Wednesday, November 29, 2017

Pu'er storage optimums, and relative versus absolute humidity

Pu'er storage:  optimum humidity

I just watched a nice video by Scott Wilson (a Yunnan Sourcing vendor owner) claiming that people might prefer pu'er stored in either wetter or drier conditions (Hong Kong / Malaysia versus Kunming, wetter and drier, respectively).  It almost goes without saying that some people prefer varying storage conditions, younger versus more aged / fermented sheng, and different aspects, region-specific traits, and aging related characteristics beyond all that.  Or potentially some people have broad preferences, and they're not fixed on one optimum narrow range.

Steepster tea group discussion approached all this from an unusual direction  Someone asked why we aren't concerned with absolute versus relative humidity instead.  It's an interesting question, and one that takes some work to make sense of (unless someone is a meteorologist), so lets start with his following introduction example.  It's a bit of a stretch but I will drag this review of relative versus absolute humidity measurement back to the scope of what broader storage parameter selection means, but it's better to get into the meteorology side of things first:

Here in Stockholm right now the relative humidity is 80% at 7 degrees C and 998 mm Hg pressure. That makes the absolute humidity 0.006 kg/m3.  In Hong Kong today the relative humidity is “only” 73% but at 24 degrees C and 1014 mm Hg pressure which works out to an absolute humidity of 0.016 kg/m3... 

So if I was comparing humidity with someone in HK using only relative values it would seem like my tea is stored “wetter” than his, while in actuality his tea is stored at almost three times the humidity mine is.

Of course one issue is that if your tea is actually stored at 7 degrees that alone could potentially be a problem.  Aging is occurring due to biological factors, from the activity of specific bacteria and fungus, and I'm certainly not clear on the temperature window for them to thrive or just continue living but 7 C is getting down there (45 F).  Fermentation might well stop and not restart with exposure to that (or much over 38 C / 100 F).  But then I've not really ran across temperature range quoted for what will ruin pu'er related to aging potential.  I've seen speculation about that sort of thing, comments about how falling outside the range will lead to specific tea-aspect outcomes, but those are hard to evaluate and impossible to keep track of.

I did run through some related background in a post once, about what relative versus absolute temperature means (in a post about shou mei, kind of strange), where I showed a psychrometric chart showing how this works (the inter-relation of temperature range, absolute and relative humidity):

a bigger version of this chart follows

Those are actually the ASHRAE data center environment guidelines shown (optimum ranges for computer equipment environments, housing servers, presented for sea level pressure), if that sort of thing is of interest.  I could say a lot more about that part but I'll stick to tea here.

I've written a little about pu'er storage before (here) but focused more on the issue of air-flow, related to pros and cons of storing tea cakes in ziplock bags.  I referenced some background there, different perspectives, including a TeaDB blog article that defines "dry storage" as 60 to 70 % RH.  Things are never that simple, as this question of absolute versus relative temperature raises, but we can go with 60 to 80% as a typical storage range, for purposes here.  It's not completely clear to me at what humidity range mold thrives, and there may be no simple answer to that but "over 80%" tends to get mentioned.

Let's assume that a pu'er storage window is between 20 and 30 degrees C (68 and 86 F; kind of a normal indoor living environment range for people), and from 60 to 80 % relative humidity.  Again, part of the point here is to review if "relative humidity" makes sense or not, and postulating a starting-point range will help us sort that out.  That envelope looks like this (outlined in blue):

In terms of absolute moisture content (the graph bar on the right) it ranges from around 8 grams of water per kilogram of dry air up to around 22 (g/kg).  It seems like a good bit of range.  But what does that mean, beyond being the proportion of water in the air by weight (beyond the actual definition)?  It makes one consider to what extent air pressure and expansion changes things (the factor that causes hot air balloons to rise, so that kg of air varies in the space it might take up), but it seems as well to keep this review on the simple side, to not introduce extra factors.

In that thread discussion I rambled on a good bit about framing a context, and how I interpreted it, and why I chose 20-30 C and 60-80 % RH as normal-range parameters to review what absolute humidity was in a working range.  I had touched on some of that background in that earlier post about pu'er fermentation, but again that wasn't the main point there.  I'll quote my own comment in the Steepster discussion, but it doesn't work well to edit it down to a few sentences that are the main point (I ramble on a lot everywhere).

It’s a problem that there really isn’t one definitive, well-agreed guideline for any of this, a range of absolute optimum and also what lies beyond that as acceptable. In a Tea Addict’s Journal article (written by an author regarded as one authority on this topic, who is familiar with other inputs) he cites a range of 20 to 30 C as being typical / acceptable (70 to 85 F, roughly). Although he doesn’t state a clear humidity range in the blog article for wet and dry storage optimums but we could rough that out in a typical range of 60 to 80 % RH; much below that and tea won’t age, just above that it may well mold, and towards the top and bottom level wet and dry storage are represented. Mind you I’m not intending this as a last word; I mention specific values for the sake of further discussion, not final definition.

Of course the premise here was that RH means nothing, that between 20 and 30 C (65 and 80 F) the absolute humidity (grams / kilogram of dry air) shifts based on both temperature and RH... 

I’m not sure that’s actually being implied, though. My take is that an envelope of optimum range is being described between those parameters..  Per my understanding these functional ranges are derived from observing natural temperature and humidity ranges in natural and controlled environments where pu’er aging has been successful. Identifying one optimum where the related bacteria and fungus thrive is a different type of project, but then per some input...  periodic natural environment variation is sometimes considered positive, or maybe even optimum.

...The right kind of biologist could move off tea-related research and extrapolate more general micro-biological optimum environment parameter research to say more. Any theoretical optimums would still need to be filtered related to experienced tea-storage practice...

...We are also concerned with limiting growth of fungus that doesn’t taste good, and other optimum characteristics, which is all going to need to be judged subjectively in the end (tasted, not lab-tested, or both, at a minimum)...

...Related to the central point here note that if we really did want to move to discussing absolute moisture content in air instead (stated in grams of water in kg of dry air, or in dew point temperature levels, how that is phrased in meteorology) we couldn’t work from the same terms used in all those discussion, which is relative humidity. 

It's not very conversational, as tea group discussion goes, and a little chopped up to limit the length, but I guess that brings the main points across.  Note that I'm not really concluding much there, just talking around the background some.  A main problem with using absolute versus relative humidity is that everyone else is discussing relative humidity; it introduces translation problems.  Which I will work on fleshing out here, a little.

It's interesting to also consider that against what local environment conditions are like where tea is typically stored, in places like Hong Kong, Malaysia, and Kunming.  Before doing that, let's take that original question more seriously; what if we switch from relative humidity as a baseline to absolute humidity?  Let's use a similar approach, and suggest a range of 20 to 30 C for temperature (65 and 80 F; assuming typical room temperature storage), and go with 10 - 20 grams of water per kilogram of dry air as a starting point range instead, chosen to describe a similar environment window.

Any problems with that?  It seems likely that range on the upper / left side of the blue box (inside the orange control-limit range), above 80 % RH, may be too wet, that the tea really might mold stored in that range.  The lower-right corner of that orange box represents conditions at 30 degrees C, 10 grams water (moisture) per kg air, and under 40% relative humidity.  That would probably feel dry to a person, and possibly to a micro-organism living out a normal life-cycle that just happens to also be causing tea to ferment.  Or maybe not; what do I know about bacteria and fungus.

Optimum storage considered related to conventional regional storage

Leaving aside the in-depth absolute humidity concern, let's consider the typical range of experienced climate in Hong Kong, one of the main places considered for natural "wet" storage:

Temperatures do dip slightly below 20 in the coolest season but fall within that 70-80% RH range for ideal "wetter" storage.  Let's check one for Malaysia (this version from Penang, in the North there, but I get the impression weather is a bit consistent in Malaysia):

So much for those people experiencing seasons.  The temperature couldn't be much more consistent but the slight variation in humidity does imply precipitation changes, which are shown in that related precipitation summary graph:

Interesting.  Our rainy season here in Bangkok is May through October (half the year, but often rainier at the end of that).  The "cool" season is a slight dip in temperatures in December and January, and then it really is a good bit hotter in March and April.  It doesn't completely match up with there but a lot of that overlaps.  It should; Bangkok isn't that far from KL (Malaysia is the next country down), and it's even closer to Penang.

The point?  These places are where pu'er is supposed to age best, but only if one prefers relatively wetter storage.  The natural environment range is actually relatively narrow, because they're in similar places as local climate goes.

Let's take a look at Kunming, the place Scott (of YS) initially suggested might represent good dryer storage:

Temperatures run cooler but not cold; the lowest outdoor maximum is 14 in December and January, or 57 F.  Of course a warehouse environment could be controlled further, if a temporary drop below some minimum was a concern.  Humidity dips into the 50s for two months but is more typically in the 70s (RH, of course).  If the indoor warehouse-storage goal there is to keep temperature between 15 and 20 (C, more like 60-70 F), and humidity between mid-50s and mid-70s (RH) the building enclosures leveling out daily variations would accomplish most of that.  It could just get a little cool in parts of December and January, and April is the only month it might dip below that bottom limit for humidity (52 is an average).

Special concerns related to heated indoor air

In US and European indoor environments the main concern relates to outdoor cold air not containing much humidity, so that after you heat that air indoors it becomes quite dry (the air would naturally mix).  The most moisture that air can even possibly hold at 12 C (54 F), the dew point at that temperature, is right at the lowest point of that pu'er window I'd defined earlier on that psychrometric chart, around 8 or 9 grams of water per kilogram of dry air.  But again, I'm not claiming that is an informed optimum range, just a rough working version of that sort of thing.  As you approach water freezing temperature air doesn't hold much moisture at all; it simply can't.  That's why winter-time outdoor cold air can so frequently feel very wet (kind of counter-intuitive, that part); because if a cooling trend is occurring it's really easy for air to be at or near a dew point, where fog occurs, and the moisture naturally condenses out of the air (more on that here).

Some of the outdoor weather range for Kunming is slightly below that 12 degree level I just mentioned, at least in terms of daily minimum values.  If you heat air at or below 10 C / 50 F back up to 70 F (room temperature) without adding more moisture regardless of the starting RH level the final indoor humidity would be very low, well below 50%.  The dew-point level for air at 0 C (32 F; did I really need to convert that?) is below 4 grams per kg of dry air, an absolute level of moisture that relates to under 20% RH at 20 C (68 F).

Of course air in Kunming pu'er storage warehouses isn't shifting about related to temperature or humidity, so the case is a little different than for someone's home, where air mixes freely.  I'm really onto to talking about issues with Northern latitude home storage; natural China-region storage was the last section.  Or the far South, there's just less people down there.  I just looked that up and the South of Australia and South Africa don't extend as far below the equator as Philadelphia is North of it (both of those to -38 degrees and -34, respectively, and Philly is at 39).  Only Chile and Argentina extend below those levels (of populated areas; obviously researchers in Antarctica are using pumidors).

it doesn't snow much in Kunming, but it did in 2013 (photo credit)

Pulling it all together

Unless I'm summarizing it wrong Scott is saying that people claiming you need some controlled, narrow-range optimum for pu'er storage doesn't match his own experience, that it still ages across a range of environmental conditions, the process just changes.  But when he's saying he likes "dry" Kunming storage that's still based on an outdoor humidity level that is above 60 % (RH) for all but two months, and per the one reference graph shown at an average of 57 for one of those months.  We're not seeing equivalents to US outdoor temperatures or humidity in those places, and the real concern would be indoor, adjusted-temperature humidity values.

Scott doesn't drop into talking about environment specifics much in that video, probably because that gets so complicated, but he does mention at one point that as long as relative humidity is above 50% the pu'er should still be ok.  He makes a good related point:  it's not really pleasant for people to live in indoor environments at very low humidity levels either, so in a sense it's about maintaining conditions that are comfortable for a person and also ok for stored tea.

I lived in the mountains in Colorado for quite awhile, where it was below zero C (under 32 F) for about half the year, where literally thinner air--less of it; I lived at over 7000 feet of altitude--made conditions worse, and it typically felt really dry indoors.  Stored pu'er might die, without adding environment controls.  Of course to really investigate what was going on up there for local climate checking out psychrometric charts for higher elevation (lower air pressure ranges) would make sense, but it seems as well to keep this simple, to not pursue that tangent.

in an earlier life

With the proper environment adjustment (use of a "pumidor") someone in any natural environment could move on from concerns about natural conditions and consider instead where to set the controls.  I have no idea, but I guess these natural environment ranges (the local climates) where storage is considered to be good to ideal for wetter and dryer storage would provide a starting point.  That was sort of a main theme in what I said in that thread discussion, and the point of that section of location graphs; we may not know storage conditions ideals for pu'er fermentation but a best storage range isn't a total mystery.

A second part of that discussion comment was about how that discussion prompted me re-read a number of old references on this subject.  I won't go further with speculating about these issues but I will leave off with mention of that reading list:

The Tea Addict's Journal "Ideas of Proper Puerh Storage" post (that I already mentioned):

A Tea Chat forum discussion of "Pu're Storage in Low Humidity":

A Cha Dao reference blog series on aging pu'er (the last article in a series of five, which links to the others)

This last reference is a research article reviewing the measured levels of biological activity in pu'er (bacteria and fungus causing fermentation in sheng and shou).  It doesn't go into environment factors, but is a good reference for that context of concerns (keeping those organisms alive, but then in conditions where they can thrive too much other fungus that grows as mold also might):


  1. A little bit head hurts reading this.
    The test is in the taste. I would say 100% of the Kunming aged teas I have tried have tasted dried out and with not much aged character, even at 10 years +. My own favourite is Penang stored tea, although it is geographically close to Bangkok it has its own unique climate. I just spent 6 months there and it is a paradise for puer lovers, lots of tea shops and tons of old tea.
    Off topic if I may re: your Liu Bao posts. Have you had a look at, by far the most extensive range of Liu Bao online and some interesting peasant style hei cha you don't see anywhere else.

    1. Cool to approve a post; amazing how consistently every comment is just spam. I've heard of Chawangshop but haven't tried their teas. After Yunnan Sourcing and White 2 Tea that's the next level of shop that gets recommended, and there's not much else on that level. I've tried sheng stored different places and the difference is notable but I can't really separate factors so well I want to pass on advice about one being better, or warn off people about Kunming storage. I've been to Penang but not in awhile; if I make it back I'll check for shops.