What the volatile fraction is
A very small share of a tea’s mass and nearly all of its character. Hundreds of individual volatile compounds have been identified across tea types, from several chemical families — terpene alcohols such as linalool and geraniol associated with floral and citrus qualities, the six-carbon aldehydes and alcohols associated with the green, cut-grass character of fresh leaf, methyl salicylate with a wintergreen quality, aromatic compounds such as indole and the jasmine lactones that appear prominently in oolong processing, and a wide range of products from the breakdown of carotenoids associated with fruity, violet and woody notes. Almost none of these is present at a level that would matter to anything except a nose, and the nose detects several of them at extraordinarily low thresholds. This is what makes aroma both dominant and precarious. The imbalance between mass and importance is worth dwelling on, because it explains a great deal about tea. A fraction that makes up a vanishingly small part of the leaf supplies most of what distinguishes one tea from another, which is why processing decisions that barely change the composition in any measurable sense can transform the product entirely — and why storage that loses a small quantity of material can make a tea unrecognisable while leaving almost all of it intact.
Most of it is made, not found
Fresh tea leaf does not smell much like tea. A large part of the aromatic fraction exists in the growing leaf in bound form — attached to sugars as glycosides, without aroma — and is released by the leaf’s own enzymes during withering and bruising. Further compounds are generated during oxidation as other leaf components are transformed, and further ones again during firing and roasting through the reactions between sugars and amino acids. This is why processing is the dominant influence on aroma: the maker is not preserving a smell that was in the leaf, they are running a set of reactions that produce it. It is also why the same leaf can become a green, an oolong or a black tea with entirely different aromatic profiles, and why the descriptions of processing steps in this catalogue are effectively descriptions of aroma generation. This also explains why withering is treated as a serious craft step rather than as drying. A long, controlled wither gives the leaf’s own enzymes time to release bound aromatics before anything else happens, and the difference between a carefully withered tea and a rushed one shows up in aroma more than anywhere else. White tea, which is essentially a long wither and a dry, is the clearest demonstration that the step is generative rather than preparatory.
Why oolong is the aromatic category
Oolong manufacture involves repeated gentle bruising of the leaf edges over hours, alternating with rest periods, which runs partial oxidation slowly and unevenly while enzymes continue releasing bound aromatics. The result is a much larger and more complex volatile fraction than either a green tea, where the enzymes were destroyed early, or a black tea, where oxidation ran fast and to completion. This is the mechanism behind the reputation of Tie Guan Yin, Dan Cong and high mountain oolong for aroma, and behind the extraordinary range of aroma types the Phoenix growers classify formally. It is also the reason oolong is the most labour-intensive category to make and the one where a maker’s judgement is most visible: the whole process is a sequence of decisions about when to bruise and when to wait. Roasting then adds a second, entirely separate aromatic layer on top of the first, which is why the oolong category spans such an enormous range. A green, unroasted high mountain tea and a heavily charcoal-roasted Wuyi tea are both oolongs and share almost nothing aromatically, because one is showing the products of the bruising-and-oxidation sequence and the other has covered those with the products of prolonged heat. Reading roast level alongside oxidation is the only way to place an oolong from its description.
Why they leave first
Volatility is exactly the property that makes a compound smellable and exactly the property that makes it depart. It departs during storage, which is why staleness reads first as a loss of top notes and only later as anything else, and why a stale tea is boring rather than objectionable. It departs during brewing if the water is too hot, driven off in the steam before it reaches the drinker — which is one of the reasons very hot water damages a delicate tea beyond simply over-extracting it. It departs across infusions, since the most volatile material comes out first, which is why the opening steeps of a fragrant oolong are the aromatic ones and the later steeps trade aroma for body. And it is the part of a tea that heat during transport and display quietly removes. There is one exception that proves the general rule. The heavier, less volatile aromatic compounds — the ones associated with wood, camphor and dried fruit in aged teas — persist far longer, which is why an aged tea’s aroma is stable where a fresh one’s is not, and why the aromatic profile of a category shifts towards those notes with time whether or not anyone wanted it to. Fragility is a property of particular compounds rather than of aroma as such.
Why the same note appears in unrelated foods
Because the compounds are not unique to tea. Linalool occurs in many flowers and herbs; the compounds associated with the violet and stone-fruit qualities of some black teas occur widely in fruit; the roasted, nutty notes of a fired tea come from the same class of reactions that produce them in bread, coffee and cooked meat. This is the mechanical answer to the question that confuses people most about tasting notes: when a note says a Dian Hong tastes of sweet potato, no sweet potato is involved and no flavouring has been added — the leaf produced compounds that overlap with those in the reference food, and the note names the nearest familiar thing. Understanding that removes both the suspicion that something has been added and the temptation to read a note as an ingredient list. The overlap also explains why flavoured tea works at all, and where the line falls. A bergamot-flavoured black tea supplies compounds that a black tea does not produce; a tea described as having citrus notes is producing related compounds itself. The ingredient list distinguishes them, and the cup usually does too, since an added flavouring front-loads and a leaf’s own aromatics persist across infusions with the tea.
The honest limits of compound-level explanation
It is tempting to describe a tea by naming its principal volatiles, and it does not work well. Perception depends on the combination and on the relative levels rather than on presence, and a compound that dominates a profile at one level can be unnoticeable or unpleasant at another. Detection thresholds differ by orders of magnitude between compounds and between individuals. Interactions between compounds change what is perceived. And analytical identification tells you what is present, not what is smelled. So the useful statement is directional: this family of compounds is associated with this kind of note, this processing step generates that family, and this handling loses it. Naming a molecule and asserting that it is why a tea smells as it does is the aromatic equivalent of an oxidation percentage. There is a further limit specific to how the data is generated. Analytical methods sample what can be extracted and detected by a particular technique, and different techniques recover different profiles from the same tea, so published inventories are partly artefacts of method. This is not a reason to dismiss the chemistry; it is a reason to read a list of identified compounds as a description of what an instrument found rather than as a description of what a nose experiences.