The panel stressed that ensuring water quality standards requires a solid understanding of analytical practices. This presentation reviewed the data needed to support science-based evaluations of groundwater and water wells throughout the Appalachian basin. It also featured a review of emerging technologies and methodologies to assess water quality, as well as an analysis of pre drill water sampling and groundwater assessments.
There is an emerging trend toward requiring sampling at lower and lower detection limits. At very low concentrations, the available analytical methods generate interference and the detections are less certain. For some methods of analysis and analytes, it is possible to achieve solid levels of certainty at standard detection levels. For newer analytes and methods, however, the rate of false positives is higher. Because shale gas is an emerging industry, and the analytes of concern are not well as known, there is a higher rate of false positives. That means that labs are reporting detections of substances they are not really detecting. The rate of false positives can be as high as 25%. There is also an issue with false negatives, but it is less common. These analytical methods will need further development so that the industry, the government and the public can have greater confidence in water quality sample results.
With respect to radionuclides in groundwater, background concentrations can vary from place to place and time to time. The authors of one study reported that they had taken background samples of 7 wells, once a day for 25 days. The wells were then sampled weekly for 4 weeks. Another 21 wells were sampled once per week for 5 weeks. The wells were in fractured bedrock. Total dissolved solids in the wells went as high as 4300 ppm. Various radionuclides were discovered in the wells in widely varying concentrations. The rate of day to day variation was very high. Stressing the wells with high pumping rates resulted in higher radionuclide levels because it stirred up sediment. The vast majority of the radionuclide activity was related to the amount of sediment in the well. The amount of radium in various wells also varied greatly with different analytical methods. This clearly has implications for water quality assessments. Not all change is a result of human activity. Better background sampling is key to accurately assessing human impacts on water quality.
Methane also occurs naturally in groundwater. Extensive pre drill sampling in the Marcellus Shale has demonstrated sometimes high levels of background methane. In a review of 20,000 samples taken before pad construction, methane was detected in 5600 samples, with the highest detected concentration being 500mg/l. A significant number of samples were above 7mg/l, the Pennsylvania action level. Methane appeared to be inversely related to depleted sulfate levels. Depleted sulfate was used as a surrogate for a low oxygen level. High methane was associated with high sulfate. Methane levels also increase with well depth in some areas. The understanding of how methane varies naturally is still developing.
There are now technologies that allow for real time measurement of methane levels in well water. Methane and other gasses are measured in the well headspace. The measurements are taken once a minute. The equipment is linked to a cell modem, and the data is automatically uploaded to a central server for analysis. A pilot study using this method shows intraday variability in background methane levels in residential wells. The methane levels rise along with water use and ambient temperature and humidity. Rainfall does not appear to affect methane levels, but methane did seem to vary inversely with barometric pressure. This is a pilot study. More work needs to be done, but the ability to collect real time data has given rise to a new level of understand as to what factors impact background methane levels.