 In one Flower Mound neighborhood near a natural gas well and compressor site, residents constantly complain about odors and symptoms like burning eyes and sore throats from fumes. But the state's primary environmental regulator, the Texas Commission on Environmental Quality (TCEQ) doesn't seem willing to figure out what is causing the health problems. Even if TCEQ could nail down the problem, it's not clear they would be able to force the facility's operator to fix it.
Even though the agency frequently responds to complaints, and has on occasion conducted air quality tests, they will quite often not even take field measurements for the sulfur compounds most likely causing the odors, and never test for these compounds in the samples they have analyzed in the laboratory.
The common theme to TCEQs investigations seems to be that someone will report an odor and some physical symptoms. TCEQ will then send an investigator out, who will take notes as to what he smells, which is quite often nothing, or a faint odor. The investigator will then check a portable analyzer for volatile organic compounds, and perhaps another analyzer for hydrogen sulfide - one of several dangerous sulfur compounds often associated with oil and gas production.
The investigator may then take a 30 minute sample using a vaccum container called a summa canister, which sucks in air over a given period of time for later laboratory analysis. The steel canisters are good for analyzing hydrocarbons, but are not useful for analyzing sulfur compounds because the sulfur compounds can react with the metal.
The lab reports from the summa canister will show elevated levels of VOCs, but all will be below TCEQ's short-term effects screening levels (ESLs), which are the one-hour exposure amounts below which a person is not expected to suffer any health effects. Most of the VOCs will be below long-term ESLs. But in some cases, as shown below, those numbers can be exceeded, but the test procedures are just not sensitive enough to quantify it to the standard of proof TCEQ is looking for.
We decided to follow just one of the many complaints about this site through the process of investigation, to see how it works.
 Sam Wilson Compressor Station, Flower Mound
On Thursday, September 23rd, 2010, a Flower Mound mother of two, and breast cancer survivor that we'll call "Jane" for this story drove along F.M. 1171 past the Wilson Compressor Station and smelled something that wasn't quite right. She called TCEQ to complain of the strong odor coming from the Scenic Road site with a compressor station and 11 natural gas wells known as the Sam Wilson unit. The site also contains 13 storage tanks and two glycol dehydrator units.
Operated by the Williams company, the compressor station takes gathering lines from wells in the area, compressing the gas into a nearly 7 inch pipeline that then feeds into a 24 inch Atmos gas transmission line crossing Flower Mound and Southern Lewisville.
 The Investigation
TCEQ had sent investigators to the site at least three times prior to this in 2010: March 4th, May 4th, and June 23rd. Because of special rules implemented recently for the Barnett Shale, complaints about natural gas sites require a response from TCEQ within 12 hours.
When TCEQ environmental investigators Aaron Houston and Annette Maxwell responded to the priority call, they would later report that they could smell a faint hydrocarbon odor. Using a portable toxic vapor analyzer (TVA) the investigators obtained maximum readings of 53 parts-per-million (ppm) on the amount of volatile organic compounds (VOCs) or hydrocarbons in the air, most of which would be methane. The normal ambient level of methane in the air would be only about 1.8 ppm.
The TVA doesn't distinguish between the individual compounds that make up the total VOCs, so in order to determine what is in the air, a sample must be taken back to the laboratory. According to TCEQ, their investigators also carry Jerome analyzers which can detect hydrogen sulfide. TCEQ's mobile response teams can analyze for other sulfur compounds like mercaptans, but on this occasion, the field notes did not show that either was used.
At just after 9 p.m. - about 45 minutes after arriving, the investigators set up a Summa canister outside the gates and took a 30 minute sample. Since the winds were from the South - Southeast that evening, the canister was placed about 258 feet North-Northwest of the dehydrator on the site.
The sample would arrive at the TCEQ laboratory on the following Monday, 9/27, and be analyzed using gas chromatography / mass spectrometry. The process involves pressurizing the canister, and diluting the sample, then taking a known volume of the sample for analysis using EPA analysis method TO-15, which is designed to quantify volatile organic compounds in the air samples. The laboratory methods used do not check for methane, which is not considered a VOC, or sulfur compounds like hydrogen sulfide, carbon disulfide, or carbonyl sulfide.
Laboratory Results
The following table lists the most abundant compounds found:
| Compound | Concentration in ppbv | STESL | LTESL |
|---|
| Ethane | 15 | 10000 | 1000 | | Propane | 2.2 | 10000 | 1000 | | Isobutane | 0.81 | 2040 | 204 | | Methyl chloride | 0.55 | 500 | 50 | | dichlorodifluoromethane | 0.51 | 10000 | 1000 | | Isopentane | 0.38 | 1200 | 120 | | Acetylene | 0.36 | 25000 | 2500 | | n-butane | 0.31 | 8000 | 800 | | Toluene | 0.29 | 170 | 330 | | Methylcyclopentane | 0.20 | 750 | 75 | | 1,2, dibromoethane | 0.16 | 0.5 | .05 | | M & p xylene | 0.15 | 80 | 42 | | Carbon tetrachloride | 0.13 | 20 | 2 | | 2,2,4, trimethylpentane | 0.12 | 750 | 75 | | 1-butene | 0.11 | 360 | - | | Benzene | 0.11 | 54 | 1.4 |
None of the compounds found exceeded any short-term effects screening levels (ESLs), but a chemical called 1,2,dibromoethane had a reading that exceeded a long-term level by more than 300%. Dibromoethane, (also known as ethylene dibromide) is used as a pesticide, and it may be produced by the reaction of bromine with glycol in the glycol dehydrators used on the site to remove moisture from the natural gas. The CDC's Agency for Toxic Substances and Disease Registry has more information on dibromoethane.
On Wednesday, September 29th, a week after the investigators checked out the odor, Houston emailed Jane to let her know that they had been on site and taken samples. Although the lab work had been done by this point, Houston only had the 53 ppb TVA reading, and not the laboratory analysis results, which were signed off by the analyst on the following day. The lab results were reviewed and signed off by other TCEQ employees on October 4th and 5th.
Making Sense of the Laboratory Findings
Despite the fact that the tests indicated what would seem to be elevated levels of a harmful substance, the level was below the 0.41 ppb sample detection limit (SDL) which TCEQ lab manager Karen Bachtel explains is the level that gives them 99% certainty that the reading was not a false positive. Bachtel went on to explain that if the certainty level had been 95% then the SDL would have been lower - around 0.25 ppb. So, because of the way the test is constructed, the compound would have had to register at 80% of the short-term ESL and 800% of the long-term ESL in order for TCEQ to have the required 99% level of confidence to say that it definitely was detected.
In addition to the SDL, another number comes into play: the SQL, or sample quantitation limit, which TCEQ Quality Assurance Manager David Manis explains is the Limit of Quantization (LOQ) adjusted for additional dilutions required to analyze the sample. Manis explained that for 1,2-dibromoethane, the LOQ is 0.3 ppbv. For this sample, the SQL for 1,2-dibromoethane was 0.61 ppbv. Even though the test method used, EPA's TO-15 cannot accurately quantify levels of this substance at levels that exceed both short and long-term comparison levels, TCEQ says this is the most sensitive method used.
This is the first of a two-part series. Continue reading Part II...
Keywords
- Investigations
|
![Open in new window](javascript:CaricaFoto("http://www.whosplayin.com/xoops/uploads/img4b30623402a5a.jpg");)
More issues with ethylene dibromide detections