Journal article: Pu-erh tea tasting in Yunnan, China

My wife’s graduate studies are winding down, so I’ve been taking advantage of her university privileges to access some tea books and journal articles.  I recently read “Pu-erh tea tasting in Yunnan, China: Correlation of drinkers’ perceptions with phytochemistry,” by Ahmed et al., from the Journal of Ethnopharmacology, and my scattered reflections follow.

Fig. 1. Lab setup

The authors studied 10 samples of puerh (ripe, new raw, and >10-year-old raw; plantation-grown and agro-forested), infusing each sample 10 times in gongfu fashion.  Each infusion was described and rated by a panel of Yunnan experts, and also analyzed in the laboratory for its constituent chemical content.

I learned several things just from the article’s introduction.  The Pu, an ancestral people of the Bulang, Wa and De’ang, are considered the first cultivators of the tea plant.  I imagine the town of Puerh got its name from them, before becoming famous as a trading center for compressed tea.  What a happy accident for this variety of tea to be indirectly named for those ancient tea pioneers!  Of course, it may be instead that the Pu’s claims for discovery of tea are falsely advanced on the strength of this “evidence.”

I object to the authors’ statement, “Gongfu cha dao developed in eastern China and draws from earlier tea preparation methods documented in Lu Yu’s Cha Jing.”  This is not patently false, but it’s about as meaningful as saying “Punk music developed in Europe and draws from structures developed by Baroque composers.”  If the authors have been exposed to that particularly over-reaching kind of Lu Yu-worship, they may even deserve credit here for repeating such claims so tepidly.

The authors set out a classification of the 4 environments where puerh is grown, which mirrors those I have seen elsewhere:

  • Forests (composed of wild trees or formerly cultivated trees that have become feral)
  • Agro-forests, “forest areas (0.5–3.0 ha.) thinned for tea cultivation or swidden areas where plant regeneration is fostered….  The multi-storied vegetative structure of agro-forests provides soil fertility and pests and disease control.”
  • Mixed crop fields, created in agro-forests by “replacing associated woody plants with rice or other crop production and by pruning tea trees to increase irradiance and promote crop growth.”
  • Terrace plantations, that easy-to-hate intensive monoculture.

In the production of ripe puerh, I was interested to learn that “leaves may intentionally be inoculated with selected microorganisms such as Aspergillus sp. (Trichocomaceae)….  The microorganisms in black pu-erh oxidize polyphenol compounds more completely than the enzymatic oxidation of other black teas, and create fermentation-derived compounds known as statins.”  I would love to know more about these chemical processes.  I remember seeing the term “theabrownin” while browsing journal articles, so this may be one of the compounds created in polyphenol oxidation by microorganisms.  Statins are a class of compounds that are used medically to inhibit cholesterol formation.

Previously, I could not find the crucial distinction between green tea manufacture and puerh manufacture.  In both, the leaves are withered, fired, rolled, and dried.  If these steps are the same, then is puerh merely compressed green tea?  The most provocative claim I found in the Ahmed paper is that “the deactivation of enzymes is not as complete for pu-erh [as it is for green tea]. Consequently, pu-erh has a distinct oxidizing process and develops a smooth taste with age.”  This is to say that in green tea, the leaves are fired–in Chinese, shā qīng or “kill-green”–to kill enzymes that would cause oxidation; while in puerh, the firing is briefer, leaving some of the enzymes intact to perform a more gradual oxidation.  The satisfaction of this crucial insight lasted only a few days, until I subsequently found a post by Tea Urchin that claims that the enzyme-deactivating stage is not categorically less complete than that of green tea.  Tea Urchin claims the difference comes from the sun-drying step at the latter end of manufacture.  I’m not convinced by his technical explanations, although I’m not in a position to refute them point-for-point.  It also occurs to me there are sun-dried green teas that would confound this distinction.

The Ahmed study focused primarily on catechin content (aka polyphenols or antioxidants) and methylxanthine content (caffeine and its relatives), and how these correlate with tasters’ perceptions.  On average, agro-forested puerh in the study had much more catechin content, while terrace-grown tea had the edge in methylxanthines.  New raw tea had the most catechins, “aged” raw tea had less, and ripe had the least.  Measurements of antioxidative activity paralleled catechin content.  A taster’s favorable perception was proportional to a tea’s catechin content, likely informed by its bitterness.  Methylxanthine content was about the same across production varieties.

As I look at the results of this study, I have a familiar frustration.  The authors have analyzed a heroic amount of infusions, but it still seems like the data is not robust enough to isolate each variable in play, and therefore no binding generalizations can be made about manufacture and chemical content.  For instance, when I compare the charts, I can identify 6 infusions that are very high outliers on the methylxanthine scale.  These 6 infusions turn up in the “aged” green category and also as terrace-grown.  I’m going to guess that these 6 infusions all belong to one tea sample, since other charts indicate that methylxanthine content doesn’t vary drastically from one infusion to the next.  So which factor contributes the most to a high methylxanthine level?  Does terrace farming somehow produce high methylxanthine content?  Could methylxanthine actually increase during the aging process?  Or is it something else entirely, such as the plant cultivar used, the harvest time, or the leaf grade?

It seems to me that a longitudinal study design would be much more effective in understanding these factors.  For example, harvest a batch of single-origin tea leaves, and manufacture some of it as raw puerh, and some as ripe.  Manufacture some as green and black tea while you’re at it.  Analyze multiple samples of the different teas, and also save some of the raw puerh to be analyzed after it has aged.  This is a long time to wait, it’s true, but any solid research started now on the effects of puerh aging would not be too late!

This kind of study might require an intimacy with tea manufacturers of several different regions and their methods, that many Western food science researchers don’t possess. And to be fair, the primary goal of the Ahmed study seemed to be establishing objective chemical markers of tea quality, and then using those markers to advocate for lower-intensity agro-forest cultivation, which the authors have performed admirably.  (Evaluating raw, “aged,” and ripe puerh on the same quality scale does confound things a bit, but the averages should still be meaningful.)  But I’m sure the information is out there–I just need to keep looking!


Ahmed, S., Unachukwu, U., Stepp, J.R., Peters, C.M., Long, C. & Kennelly, E., 2010.  Pu-erh tea tasting in Yunnan, China: correlation of drinkers’ perceptions to phytochemistry. Journal of Ethnopharmacology, 132, p. 176-185. doi:10.1016/j.jep.2010.08.016


5 Responses to “Journal article: Pu-erh tea tasting in Yunnan, China”

  1. 1 Brandon February 13, 2012 at 1:55 AM

    “I object to the authors’ statement, ‘Gongfu cha dao developed in eastern China and draws from earlier tea preparation methods documented in Lu Yu’s Cha Jing.’ ”

    If this was your first hint, your second was:
    “the deactivation of enzymes is not as complete for pu-erh [as it is for green tea]. Consequently, pu-erh has a distinct oxidizing process and develops a smooth taste with age.”

    The lack of broader cultural reference and procedural know-how is not helping this study make sense of its findings.

    I guess I will agree with you that the science is not entirely satisfying, but Urchin’s onto two things – fa jiao is in part due to microbial action, which is why good shu can sort of, maybe mimic very old traditionally stored sheng puer. Perhaps if you could squint your tongue. Also, it is clear that some differences in the raw material, other than the forcefulness of shaqing, lead to better aging outcomes for puer than for green tea. Green tea certainly oxidizes fairly quickly, and oxidative enzymes remain active. (sidebar: try leaving sencha out for a few days. is it less ‘green’ for having less of a proper pan firing? I’m not drinking it either way you slice it.) If shaqing time was constant (not sure,) certainly the temperature used to shaqing puer tea is significantly higher.

    What about moisture content of the leaves? I think this is more key to post-secondary fermentation rates than anything else. I would suggest that a long, rolled leaf sencha would have worse results in a few days of open air than a more fired Korean green, who seemingly has more surface area exposed. Should we taste?

    Now we get into the tricky bit. If moisture is removed significantly in shaqing, it is put right back in by steaming. I recently heard of a high-compression brick meant for aging, by having all of the intended moisture locked inside – no outside influence needed, provided you don’t handle it so poorly that it dries out. Zhou Yu certainly saw a better outcome for aging tea with a moist center. This is a good stage for a balanced mixture of enzymatic and microbial action, and if I had to place bets, is where focused research on improving long term storage outcomes would land.

    But what makes me nervous about the context imposed by the authors on their lab work is that this is relatively new thinking, or at least newly popular.
    Yunnan Puer tea was not popularized or spread by ethnic minorities in Yunnan, even if they all got dressed up for photo ops at once. It was spread amongst the more Northern Chinese only after intensive interest from Hong Kong and Taiwan. The return to bingcha made of single estate, old arbor teas is a Taiwanese invention, I believe dating to the very end of the 1990s.

    So while the ethnic customs of the minorities of Yunnan is extremely interesting history, I think the tea drinking and tea aging habits of Hong Kong and later Taiwan are more important.

    From the source:

    So for our last points, 10 year aged tea of any stripe should not be “smooth,” or should at least contain an appreciable bitter component. If not it had wussy leaves or some other fixing. 20 year tea stored under the newer method also maintains some bitterness.

    And prior to the end of the start of the 1990s, aging puer was almost always of the type described by Professor N. Which is all about “cooking” the tea at a slower rate in an artificial environment of high moisture and fairly high heat, and then taking the next several years to “remove the storage” by resting in a more natural environment. With all this investment made into the finishing of the tea, is the long forgotten shaqing stage really worth arguing over? I guess to an ethnowhateverist, it seems like a big value add by the locals. You can definitely ruin the tea by conducting this step wrong, but there is a lot of work to go to make aged puer.

    • 2 Brandon February 13, 2012 at 2:02 AM

      I guess what I should have said more succinctly is, the authors case would be more compelling if they had stuck to late model tea, and not mixed in the other teas. Variables compound, and confound.

      The first sentence, last paragraph escaped my copy editor’s glance, I had started to write end of 1980s, and replaced it with 1990s, because good ole Vesper is mostly cooking the books when he dates his Ba Ba Qing Bing to something before 1991.

      • 3 plucktea February 13, 2012 at 6:47 PM

        I think the “ethnic minority” part of the Ahmed article is more limited than you imagine, but you do well to dispel any possible interpretation of puerh as “ancient native wisdom.”

        In fact, I’d say the whole article is more limited than you’d imagine; roughly, it’s only saying, “quality puerh is characterized by high catechin levels–which are good for you btw–and those tend to come from agroforests rather than plantations.” The rest is context, and academic base-covering. It just happens that several disparate nuggets were relevant to my learning at the moment.

        So though it isn’t the focus of the article, there’s a lot I need to learn about the details of the oxidation process. I think it’s important to remember that enzymes are catalysts, which is to say that a reaction may still happen without them albeit in a limited fashion. I get a little confused when you refer to moisture in the context of shāqīng, since you kill enzymes with heat, regardless of moisture level. I do understand that it plays a big role in post-fermentation.

        I’m probably trying to make theoretical distinctions that you don’t find useful. Are different kinds of green tea more different from each other than they are from puerh? Is compression and post-fermentation the only functional difference that separates green tea and puerh? Historically, the cultivar and the geography have separated these two, but those days are fading fast–how long before East China makes a native shēngbĭng, or Taiwan, or India?

  2. 4 Patricia February 13, 2012 at 10:33 AM

    It sounds like Evan needs to work in a lab somewhere to get this sorted out. Is it true that tea is enveloped in such shabby science? Could a life be enough to isolate and answer these research issues? When tea is analyzed in this way, does it lose its romance (i.e. don’t think I’d want to see an opera about chemical compounds)? [snark-snark from a non-tea aficianado.]

  3. 5 David Galli March 14, 2012 at 8:51 PM

    And while you’re working in that lab there, Evan, I would love to know more about the following:

    1) What microbes are present on puer (e.g., Da Ye) leaves PRE-processing?

    2) What microbes are present in a raw cake immediately POST-processing (i.e., what survives processing)?

    3) What microbes are present in middle-aged, or even old cakes (i.e., does a cake gather more microbes, after whatever processing kill-off might take place)?

    4) How much microbial variation is there from location to location (as I understand it, some beers, cheeses, sourdoughs, etc., developed their distinctive natures partly as a result of ambient, wild yeast strains)?

    5) What byproducts of these little critters are found in microbially-processed teas?

    6) Do said byproducts increase a drinker’s feeling of “drunkenness”?

    So… how do we get funding?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Subscribe to updates via Twitter
Drink with us at Philly Tea Meetup
Questions? General comments? email plucktea (at)

%d bloggers like this: