Every day seems to bring a new story about the crap we are leaving strewn around our environment. Not just trash but chemical trash, the kind of stuff we eat, pop as pills or dispose of and which winds up as molecules in our air, water and food. You may think you know more than you want to about this but in fact you know very little about it because no one knows much about it. For example, there are no systematic measurements of the stuff in the raw water that goes into our water treatment plants, facilities designed to handle micro-organisms and not chemicals. The EPA has a program called the Unregulated Contaminant Monitoring Rule (UCMR) that requires public water systems to look for specific unregulated chemicals in their finished water (the stuff that comes out of the tap in contrast to the raw water, the stuff that goes into the treatment plant). But the list is limited and doesn’t contain most of the organic chemicals now being found in the waste water and run-off finding its way into our sources of drinking water, whether surface supplies (lakes, rivers, reservoirs) or groundwater (wells that tap water underground). In summer 2001 the US Geological Survey took samples from raw water sources in 25 states and Puerto Rico and analyzed them for 100 targeted chemicals. They (finally) report it in a new paper in Science of the Total Environment.

There were 25 groundwater sources and 49 surface water sources represented, serving anywhere from a single family to 8 million people. This was a preliminary reconnaissance study to see what was out there in a wide range of raw water sources. I’m not sure you want to know, but if you keep reading I’ll tell you anyway.

First the “good” news. Even though very sensitive analytical methods were used, 60% of the 36 pharmaceuticals they looked for weren’t detected. However the authors note that many of these drugs are chemically altered as they pass through the environment and the absence of the parent compounds may not be indicative of the absence of their metabolites (which weren’t looked for).

And a lot of other stuff was there. Here are some interesting examples (“%” is the percent of samples in which these materials were detected).

[I]n surface water, metolachlor (53%, herbicide), cotinine (51%, nicotine metabolite), beta-sitosterol (37%, natural plant sterol), and 1,7-dimethylxanthine (27%, caffeine metabolite); and in ground water: tetrachloroethylene (24%, [drycleaning] solvent), carbamazepine (20%, pharmaceutical), bisphenol-A (20%, plasticizer), 1,7-dimethylxanthine (16%, caffeine metabolite), and tri (2-chloroethyl) phosphate (12%, fire retardant). (Focazio MJ, et al, A national reconnaissance for pharmaceuticals and other organic wastewater contaminants in the United States, Sci Total Environ (2008), doi:10.1016/j.scitotenv.2008.02.021)

The median number of detections was 4. In other words, half the sites had more than four detects and half had less. Four sites had no detects but one site had 31. 63 of the chemicals were detected at least once. Here is a figure from the paper showing, for surface water (white bars) and ground water (black bars) separately, the relative prevalence of various classes of chemical types:

You have to know what you are looking for before you can measure it. You don’t just take a water sample to “see what’s in it.” You need a shopping list ahead of time. This paper thus provides no information on other chemicals. It is a snapshot of the prevalence of these 100 chemicals in a range of raw water supplies in the US. You may wonder how likely are just these chemicals to get into your drinking water. Good question:

It is evident from these comparisons and results that natural attenuation processes alone will not likely remove or decrease all OWCs to non-detectable concentrations in surface- and ground-water sources of drinking water. For example, carbamazepine, which has been shown to be resistant to various natural attenuation and treatment processes and, was among the most frequently detected compounds for both surface-water (22%) as well as ground-water (20%) sites in this source-water reconnaissance.

Snyder et al. (2007) evaluated the efficacies of a range of drinking water treatment technologies for the removal of several targeted pharmaceuticals, personal care products and other trace contaminants. That study showed through bench as well as full-scale studies that several treatment technologies (e.g. advanced oxidation, granular activated carbon, reverse osmosis) are highly effective in removing most of the target analytes from the water phase. However, a smaller subset of analytes, including some that were found in the present study (e.g. N-N-diethyltoluamide and carbamazepine; Table 3), were not efficiently removed by any of the treatment methods tested. (Focazio et al., some cites omitted)

Despite the wealth of “new” [2001 sampling] data in this paper, its main effect is to reveal how little we know about the chemical cocktail we have shaken and stirred in our environment. Most of the detections were at relatively low levels but some of them, like bisphenol-A (see our here, here, here) have hormonal activity at extremely low levels, so this is not a guarantee of irrelevance.

Meanwhile we have spent the seven years since these samples were taken putting more and more stuff out there, used resources that might have provided more and better information about what else there was and sent it up in smoke, turned our backs on regulating chemicals in the name of “the market” and “less government in our lives,” and generally given a blank check to a government that spent it like a drunken sailor on bad stuff while spending less and less each year to keep us truly safe.

Reference: Focazio MJ, et al, A national reconnaissance for pharmaceuticals and other organic wastewater
contaminants in the United States — II) Untreated drinking …, Sci Total Environ (2008), doi:10.1016/j.scitotenv.2008.02.021