Why States Mandate Water Activity Testing for Cannabis Flower
Photo via Unsplash.
In May 2026, Missouri regulators pulled Sinse-brand prepackaged flower off shelves over Aspergillus contamination. Four months later, Juicy J's cherry limeade infused pre-rolls got hit with the same type of recall. And in late February, New York's Office of Cannabis Management yanked 55 product lots after discovering that Keystone State Testing had been reporting passing results on Aspergillus screening for lots that never actually cleared the bar. These aren't stories from the early, chaotic years of state legalization. They happened this year, in two of the more mature regulated markets in the country, which tells you something uncomfortable: mold contamination in commercial cannabis is not a solved problem, and the testing systems built to catch it are still failing in ways that put real people at risk.
Water activity is at the center of how regulators try to prevent this, and it gets confused with moisture content constantly, including by growers who should know better. Moisture content tells you how much water is in your flower by weight. Water activity tells you how much of that water is actually free to do something -- specifically, to feed mold spores and bacteria looking for a place to germinate. A sample can read fine on a kitchen scale and still carry an Aw reading that's a mold incubator waiting to happen. That distinction is why water activity, not moisture content, is what shows up in state testing statutes.
This piece walks through the actual science behind Aw measurement, the ASTM standard that defines the target range, how California, Maine, New York, and Washington handle the thresholds differently, and why -- even with all this regulatory infrastructure in place -- contaminated product keeps reaching consumers anyway.
Water Activity vs. Moisture Content: What Labs Are Actually Measuring
Every cannabis flower sample contains water in two functionally different states. Some of it is bound tightly to plant tissue -- cellulose, sugars, proteins -- and isn't going anywhere without serious energy input. The rest is loosely held, sitting in the free space between cell walls, available to migrate, evaporate, or get used by whatever mold spore happens to land on it. Total moisture content, the number you get from a simple loss-on-drying test or a cheap moisture meter, doesn't distinguish between the two. It just tells you the total water weight as a percentage of the sample. Water activity (Aw) measures something more specific: the vapor pressure of that free water relative to pure water, expressed on a 0 to 1 scale, and it's that free fraction that actually determines whether microorganisms can establish and spread.
This is why two flower samples can read similar moisture content and behave completely differently on a shelf. One, cured slowly in a stable environment, has most of its remaining water bound up and sits at 0.58 Aw -- stable, inert, safe. The other, dried too fast or stored somewhere humid, has more of that water sitting free and tests at 0.72 Aw, which is well into the range where mold spores can germinate and start colonizing the bud. A moisture meter won't catch this difference. Only a proper water activity measurement will.
The widely recognized threshold is 0.65 Aw. Below that line, most mold and bacterial species struggle to find enough available water to establish a colony -- their growth curves flatten out dramatically. Above it, particularly once you cross into the 0.70s, risk climbs fast, and you start seeing visible mold within days under warm storage conditions. This isn't an arbitrary number; it reflects decades of food science research on water activity and spoilage organisms, applied to cannabis flower.
The standard practice for actually measuring this is ASTM D8196, which specifies using a calibrated water activity meter -- a dedicated hygrometer-based instrument, not a capacitance-based moisture probe -- to get a reliable reading. For growers, the practical lesson is blunt: you can hit your target dry weight on the scale and still fail a compliance test, because the scale was never measuring the thing that actually matters. Curing environment control, not just drying time, is what determines whether your flower passes.
The ASTM D8197 Standard: Why 0.55 to 0.65 Is the Target Window
ASTM D8197 sets a narrow safe zone for dry cannabis flower, with water activity needing to stay between 0.55 (risk of breakage) and 0.65 (risk of mold), while readings of 0.68 or above trigger a mandatory retest.
ASTM International's D37 committee, the group responsible for cannabis industry standards, published D8197 -- the Standard Specification for Maintaining Acceptable Water Activity Range for Dry Cannabis Flower -- specifically to give the industry a defined target window instead of a single pass/fail line. The range is 0.55 to 0.65 Aw, and both ends of that range exist for real, practical reasons rather than regulatory caution.
The upper bound, 0.65, is the mold and microorganism ceiling discussed above. Cross it and you're giving bacteria and fungal spores enough available moisture to start multiplying, which is exactly the scenario behind the Aspergillus recalls that keep surfacing in state after state. But the standard doesn't stop at just warning growers not to leave flower too wet. It also sets a floor at 0.55, because flower that's too dry has its own failure mode: trichomes become brittle and shatter off during trimming and handling, the flower itself crumbles rather than breaking apart cleanly, and you lose both bag appeal and a meaningful amount of the cannabinoid and terpene content that was sitting in those trichome heads. Overdried flower isn't a safety problem, but it's a quality and yield problem, and D8197 treats both ends as failures worth preventing.
The standard also builds in a practical escalation path. If a submitted sample tests above 0.68 Aw, it doesn't just fail outright with no recourse -- the standard requires further drying before the batch can be resubmitted for testing. That's a meaningful detail, because it acknowledges that water activity in a cured batch isn't static; it can be brought back down with additional controlled drying time, unlike, say, a failed pesticide screen, which usually means the batch is done.
What makes D8197 worth knowing even if you're not a licensed commercial cultivator is that it's not really a lab-created number divorced from what good growers already do. The 0.55-0.65 window is close to what experienced cultivators have long targeted by feel and taste for shelf-stable, smooth-burning, long-lasting flower. The standard just puts a measurable number on an instinct that good curing has always chased: dry enough to be stable and safe, moist enough to stay intact and pleasant to smoke.
How State Regulations Differ -- and Where They Agree
Photo by National Cancer Institute via Unsplash.
States generally agree on the core numbers but diverge on scope, and the differences matter if you're trying to figure out what actually gets tested where. California's Department of Cannabis Control, under Section 5322/15717 of its regulations (revised January 2026), requires dried flower batches to test at or below 0.65 Aw, with infused edibles held to a separate, higher ceiling of 0.85 Aw -- edibles carry different water activity risk profiles because of their matrix and ingredients. Results have to be reported to two significant figures, which closes off the kind of rounding ambiguity that could otherwise let a borderline sample slide through.
Maine's rules, found in 18-691 C.M.R. chapter 5, track California's numbers almost exactly for both flower and edibles, which gives multi-state operators some consistency to design around. But Maine adds a provision worth flagging specifically: it explicitly prohibits remediation through dilution. A batch that fails can be retested, or it can go through legitimate remediation -- further drying, for instance -- but it cannot be watered down, blended with lower-moisture product, or otherwise diluted to drag the average Aw reading under the threshold. That's a direct response to a workaround regulators clearly anticipated operators might try.
New York's Office of Cannabis Management updated its testing-limits document on February 9, 2026, and the revision is notable for how far it extended water activity testing beyond flat flower. Raw pre-rolls, kief, infused flower, infused pre-rolls, non-solvent concentrates, and infused edibles all now fall under Aw testing requirements. At the same time, New York carved out exemptions for non-edible infused products, tinctures, and beverages, reasoning that those product forms don't carry the same mold-growth risk profile as dried plant material.
Washington takes a slightly different structural approach. Rather than treating the 0.65 flower threshold and 0.85 edibles threshold as standalone pass/fail checkpoints, the state ties them directly into its microbial testing requirements -- water activity functions as a triggering or contextual measurement feeding into the broader mold and bacteria screening process, rather than sitting as an independent gate. The net effect across all four states is the same practical target range, but very different answers to the question of exactly which products on a dispensary shelf were actually required to pass an Aw test before they got there.
Aspergillus and the Limits of Water Activity Alone
Hitting the 0.55-0.65 window reduces mold risk substantially, but it doesn't eliminate it, and this is where a lot of compliance programs quietly fall short of actual safety. Water activity is a measure of growth potential, not a direct test for the presence of any specific pathogen. A sample can sit comfortably inside the acceptable range and still carry a dangerous load of Aspergillus that got established earlier -- during a humid stretch mid-cure, in a storage tote that didn't get cleaned between batches, or from spores already present on the plant material at harvest. That's why California, alongside its Aw threshold, separately enforces strict species-specific Aspergillus limits as part of its microbial panel.
Aspergillus matters more than most contaminants because of who it hurts worst. It's a mold genus, and several species within it -- Aspergillus flavus, fumigatus, niger, and terreus are the ones regulators typically screen for -- can cause aspergillosis, a serious and sometimes fatal lung infection in people with compromised immune systems. For recreational users with healthy lungs, exposure might cause nothing noticeable. For medical cannabis patients undergoing chemotherapy, managing a serious chronic illness, or recovering from organ transplants, inhaling Aspergillus spores from contaminated flower is a genuine, documented health hazard, which is exactly why medical-use states have historically pushed the hardest on mold-species screening specifically.
The 2026 Missouri cases make the gap between Aw compliance and pathogen-free product concrete. On May 1, the state's Division of Cannabis Regulation recalled Sinse-brand prepackaged flower after it came back positive for Aspergillus. That September, Juicy J's cherry limeade infused pre-rolls got recalled on the same basis -- a failed Aspergillus test, not a water activity failure. In both cases, the product presumably moved through whatever water activity checks Missouri requires before reaching the point of sale. The Aspergillus contamination showed up on a separate screen.
This is the core argument for layered testing rather than a single number doing all the work. Water activity tells you whether conditions are favorable for mold broadly. Species-specific screening tells you whether a particular dangerous organism is actually present, regardless of what the Aw reading says. States that only check one of the two are leaving a real gap, and the recalls keep proving it.
When Testing Itself Fails: Lab Fraud and Uneven Enforcement
Photo via Pexels.
None of this testing architecture matters if the labs running it aren't trustworthy, and 2026 has been a rough year for that assumption. A panel at MJ Unpacked had already surfaced a warning sign out of Missouri: analysis of state Sunshine Law data showed that a large majority of all cannabis testing in the state ran through a single licensed lab. That kind of concentration is a problem on its own, but the more telling number was the failure rate. That lab's Aspergillus failure rate sat far below the failure rate detected statewide through other labs and audits. Product isn't that much cleaner when it happens to go through one particular facility -- what's far more likely is that the testing itself was inconsistent, under-rigorous, or incentivized to produce passing results for the clients paying the bills.
New York's case this past February removed any ambiguity about whether that kind of thing actually happens in practice. Keystone State Testing was found to have falsely reported Aspergillus screening results as passing for lots that had not, in fact, met the state's standards. The response was a recall covering 55 separate product lots -- not a theoretical risk, but confirmed contaminated flower that had already been reported clean and, in many cases, already sold.
Put these together and the pattern is clear: water activity and microbial testing mandates only function as intended when the labs performing them are independently audited, financially insulated from the outcome of each test, and genuinely penalized for false passes rather than rewarded for client-pleasing results. A testing requirement on paper is not the same thing as a testing system that works.
For growers, especially smaller operators relying on third-party labs for compliance, the practical implication is uncomfortable but important: a certificate of analysis showing a passing water activity and mold screen is not an absolute guarantee that your product is safe, if the lab generating it is cutting corners or has been compromised. The most reliable safeguard still sits upstream of any lab -- controlling your own drying and curing environment tightly enough that you're not depending on a testing system to catch a problem you could have prevented in the first place.
Hitting the Target: Practical Drying and Curing Moves for Growers
Photo by CRYSTALWEED cannabis via Unsplash.
Getting into the 0.55-0.65 Aw window starts well before a bud ever touches a water activity meter. A slow dry at 60-65°F and 55-60% relative humidity, stretched over a week or more, is the baseline that gets most flower into the right ballpark before curing even begins. Rushing this step with higher heat or lower humidity to save a few days almost always backfires -- you lock moisture unevenly inside denser colas while the outer leaf material dries too fast, which leaves you with a sample that reads fine on average but still has pockets of free water sitting inside the bud, exactly the kind of inconsistency that produces a surprise failed test later.
Curing in airtight jars, with routine burping to exchange air and let humidity vent gradually, lets water activity equilibrate downward slowly rather than crashing all at once. Crash-drying flower too aggressively tends to trap chlorophyll breakdown products that haven't had time to fully convert, which shows up as a harsh, grassy taste, and it's also rough on trichome structure -- brittle, degraded trichome heads don't hold terpenes or cannabinoids as well as ones that were allowed to cure out properly.
Handheld water activity meters have gotten affordable enough that there's no real excuse for home or craft growers to be flying blind here. Checking a sample before it goes to a lab, or before flower gets sold informally, catches problems while they're still fixable rather than after a batch has already failed somewhere official.
Once flower is cured and in the target range, holding it there during storage is its own task. Two-way humidity control packs set to 62-65% RH are the standard practical tool -- they buffer against both over-drying in low-humidity storage and moisture creep in humid climates, keeping cured flower inside the 0.55-0.65 band for months rather than days.
Genetics play a bigger role in this than most new growers expect. Denser, resin-heavy colas -- the kind that come from well-bred, high-trichome genetics -- hold moisture unevenly and are genuinely harder to dry through to the core without surface-drying too fast. Seedtiva's recommendation here is simple: match your drying setup, airflow, and timeline to the actual bud structure of the genetics you're running rather than applying one generic schedule to every strain. And because climate, room setup, and strain structure all shift these numbers in practice, treat 60-65°F and 55-60% RH as a starting point to dial in for your own space, not a fixed guarantee.
Water activity testing exists because mold contamination in commercial cannabis is an ongoing, documented public health risk, not a regulatory formality someone dreamed up to generate paperwork. The Missouri recalls and the New York lab fraud case didn't happen during legalization's rough early years -- they happened in 2026, in states with years of regulatory infrastructure already built. That's the clearest evidence available that this problem hasn't been solved, only managed unevenly.
The 0.55 to 0.65 Aw window defined in ASTM D8197 is worth hitting whether or not a single sample of your flower ever crosses a state lab's threshold. It's the range that keeps mold from getting a foothold and keeps trichomes intact through trim and packaging -- in other words, it's just what good flower requires, with a number attached to it.
And given how often the testing infrastructure itself has proven fallible this year, the most dependable safeguard isn't a certificate of analysis from a lab you can't fully vet. It's disciplined drying and curing at the source, done by someone who knows the genetics they're running and controls the environment tightly enough that a lab result is confirmation, not a gamble.
