METHODS
Wells were selected for water sampling according to the distribution and the age of the wells. Sufficient time (minimum 1-2 weeks) past completion or workover is needed to ensure that formation water uncontaminated by drilling and completion fluids is sampled. Two wells per township were sampled to represent each producing coal seam. From June, 1999 through May, 2000, 47 wells were sampled for water (Fig. 4). Wells sampled, the date of sampling, and other pertinent information is listed in Table 1 and was obtained from the WOGCC well files and drilling completion reports. The producing coal listed in Table 1 for each sample is the well operator’s designation (from well completion reports) and identification of coal seams or coal seam names may neither be consistent among operators nor consistent with nomenclature used by others. 

Water samples were collected following guidelines of Lico and others (1988). Wells were allowed to flow through the tubing and fittings prior to collection of water samples to ensure flushing of the sample ports and collection of a representative sample. Most wells were pumping nearly continuously so water in the well bore was constantly being replaced. Water was collected directly from the wellhead by attaching tygon tubing to a port on the wellhead tee. The pressure on the port was generally less than 60 psi. Both gas and water were expelled from the well, but the amount of gas expelled was generally small relative to the amount of water. The water was allowed to collect in 5 gallon buckets while flushing the well and into a clean, rinsed polyethylene carboy with spigot or directly to the filter setup during sampling. Clean sample bottles were rinsed with well water at least twice prior to collection of a sample. For those analyses that did not require filtering (total inorganic carbon, alkalinity, and conductivity) samples were taken directly from the tygon tubing. 

Water in the carboy was immediately filtered through a 0.1 mm polyethersulfone membrane filter utilizing a peristaltic pump, tygon tubing, and an acrylic filter holder (Fig. 45). Polyethylene bottles for major, minor, and trace cation analyses were prewashed with a mixture of 1.6 N nitric and 3.6 N sulfuric acid followed by a rinse with deionized water. Polyethylene bottles used for anions were prewashed with deionized water. Samples for major, minor, and trace cations, deuterium, oxygen, and carbon stable isotopes, and anions were collected from filtered water. The major, minor, and trace cation samples, except for mercury, were acidified with Ultrex nitric acid to a pH <2. Samples analyzed for mercury were collected in 30 mL glass bottles containing 1.5 mL of a sodium dichromate-ultrapure nitric acid mixture.

Temperature and pH were measured while the well flowed prior to collection of samples. The pH meter and electrode were calibrated using standard buffers before each measurement. Conductivity of the water was measured at the well, but measurements proved to be unreliable because of gas bubbles affecting the conductivity probe. Conductivity reported in this paper is the measured conductivity in the laboratory at 20^o C. Total alkalinity for samples was determined by titration with standard sulfuric acid as soon as possible after sample collection, generally within 8 hours of collection. Samples for alkalinity, anions, dissolved organic carbon, ammonia, and d¹³C of bicarbonate were placed on ice in an ice chest in the field and transferred to a refrigerator on return to the laboratory.

Analytical methods used in this study are described in detail in Arbogast, 1996. Major and minor cations were determined by inductively-coupled plasma atomic emission spectroscopy (ICP-AES) with duplicate samples having a mean deviation generally within 6 percent. Trace cations except for mercury and selenium were determined by inductively-coupled plasma mass spectroscopy (ICP-MS). Samples analyzed by ICP-AES and ICP-MS were analyzed using both prepared multi-element standards and standard water samples obtained from the U.S. Geological Survey National Water Quality Laboratory. Mercury was determined by two methods. Sixteen samples were analyzed using cold vapor atomic fluorescence spectroscopy having a detection limit of 0.005 mg/L (Crock, J., USGS, personal communication) and the remainder of the samples were analyzed utilizing cold vapor atomic absorption spectroscopy with a detection limit of 0.1 mg/L. Selenium was determined by hydride generation atomic absorption spectroscopy. Values of detection limits for each element are shown in Tables 2 and 3. Concentrations of anions in the samples were determined by ion chromatography using a Dionex 500 chromatography system equipped with an AS-14 anion exchange column and using a sodium bicarbonate-sodium carbonate eluent. The estimated precision for the anion analyses is +/- 5 percent except for bromide whose concentrations are near the detection limit. Estimated precision for bromide is +/- 11 percent.