I've recently come across this Guardian article which cites a story by Orb Media.

Tests show billions of people globally are drinking water contaminated by plastic particles, with 83% of samples found to be polluted

The main point is

In the first public scientific study of its kind, we found previously unknown plastic contamination in the tap water of cities around the world.

Microscopic plastic fibers are flowing out of taps from New York to New Delhi, according to exclusive research by Orb and a researcher at the University of Minnesota School of Public Health. From the halls of the U.S. Capitol to the shores of Lake Victoria in Uganda, women, children, men, and babies are consuming plastic with every glass of water.

More than 80 percent of the samples we collected on five continents tested positive for the presence of plastic fibers.

(emphasis mine)

Orb Media seems to have conducted/funded the study itself so is this a source to be trusted?

In the Guardian article they also report that

In Paris in 2015, researchers discovered microplastic falling from the air

citing a study from 2015. I don't have access to the full article, so I wonder if that sentence has any ground of truth.

I find that none of these claims particularly surprising or unlikely, but I'm wondering why this story hasn't gotten more coverage.

  • 4
    @T.Sar: Plastic#Toxicity. Microplastics#Biological integration into organisms. "Many additional researchers have found evidence that these fibers had become chemically-associated with metals, polychlorinated biphenyls, and other toxic contaminants while in water. The microplastic-metal complex can then enter humans via consumption. It remains unclear how much of an impact this has directly on the health of humans, but research on this issue continues."
    – DevSolar
    Commented Sep 26, 2017 at 13:28
  • 16
    (ctd.) So the question has merit, and ridiculing plastics as "just a way to structure carbon" is the bad science here. So is gasoline, and we don't want that in our water either.
    – DevSolar
    Commented Sep 26, 2017 at 13:34
  • 19
    There is how much microfibre in 83% of tapwater? 1 part per trillion? 50%? Commented Sep 27, 2017 at 13:35
  • 3
    @GrimmTheOpiner The article in question found that this number ranged from 0 to 57 particles per liter, with a mean value of 4.34 particles per liter. Also it's worth to note that they barely have anything of "micro" in them - they are big particles, ranging from 0.1mm to 5mm in size. They are more comparable to big (visible) pieces of dust than microscopic particles. Still according to their report, 99.7 particles were plastic fibers (like poliester and similar synthetic fabrics).
    – T. Sar
    Commented Sep 27, 2017 at 14:30
  • 1
    @GrimmTheOpiner That's why this article is a lot of noise for nothing. Those particles appear to be too big to be harmful. The article says they exist, yes, but not how exactly they can damage health. I would gladly change my position regarding this issue if someone show me that those particles can actually damage human health, but otherwise I remain skeptical.
    – T. Sar
    Commented Sep 27, 2017 at 15:01

2 Answers 2


The study by Orb Media was subject for articles in German news website Spiegel Online (commercial) and WDR (public service broadcasting). Both articles, quoting staff from the German Federal Environmental Agency, agree on criticism of the study:

  • The study does a "plastic / no plastic" rating, without actually going into how many particles were found.
  • Samples were transported in PE (polyethylene) containers, which could very easily lead to contamination of the samples (e.g. abrasion from the container cap).
  • Samples were evaluated manually by microscope, which is no longer considered "state of the art", and introduces further potential sources for contamination. (Cotton fibres from lab coats, for example.)

Either article agrees, though, that microplastics are much too common and widespread already, that they are a problem that needs to be addressed, and that the subject needs further study.

As for "why hasn't this seen more coverage", the WDR article gives a kind of explanation (translation mine):

{Quoting staff from the German Federal Environmental Agency} "Somehow finding plastic is not a problem. But how do I classify it?" To better determine number and type of the plastic particles, worldwide standardized chemical-analytical procedures should be developed.

Via the Deutsche Institut für Normung (DIN), the German FEA supports corresponding efforts. "If we have those, we can better classify many of today's findings."

So yes, plastics are there, they are a problem, but we don't really know how to test for them and how to classify them, yet.

Sorry for linking German-language articles, but I can judge the merit of the sources much more easily in my native language, and it's a very emotional subject for some, leading to highly-vocal participants in the discussion I would rather not quote.

  • 57
    Science is science, you don't need to be sorry because it is in another language. Have an upvote. Commented Sep 26, 2017 at 15:31
  • 57
    @1006a PE = polyethylene, a plastic (the most common, in fact). Ironic that they would use plastic containers when transporting water to be tested for plastic contamination.
    – BlueBuddy
    Commented Sep 26, 2017 at 15:43
  • 9
    I thought the phrase "tested positive for the presence of plastic fibers." seemed too vague. A binary test that seems uncontrolled for contamination is definitely not the most useful or reliable statistic.
    – JMac
    Commented Sep 26, 2017 at 17:05
  • 11
    "without actually going into how many particles were found." And that's the real killer here. Pretty much anything is allowed in water (and not that harmful for you) as long as it's in very low concentrations. Without knowing whether they found 20 parts per million or 0.02 parts per million, the report is useless. Pretty much 100% of the available water supply is contaminated with something if you don't take concentrations into effect.
    – Mast
    Commented Sep 27, 2017 at 7:42
  • 5
    Reading the article in full, I came across this: "All but two plastic particles were fibers ranging in size from 0.1 mm to 5 mm in length" and "with an overall mean of 4.34 plastic particles per liter of water". Those "particles" are extremely big and extremely few from a contaminant viewpoint - far bigger than your usual home dust. Before assuming that this is a problem, I would like to see evidence that plastic of this size is actually harmful to the body.
    – T. Sar
    Commented Sep 27, 2017 at 12:17


Of course there exist locations in the world where microplastic will be found in the water -- given the standards in some parts of the world, this cannot be doubted -- but overall the claims are most certainly (in the case of Germany provably) wrong, and intentionally misleading.

The article you linked to is not a study but a sensational article which fails to provide any hard facts or sources (although there is a "study" from the same source as well, hard to find as it is).

The articles in Der Spiegel and WDR mentioned in another answer actually doubt the validity of the forementioned study, citing an expert from the Federal Environmental Agency who provides a couple of excellent arguments why the study results are (at least for Europe and in particular Germany) doubtful. The most compelling argument is that 80% of our drinking water comes from deep ground water, which, by passing through sediment, has been filtrated to the point of being naturally sterile.

As a funny anecdote, Der Spiegel cites The Guardian and BILD as sources, both of which are not precisely well-known for high-quality journalism or factual accuracy. There exists the well-known proverb: "BILD talked with the dead" to account for their habit of publishing unbelieveable and completely false stories (followed by tiny dementi on page 6 or 7 in the next issue) regularly.

The actual study done by ORB Media disqualifies itself in the abstract by making a claim that is outright impossible for Germany:

Most these particles [sic] were fibers (99.7%), between 0.1 - 5 mm in length.

The law ("Trinkwasserverordnung" in Germany, but other EU countries have similar laws) requires a threshold of zero on several germs (E. coli among them), with technical filtration if need be. Which effectively means unless you already have exclusively deep ground water filtered appropriately by nature, water suppliers must run their water though a filter none larger than 1.2µ to comply with the law. There are many more requirements (all of which I will not go into detail), among them a maximum cloudiness, which also factually implies filtration on a single-digit micrometer scale.
Some sources for actual ("in use") filter sizes for preparing surface water as drinking water in the 0.1 to 3µm range and 5µ/1µm prefilters followed by nanofiltration: [1], [2]

The supplier regularly needs to provide evidence of compliance, and penalties are harsh, to say the least.

Additionally, the overwhelming majority of households (and public buildings) have additional filters installed in the basement meant to hold back the odd scrap that might come from the water pipe. Those are not meant to hold back bacteria, but have a pore size anywhere from 5µ to 20µ.

Thus, the claim of having found particles 0.1-5mm in size (which is 100 to 5,000µ) in roughly 3/4 of the samples is, disregarding the possibility of having contaminated the samples, beyond "unlikely" but outright ridiculous.

The "Methods" section states, among other things, that samples were in some cases collected by scientific partners and in some cases by non-scientific partners (read as: laymen), as well as Orb staff and volunteers (again: laymen) using plastic bottles which were then sent across the globe by mail. That methodology provides for an excellent opportunity to contaminate the samples.

Evidence for microplastic was obtained by colorizing the residue on a 2.5µ filter with a pigment which binds to organic material. Thus followed prodding residue with a stainless steel spatula, if it didn't break, plastic was assumed. Again, the methodology is somewhat doubtful as hard evidence.

  • 1
    In any case, anything single- or double- digit micrometer renders claims about having found millimeter-sized particles pretty much absurd.
    – Damon
    Commented Sep 28, 2017 at 12:26
  • 1
    They might get rid of the pathogens in other ways in addition to sedimentation and filtration such as adding chlorine or boiling / ozone (the last two are retarded examples probably not used for german groundwater)
    – daniel
    Commented Sep 28, 2017 at 12:33
  • 1
    I'm uh, not sure Turbidity has a quick implication for water particulate size (I'm not sure thats even the right word) since the plastics here could be see through and have no resulting effect on the NTU number.
    – daniel
    Commented Sep 28, 2017 at 13:14
  • 3
    You mention multiple rounds of filtering, up to and inside the building of use. Do those buildings routinely use non-plastic plumbing? In the U.S., PVC plumbing is relatively common, so none of this filtering will prevent plastic exposure over the last several meters to the tap. (I'm not claiming PVC particles are entering the water. I'm not claiming anyone has reported PVC detections. I'm not claiming any relevant toxicity. I'm just saying that no amount of filtering changes the composition of the plumbing.) Commented Sep 28, 2017 at 16:28
  • 3
    @Damon: The filter you mentioned would let through long particles that are significantly less than 0.1µ thin, albeit with much lower probably than a particle that is simply a 0.1µ diameter spheroid. Commented Sep 28, 2017 at 16:50

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .