Mapping Drug Use by Testing Sewer Water

I wrote about this in 2007, but there's new research:

Scientists from Oregon State University, the University of Washington and McGill University partnered with city workers in 96 communities, including Pendleton, Hermiston and Umatilla, to gather samples on one day, March 4, 2008. The scientists then tested the samples for evidence of methamphetamine, cocaine and ecstasy, or MDMA.

Addiction specialists were not surprised by the researchers' central discovery, that every one of the 96 cities -- representing 65 percent of Oregon's population -- had a quantifiable level of methamphetamine in its wastewater.

"This validates what we suspected about methamphetamine use in Oregon," said Geralyn Brennan, addiction prevention epidemiologist for the Department of Human Services.

Drug researchers previously determined the extent of illicit drug use through mortality records and random surveys, which are not considered entirely reliable. Survey respondents may not accurately recall how much or how often they use illicit drugs and they may not be willing to tell the truth. Surveys also gathered information about large regions of the state, not individual cities.

The data gathered from municipal wastewater, however, are concrete and reveal a detailed snapshot of drug use for that day. Researchers placed cities into ranks based on a drug's "index load" - average milligrams per person per day.

These techniques can detect drug usage at individual houses. It's just a matter of where you take your samples.

Posted on July 23, 2009 at 6:09 AM • 57 Comments


Paul WiedelJuly 23, 2009 7:14 AM

This kind of testing really only works at an aggregate level.

Wastewater can be sampled at a house level, but the practice of collecting testable and legally admissible samples would be a tremendous challenge. Collecting samples that would hold up in court would be a logistical nightmare.

Taking a water sample from an individual sewer hook up would logistically be very challenging for most houses. But let's assume that a house of interest is next to a manhole cover.

Most household wastewater does not contain human waste. It's mostly waste from showers/dishwashers/laundry, etc. Most of the water that would pass through a sewer hook up would give a negative.

Taking a point sample would likely result in a lot of false negatives.

False positives from contamination is another issue. Sanitary sewers are de facto storm sewers in just about every community. That is to say that there is enough leakage in the sewer that a rainfall will raise the water level in the sewer to contaminate the hook up with traces of waste from other houses. Even 'clean' households will have traces of illicit substances.

Letting a sample soak would leave that sample susceptible to contamination from rain or even a blockage in the sewer.

Testing at the plant or at neighborhood sewer lift stations is far more practical than at houses. It also does a decent job of protecting individuals' privacy.

pegrJuly 23, 2009 7:27 AM

Thanks for the heads-up, Bruce! I'll be peeing in the yard from now on! ;)

John MooreJuly 23, 2009 7:42 AM

They don't say if they used internal controls such as testing for alcohol, birth control, or other prescription medications whose sales could be correlated with waste water levels to calibrate overall drug use. They just seem to be looking at relative levels between communities. One could have done a similar study with cocaine and meth using a gas chromatograph/mass spectrometer with an ion gun sampler on dollar bills ( URL: ) from convenience store receipts. Relative trace levels on dollar bills could likely be used to determine drug use at the individual level or map drug distribution patterns in a more granular fashion (users to distributors). What happens to privacy when a GC/MS costs a $1000 per unit on a handheld device and your employer sniffs your employee badge for drugs?

Bill's Not HereJuly 23, 2009 7:48 AM

Just another reason not to hook up to city sewer services. Don't ya just love living in da South.

bobJuly 23, 2009 7:52 AM

I had been wondering how long this would take. Next they'll have a little robot that can accurately navigate to the sewer entry point of any given house, sample the output and provide probably cause to break in and seize everyone/everything within it.

I suspect the courts will treat this the same as refuse and consider that once you put something in the sewer it is public information and can be gathered without a warrant.

Andy SJuly 23, 2009 8:00 AM

I don't see any value in sampling individual houses: you get other household members and the one you suspect may have not gone there...

Also, it's not a crime. If it were, say you have a parole case, then you can administer a urine test at the time of your choosing which is properly identifiable.

It's similar as why datamining works for marketing not terrorism

John HoughtonJuly 23, 2009 8:04 AM

And how much of that Meth was prescribed to people with ADHD I wonder? Even if there are ways to chemically differentiate it, a fair amount of amphetamine abuse is via prescription drugs.

Bob, until there is a way to directly tie it to a person's DNA there is a defense of it being a visitor using the toilet. Making the test cheap enough and reliable enough for forensic use isn't going to happen any time soon.

Also, there are states that don't allow thermal imaging or air surveillance to look for drugs.

dugJuly 23, 2009 8:18 AM

Interesting that the saturation level is such that they did not get false negatives in the survey, If the sample was daylong, how big was the jar they carried it to the lab in. Apparently when a drug is classified schedule I , it becomes so soluable that it can be detected in a huge volume of water all of which contains a molecule of the substance.

DavidJuly 23, 2009 8:35 AM

A long time ago, living in a poorer neighborhood than I do now, I found a gold-looking razor blade in a snowbank. It had some white stuff adhering to it, which I washed off thoroughly, making sure I didn't accidentally ingest any of it.

Therefore, I probably dumped a fair amount of cocaine into the sewer. If so, it's indicative of drug use in the neighborhood, but not my home.

As far as I can tell, I wasn't doing anything illegal, assuming that something tossed into a snowbank is abandoned property. I certainly would not have welcomed a raid from a narcotics force, conducting a very painstaking search of my house.

paulJuly 23, 2009 8:36 AM

@John Houghton

For people living in public housing, the "visitor" defense is not available -- you can be thrown out of your apartment if anyone there uses drugs, whether they live there or not.

In particularly defendant-unfriendly jurisdictions I could see this being used to get probable cause on a bunch of suspects. But yeah: spetic tanks and composting toilets are the wave of the future.

WallyJuly 23, 2009 8:50 AM

Is this cool or scary? From a drug-use researcher's point of view, aggregate demographic data of this nature (even with false positives) would be awesome. But from a civ-lib standpoint, it seems to be illegal search. I guess if it were far enough downstream that you couldn't identify individual houses...

Big KJuly 23, 2009 9:01 AM

I don't think any of this is meant to be used in court, it seems to me that it's from more of a research/public health angle that a criminal prosecution one.

TSJuly 23, 2009 9:16 AM

If police can legally search your trash, why not the sewer? Once it leaves your immediate possession, is there any expectation of privacy? It is going into a public sewer system.

On the other hand, in urban areas, you could have hundreds of apartments in a MDU all sharing a single sewer line. Since it's impossible to tell which apartment the drugs are coming out of without going into the apartment and tapping private lines, that may require a warrant.

Unless they also pull DNA and match to known DNA records, which I expect would be possible.

kashmarekJuly 23, 2009 9:18 AM

By indicating an average per person use of an illegal drug in a community (or area), they have made everyone in that area a criminal. To avoid that broad brush generalization, they have to back it up with statistics on the number of drug users in that community and adjust the average to that population.

TimHJuly 23, 2009 9:23 AM

Don't the sewage farms check for human blood markers in the input as matter of course, and also LEOs can backtrack the piping to see which house the chop 'n' flush of an (un)loved one came from?

clvrmnkyJuly 23, 2009 9:29 AM

A similar survey was done in Scotland, though they were looking at aggregate use of mood elevators. Apparently, some of these sorts of drugs are excreted unchanged in human waste, resulting in measurable concentrations in treated water.

The concern was that the general population was getting exposed to these drugs in food and water as a side effect of the booming prescription drug trade (legal or illegal.)

I can't find the link, sorry.

Davi OttenheimerJuly 23, 2009 9:53 AM

It would be much much more interesting if they focused on pervasive health issues for these maps.

I mean illicit drug use is a risk, but it's minor compared to the effect of pollutants running through every tap and into the sewers.

I remember something like this done in Ann Arbor to show toxins in tap water that made it into the waste system.

Here's a typical tapwater study:

"An Environmental Working Group analysis of tap water tests from 1998 through 2002 shows that customers of Ann Arbor Water drank water containing up to 11 pollutants."

So that's the input. Now where's the output measure?

It also seems to sometimes come up with regard to neighborhoods near WalMart

And if we're talking about water contamination in America, we have to mention China

Of course mapping sewer water toxicity is good for general health of an area, but if investigators really want to hone in on a suspect they should consider deploying sensors like those developed in Japan.

"Japanese lavatory engineers, in an effort to put the physician back in the bowl, have devised medical toilets such as Toto's "WellyouII," which "automatically measures the user's urine sugar levels by making a collection with a little spoon held by a retractable, mechanical arm."

I like to call this Sewer Harm Information Technology.

mooJuly 23, 2009 9:54 AM

Neal Stephenson's 1988 novel "Zodiac" included as part of the plot, the main character (a hands-on environmentalist) drove around Boston in a van "manhole diving" (taking samples from the sewers and chemically testing them on the spot) in order to narrow down the source of a deadly contaminant that had been flushed down somebody's toilet.

Trichinosis USAJuly 23, 2009 9:56 AM

@ John Houghton:

While amphetamines and amphetamine derivatives are prescribed to people with ADHD, methamphetamine is not. The difference is similar to that between methadone and heroin. I don't think they even prescribe dexidrine to people with ADHD anymore unless the other medications don't work.

Petey BJuly 23, 2009 10:06 AM

@TimH "Don't the sewage farms check for human blood markers in the input as matter of course"

please do go on.

mcbJuly 23, 2009 10:12 AM

Back in Si valley before the dotcom bubble popped there was at least one investigations outfit offering to use field drug tests to determine whether cocaine was being used by a company's employees. Said they'd start by swabbing the copy machine start buttons and the coffee pot handles and work their way backward to your subject/suspect's desk. Would have been quite the racket if they'd got anyone to bite. We politely declined their overture.

BodiJuly 23, 2009 10:30 AM

Looking for blood markers is pointless, you'll always find blood. Animal blood from normal cooking and human blood since at any time about 20% of all fertile females are discharging some blood, some of which will end up in the sewer. Add to that all the cuts and scrapes being washed off...
I don't think there's any realistic way to distinguish this background blood level, which would be quite variable, from a blood level spike caused by someone's entire blood volume being flushed down a drain... and the cost would be outrageous against the limited probability of accomplishing any useful goal. Is there even a blood marker detection system that could analyze an effluent flow in real time?

Occam's RazorJuly 23, 2009 10:35 AM


"A long time ago, living in a poorer neighborhood than I do now, I found a gold-looking razor blade in a snowbank. It had some white stuff adhering to it, which I washed off thoroughly, making sure I didn't accidentally ingest any of it."

1) What did you want the used razor blade for?
2) How does one accidentally ingest rinse water?

BobWJuly 23, 2009 10:48 AM

If they have a search warrant they don't need a manhole. The sewer cleanout plug will work.

The problem is getting the samples without being seen.

Clive RobinsonJuly 23, 2009 10:57 AM

@ TimH,

"Don't the sewage farms check for human blood markers in the input as matter of course"

Probably not do you have a lady in your life?

If so ask her why the sewers would have quite high concentrations of human blood in them...

Oh and do you shave?

Then there are people with hemeroids and other lower intestinal tract bleeding...

However if you realy must get rid of large quantities of human blood first mix it with "biological washing powder" (or whatever your local stain digesting laundry powder is called) then after a day or so mix it with large quantities of house hold bleach (in a well ventilated area).

The resulting liquid should be reasonably ok to flush...

neighborcatJuly 23, 2009 12:50 PM

So how long before some clever market-demographics research firm uses this technique to track product use or insurance company uses this technique to refine their risk maps?

While there is some potential for law enforcement abuse of this method (individual houses could easily be sampled from yard clean-outs, leading to probable-cause search), it would be expensive. As far as covert sampling techniques go, I'm much more concerned about this:


LuxJuly 23, 2009 12:58 PM

Paul W: you make some good points, but I doubt anyone would seek a conviction on the basis of sewage evidence alone. But probable cause to obtain a warrant? It seems plausible on its face that a home under sewage surveillance on suspicion of drug-related crimes that returned a positive result in sewage tests could translate into a search warrant being issued.

AndrewJuly 23, 2009 1:12 PM

Think of the possibilities for getting search warrants aimed at septic tanks!

mcbJuly 23, 2009 2:39 PM

@ so what

"This is only a problem if you do drugs."

Thanks for reminding us the Bill of Rights is only important to criminals.

If you got nothing to hide you won't mind the cops, your insurance company, or some total stranger running your urine through a mass spectrometer...

antonJuly 23, 2009 2:52 PM

Although the focus here is more on illicit drugs, there is all sorts of other intresting stuff in sewage. I believe hormones are not biodegrabale and there are tons of this stuff going down rivers.

So if you don't want to grow bigger breasts, don't drink recycled water.

ReallyJuly 23, 2009 3:00 PM

@So what: seriously? New to this blog? I thought we were past the "you only have something to worry about if you are guilty" argument.

Clive RobinsonJuly 23, 2009 3:45 PM

@ anton,

"So if you don't want to grow bigger breasts, don't drink recycled water."

In the UK you have little choice in some places.

For instance people living in Essex drink water from the River Thames.

Up river from them is the outflow from various sewage plants there is one from Berrylands sewage plant that joins the Thames at Kingston. A bit further up river at Surbiton (between Hampton Court and
Kingston) is another water treatment works.

It has been estimated that in each glass of Thames water in Essex is the by products of having been through seven individuals bladders in succession.

And artifical hormones as used in "The Pill" appear to make it through the water treatment process.

Which might acount for why Essex men are the least fertile in England and the average bra size for under 20's in Essex is two full cup sizes bigger than it is in the Cotswalds where the river Thames rises....

Apparently a similar trend in bra size is true of all long English rivers...

StephanieJuly 23, 2009 9:30 PM

Thanks Mr. Schneier for your interesting articles and comments. I enjoy reading these pieces. I wonder if hiring more police would be more useful than these sorts of tests? I think it would be safer to put these folks on patrol with two sets of eyes and ears like in the old days with partners. People watching out for their neighbors are perhaps better indicators than a sewer test. I'd rather see an investment in police protection, increased patrols, more officers on the job, better support for them in the field, than some lab test that can be tossed out in a courtroom. If you are a homeowner and have a meth/drug dealer next door,you want cops around.

Clive RobinsonJuly 23, 2009 10:32 PM

@ Alex,

"Clive Robinson scares me."

Tell me how is life in and around the Royal Free?

Clive RobinsonJuly 24, 2009 5:30 AM

@ eoj,

"Don't drink recycled water? It's all recycled"

Yes that's true, but...

It's what you get "for free" as it where such as heavy metals, carcnogens, and other "addatives" etc.

In the UK around London in particular we still use Victorian methods of water purification and a lot of the sources are not springs artesian wells etc where the water is virtually drinkable without further processing. A lot comes from rivers etc where the same Victorian treatment of sewerage is used.

In France they tend to use UV and Ozone to purify the water and make people purify their effluent at source.

Joseph Bazeljet would recognise most if not all the ways we treat water in London as it has not changed since he sat down to dine with Queen Victoria...

bobJuly 24, 2009 6:02 AM

Lets enable 3,000,000 people to find society-improving work, return $750,000,000,000 out of federal and state budgets to the taxpayers to whom it belongs and put "free" back in the last line of the national anthem - legalize drugs. That way noone will care whats in your sewer line. Well, other than the organization that treats it - they will want to make sure there arent Q-tips in it. Of course, organized crime would never allow drugs to be legalized, nobody benefits more from the current system than they do.

ToxicJuly 24, 2009 3:31 PM

Maybe they should have gone and tested the water coming out of the pipe in St. John's park to see what's in it. It's dumped right into the river straight from a sewer(Chicago style). Portland doesn't treat and recycle their water. They have no problem letting big business pollute the waterways that are partially used for drinking water later. Of course, the little neosocialists that are in Portland schools have a better agenda trying to paint anyone who doesn't agree with their statist views as a "tweaker". Hence the reason for the study( if you can call bad science that). More to do with pushing an agenda for more government intervention in citizen's lives. Expect some garbage to come out Teddy's mouth and some new laws passed soon. After all. This is the city that locks up spray paint in a cage at the store and makes you have to give your license to a store clerk so they can write down all your personal info before you can buy it. All in the name of fighting spray tagging. That's the country we live in now.

RogerJuly 24, 2009 10:17 PM

> Don't drink recycled water? It's all recycled - see

This is what is known as a "semantic quibble". While it is true that all water is recycled in some sense, the term "recycled" is here being used in two distinct senses.

In particular, the water cycle involves (natural) distillation, which is a far more complete purification process than what occurs in any practical artificial recycling plant [1]. While we could build wastewater recycling plants that used distillation, then the energy requirements would be about the same as flash desalination of sea water, while the failure modes are more dangerous [2] and other operating costs and environmental burdens would be considerably higher [3].

1. It removes *all* organisms, all complex biological molecules -- even ones that haven't been discovered yet -- and all non-volatile contaminants, whilst reducing the concentration of volatile contaminants that are not more volatile than water. It is not particularly effective at reducing the concentration of extremely volatile contaminants but fortunately the ones of practical concern are few enough in number that they can be specifically monitored for.
2. The worst case failure mode in either case is for feedwater to contaminate the clean water output. For desalination this results in the supplied water becoming slightly brackish, which is annoying but not dangerous. For wastewater recycling, it results in the supplied water becoming toxic. This is not a hypothetical problem; such accidents have actually occurred in cities that use wastewater recycling, resulting in enormous numbers of people being made seriously ill. So far to date there are no confirmed deaths, though.
3. On the discharge side of a desalination plant we have brine, which can be diluted with seawater then safely returned to the sea without environmental impact. Whereas on the discharge side of a wastewater recycling plant we have a concentrated, toxic sludge that must somehow be disposed of, adding considerable cost and environmental hazard.

TruePathJuly 25, 2009 4:30 AM

Seems to me the claim that this measures illicit drug use is questionable.

Yes, they measure drug metabolites but have they checked to make sure those metabolites only occur as a result of in vivo metabolism? Might they not arise from the interaction of flushed drugs and drug precursors with human sewage? If so the presence of something like a meth lab in a town could heavily bias the results.

One might also note that meth (and cocaine) are actually schedule II drugs and can be legally prescribed for certain disorders (narcolepsy for meth and I think certain kinds of eye surgery for coke). Still I suspect that such legal use is rare enough and at such low levels that it's probably not significant.

TruePathJuly 25, 2009 4:34 AM


>It has been estimated that in each glass of Thames water in Essex is the by products of having been through seven individuals bladders in succession.

I suspect if you do the math all water we drink contains molecules that have passed through hundreds or thousands of people. There are a fuckton of H20 molecules in a glass of water.

mooJuly 26, 2009 10:10 AM


This is only tangentially related, but your comment made me think of it.

Most U.S. currency is contaminated with trace amounts of cocaine, which can't rub off or anything.

But, even those trace amounts can trigger drug sensors. And I recall reading an entertaining article (written in the late 80's) about money seizures at airports--people trying to go through security to get on their flight with some cash (maybe a few hundred or thousand dollars) would get caught by the drug sensors and they would seize the cash, then they would battle the system for years without success, trying to get it back. This was happening to totally innocent people, not drug mules (the article was following the case of a farmer who would go to a sales show with cash to buy equipment, and they seized like $7000 from him which he was never able to recover). So basically *any* cash that they detected in the airport, could set off their drug sensors and cause the security apparatus to descend on you and treat you like a hardened criminal.

This is not the article I was thinking of, but its in a similar spirit:

EponymousJuly 27, 2009 9:20 AM

Sure it could turn out to be accurate data...once they tare all the false positives from the metabolism of other compounds that look like meth use such as PEA HCL, ephedrine, pseudoephedrine, tyramine, etc...

Peter E RetepJuly 29, 2009 5:30 PM

Now passed-on Mayor Bradley of L.A. wanted to design his own Golden Parachute
as chief of the to-be-created California State Water Police,
to be created on the lines of the bureaucracy of the Ottoman Empire,
including arming the otherwise unemployables with water keys
and authority to shut down water to businesses on the spot,
unless they paid fines on the spot,
but he was stopped by Joan Milke Flores.
Now, if he had only known of this possibility - - -

daveAugust 17, 2009 11:02 AM

not sure why they think this works. aren't drugs often disposed of in wastewater systems? the "positives" could just as easily be from legitimate sources that are eliminating either expired or confiscated product.

DaveAugust 18, 2009 12:09 PM

Actually, this is old news. The DEA used this method in Portland way back in the 70's to narrow down the locations of drug manufacturing sites. They regularly tested the sewers all around town in this effort and were somewhat successful in pinpointing the locations of a good number of the "cooks" this way. I know, they were after me too...

PatrickNovember 9, 2009 5:03 PM

This has also been done in Oslo, the capital city of Norway.

According to the newspaper VG in May 2007, the "NIVA" (Norwegian Institute for Water Research) had tested the city's sewage, and calculated a daily use of 8k doses cocaine in the city.

The following year, 2008, the same paper tells about the institute sampling sewage water again fore cocaine, and hoping to develop tests for amphetamine, heroine, cannabis and, believe it or not, alcohol.

Just 5 days ago, the topic is in the news again. Researchers have now found that there is more (ab)use of cocaine in the western part of the city than in the eastern. The research is also mentioned in the NRK (Norwegian National Broadcast) TV program "Schrödingers katt" (Schrödinger's Cat):

Leave a comment

Allowed HTML: <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre>

Photo of Bruce Schneier by Per Ervland.

Schneier on Security is a personal website. Opinions expressed are not necessarily those of Co3 Systems, Inc..