Sign up to receive email updates, action alerts & health tips from EWG. [Privacy]

Chapter 2. Methodology

Weed Killers By The Glass: Chapter 2. Methodology

September 1, 1995

Weed Killers By The Glass presents the results of drinking water testing in 29 communities throughout the Midwest, the Chesapeake Bay and Lousiana. Participants in the study collected tap water samples every third day during the peak pesticide contamination period from May through August, providing the first-ever extensive study of pesticide contamination in finished drinking water at sampling sites throughout the Midwest and at other selected locations (such as Baltimore and New Orleans).

 

Communities Selected For Participation

Twenty-nine communities, with a total population of over 7.5 million people, were selected for participation in this study (Table 4). Communities were selected based upon previous indications of contamination, the willingness and availability of project participants, and the need to ensure geographic diversity in regions tested. The fact that a community is not included in the study does not indicate that there are no weed killers in their drinking water. In fact, in many smaller corn belt communities that rely on rivers or reservoirs for drinking water, herbicide contamination problems are almost certainly worse than the problems identified in the larger communities where we sampled. For example, of the 38 communities reporting a violation of EPA's current atrazine standard, 31 were small community water systems serving less than 3,300 people.

 

Table 4. Communities where tap water was tested for weed killers.

 

Akron, OH Johnson Co. WD 1 (Mission), KS
Alliance, OH Kansas City, KS
Baltimore, MD Kansas City, MO
Bettendorf, IA Lawrence, KS
Bowling Green, OH Mankato, MN
Cedar Rapids, IA Memphis, TN
Columbus, OH Minneapolis, MN
Danville, IL Muncie, IN
Decatur, IL New Orleans, LA
Des Moines, IL Omaha, NE (Missouri R.)
Ft. Wayne, IN Omaha, NE (Platte R.)
Granite City, IL Richmond, VA
Indianapolis, IN Springfield, IL
Iowa City, IA St. Louis, MO
Jefferson City, MO Topeka, KS

 

Sampling and Testing Methodology

In all communities, testing began on either May 15 or May 25, and will continue into August. The results presented in this report represent approximately the halfway point in sample collection. All samples were of municipally treated tap water, collected at a kitchen or bathroom sink. No home water treatment units (softeners, filters, etc.) were used on taps that were sampled for pesticides. Cold water was used for all samples, and was allowed to run from two to three minutes before sample collection.

Specially prepared sample jars were provided by the laboratory with appropriate cleaning and pretreatment. Samples for immunoassay analysis were collected every third day in special sample containers, stored in a refrigerator, and after a fourth sample was collected, mailed in a special shipping container to the University Hygienic Laboratory, Iowa City, Iowa.

All samples were refrigerated at 4° C upon receipt at the laboratory until analyzed. Additionally, a liter sample was collected from each location once per month, mailed to the laboratory, and refrigerated upon receipt. The liter samples from each location were extracted for herbicide analysis by gas chromatography (GC) methods. The GC samples were extracted within seven days of receipt at the laboratory.

Analysis of water samples was done under contract with the University Hygienic Laboratory (UHL) in Iowa City, Iowa. The UHL is a nationally respected environmental and public health laboratory that has been in operation for over 90 years, and routinely performs analytical work for the US EPA, other federal and state agencies, public water suppliers, and industry. The Hygienic Laboratory follows strict Quality Assurance and Quality Control guidelines, and is certified to perform environmental analysis ranging from drinking water to hazardous waste sites.

Immunoassays were performed for atrazine and cyanazine using standard commercial kits, and all analyses were conducted in the controlled conditions of the professional laboratory. The EnviroGardTM Triazine and Cyanazine Plate Kits, marketed by Millipore, Inc., were used for the determination of atrazine and cyanazine respectively. Procedures provided by the kit manufacturer were used with additional quality control measures established by the laboratory to help ensure accurate results. Following the manufacturer's instructions, samples were analyzed within seven days of receipt at the laboratory. All samples were analyzed in duplicate, and the results were normalized against control standards and averaged. Results from both analyses were required to be within specified quality control limits or the sample was re-analyzed.

Quantification limits set by the laboratory through validation studies are 0.1 ug/l (ppb) for the triazines, and 0.25 ug/l for cyanazine immunoassay methods respectively. Previous work has shown that these tests provide reliable estimates for the concentration of atrazine and cyanazine, but they are also responsive to other related compounds.

Gas chromatographic (GC) analysis provides more specific identification of individual compounds. The GC analysis was performed according to Method 507 --Determination of Nitrogen- and Phosphorus-Containing Pesticides In Water By Gas Chromatography with a Nitrogen-Phosphorus Detector -- developed by the U.S. Environmental Protection Agency. This method allows the identification and quantification of many different herbicide compounds such as acetochlor, alachlor, atrazine, butylate, cyanazine, metolachlor, metribuzin, simazine, and trifluralin. In addition, although desisopropylatrazine and desethylatrazine, the metabolites (breakdown products) of some of these compounds, are not specifically included in the method, they can be determined by Method 507. Laboratory established quantification limits for compounds determined by Method 507 are 0.1 ug/l (ppb) for each compound. For confirmation purposes, the laboratory uses two dissimilar chromatography columns for recognition and quantification of each compound. Instrument calibration is performed routinely using appropriate methods for many of the herbicides commonly applied in the Midwest.