Modern laboratories can detect substances in water at extraordinarily low concentrations.
A result may be reported as:
0.5 mg/L
0.5 µg/L
or
0.5 ng/L
At first glance, those numbers look almost identical.
Chemically, they are not.
Each step represents a difference of one thousand times.
And when drinking-water reports begin discussing micrograms or nanograms per litre, it is easy to make one of two mistakes:
“The number is tiny, so it cannot matter.”
or
“The laboratory detected it, so it must be dangerous.”
Neither conclusion is scientifically sound.
The concentration has to be interpreted together with the identity of the substance, its toxicology, the duration of exposure, applicable regulatory or health-based values and the quality of the analytical measurement.
A laboratory result tells us what was measured.
Risk assessment tells us what that measurement means.
What Is a Trace Contaminant?
There is no single universal concentration at which every substance suddenly becomes a “trace contaminant”.
The term is generally used for substances occurring at very low concentrations. Germany’s Federal Environment Agency describes trace substances, or micropollutants, as mostly anthropogenic chemicals occurring in waters at very low concentrations. Examples include residues from pharmaceuticals, pesticides, biocides and industrial or household chemicals.
In technical water literature, these concentrations are frequently discussed in the microgram-per-litre and nanogram-per-litre range.
But the word trace describes the amount.
It does not automatically describe the risk.
A substance can occur at a very low concentration and still be relevant because it is biologically potent.
Another substance may be present at a much higher concentration without presenting the same toxicological concern.
Small concentration does not mean irrelevant.
But:
detectable concentration does not automatically mean dangerous either.
mg/L, µg/L and ng/L: The Scale Explained
The easiest way to understand trace concentrations is to start with one litre of water.
Milligram per litre — mg/L
One milligram is:
0.001 gram
So:
1 mg/L = 1 milligram of substance in 1 litre of water
Microgram per litre — µg/L
One microgram is one thousandth of a milligram:
1 µg/L = 0.001 mg/L
or:
0.000001 gram per litre
Nanogram per litre — ng/L
One nanogram is another thousand times smaller:
1 ng/L = 0.001 µg/L
or:
0.000000001 gram per litre
The complete relationship is:
| Concentration | Equivalent |
|---|---|
| 1 mg/L | 1,000 µg/L |
| 1 mg/L | 1,000,000 ng/L |
| 1 µg/L | 1,000 ng/L |
| 0.1 µg/L | 100 ng/L |
| 0.01 µg/L | 10 ng/L |
The decimal point therefore matters enormously.
1 mg/L and 1 ng/L differ by a factor of one million.
Why Water Reports Use Different Units
Not every drinking-water substance occurs at the same concentration.
Major minerals can naturally occur at concentrations of tens or hundreds of milligrams per litre.
Trace organic chemicals may be measured in micrograms or nanograms per litre.
The EU Drinking Water Directive illustrates the scale very clearly.
Its chemical parametric values include, for example:
| Parameter | EU parametric value |
|---|---|
| Boron | 1.5 mg/L |
| Arsenic | 10 µg/L |
| Benzene | 1.0 µg/L |
| Sum of PFAS | 0.10 µg/L = 100 ng/L |
| Benzo(a)pyrene | 0.010 µg/L = 10 ng/L |
These are all drinking-water parameters.
But their permitted concentrations differ by orders of magnitude.
That tells us something fundamental:
The unit itself does not determine whether a concentration is acceptable.
The substance does.
1 µg/L of One Chemical Is Not Equivalent to 1 µg/L of Another
Suppose a laboratory detects two substances at exactly:
1 µg/L
Chemically, the measured mass concentration is identical.
Toxicologically, the meaning may be completely different.
One substance may have a health-based value far above that concentration.
Another may have a regulatory value close to it.
A third may require evaluation using a precautionary value because toxicological information is still incomplete.
This is why water-quality assessment is substance-specific.
WHO derives drinking-water guideline values using toxicological evidence, exposure assumptions and safety factors. For many threshold chemicals, the assessment begins with a tolerable daily intake — an estimate of how much of a substance can be ingested over a lifetime without appreciable health risk, including a margin of safety.
So:
concentration without substance identity tells us very little about risk.
Dose Matters, Not Just Concentration
A concentration tells us how much substance exists in a particular volume of water.
Human exposure depends partly on how much of that water is consumed.
For example, purely as an arithmetic illustration:
If water contains:
0.1 µg/L
and a person drinks:
2 L/day
the amount ingested from that water would be:
0.2 µg/day
That calculation alone still does not tell us whether the exposure is safe or unsafe.
For that, we need toxicological information.
WHO uses default assumptions such as 2 litres of drinking water per day and 60 kg body weight for adults when deriving many drinking-water guideline values. Different assumptions can be used where children or infants represent the more sensitive population.
Health relevance therefore depends on more than the number printed after “µg/L”.
It can involve:
concentration × water consumption × duration of exposure × toxicity
plus exposure from other sources where relevant.
Exposure Duration Changes the Question
A concentration measured once does not necessarily represent the same health question as a concentration present every day for decades.
Some chemical guideline values are designed primarily to protect against long-term exposure.
WHO explains that its guideline values generally represent concentrations that do not create significant health risk over a lifetime of consumption.
That is why a small, temporary exceedance of a chronic guideline value does not automatically mean that an acute poisoning event is occurring.
WHO specifically notes that small exceedances of many chemical guideline values for short periods would not normally constitute a health emergency.
This does not mean exceedances should be ignored.
It means they should be interpreted according to the toxicology of the specific substance and the exposure scenario.
Detection Is Not the Same as Risk
Modern analytical technology has changed water science dramatically.
Substances that could not realistically be measured decades ago can now be detected at extremely low concentrations.
Germany’s Federal Environment Agency notes that improved analytical methods are one reason trace substances are increasingly detected in aquatic environments.
This creates an important communication problem.
A headline may say:
“Chemical X detected in water.”
But scientifically, that statement is incomplete.
We still need to know:
At what concentration?
With what analytical certainty?
Compared with which reference value?
What is known about the substance’s toxicity?
Was it a single measurement or a recurring concentration?
Detection answers one question:
Was the substance measurable?
It does not automatically answer:
Does this concentration represent a health risk?
What Is a Detection Limit?
Every analytical method has limitations.
An instrument cannot simply measure infinitely small amounts with perfect certainty.
The limit of detection describes a concentration at which the method can establish that the substance is present, although the amount may not yet be measurable accurately enough to report as a reliable numerical concentration.
The German Environment Agency defines the detection limit accordingly: it is the concentration from which the presence of a substance can be demonstrated, but at which an exact numerical value cannot yet be stated with sufficient accuracy.
Below that level, the analytical signal becomes too difficult to distinguish reliably from background noise.
Detection Limit vs. Quantification Limit
There is another important term:
LOQ — Limit of Quantification.
The LOQ is generally higher than the detection limit.
At this concentration, the laboratory can not only say:
“We detected something.”
It can quantify the substance with an acceptable level of accuracy and precision.
This distinction matters when reading laboratory reports.
You may encounter results such as:
< LOQ
This generally means the substance could not be reliably quantified above the method’s defined quantification threshold.
It does not necessarily mean:
absolute zero molecules are present.
It means:
the analytical method cannot reliably report a concentration above that specified threshold.
EU Drinking-Water Analysis Has Performance Requirements
The EU Drinking Water Directive does not simply establish chemical limits.
It also specifies requirements for the analytical methods used to demonstrate compliance.
For many regulated chemical parameters, the analytical method must be capable of measuring the relevant parametric value with a limit of quantification no greater than 30% of that parametric value, alongside requirements for measurement uncertainty. Methods must also be validated according to recognised quality standards.
That is important.
A legal limit has little practical value if the laboratory method cannot measure sufficiently below it.
The regulation therefore connects:
the regulatory number
with
the analytical capability required to evaluate it.
PFAS Show How Small Modern Measurements Have Become
PFAS provide a useful real-world example.
Under the EU Drinking Water Directive, the parametric value for the Sum of PFAS is:
0.10 µg/L
which is:
100 ng/L.
The European Commission’s technical monitoring guidance specifies that the corresponding quantification limit should be no greater than:
30 ng/L
for the Sum of PFAS parameter.
Thirty nanograms per litre is:
0.00003 milligrams per litre.
That is the scale at which modern drinking-water regulation and analytical chemistry now operate.
Some Monitoring Goes Down to 1 ng/L
EU monitoring can reach even smaller concentrations.
The EU drinking-water watch list includes 17-beta-estradiol, for which the Commission established a guidance value of:
1 ng/L
and a required analytical limit of quantification of 1 ng/L or lower for watch-list monitoring.
This is a watch-list guidance value rather than one of the standard parametric limits in Annex I of the Drinking Water Directive.
But it illustrates how sensitive modern analytical methods have become.
Nanograms per litre are no longer theoretical laboratory numbers.
They are part of real water monitoring.
Better Detection Can Make It Look Like Pollution Is Increasing
There is an important statistical trap here.
Imagine a substance has existed in a river at:
20 ng/L
for many years.
An older laboratory method could only detect concentrations above:
100 ng/L.
Every historical sample might therefore have been reported as:
not detected.
A new method has a quantification limit of:
5 ng/L.
Suddenly the substance appears in nearly every sample.
Did contamination suddenly increase?
Not necessarily.
Analytical visibility increased.
This is why trends in trace contaminants should ideally be interpreted together with changes in analytical methods and reporting limits.
Better measurement can reveal contamination that was previously invisible to the instruments.
It does not necessarily mean the substance was previously absent.
“Not Detected” Does Not Mean Absolute Zero
This deserves its own distinction.
When a laboratory reports:
not detected
or
below detection limit
it does not prove that absolutely no molecules of that substance exist in the water.
It means the concentration was below what that particular validated method could reliably detect under those analytical conditions.
Likewise:
<0.01 µg/L
is more informative than simply writing:
zero.
Scientific measurement always has a resolution.
The reporting limit tells you something about that resolution.
Why Legal Limits Differ So Much
Why can one parameter have a value in mg/L while another is regulated in ng/L?
Because different substances have different properties.
Regulatory and health-based values can consider factors such as:
- toxicological potency
- target organs or biological effects
- chronic versus acute exposure
- carcinogenicity
- sensitive populations
- exposure from food and other sources
- analytical capability
- achievable treatment
- precautionary policy
WHO’s current drinking-water guidance explains that guideline values are developed from health evidence but may also need to account for practical monitoring and treatment considerations.
A lower number therefore does not mean regulators simply “found more contamination”.
It often means the substance is evaluated according to a different toxicological and regulatory framework.
A Limit Is Not Necessarily a Biological Cliff
Another common misunderstanding is to imagine a regulatory value as a line where:
0.99 = completely safe
and
1.01 = immediately dangerous.
Risk assessment is generally not that simple.
Many regulatory limits include conservative assumptions and safety margins.
WHO drinking-water guideline values are intended to provide a high level of protection over long-term consumption.
That does not mean exceeding a limit is unimportant.
Legal requirements must be respected and exceedances require assessment and action.
But a slight exceedance of a chronic value is not automatically equivalent to an immediate health effect.
Regulatory non-compliance and acute toxicity are not synonymous.
Germany Uses Precautionary Values for Substances Without Full Toxicological Data
Trace-contaminant science creates another challenge:
What happens when laboratories can detect a substance before toxicology has produced a complete health-based limit?
Germany uses the concept of the Gesundheitlicher Orientierungswert, or GOW — a health-related orientation value.
The Federal Environment Agency developed this framework for substances that are not yet fully toxicologically assessed.
A general precautionary GOW of 0.1 µg/L is used in many cases, while lower or higher orientation values can be derived depending on available toxicological information.
Importantly, the UBA states that exceeding a GOW does not automatically mean that a health effect or danger exists, because the values are deliberately precautionary.
That is a perfect example of the distinction between:
measurement
precaution
and
demonstrated health risk.
So Is 0.1 µg/L Safe?
Not as a universal rule.
This would be exactly the wrong way to interpret the GOW concept.
For some incompletely assessed anthropogenic substances, 0.1 µg/L can serve as a precautionary orientation level in Germany.
But other substances have their own legally established or toxicologically derived values.
Some are higher.
Some are lower.
The EU drinking-water value for benzo(a)pyrene, for example, is:
0.010 µg/L
or:
10 ng/L.
The appropriate benchmark must therefore always be substance-specific.
There is no universal concentration below which every chemical is automatically harmless.
“Present” and “Relevant” Are Different Questions
Imagine three analytical findings:
Substance A
5 mg/L
Substance B
5 µg/L
Substance C
5 ng/L
The temptation is to rank them:
A = most concerning
B = medium
C = least concerning
That ranking could be completely wrong.
Substance A might be an ordinary mineral at an unremarkable concentration.
Substance B might be a regulated contaminant below its permitted value.
Substance C could be a highly potent substance requiring much closer evaluation.
Mass concentration does not equal toxicological importance.
The correct sequence is:
What substance? → What concentration? → What reference value? → What exposure? → What does the evidence say?
Trace Contaminants Are Also an Environmental Question
Human drinking-water safety is not the only reason low concentrations can matter.
Aquatic organisms are continuously exposed to chemicals in rivers and lakes.
Some substances can affect organisms at concentrations that may be very low.
Germany’s Federal Trace Substance Centre therefore assesses certain chemicals as “relevant trace substances” when they can adversely affect aquatic ecosystems at very low concentrations and/or negatively affect drinking-water production.
This creates another important distinction:
environmental relevance is not automatically identical to drinking-water health relevance.
A concentration of concern for an aquatic organism is not necessarily the same concentration used for assessing human drinking-water exposure.
Different endpoints require different reference values.
Why Source-Water Concentrations and Tap-Water Concentrations Must Not Be Confused
Headlines also sometimes compare measurements from rivers or groundwater directly with drinking-water limits.
That can be misleading.
A water utility may take raw water from:
groundwater,
a river,
a reservoir,
or bank filtrate.
That water may then undergo several treatment stages before it becomes drinking water.
A concentration measured in the environment therefore does not automatically equal the concentration at the consumer’s tap.
The measurement location matters.
Always ask:
Was the sample taken from surface water?
Raw water?
After treatment?
The distribution network?
A household tap?
Without that information, the concentration lacks context.
One Sample Is Also Not Always the Whole Picture
Concentrations can change over time.
Rainfall can affect rivers.
Agricultural applications can create seasonal patterns.
Groundwater concentrations may change slowly.
Building plumbing can affect certain metals depending on stagnation time.
Industrial or wastewater inputs can fluctuate.
One laboratory result is therefore a measurement of:
a specific sample, collected at a specific place, at a specific time.
It is valuable information.
But it should not automatically be assumed to describe every litre of water at every future moment.
Where a contaminant is important, repeated or strategically designed sampling may provide a better picture.
Why Extremely Small Numbers Still Matter for Water Treatment
From a filtration perspective, the difference between mg/L, µg/L and ng/L is important for another reason.
A treatment technology does not only need to interact with the target substance.
It needs to reduce it sufficiently from the starting concentration to the required target concentration.
For example, reducing a contaminant:
from 10 µg/L to 1 µg/L
is a different analytical objective from reducing it:
from 100 ng/L to 10 ng/L.
In both examples the reduction is 90%.
But confirming that performance requires analytical methods capable of reliably measuring the corresponding concentration range.
This is why filtration testing and laboratory analysis belong together.
A claim is only as useful as the measurement used to verify it.
Percentage Removal Can Also Be Misleading Without Concentrations
Consider a filter advertised as providing:
90% reduction.
That sounds impressive.
But what does it actually mean?
If the inlet concentration is:
100 µg/L
a 90% reduction would leave:
10 µg/L.
If the inlet concentration is:
1 µg/L
the same percentage would leave:
0.1 µg/L.
The percentage is identical.
The resulting concentration is not.
So meaningful treatment data should ideally answer both:
How much was removed?
and
What concentration remained?
This is particularly important when a regulatory or health-based target has to be met.
More Sensitive Analysis Is a Good Thing
It can sometimes feel unsettling that laboratories are finding more and more substances in water.
But greater analytical sensitivity is not inherently bad news.
It allows scientists and regulators to:
identify pollution earlier,
track environmental pathways,
evaluate treatment,
recognise long-term trends,
and investigate emerging substances before concentrations become larger.
The Federal Environment Agency explicitly links the increasing detection of trace pollutants partly to improvements in analytical methods.
Being able to measure something is the first step toward understanding it.
But Measurement Needs Interpretation
The opposite mistake is to treat every detected molecule as automatically dangerous.
If analytical instruments become one thousand times more sensitive, they will inevitably identify substances that previously went unreported.
Science must therefore become better at communicating what those results mean.
A useful laboratory result should ideally be interpreted using:
the substance name
the measured concentration
the analytical reporting limit
the applicable legal value, where one exists
a recognised health-based or precautionary value where appropriate
and
the sampling context.
Without those pieces, a number can create more confusion than understanding.
A Better Way to Read a Water Analysis
When you see a trace contaminant on a laboratory report, do not stop at the substance name.
Ask:
1. What is the unit?
mg/L?
µg/L?
ng/L?
2. What was actually measured?
For example:
0.05 µg/L
or:
50 ng/L
Those are the same concentration.
3. What is the analytical limit?
Was the value quantified reliably?
Is it reported as below the LOQ?
4. Is there a legal parametric value?
If yes, compare the measurement with the correct regulatory benchmark.
5. If there is no legal value, is there a recognised health-based or precautionary value?
In Germany, the UBA’s GOW system is one example for incompletely assessed substances.
6. Where was the sample taken?
River?
Raw water?
Waterworks?
Household tap?
7. Is this one sample or part of a trend?
That is when a number starts becoming useful information.
Tiny Numbers Need More Context, Not More Fear
Trace analysis is one of the most powerful tools available in modern water science.
It allows us to observe substances at concentrations far below what the human eye, nose or taste could ever detect.
But sensitivity creates responsibility.
A laboratory can tell us that a substance is there.
That does not automatically tell us whether the concentration is toxicologically relevant.
The correct interpretation combines:
chemistry
analytical science
toxicology
exposure
and
regulation.
At Klar2O, this distinction is fundamental to understanding modern water treatment.
The goal should not be to become alarmed whenever an instrument detects a molecule.
Nor should extremely low concentrations simply be dismissed.
The better question is:
What was measured, at what concentration, compared with what scientifically appropriate benchmark?
Because in trace-contaminant science, the number matters.
But the context around the number matters even more.
Sources
World Health Organization, Guidelines for Drinking-water Quality: Fourth Edition Incorporating the First, Second and Third Addenda, June 2026.
World Health Organization, Chemical aspects of drinking-water quality, including guideline-value derivation, tolerable daily intake and exposure assumptions.
European Parliament and Council, Directive (EU) 2020/2184 on the quality of water intended for human consumption.
European Commission, Technical Guidelines Regarding Methods of Analysis for Monitoring PFAS in Water Intended for Human Consumption, 2024.
European Commission, Drinking-Water Watch List — Implementing Decision (EU) 2022/679.
German Environment Agency, Health-Related Orientation Value (GOW).
German Environment Agency, Drinking-Water Guideline Values, updated January 2026.
German Environment Agency, Trace Substances in Waters.
German Environment Agency, Federal Trace Substance Centre.
German Environment Agency, Detection Limit.