, biodegradation can alter such products to polar components thereby influencing the conductivity. The conductive nature of mature plumes has been described from field surveys (Bermejo et al., 1997; Atekwana et al., 1998).

Typical GPR responses are recognised on a radar scan in the field by "shadow zones". Here the presence of hydrocarbons appears coincident with a collapse in amplitude of the electromagnetic pulses and the GPR line scan pattern appears locally chaotic. These zones have also been reported as relatively more conductive compared to their surroundings, using other geophysical methods. Others have also observed the GPR response as a bright reflective layer, from the influence of a more resistive zone (Atekwana et al., 1998). It has been inferred that the more resistive horizons may typically relate to recently spilled hydrocarbons while the more conductive are older in age.

The 200 surveys undertaken to date using Enviscan have indicated that each plume is associated with areas of collapsed amplitude. It is considered that the GPR technique is sufficiently sensitive to changes occurring at the microscopic scale, that electrical reflectors are haphazard as a result. This could create the observed chaotic signatures and loss of energy creating a shadow zone. Site observations suggest that changes in apparent conductivity that result from biodegradation, are mostly stronger than variation in geology. Where a site is, however, characterised by significant changes in geology, then the field instrument gain settings can be used to differentiate between the two.

In order to provide a meaningful geophysical interpretation, there is a need to understand the inter-relationship with the geological, hydrochemical, hydrogeological and contaminant migration processes that could affect it.

There are a number of technical advantages and also limitations associated with this non-intrusive technique, the advantages being:

• •

  minimal site operational disruption;

• •

  its continuous data collection provides a comprehensive data set (comparing some 10,000 data points for a typical petrol station survey compared to some ten boreholes);

• •

  greater certainty for any environmental risk assessment;

• •

  greater control of any remediation programme.

Limitations include those generally associated with GPR such as:

• •

  depth of investigation typically down to 10m depth;

• •

  influence of thick clay at ground level affecting depth of penetration;

• •

  heavily reinforced concrete masking ground below it;

• •

  loss of ground penetration during rainfall, particularly on concrete surfaces.

Other influences caused by above and below ground utilities, vegetation and type of surface terrain all contribute to the need for a detailed analysis of the data in order that any false positives or false negatives are not created due to misinterpretation. Calibration of the Enviscan data using borehole data minimises this, as well as being able to quantify the contaminant distribution across a site.

The Enviscan survey cannot differentiate between the types of hydrocarbons. If the source of contamination is not already known, then laboratory analysis of borehole samples would be required.

At present most of the 200 case studies undertaken related to hydrocarbon contamination only. It is now being expanded for other types of chemicals as part of an on-going evaluation. One survey conducted over an area contaminated with chlorinated solvents has been successful and this will be one area of future development. Other authors (Brewster and Annan, 1994; Carpenter et al., 1994) have also investigated the use of GPR for similar solvent chemicals.