Reviewed by Nani Frenkel, chief editor.
Everything that makes cannabis smell like anything is trying to leave. Some of it can be caught on the way out, and some of it can be put back.
Terpene preservation is a different problem from cannabinoid preservation, even though the two share a container. Cannabinoids are large, effectively non-volatile molecules whose profiles change chemically over time, principally through decarboxylation and oxidation. Terpenes are small and volatile. They also oxidise, but their main early loss route is more straightforward: they escape into the air. This article is about that volatile fraction — where it goes between harvest and consumption, and what can be done about it.
For general practice on drying, curing and keeping flower in good condition, our guide to drying and curing methods covers the harvest end in more detail.
At what temperature do terpenes start to degrade?
There is no single temperature at which terpene loss begins, and the widely circulated charts implying otherwise are the main reason people believe there is.
Those charts list boiling points — the temperature at which a pure compound boils at a given pressure. A boiling point is not a threshold below which nothing happens. Monoterpenes evaporate from cured flower at room temperature, continuously, which is why a jar smells strongest the first time it is opened and progressively less thereafter. Loss is a rate, not a switch, and the rate rises with temperature, airflow and surface area long before any listed boiling point is reached.
The charts are also frequently wrong on their own terms. Values get copied between sources without provenance, some are measured at reduced pressure rather than atmospheric, and at least one commonly reproduced figure appears to be a melting point mislabelled as a boiling point. We have written separately on why most terpene boiling point charts are misleading.
The practical version is simpler. As temperature rises, terpene loss accelerates. Any process that applies deliberate heat — and decarboxylation is the obvious one — will drive off a substantial part of the volatile fraction as a cost of doing what it does. If terpenes matter to the finished product, they have to be either protected from heat or captured separately from it.
Oxidation is the second mechanism and it needs no heat at all. Terpenes react with atmospheric oxygen over time, which changes the profile rather than simply thinning it: some products of that reaction are themselves aromatic compounds that were not in the fresh material.
Where are terpenes lost between harvest and jar?
Most of the loss happens before anyone opens a container.
Drying and curing remove water, and they remove volatiles at the same time, because the conditions that carry moisture out of plant material can carry aroma compounds with it. Temperature, duration, airflow and pressure all matter. More heat and airflow accelerate evaporation, but a long, cool dry also gives volatiles more time to escape, and even freeze-drying can cause substantial losses. There is no universally superior method. This is still one of the largest intervention points in the whole chain, and the one furthest from the consumer.
Handling costs more again. Grinding increases surface area dramatically and accelerates evaporation. Repeatedly opening a container replaces terpene-saturated headspace with fresh air, resetting the gradient that was slowing further loss.
Freezing fresh material immediately after harvest sidesteps much of the loss associated with drying. That is the logic behind live resin: preserve the volatile profile until extraction rather than exposing it to days of moving air.
Does stored cannabis lose potency?
Over the first weeks and months of controlled storage, total cannabinoid content can remain fairly stable even while the cannabinoid profile changes continuously.
A 2024 study on nitrogen-modified atmosphere packaging tracked canned inflorescence over 74 days and found no significant change in total cannabinoids, either over time or between packaging treatments. What did change was the composition. THCA declined while neutral THC rose sharply under every condition tested, so decarboxylation proceeds during ordinary storage without deliberate heating. A separate 2025 study on postharvest drying and curing in industrial hemp found total cannabinoid content was likewise not significantly changed by the treatments, while decarboxylation again occurred throughout.
So across these experiments the total was fairly robust while the ratio was not. Over longer storage, THC itself can degrade, so this should not be read as evidence that potency remains unchanged indefinitely — earlier work tracking cannabis over years found exactly that decline.
Sample age matters most when labels report acidic and neutral cannabinoids separately. Total-THC compliance calculations are less affected by early decarboxylation, because they already account for THCA, although longer storage and oxidation can eventually change the total as well.
Terpenes behave nothing like this, which is the whole reason they need separate treatment. There is no conserved total. What leaves is gone.
Does nitrogen packaging protect terpenes?
Displacing oxygen with nitrogen is an obvious idea, and the study that tested it directly did not find the benefit its authors expected.
The 2024 work compared nitrogen-modified atmosphere packaging against ordinary atmospheric storage, both inside sealed cans, and found no significant preservation of either total cannabinoids or total volatile terpenes. The authors had hypothesised otherwise.
The detail underneath the null result is the interesting part. Individual compounds diverged rather than moving together. Myrcene held up under nitrogen while declining by a third in air, which is the result the hypothesis predicted. Limonene did the opposite, along with both pinenes, camphene and terpinolene, all of which fell only under nitrogen. At the level of total terpenes, gains and losses among individual compounds obscured one another.
Grouped by class, the pattern is sharper still. Monoterpenes fell further under nitrogen than in air, while sesquiterpenes rose further, shifting the ratio between them. Since monoterpenes carry much of what the nose registers first, a treatment that leaves the total unmoved can still move the aroma.
There is a second reading of this study worth noting. Total cannabinoids and total terpenes held steady in both arms, which contradicts earlier work reporting roughly half the terpene content lost within a month. The authors attribute the difference to the canning itself, suggesting that sealing against gas exchange may do the preserving and the nitrogen adds little. If that holds, the practical lesson is about the container rather than the gas inside it.
Either way, the framing matters. Terpenes are not one substance with one set of storage requirements. A condition that protects one compound can cost another, and a total-terpene number can stay flat while the profile underneath it shifts substantially. Anyone evaluating a storage method on a single summary figure may be measuring the wrong thing.
Which extraction method keeps the most terpenes?
Extraction forces a choice, because the conditions that recover cannabinoids efficiently are not the conditions that preserve volatiles.
A 2024 comparison of three methods on hemp found the trade-off clearly. Hydrodistillation produced the greatest variety and concentration of terpenes, with beta-caryophyllene dominant at 25 to 42 mg/g of extract. Supercritical CO₂ produced the greatest variety and abundance of cannabinoids but relatively few terpenes. Ethanol extracts contained the most acidic cannabinoids, again with little terpene content. These are concentrations within the resulting extracts rather than total recovery from the plant — hydrodistillation yields a far smaller amount of extract overall.
One caveat has to travel with those numbers: the authors are explicit that changing operating conditions changes the profile. Temperature, pressure, solvent ratio and run time all move the result, so these are outcomes for the parameters tested rather than fixed properties of each method. A differently tuned CO₂ extraction is a different experiment.
The alternative to choosing is to stop treating it as one process. A 2021 study demonstrated a microwave distillation method that isolated terpene-rich essential oil from five hemp cultivars. It did not test recombination, but it illustrates the first half of an approach already familiar to processors: capture the volatile fraction separately, then blend selected compounds into the cannabinoid extract later.
Can terpenes be added back?
Terpenes can be reintroduced, and the practice is widespread, but what it produces is a reconstruction rather than a restoration.
The most directly relevant published work is a 2020 study in the Journal of Cannabis Research on preserving and augmenting volatile terpenes in cannabis inflorescence. From the dose-response relationship, the authors estimated that an external terpene source equivalent to 1.18% of the flower’s weight would be enough to maintain its initial total terpene content for six weeks. That would maintain the measured amount, although not necessarily the original molecules. More striking, the same approach was used to adjust the profiles of two separate harvests toward each other, reducing the difference between them by 39.5% — batch-to-batch standardisation, which is a real commercial problem for producers selling a named cultivar that tests differently each run.
The context matters. The study was funded by Vireo Health, all five authors worked for and held shares in the company, and one was named on the provisional patent covering the system. It is useful proof of concept from a financially interested group, not independent verification, and it has not to our knowledge been replicated by a neutral group.
Beyond that study, most of what happens commercially is industry practice rather than published research. Terpenes are reintroduced into vape formulations, edibles and flower using either botanical terpenes — the same molecules sourced from other plants — or cannabis-derived terpenes recovered from extraction. Blends are frequently sold as strain-inspired profiles. There is very little peer-reviewed work evaluating any of this, and claims about it should be read accordingly.
Botanical sourcing raises one technical issue worth knowing. Many terpenes are chiral, existing as two mirror-image forms that can smell noticeably different and behave differently in the body. Limonene is the standard example. A botanical terpene may supply a different enantiomeric ratio than the cannabis-derived version, so a profile can match on the compound list and still not match in practice — a distinction covered in our piece on enantiomers.
What a reconstructed profile cannot capture
The limit on reintroduction is a measurement limit, and it is easily stated: you can only rebuild what you measured.
Cannabis produces well over a hundred terpenes. Commercial panels commonly report only a few dozen, and formulations may be built from fewer still. Every compound outside the panel is invisible to the reconstruction, including trace compounds that may contribute disproportionately to aroma, and including any non-terpene volatiles — the sulfur compounds among them — that shape how cannabis actually smells.
A reconstructed profile is therefore a sketch drawn from the compounds someone chose to measure. It can be a good sketch, and for standardisation purposes it can be good enough. It is not a copy, and what a terpene panel covers sets the ceiling on how close it can get.
That is the honest frame for terpene preservation as a whole. Preservation keeps what the plant made. Reintroduction adds back an approximation of what was lost, built from a partial list. The two are not equivalent, and only one of them is limited by an analytical menu.
For the compounds themselves and what the research says about each, start with our complete guide to cannabis terpenes.
References
- Bueno J, Leuer E, Kearney M Jr, Green EH, Greenbaum EA. The preservation and augmentation of volatile terpenes in cannabis inflorescence. Journal of Cannabis Research. 2020;2(1):27. doi:10.1186/s42238-020-00035-z
- MacLaughlin LL, MacDonald MT. Is nitrogen-modified atmosphere packaging a tool for retention of volatile terpenes and cannabinoids in stored Cannabis sativa inflorescence? Journal of Cannabis Research. 2024;6:42. doi:10.1186/s42238-024-00253-9
- Baek Y, Grab H, Chen C. Postharvest Drying and Curing Affect Cannabinoid Contents and Microbial Levels in Industrial Hemp (Cannabis sativa L.). Plants. 2025;14(3):414. doi:10.3390/plants14030414
- Chacon FT, Raup-Konsavage WM, Vrana KE, Kellogg JJ. Effect of Hemp Extraction Procedures on Cannabinoid and Terpenoid Composition. Plants. 2024;13(16):2222. doi:10.3390/plants13162222
- Micalizzi G, Alibrando F, Vento F, Trovato E, Zoccali M, Guarnaccia P, Dugo P, Mondello L. Development of a Novel Microwave Distillation Technique for the Isolation of Cannabis sativa L. Essential Oil. Molecules. 2021;26(6):1588. doi:10.3390/molecules26061588
