Every year a national monitoring programme publishes a table showing that pesticide residues were found on some large fraction of the fruit and vegetables tested, and every year that number is reported as though it were a measure of risk. It is not. It is a measure of how good the instruments have become.
The distance between “a residue was detected” and “a residue was present at a concentration that matters” is enormous, and almost nothing in the way these results reach the public helps a reader cross it. A modern residue laboratory can find a few hundred different compounds in a single extract, at concentrations far below any level that regulators consider relevant, in a sample that would have been reported as clean twenty years ago using the same regulatory framework.
Understanding what these numbers mean requires following the sample: how it is chopped and extracted, how the instrument decides that a particular molecule is present, what the legal limit it is compared against was actually derived from, and what a monitoring programme is sampling for in the first place. None of these steps is intuitive, and each one changes the meaning of the final figure.
Key takeaways
- Multiresidue methods screen one extract against hundreds of compounds at once, which is why detection rates rise as instruments improve.
- The maximum residue limit is a marker of correct agricultural practice, not a toxicological safety threshold.
- Exceeding an MRL means a crop was probably treated incorrectly; it rarely means the food is unsafe to eat.
- Washing removes surface residues but does nothing to compounds absorbed into plant tissue.
- Organic production permits several pesticidal substances, so organic samples can and do return detections.
How Multiresidue Screening Methods Work
A residue laboratory does not test for one pesticide at a time. It runs a single prepared extract through one or two instruments and asks, in effect, whether any member of a long predefined list is present. That list commonly runs to several hundred active substances and their relevant breakdown products.
The instruments are chromatographs coupled to mass spectrometers. Chromatography separates the components of the extract in time: the mixture is pushed through a column, compounds interact with the column material to different degrees, and each emerges at a characteristic moment called its retention time. Gas chromatography handles volatile and thermally stable compounds, the older organochlorine and organophosphate chemistries among them. Liquid chromatography handles everything too polar or too fragile to survive being vaporised, which is most of what has been registered in recent decades.
The mass spectrometer then identifies what emerges. In the configuration used for routine screening, the instrument selects a parent ion of a specific mass, smashes it in a collision cell, and watches for particular fragment ions. Each pesticide is monitored through at least two of these parent-to-fragment transitions. A confirmed identification requires the right retention time, the presence of both transitions, and the correct intensity ratio between them. That combination is what makes a result specific rather than merely suggestive.
The consequence worth holding on to is that the analyst never decides to look for a particular compound based on suspicion. The method looks for everything on the list in every sample, and anything outside the list is invisible no matter how much of it is present. A newly registered substance, or a compound that behaves badly in the chosen extraction, will not appear in the results simply because it was never asked about.
The QuEChERS Sample Preparation Approach
Before any instrument sees the sample, the pesticides have to be moved out of the food and into a solvent, and the food itself has to be left behind. This is the step that historically consumed most of a laboratory’s time and most of its solvent budget, and its replacement is the reason residue testing became cheap enough to do at scale.
The approach now used almost universally goes by an acronym built from the words quick, easy, cheap, effective, rugged and safe. The principle is deliberately crude. A homogenised portion of produce is shaken with acetonitrile. Salts are added, which forces the water and the acetonitrile to separate into two layers and drives the pesticides preferentially into the organic layer. A buffering salt is usually included, because several common pesticides degrade rapidly if the extract drifts too acidic or too alkaline.
An aliquot of that organic layer is then cleaned up by shaking it with small quantities of sorbent powders rather than passing it through a column. A primary secondary amine sorbent removes organic acids and sugars. A bonded carbon sorbent strips out fatty material. Graphitised carbon removes pigment, which matters enormously for spinach, kale and other intensely green matrices that would otherwise foul the instrument.
Each cleanup choice costs something. Graphitised carbon is excellent at removing chlorophyll, and it also retains flat, planar pesticide molecules, so laboratories that use it must either accept reduced recovery for that group or compensate for it deliberately. This is a general truth of sample preparation: every material that removes interference removes some analytes too, and the method is a negotiated compromise rather than a clean separation.
Maximum Residue Limits and Their Basis
Here is the most widely misunderstood object in the whole field. The maximum residue limit is not a safety threshold. It is a trading standard.
An MRL is set by conducting supervised field trials in which a crop is treated with a pesticide exactly according to its approved instructions: the approved dose, the approved number of applications, and the approved interval between the last spray and harvest. Residues are then measured in the harvested crop, and the limit is set near the upper end of what those correctly conducted trials produce. The limit therefore answers a single question: was this crop treated the way the label says it should be treated?
The health-based values are separate and are established quite differently. Toxicological studies establish a dose at which no adverse effect is observed, and safety factors are applied to that figure to allow for differences between species and between individuals. The result is an acceptable daily intake for lifetime exposure, and for compounds capable of causing acute effects, an acute reference dose for a single meal. These are the numbers that describe risk.
| Term | What it is derived from | What it tells you | What it does not tell you |
|---|---|---|---|
| Limit of detection | Instrument and method sensitivity | The compound is present | Anything about quantity or risk |
| Limit of quantification | Lowest reliably measurable level | A number can be reported | Whether that number matters |
| Maximum residue limit | Supervised field trials at approved use | Whether the crop was treated correctly | Whether the food is unsafe |
| Acceptable daily intake | Long-term toxicology with safety factors | The lifetime exposure judged tolerable | Anything about a single meal |
| Acute reference dose | Short-term toxicology with safety factors | The single-meal exposure judged tolerable | Anything about chronic exposure |
Because MRLs are anchored to agricultural practice rather than toxicology, they usually sit far below the intake that would raise a health concern. Exceedances are consequently treated as compliance failures first and health questions second. Regulators do assess whether a specific exceedance approaches the acute reference dose, and occasionally it does, which is precisely when a product recall follows rather than a warning letter.
Detection Versus Exceedance in Reports
Detection rates and exceedance rates move in opposite directions over time, and conflating them produces most of the confusion in public reporting.
Detection rates rise as instruments improve. A mass spectrometer purchased today will find residues in samples that an instrument from a decade ago would have reported as free of residues, and the food has not changed. When a monitoring report states that residues were found in the majority of samples, that sentence is largely a statement about the analytical method, not about agriculture.
Exceedance rates behave differently. They reflect whether growers are following label instructions, whether the correct pesticide was used for that crop, and whether the interval before harvest was respected. They tend to be low in established regulatory systems and to cluster in particular crop and origin combinations rather than spreading evenly.
A third category causes disproportionate trouble: the technical exceedance. Many pesticides have no approved use on a given crop in a given jurisdiction, and where no use is approved, the limit defaults to the lowest level that can reliably be measured. A trace arriving through spray drift from a neighbouring field, or through a shared piece of harvesting equipment, then produces a formal exceedance at a concentration that is toxicologically trivial. These cases are legally real and biologically meaningless, and they are counted in the same column as everything else.
The useful habit when reading any residue figure is to ask three questions in order. Was anything detected at all? If so, at what fraction of the relevant limit? And is that limit a toxicological one or a good-practice one? Almost every alarming headline collapses under the second question.
Washing, Peeling and Residue Reduction
Whether household preparation removes a residue depends entirely on where the residue is, and that depends on how the pesticide was designed to work.
Contact pesticides stay on the outside of the plant. They are meant to sit on the surface and affect insects or fungi that touch them, and they are consequently the ones most amenable to physical removal. Rinsing under running water, with mechanical rubbing, removes an appreciable share of surface residues. Soaking is less effective than rubbing, because the limiting factor is usually adhesion rather than solubility. Various additives have been promoted for wash water; a dilute alkaline solution appears to help with some compounds by accelerating their breakdown, but the effect is compound-specific and modest compared with simply rubbing the surface.
Systemic pesticides are the problem. These are absorbed by the plant and distributed through its tissues, which is the entire point of their design, since it protects new growth and parts the spray never touched. No amount of washing reaches them. Peeling removes them only to the extent that they concentrated in the skin, which for many systemic compounds they did not.
Peeling reliably removes more residue than washing, and it also removes fibre and a share of the micronutrients concentrated near the skin. For most people the nutritional loss outweighs a residue exposure that was already far below any threshold of concern. Cooking degrades heat-sensitive compounds and can concentrate stable ones as water is driven off, so its net effect varies by compound and by method.
Organic Produce and Permitted Substances
Organic certification does not mean an absence of pesticides. It means restriction to an approved list, which is dominated by substances of natural origin or of long-established use, together with a prohibition on most synthetic compounds.
Several permitted substances are genuinely pesticidal and genuinely detectable. Copper compounds are widely used as fungicides in organic viticulture and horticulture and accumulate in soil. Sulphur is used against mildew. Pyrethrins extracted from chrysanthemum are potent insecticides. Spinosad, produced by a soil bacterium, is used against a range of insect pests. Where a laboratory includes these in its screening list, organic samples return detections at rates that surprise people who assumed organic meant untreated.
Organic samples also pick up residues they were never treated with. Spray drift from adjacent conventional fields, shared harvesting and packing equipment, storage in facilities that previously held treated produce, and persistent compounds still present in soil from historical use all produce low-level detections. Certification schemes generally handle these through documentation and investigation rather than through a numerical threshold, since the analytical result alone cannot distinguish drift from deliberate application.
The important analytical point is that a residue result carries no information about intent or provenance. The mass spectrometer reports a molecule at a concentration. Whether that molecule arrived by approved organic application, by drift, by contamination in the supply chain, or by fraud is a question for auditors, records and sometimes isotope work, not for the screening method.
Interpreting Annual Monitoring Programme Data
Monitoring programmes generally run two quite different sampling streams, and reports that merge them are difficult to interpret.
Randomised surveillance sampling aims to estimate what is actually on the market. Samples are drawn according to a plan weighted by consumption, so commonly eaten items appear more often, and the resulting figures can reasonably be read as representative. Enforcement or targeted sampling does the opposite on purpose. It concentrates on crop and origin combinations with a history of problems, on consignments flagged at import, and on suppliers under investigation. Exceedance rates in targeted sampling are higher by design, and quoting them as though they described the general food supply is simply an error.
Programmes also increasingly report a dietary exposure calculation alongside the raw detection figures, combining measured residue concentrations with consumption data to estimate what an actual eater would ingest. This is the part of the report that speaks to risk, and it is almost always the part that goes unquoted, because its conclusions are undramatic.
Two further limitations deserve mention. Screening lists differ between programmes and between years, so trends in detection rates partly reflect changes in what was looked for. And the treatment of multiple residues in a single sample remains an area of active methodological work; finding several compounds together is common, and assessing their combined effect requires assumptions about whether they act on the same biological target, which for unrelated chemistries they generally do not.
Frequently asked questions
Does a detected residue mean the food is unsafe?
Almost never on its own. Detection means only that the instrument found the compound above its limit of detection, which for modern mass spectrometry sits far below the concentrations regulators consider relevant. The meaningful question is what fraction of the maximum residue limit the measured concentration represents, and beyond that, what fraction of the acute reference dose a realistic portion would deliver. Most detections in routine monitoring sit at a small fraction of the legal limit, which itself usually sits well below the health-based value.
Why do detection rates keep rising if regulation is working?
Because sensitivity keeps improving. Each generation of chromatography and mass spectrometry lowers the concentration at which a compound can be confidently identified, and screening lists expand to include more substances and more breakdown products. A sample that returned no detections a decade ago may return several today with no change in how it was grown. The figure to watch across years is the exceedance rate, which reflects agricultural practice rather than instrument performance, and which has generally been stable or falling in established regulatory systems.
Is it worth buying organic to reduce pesticide exposure?
Measured residue exposure from organic produce is generally lower for synthetic compounds, though not zero, and organic production uses its own pesticidal substances that are detectable when laboratories look for them. Whether that difference matters depends on the starting exposure, which for conventional produce in a well-regulated market is already far below the intakes judged tolerable. There are other reasonable arguments for organic purchasing, concerning farmworker exposure, soil management and biodiversity, that do not depend on the residue exposure of the eater.
Can a laboratory tell where a residue came from?
Not from the residue result alone. A screening method reports which molecules are present and at what concentration; it says nothing about whether a compound was applied deliberately, drifted in from a neighbouring field, or was picked up from shared equipment. Investigating origin requires treatment records, audit trails and sometimes analysis of surrounding soil or of other consignments from the same source. This is why certification bodies treat an unexpected detection as the start of an investigation rather than as a verdict.
What does it mean when a report says a residue was below the limit of quantification?
It means the compound was identified with enough confidence to say it was there, but at a level too low for the method to attach a reliable number to. Laboratories distinguish between the detection limit, where presence can be recognised, and the quantification limit, where a concentration can be measured with acceptable precision. Results between the two are usually reported as traces without a figure. For regulatory purposes these are typically treated as compliant, and for exposure calculations they are handled with explicit assumptions about whether to count them as zero or as sitting at the reporting limit.
The reasonable position for a shopper is unglamorous. Residue testing is one of the more mature and better-controlled areas of food surveillance, the limits being enforced are conservative by construction, and the difference in measured residue exposure between the produce you might buy and the produce you might avoid is small next to the difference between eating vegetables and not eating them. Read the exceedance column rather than the detection column, check which sampling stream produced it, and treat any figure that omits a comparison against a limit as a number without a meaning attached.
This is education, not medical advice. Laboratory results only carry meaning alongside your symptoms, history and examination. Talk to a qualified clinician about your own results before changing anything about your care or supplements.
