Methyl Tertiary-Butyl Ether (MTBE) was mandated by the US Environmental Protection Agency as a gasoline additive in 1979 to reduce air pollution and is part of a reformulated gasoline to increase oxygen in the gasoline causing it to burn more cleanly in automobiles. Gasoline containing MTBE is also distributed in Canada, however as of late 2000, MTBE is no longer sold in British Columbia. MTBE is highly soluble and can move rapidly through soils and aquifers. Concern over MTBE contamination of ground water in British Columbia was raised during the November 2000 Legislative Report of Proceedings from the Public Accounts Committee. As part of the Provincial Government’s response over this concern the Ground Water Section of the Water Protection Branch conducted a ground water sampling program between June and November 2000. A total of 76 water samples were collected from 58 domestic wells, observation wells and piezometers located in the Lower Mainland, Vancouver Island, and the Interior of B.C.. Most wells sampled were completed in sixteen shallow, unconfined, highly vulnerable, sand and gravel drinking water aquifers that are regularly monitored by Ground Water Section as part of the Province’s Observation Well Network and Ambient Ground Water Quality Monitoring Network. The sampling did not target specific sites contaminated by hydrocarbons. Of the 76 samples analyzed by Analytical Service Laboratories Ltd. all MTBE results were <1.0 part per billion (ppb) except one well, domestic well "U" located in the Abbotsford-Sumas Aquifer. This well had a slight detect of 2.0 ppb MTBE, was re-sampled in December 2000 by Regional Water Management staff and MTBE was confirmed at 3.0 ppb, well below the British Columbia drinking water guideline of 20 ppb based on taste and odour thresholds. Quality assurance / quality control (QA/QC) procedures including taking duplicate samples were followed each time the portable pump was used. The pump and discharge hose were decontaminated by pumping distilled water through the system and submitting this washed sample to ASL for analysis. All QA/QC results were negative. A field survey was conducted in February 2001 to identify potential sources of contamination and to sample wells located approximately 1 km radius of well "U" where MTBE was detected. The majority of surrounding residences were, however, connected to the City of Abbotsford water supply and only 2 residences were using well water for domestic use. These wells, including domestic well "U", were re-sampled in February 2001 and MTBE results were <1.0 ppb MTBE for all three wells suggesting that the very low levels of 2.0 and 3.0 ppb MTBE may have naturally flushed through the aquifer. Possible sources of MTBE included nearby gasoline storage tanks on private properties. Well owners should protect their water source from MTBE contamination by inspecting the immediate area around their well and pump house and relocating or safeguarding any potential contaminant sources such as gasoline storage tanks a safe distance from their water source.

1. Occurrence and Physical Characteristics of MTBE

Methyl tertiary-butyl ether (MTBE) is a colourless, volatile organic compound (VOC) produced almost exclusively for use in gasoline. It is made of methanol and a by-product of the oil refining process. Oxygenates (MTBE, Ethanol) were mandated by the USEPA in certain areas as a gasoline additive in 1979 to reduce air pollution and is part of a reformulated gasoline to increase the amount of oxygen in gasoline causing it to burn more cleanly in automobiles. Such reformulated gasolines typically may contain up to 15 percent of MTBE to reduce vehicle carbon monoxide emissions. MTBE has been reported to occur in Western Canada gasoline at an average level of 0.44 percent by volume for all brands, (Tushingham, Hodge, and Roberge, 2000).

MTBE is lighter than water and has a high solubility in water. The specific gravity of MTBE is 0.740 grams/cm³ (Aldrich, 1994) whereas water has a specific gravity of 1.00 gram/cm³. Because MTBE has a high solubility in water, it can readily dissolve in ground water and be transported as a dissolved species by ground water flow. Table 1 summarizes some basic physical properties of MTBE and another component of gasoline (benzene) which is included for comparison. MTBE is resistant to microbial decomposition and is difficult to remove in water treatment. It’s turpentine taste and smell renders water undrinkable at even 30 parts per billion (ppb).

MTBE concern was raised in the November 2000 Legislative Report of Proceedings from the Select Standing Committee on Public Accounts. Concern over MTBE contamination to ground water was discussed on several occasions including the potential hazards presented to rivers and aquifers from gasoline pipelines crossing the Province and transport of MTBE in concentrated form by truck and rail. The importance and results of the province wide sampling program being carried out were also discussed during these proceedings.

Common sources of MTBE are leaking underground fuel storage tanks, spills at gasoline storage sites and storm water runoff from roads and parking lots.

Table 1. Physical Properties

	MTBE	Benzene
Molecular Weight	88.15	78.11
Specific Gravity (g/cm³)	0.740	0.874
Solubility (mg/L) at 20°C	4700	1780

2. Objectives of MTBE Sampling Program

In order to survey how widespread MTBE might be present in the ground water (and aquifers) in British Columbia, sampling was carried out in conjunction with the 2000 / 2001 Observation Well Network and Ambient Water Quality Network sampling programs.

Briefly, the Observation Well Network is comprised of about 160 wells located in major ground water areas of British Columbia. Each year about 25 of these wells are sampled for inorganic chemical analysis. For further information on the Observation Well Network see the Fact Sheet (Appendix A). The Ambient Water Quality Network was initiated in the late 1980’s and comprises about 45 domestic wells and 40 piezometers that are sampled once a year for inorganic chemical analysis to monitor non-point source contamination of nitrate in three regions: Lower Mainland, Osoyoos, and Grand Forks.

Whenever an observation well or domestic well was sampled for regular sampling, MTBE sampling was included. This sampling program was conducted to provide a preliminary survey to check for the presence of MTBE in ground water.

Site-specific sampling of ground water beneath known hydrocarbon contaminated sites was not carried out during this preliminary sampling program. The Ground Water Section sampling program was one component of a number of components to the provincial government response to concerns over MTBE contamination. Preliminary sampling for MTBE was also undertaken by the Ministry of Health Services and local health regions for a number of community water supply systems. Other work has been preparation of a Fact Sheet on MTBE completed in the spring of 2000 (Appendix B) to raise public awareness and the development of Ambient Water Quality Guidelines for MTBE. The MTBE guideline for drinking water in British Columbia is set at 20 ppb based on taste and odour thresholds.

This report presents the Ground Water Section’s MTBE sampling procedures and sampling results and provides recommendations to well owners for protecting and safeguarding their water supplies by identifying and removing potential contaminants (i.e. gasoline storage tanks) a safe distance from the well.

3. MTBE Sampling Program and Site Selections

The sampling program included sampling of all established domestic wells within the Ambient Water Quality Network Program in the Lower Mainland, Osoyoos, and Grand Forks, specific observation wells and select irrigation wells in Keremeos. The majority of the wells sampled are shallow, dug and drilled wells and are located in highly vulnerable, unconfined sand and gravel aquifers except observation well No. 65 (located in Saanich, just north of Victoria) which is completed in a bedrock aquifer (Table 2).

During the summer and fall of 2000, the Ground Water Section, Ministry of Environment, Lands and Parks (now Ministry of Environment) collected a total of 76 samples from 58 wells located in sixteen aquifers in the Lower Mainland, Vancouver Island and the Interior of B.C. (Figure 1). For privacy reasons, domestic well owners’ names have been abbreviated as shown in Table 2 and throughout the text. All wells sampled are shown in Table 2. A photograph of a typical domestic well sampling site is shown in Appendix C.

4. MTBE Sampling Procedures

Domestic wells were sampled by turning the inside or outside water tap closest to the well on and filling 2-40 ml purge and trap glass vials. The sample vials contained a small amount of sulphuric acid as a preservative. The sampling locations have been previously established and tagged during the regular sampling program.

Sampling of observation wells and piezometers was carried out by installing a portable submersible pump, pumping the well or piezometer for a specified period (2 to 5 well volumes if possible) through a hose and then collecting and shipping the samples in the same manner as the domestic well sampling. Specific conductance and water temperature were taken at regular intervals during pumping to determine the optimal time for sample collection.

As a quality control measure and to minimize cross contamination by the pumping equipment between wells, decontamination procedures were carried out. Pump decontamination procedures have also been included in Appendix D.

5. Land Use Survey

Prior Prior to sampling a well, a visual inspection around the vicinity of the well was carried out to locate and identify any potential MTBE sources. MTBE sources such as gasoline stations, gasoline storage tanks and utility storage sheds were noted. This inspection was preliminary and carried out with the intent of obtaining more detail if a particular well showed the presence of MTBE after initial sampling. If MTBE was detected in a well then a detailed land survey in the vicinity of the wellhead would be carried out when the well was re-sampled to verify the presence of MTBE. The general land use around each well sampled is indicated in Table 2.

6. Analytical Service Laboratories Ltd.

All samples were forwarded by courier to Analytical Service Laboratories Ltd. (ASL) in Vancouver for MTBE analysis. The purge and trap 40 ml vial samples were analyzed by a Gas Chromatograph Mass Spectrometer with purge and trap sample introduction. ASL was able to analyze and detect MTBE to <1.0 ppb. The turn around time between sample submission and reporting of results was about 10 working days.

7. MTBE Sampling Results and Discussion

All MTBE results as shown in Table 2 were <1.0 ppb except for domestic well "U" (2 ppb) located in the Abbotsford area (east of Vancouver).

Well "U" was re-sampled in December 2000 to check the 2.0 ppb result of November 2000. The follow-up sample contained 3.0 ppb MTBE. In February 2001, a field survey was carried out and residences within approximately 1 km radius of well "U" were contacted. Only two other residences were found to be using private well water for domestic use within the surveyed area. The remaining residences are connected to the City of Abbotsford water supply. The three wells were sampled and samples were submitted to ASL for MTBE analysis. The three sample results were <1.0 ppb MTBE including domestic well "U". The fact that MTBE was not detected in "well U" in the February 2001 sample suggests that the low levels of 2.0 and 3.0 ppb MTBE may have naturally flushed through the aquifer. During the field survey of well "U" some potential sources of MTBE were noted. Gasoline and diesel fuel storage tanks on a wooden pedestal were noted located approximately 60m (200 ft) east and slightly down slope of well "U". It was further noted that the pedestal was located on a coarse gravel base and not a concrete or spill containment pad. A gasoline tank on wooden pedestal also on a gravel base was located directly behind the residence. A buried gasoline tank (s) with two hand pumps at surface was noted beside the driveway of a neighbouring property. The hand pumps were padlocked.

This was a preliminary sampling program. It was carried out to determine if MTBE contamination was prevalent in British Columbia and if it was present in shallow unconfined vulnerable aquifers in British Columbia that are regularly monitored. Results should not be interpreted to indicate all similar aquifers are necessarily safe from MTBE contamination nor that MTBE is not present in the aquifer elsewhere. Any future sampling should also target wells located near industrial or commercial areas where the possibility of detecting MTBE is more likely to occur.

Quality assurance and quality control measures (QA/QC) were taken for those observation wells that were pumped using the portable Redi-Flo pump. Following sample collection, step-by step decontamination procedures as shown in Appendix D were taken. QA/QC sample results indicated that use of the portable Redi-Flo pump did not have any effect on the results.

8. Conclusions and Recommendations

The MTBE Preliminary Sampling Program included sampling established domestic wells in the Ambient Water Quality Program Network and specific Observation Wells. The sampling Program did not specifically target any known hydrocarbon contaminated sites. The purpose of the sampling program was to survey a number of aquifers across the province for the presence of MTBE.
Fifty-eight wells located in the Lower Mainland, Vancouver Island, and the Interior of B.C. were sampled for MTBE during the summer and fall of 2000. A total of 76 samples were submitted for analysis.
The sampling program covered a total of sixteen drinking water aquifers. Most wells sampled were shallow, dug and drilled wells and located in unconfined, highly vulnerable aquifers. All wells sampled were from surficial aquifers except observation well 65 (Saanich) which is located in a fractured bedrock aquifer.
Whenever an observation well was pumped, the pump and hose were washed (decontaminated) using distilled water to identify if cross-contamination of the pumping equipment was an issue. The washed samples were also submitted to ASL Ltd. for analysis as a quality assurance / quality control measure. All QA/QC sample results were <1.0 ppb.
Of the 76 samples submitted to ASL Ltd. all results were < 1.0 ppb except domestic well "U" located in Abbotsford which had an initial detection of 2.0 ppb MTBE. This well was re-sampled in December 2000 by Regional Water Management Staff and MTBE was confirmed at 3.0 ppb. In February 2001, a field survey was carried out and residences within approximately 1 km radius of this well were contacted. Only two other residences were found to be using well water for domestic use within the surveyed area. The remaining residences are connected to the Abbotsford City water supply. The three wells were sampled and submitted to ASL for analysis. The three sample results were <1.0 ppb including domestic well "U". Based on these results it would appear that the November and December 2000 MTBE results of 2.0 ppb and 3.0 ppb respectively may have naturally flushed through the aquifer. It should be further noted that MTBE levels of 2.0 or 3.0 ppb are just above the detection limit and below the MTBE guideline for drinking water in British Columbia set at 20 ppb based on taste and odour thresholds.
During the February 2001 field survey of well "U" some potential sources of MTBE were detected in close proximity to the well including a nearby gasoline and diesel tank located on a wooden pedestal over coarse gravel. This survey was carried out at this time with the intent of identifying potential sources of ground water contamination, and not to assess these sources more fully to determine if they were indeed contaminating the underlying soils and aquifer.
Based on the results obtained from this survey and preliminary results of independent sampling by local health authorities of a limited number of high risk community wells (Egan, pers comm, 2001), MTBE in ground water does not appear to be a problem at this time.
The Ministry of Environment (MoE) should continue to sample a limited number of high risk observation wells completed in unconfined, highly vulnerable aquifers for MTBE in 2001/2002. Results should not be interpreted, however, to indicate that all similar aquifers are safe from MTBE contamination nor that MTBE is not present locally in the sampled aquifer elsewhere. Any future sampling should include wells located near industrial or commercial areas where the possibility of detecting MTBE may be increased.
Private well owners and purveyors should protect and safeguard their water source by inspecting the immediate area around their well and / or pump house and identify and relocate potential contaminants such as petroleum products and gasoline storage tanks at least 30m (100 ft) distance from their water source and providing safeguards such as spill containment facilities to contain any leaks. In some cases this distance may need to be increased depending on specific site conditions.

9. References

Aldrich, 1994. Aldrich Chemical Company Catalogue/Handbook. Aldrich Chemical Company, P.O. Box 14508, St. Louis, MO. 63178-9916.

Egan, L., pers comm., 2001.

Tushingham, Hodge, and Roberge, 2000. Benzene in Canadian Gasoline: Report on the effect of the Benzene in Gasoline Regulations 1999. Environment Canada.

Figure 1: Map of MTBE Sampling Sites

Appendix A

Ground Water Observation Well Network in British Columbia

Appendix B

MTBE in Ground Water (PDF: 105 KB / 2 pages)

Appendix C

Photograph of Typical Domestic Well Sampling Site

Appendix D

Step by Step Decontamination Procedures for MTBE Sampling from Observation Wells and Piezometers

Fill a clean 21 litre (5 gallon) plastic bucket with distilled water and place the pump inside the bucket washing the outside of the pump and pump sleeve.
With the pump fully submersed in the 21 litre (5 Imperial gallon) bucket, pump and flush the pump and hose for approximately 5 minutes. Reduce the flow to about 0.02 to 0.04 L/s (1/4 to1/2 Igpm) and top the bucket up with distilled water while pumping to ensure that the bucket is always full.
After about 5 minutes reduce the flow further and collect another 2-40 ml vials ensuring that the vials do not overflow and there is no head space in the vial. These vials are submitted to ASL and analyzed for MTBE.