Whether cannabis passes a pesticide test is decided during cultivation. This part explains how laboratories test for pesticide residues, how to test your own crop before harvest, how to prevent residues with integrated pest management, and why cleaning up a contaminated batch rarely works.
Some background in brief. Drying concentrates residues by about 3–5 times, and extraction can raise them several times more. Smoked or vaporised cannabis carries residues straight to the lungs. Limits differ sharply between the European Union, Canada, California and Thailand. Part 1 covers these points and the compounds that fail most often.
This is Part 2. Part 1: Pesticide Residues on Cannabis: Why They Matter, Limits by Market and the Compounds That Fail Most Often (Part 1 of 2)
4. How Laboratories Test for Pesticides
The lab report that comes with a test result is called a certificate of analysis (CoA). How to read one is explained in How to Read a Cannabis Certificate of Analysis. This section covers only what affects pesticide results.
4.1 Two instrument methods cover all pesticide classes
No single instrument can measure every pesticide. Accredited laboratories therefore combine two methods. A 2018 review by Leghissa, Hildenbrand and Schug in the Journal of Separation Science gives a general overview of analytical methods for cannabis.
- LC-MS/MS (liquid chromatography with tandem mass spectrometry). The sample is separated in a liquid, then each compound is identified by its mass in two stages. This method suits polar compounds and compounds that break down when heated. Examples are organophosphates, carbamates, neonicotinoids and systemic fungicides.
- GC-MS/MS (gas chromatography with tandem mass spectrometry). The sample is separated as a gas, so a compound must survive heating. This method suits non-polar, heat-stable compounds such as pyrethroids and organochlorines.
The standard sample preparation is QuEChERS (“quick, easy, cheap, effective, rugged, safe”). It has three steps: extraction with the solvent acetonitrile, separation with salts, and a cleanup step. The method is checked by its recovery, the share of a known amount of added pesticide that the method finds again. Mean recoveries of 70–120% are the usual accepted range, according to SANTE/11312/2021. That document is the European Commission’s guidance on quality control for pesticide-residue laboratories.
What to do: When you ask a laboratory for a quote, ask whether its pesticide panel uses both LC-MS/MS and GC-MS/MS. A panel that uses only one method will miss part of the pesticide classes.
4.2 False positives and confirmation
A screening result above an action level (the concentration at which a product fails) should be confirmed before anyone acts on it. The confirmation should be a second, independent analysis. Common causes of a false positive are:
- Matrix interference. Terpenes, cannabinoids and chlorophyll from the plant itself can come out of the instrument at the same moment as a pesticide. This happens with methods of low selectivity.
- Laboratory contamination. Pesticides used to control pests in the laboratory building can get into samples.
- Misidentification. In MS/MS, each compound gives a typical ratio between its fragment ions. The ratio in the sample should be within ±30% (relative) of the average from the calibration standards in the same run (SANTE/11312/2021). If it is not, the identification is doubtful.
What to do: If a result shows a compound you never applied, ask for confirmation by a second, independent analysis before you destroy anything. Bring your spray records to that conversation (see section 5).
4.3 Measurement uncertainty near the limit
Every measurement has a margin of error. SANTE/11312/2021 sets a default expanded measurement uncertainty of 50% for enforcement decisions, unless a laboratory shows that its own value is lower. Under that default, a result of 0.02 mg/kg could really be as low as 0.01 mg/kg. So a result has to be above about 0.02 mg/kg before it is clearly over a 0.01 mg/kg limit.
Some systems ignore measurement uncertainty. Under those systems, identical lots can pass in one laboratory and fail in another.
What to do: Ask the laboratory to report its measurement uncertainty with every result close to a limit. Use that figure when you decide whether a lot really failed.
4.4 Testing before harvest as insurance
An internal screen for many residues at once, run 4–6 weeks before harvest, leaves you time to act. You can still change the plan or separate an affected lot. Finding contamination after packaging and distribution costs far more.
- Use laboratories accredited to ISO/IEC 17025, the international standard for testing laboratories.
- Order a panel that covers every compound applied during the crop cycle, plus common environmental contaminants.
- Combine at least 5 sub-samples per lot. Record where in the room or field each one was taken and the growth stage of the plants.
Translator’s note: What does a grower gain from a test before harvest? You find out whether a lot is clean while there is still time to separate it, instead of after it has been sold. Before sending a sample, ask the laboratory to show its accreditation scope and check that pesticide residues in plant material are on it. The forum guide Testing Cannabis Flower Before Export: Thai Laboratories covers Thai laboratories in more detail. To start small, test one lot from your next harvest 4–6 weeks before the cut, and compare the result with your spray records.
5. Prevention: Integrated Pest Management in Brief
The most reliable way to pass pesticide testing is to use as few pesticides as possible. Choosing compounds that “test well” is not a substitute. Integrated pest management (IPM) combines exclusion, monitoring, biological control and targeted sprays. It is covered in its own guide. This section lists only what matters for residues.
5.1 Keep pests out first
- Quarantine and inspect incoming clones before they enter the main room.
- Fit screens over air intakes.
- Keep rooms cool enough to slow spider mite breeding.
- Use sticky traps to monitor pest numbers.
Translator’s note: In Thailand, clones bought from another farm are a common way pests enter. They can also bring hop latent viroid (HLVd) or turn out to be the wrong strain. You also do not know what was sprayed or drenched on the mother plant, and a systemic product can stay in the cuttings. Ask the seller for spray records, and keep new clones apart until you have inspected them.
Translator’s note: Rooms in Thailand often sit at 30–35 °C without air-conditioning, so keeping them cool costs electricity. A power cut or air-conditioner failure lets the temperature climb, and mites can build up quickly. After any cooling failure, check the undersides of leaves more often over the following week.
What to do: Set up a quarantine corner for new plants, and inspect leaf undersides with a magnifier before anything enters the main room.
5.2 Biological control
Biological control means releasing natural enemies of a pest. Match the agent to the pest:
- Phytoseiulus persimilis and Neoseiulus californicus, both predatory mites, target spider mites. See Biological Control of Spider Mites and Thrips.
- Amblyseius swirskii mainly eats thrips and whitefly. It is weak against the two-spotted spider mite (Tetranychus urticae). The forum compares it with P. persimilis in hot rooms in swirskii vs persimilis at 30–32 °C.
- Steinernema feltiae, a beneficial nematode (microscopic worm), targets fungus gnat larvae in the growing medium.
- Bt (Bacillus thuringiensis, a bacterium) works only on caterpillars. See the Bt programme for budworms.
Natural enemies work best when they are introduced early, before pest numbers build up. Check what local suppliers can provide.
Translator’s note: We could not confirm which of these agents are sold in Thailand, or whether each one is registered or permitted for this use. Ask the official distributor in Thailand to show the Thai registration or permit, or ask growers in the Plant Infirmary section what they have been able to buy and how well it worked.
What to do: Choose the agent for the pest you actually see on your sticky traps and leaves. Release it at the first sign of that pest.
5.3 Inputs
- Use only products permitted under your certification (Thai GACP or the rules of your target market) and registered for that use. The forum guide What You May Legally Spray Under Thai GACP explains how to check this. In Thailand, that means products registered with the Department of Agriculture (กรมวิชาการเกษตร), used according to the label.
- Spray only when monitoring shows a need, not on a fixed preventive schedule.
- Follow the pre-harvest interval (PHI) on the registered label. This is the minimum number of days between the last application and harvest.
- Do not apply sulfur to flowering buds. A contact product stays on the surface it hits, and on a bud that surface is the harvest.
- Avoid routine neem on flowers.
What to do: Before each spray, write down what your monitoring showed. If you cannot name the pest and its numbers, do not spray.
5.4 Records
For every application, record:
- date
- product and active ingredient
- concentration
- who applied it
- conditions at the time
- which batch was treated
Contractors should give you the trade name, active ingredient, rate, date and label PHI. GACP inspectors review these records. If a lot fails, the records also let you trace what caused it.
Translator’s note: If you hire a spray crew, or a neighbour sprays your plants, the same record is needed. Outdoor growers next to rice, fruit or vegetable fields can also receive drift from neighbours’ sprays. Note any spraying you see nearby, with the date, so that an unexpected test result can be explained.
What to do: Start a spray log today, using the fields above, and fill it in on the day of each application.
5.5 Growing media
Coco coir, peat and bark mixes can carry residues of their own. Ask the supplier for a pesticide-residue CoA before you buy a batch.
What to do: Keep the CoA of each growing-medium batch with your crop records, so you can rule the medium in or out if a lot fails.
6. Remediation: What It Can and Cannot Do
“Remediation” means processes that reduce residues in material that is already contaminated. Options exist, but they are limited and each works only for certain compounds. Many regulated markets do not accept remediation at all.
6.1 First step: separate the lot
When contamination is found, the most important step is to separate the lot immediately. Keep it away from clean lots, and keep it traceable so you can find the source.
What to do: Label the affected lot, move it to a separate locked area, and record the date and the reason.
6.2 Before harvest
Growing longer. Some compounds break down quickly in living tissue. For those, delaying harvest may bring residues below the limit. This works only if the compound’s breakdown in cannabis tissue is documented. Do not work it out from the PHIs on food-crop labels. Re-test before you harvest.
Removing affected plants. Contamination is sometimes limited to one area. For example, a contractor may have applied an unauthorised product in one section. In that case, destroy those plants and test the remaining lots separately. Part of the crop may then be saved.
What to do: If you suspect one area, sample it as a separate lot so a positive result does not condemn the whole crop.
6.3 After harvest
Washing. Washing with water removes some residues that sit on the surface and dissolve in water. It does little against systemic compounds, because they are already inside the tissue.
Activated carbon. Running crude extract through activated carbon can reduce some residues. It can also remove cannabinoids. Its effect must be validated separately for each contaminant, and regulators rarely accept it as remediation.
Distillation. Distillation removes many other contaminants. Some pesticides, however, evaporate at temperatures close to those of cannabinoids, so they can carry over into the distillate. Distillate must still be tested. It is wrong to assume that extraction removes all pesticides.
Dilution is not remediation. Blending contaminated material with clean material to bring the average below the limit is generally prohibited. It can count as adulteration.
What to do: Never plan to “fix it after harvest”. If a lot fails, test any processed product made from it again before deciding anything.
6.4 Disposal
This applies to material that cannot be brought within limits.
Thailand, controlled-herb material. Cannabis flower is a controlled herb, so rejected flower held under a licence is not ordinary farm waste.
- Licence holders must keep the required records on forms PT 27 and PT 28.
- Missing or incomplete records can lead to licence suspension. Under Section 52 of the Thai Traditional Medicine Protection Act B.E. 2542, one suspension order can last up to 90 days.
- Before you destroy rejected flower, notify the licensing authority as your licence conditions require.
- Record the lot, quantity, reason (attach the failed CoA), method, date and witnesses, so that your stock records match.
- Never sell, give away or blend failed material.
Composting is suitable only for material you are permitted to compost. Keep compost made from contaminated material on your own site, so the contamination does not spread.
Incineration at adequate temperatures destroys most organic pesticide molecules.
What to do: Before destroying anything, fill in the PT 27/PT 28 records, notify the licensing authority, and keep copies of the failed CoA and the destruction record together.
6.5 Why prevention beats remediation
No post-harvest technology reliably brings every pesticide class below action levels. Several costs all point toward clean growing:
- the cost of testing
- the cost of processing
- the yield lost along the way
- the risk that the final product still fails
In regulated markets, a failed test usually means the lot is destroyed, not given a chance at remediation.
What to do: Put your budget into monitoring, biological control and a test before harvest, rather than into post-harvest cleanup.
Conclusion: What to Do on Your Farm
Pesticide residue compliance is decided during cultivation. Drying and extraction concentrate residues, and inhalation raises the exposure concern. Limits differ between the EU food and medicine rules, Health Canada’s list, California’s action levels and Thailand’s mix of controlled-herb rules, Thai GACP and food MRLs.
The compounds that fail are predictable:
- myclobutanil
- bifenazate
- neonicotinoids
- abamectin
- PBO from pyrethrin products
- spinosad applied too close to harvest
Remediation is limited, works only for certain compounds and is often not permitted. Clean cultivation is the only strategy that works commercially. In practice:
- Use only permitted and registered inputs, and follow the label PHI.
- Prevent problems with IPM and biological control, and spray only when monitoring shows a need.
- Keep a record of every application, at batch level.
- Run an internal test at an ISO/IEC 17025 laboratory 4–6 weeks before harvest.
- If a lot fails, separate it, confirm the result, and dispose of it with full records.
If you have a test result or a pest problem to discuss, post it below this topic so other growers can learn from it.
References
- European Commission. (2021). SANTE/11312/2021: Analytical Quality Control and Method Validation Procedures for Pesticide Residues Analysis in Food and Feed.
- Leghissa, A., Hildenbrand, Z. L., & Schug, K. A. (2018). A review of methods for the chemical characterization of cannabis natural products. Journal of Separation Science, 41(1), 398–415.
- Health Canada. Mandatory cannabis testing for pesticide active ingredients – List and limits
- Ministry of Public Health (Thailand). Notification Re: Controlled Herbs (Cannabis) B.E. 2568