Designer mineral profile water for brewing tea

 

In the old form of this same topic discussion people would speculate about how individual minerals in tea brewing water affect compound extraction and tea character, and then map their own preferences to specific versions of bottled water, or discuss using local spring water, often referencing colorful stories about secret sources.  Times have changed, and it's on to researching more ideal mineral profiles for water.  These are often presented as either one general recipe working out well enough for all teas or else identifying and tailoring mineral blend recipes for individual tastes, purposes, and tea types.  Fascinating stuff!

One generality isn't made clear in all that I'll summarize or discuss here:  per my understanding some minerals help support other compound extraction from tea, it's not that you are tasting those minerals directly in the water / brewed tea.  Maybe that comes into play, maybe more related to an input like silica than minerals like calcium and magnesium, but I haven't really ran across a complete account of these sorts of cause and effect sequences, what causes what, or rather how it works instead.

Let's start with that history first; it might be of interest to some, as it was to me.  Tea Chat was a main older discussion forum, and there are lots of threads on this topic, but here is a main one from 2011.  I think even back then people were clear on low levels of minerals potentially working out for some tea types, but in general with a moderate level of calcium and magnesium helping extract positive compounds better.  Then "hard water," versions with really high calcium content, or else just high total dissolved solids (TDS), total mineral content, wouldn't be regarded as favorably.  Lots of calcium can even cause a scum to form on the tea, partly a cosmetic issue in addition to an experiential problem.

I wrote about this subject based on trying out Volvic in comparison against my local filtered tap water in this post, the second of two.  There were a lot of related discussions about that topic around then in a Facebook group that I moderate, here.  For people mapping preferences to standard bottled types, or local water sources, then relying on trial and error and hearsay input, that review and discussion only ever went so far.

Tea Curious took a next step, when they developed an "open source" water mineral recipe for "ideal water" back in 2020, here, a page that also works as an introduction to lots of the related theme and factors.  Rie Taluli was involved in that development; maybe her name rings a bell, as one of those known tea people, someone I've enjoyed talking to online.  I can't judge any of these sorts of recipes or developments myself, so the theme here is about passing on a summary, not leading towards that level and type of evaluation.  

A second main development step gives us a lot more to go on, which is well-documented across lots of clear background, related to what Empirical Tea / Empirical Water (related sites) have been developing, and what the Tea Curious blog has been discussing.  It might sound like I'm trying to establish a chronology of development more than I really intend; to me it's just interesting taking note of these milestones along with main publicly available references.  Let's check into that, before the main discussion of what Empirical Tea / Water has developed, with limited comparison to the Tea Curious ideas.  The Empirical Tea site lists this for a history:

Hello. My name is Arby, and I’ve been drinking good tea since February 2018.

Over this (short) time, I’ve become more and more obsessed with the science of what consumers can do at home to bring the best out of their tea. This includes experimenting with tea storage, tinkering with custom water recipes, and more. I will be using this platform to provide updates on things I’ve learned and experienced in the realm of tea.

As of 2021, I have decided to dedicate my life to studying tea academically. As I continue to learn and improve as a researcher over the years, expect to see better-designed experiments.

As far as that online site history per the Wayback Machine scan that site might've been set up in October of 2020, or at least that's their first backup snapshot, so he was putting developed ideas out there even earlier (than that later date).  To me it doesn't matter so much that he's 5 years into exploring better tea, although I get it that some people might like to judge their own personal tea history with others in different ways.  In discussing the early development of the Truth Serum formula Arby said that he wasn't aware of the Tea Curious work at that time, but it probably wouldn't have made much difference either way.  

In a main reference, a Crimson Lotus podcast discussion,  Arby's friend and author of the Tea Secrets blog joined him.  Checking that blog timeline the earliest post on water review was in May of 2021, but for sure his exploration didn't start then, for how well developed that first listed post is.  Both of them have been at this awhile.  

This isn't really intended as a claim that they followed Tea Curious in timing, but maybe, or any sort of guess about the Empirical Water and Tea Secrets histories.  These are just dates this content identifies, which we can take as one form of implied online history.  I didn't try to identify if anyone else has been exploring and developing these themes over the past few years; if so my apologies to them for any potentially implied lack of credit, which was not intended.  These sources are simply enough to tell a bit of a related story, to cover a sampling of some core themes and ideas, with the limitation here that I won't be able to cover detailed findings.

Empirical Tea / Empirical Water research, presented findings, and public access recipes

People interested in following this up should start with that podcast blog intro and then spend a couple of hours browsing those three sites, but of course I'm still going to summarize parts of it.

One main theme is development of a "Truth Serum" version of a water mineral recipe, described as including appropriate mineral content for extracting a lot from a tea for the purpose of evaluation.  Recipes for three different mineral blends are here, that one plus and older version and one more special purpose type, designed for a certain outcome.  

Would it be easy to grab the related equipment and buy the food-safe mineral components to make those recipes, adding that to either distilled or low mineral content reverse osmosis processed water?  Surely not.  This Tea Secrets post probably works as well as others related to basic context, what is involved, related to range of minerals and equipment.  This is a recipe there of an updated Truth Serum version, which will help support comparison of that to the Tea Curious recipe version in a next section.  

One starting point might be reading up on Tea Secrets blog posts reviewing different commercial water versions, with commentary and judgement about them, how well they perform in relation to brewing different tea types.  The next step, re-creating these blends, would be a bit more involved.  Empirical Tea sells a mix version they've created; that might work as an interim step, to judge if the next levels of review and effort are of interest or not.

Jumping around a bit between related themes, this Truth Serum version was intended for evaluation, as presented, and other mineral mixes could suit others better for specific tea type preparation.  Personal preference would define "better" in this case, a running theme across all tea experience.  Let's compare two tea brewing water mineral content recipes, which does circle back to that idea.

Before that one related clarification, there is an earlier and more recent Truth Serum versions listed on the related sites in these two places:

1. https://empiricaltea.com/water-recipe-truth-serum/

2. https://empiricalwater.com/products/truth-serum

Comparing the Tea Curious and Truth Serum versions

Better interpretation of these would definitely involve embracing a learning curve, including the Tea Secrets blog, Empirical Water content, and Tea Curious page.  To keep going one could review a listing of other tea writers' reviews and thoughts on their work and specific recipes, which would require a good bit of review time expenditure.

Let's keep this simple, and check out the listings of final mineral amounts, similar to those we definitely see on the back of bottled water versions, starting with the Tea Curious version:

Together, this leaves us with a water with (roughly) 7 PPM calcium, 2 PPM magnesium, 5 PPM sodium, 5 PPM potassium, 11 PPM chlorine, 10 PPM sulfate, and 14 PPM carbonates and bicarbonates: or around 58-59 PPM, right in our target range after the slight variation you'll get in even the most accurate home scales. 

By now you might be wondering why this water is so dang light, especially compared to the targets specified in the coffee industry, but this just seems to be what tea prefers: as a leaf rather than a fruit + roasted bean, it relies on slightly different mechanisms to get to the ideal brew...

Note that this single webpage cited contains background framing, description of the role of different minerals, how to go about creating the same final mix, and links to related sales offerings (you can buy a mix from them, or a form of food grade, soluble silica--best to let them describe what that's about).

The values do sound low, at least in relation to discussions I've seen about ideal ranges for calcium and magnesium content.  But then hearsay I remember from discussions from years ago is of essentially no value at all, even though a little on that is in one of those old blog posts.  Let's check a Truth Serum version (cited as an updated version in a Tea Secrets post in 2021):

These actually match more than one might expect, given the Truth Serum version is presented as including a mix designed to aggressively extract more compounds, part of the "experiencing the whole truth" of a tea version theme, versus the Tea Curious version being made for more general use.

That Tea Secrets Water Guide post unpacks how these affect outcome individually (in summary form; one gets the sense the depth of those interactions goes on and on), with a sample from that showing how it tends to read:

Hardness to Alkalinity Ratio – The ratio of hardness to alkalinity. The higher this value is, the more acidic/bright/vibrant the resulting tea is. The lower it is, the smoother/thicker/darker it is. What’s the optimum value? It depends! But probably above 0.8 and below 1.5.

Right, it's not easy to interpret without taking a few passes through it all to piece the parts together.  This is also why I would recommend starting with that podcast discussion version, to hear the backstory first, and some of these ideas explained, mixed in with a lot of other related discussion.

From here one might cross-compare this with lots of information on common bottled water sources, either mainstream often-recommended versions or else what Tea Secrets posts review, which seem to be two sets that don't overlap much.  In discussion of good alternatives in that podcast the brand suggestions didn't sound so familiar, but then I avoid drinking water from any kinds of bottles, for the obvious reason, it's just not necessary and it creates waste.  

Awhile back in those rudimentary water input review posts I compared local tap water (which had been tested for mineral content in a snapshot source I found, but which would surely vary a lot day to day) with Volvic; here is that label:

Note that parts per million is mg / liter; the metric system is just great.  What is the effect of all those bicarbonates going to be?  No idea, but this is where input from that Tea Secrets blog can pass on at least one well-informed person's opinion (the first and last table lines):

I'm still not sure after reading that, but with a good bit of experiential exposure I'm sure it would become clearer.

It's interesting how one might arrive at a too-hasty summary conclusion, contrasting a water version designed for essentially maximizing extraction (to an extent; it must not be that simple), and one designed as a broad-scope optimum.  It would be easy to think dialing up most mineral content would work better for maximum intensity, perhaps even better performance when tasting milder teas, and that lower mineral content may well be better when tea intensity is high to begin with, a very common theme in sheng pu'er experience.  Probably far more nuanced conclusions would follow from extensive experience, a mapping between those mineral input functions, a tea type and version's character, and personal preference, one that might never completely sort out. 

Comparing this to my own limited water version trial experience

Back in those two trial sessions that I wrote blog posts about, before all this was developed, two main points stood out, which could help place all this.  That evaluation and those findings are so rudimentary compared to this developed take that maybe I shouldn't go there, but why not?  It's relevant input, even if limited in scope and review input basis.  Summarized general conclusions:

1.  round by round aspect comparison results were at times just different, not clearly better or worse, since the set of aspects just varied slightly in a few different ways.

2.  in the second set of two version trials one water type seemed clearly superior in early rounds of tasting, related to overall experienced results, and not as positive as the other in the later rounds.

No need to read too much into that; it was a cursory account of very little actual experience.  But in none of the three trials cases did it summarize to "this water version was much better."  Citing an example of comments relating to #1 can clarify that:

[Alishan light rolled oolong review]

Volvic: pleasant; sweet, rich, soft, intense but not overly so. Floral range is slightly different, but the difference is so subtle it is hard to pin down.  The point here isn't to derive a flavor-list as much as specify differences anyway.

Tap water: more intense; it pops a little more. A touch more vegetal edge goes along with that, and more mineral. It comes across as creamier. These are definitely different, but again which is better would be a judgment call. The higher intensity is nice but a touch more vegetal range comes with that.

Then the next post, reviewing two different teas in two tastings (pretty good Wuyishan Rou Gui and Lao Man E sheng pu'er), based on the same water type comparison, draws the other conclusion cited:

It's crazy that two main shifts occurred in difference, varying a lot in early and later rounds.  Volvic brewed tea was clearly much better over the first two infusions, and clearly not as positive in the last, probably more even in the fourth, with tap water working just a little better in that round too.  I can't really unpack or explain that.  Somehow a shift in flavor and feel proportion across rounds favored the Volvic earlier on, and then the filtered tap water later (which probably contains a lot less mineral).

Note that a reference source testing local Bangkok water supply samples probably did point towards a general mineral distribution range my tap water might be in at any given time, with that backed off a bit by filtering it at home, but those test results were only a snapshot, so not accurate over time.

Of course I wouldn't expect that to be a universal generality, across all tea types, that this pattern would just keep repeating comparing these water versions.  It's tempting to guess that Volvic had a generally higher mineral content and extracted a lot more tea compounds earlier on, since some minerals support tea compounds extraction, then that these just weren't available for further extraction in later rounds, so tap water gave better results later on, but that may or may not be the case.  This is more just an account of what I experienced, versus a probable explanation of inputs and outputs.

I will probably follow up and try some designed-for-purpose water type in later trials, not to the extent others cited have, not in the form of years worth of experimentation.  But let's touch a little on how this relates to my philosophy on tea.

I'm not trying to optimize my tea experiences.  I don't measure dry weight of teas, although the wetted tea level in gaiwans always ends up where I expect it to.  After limited experimentation with clay pots--I only own three, probably not good versions of any type--I switched back to using gaiwans instead, which provide suitable results.  For using a thermos for brewing water limited heat loss over time would cause minor variation in brewing temperature, which I'm ok with.  I can't spend a lot pursuing the highest quality versions of teas, and I don't regret that, it's fine drinking what I get to.  I like basic, or probably above average quality and novelty of style teas, and prefer to keep the tea experiences simple.  

For reviews I won't eat any food with the teas, and try to set aside an hour or longer to just focus on the experience in order to describe it clearly.  For daily drinking I will eat a light breakfast with teas, without necessarily pairing optimum foods with versions.  I'll often have toast and tropical fruit as a breakfast (I live in Thailand), and I don't think that's necessarily ideal pairing for any tea types.  Related to ceremonial brewing forms I'm essentially Buddhist, not that I do much with most labels, but I don't see most of my tea experience as some sort of exercise in mindfulness, any more than the rest of my life.

So in conclusion I want to express much gratitude and respect to the creators of the ideas I've summarized, and maybe haven't really done justice to.  They've done fantastic work, genuinely researching and communicating findings openly.  I don't see how any tea enthusiast couldn't appreciate and relate to this work, even without it leading to their own water source exploration.  

I probably will follow up with further experiential review myself, I'm just in an unusual life phase just now, so maybe not for another couple of months, or this could be sidelined much longer.  In the end I'm most likely to drink filtered tap water, probably just in Honolulu instead of here (Bangkok), within another month, but it would still be interesting exploring this subject on a much deeper level.

Ketlee 2020 Manipur (Indian) Spring Wild White Tea

 

Reviewing the last of a set of samples of interesting and novel Indian teas, provided by Susmit of Ketlee for review (many thanks!).  Earlier posts covered trying Indian sheng (a variation of Yunnan pu'er), Sikkim and Manipur origin Indian oolongs, with this a wild origin white.  We met Susmit in a meetup discussion described here, talking about changing Indian tea styles, and shifts in awareness and demand.  I'll add his description before posting this, and get right to the tasting part.

2020 Spring Wild White Tea (the Ketlee site listing)  ($7 per 25 grams)

Grown in the wilderness of Manipur, this tea from very old plants are mother nature's gift to us. It has been one of our most awaited teas yet, after the last years harvest sold out way earlier than expected!

The tea starts floral and fruity with hints of fresh coriander leaves and spearmint. The spice notes are soothing and present just on the finish. The fruit notes are reminiscent of litchi, green apples and kiwi. The flowery notes are dominant in yellow flowers with a hint of honeysuckle. There is a hint of sandalwood in the later steeps which blends seamlessly with the flowery character. The liquor coats your mouth and is extremely silky.

It's cool the way that the general impression and flavor categories are almost a complete match with the review that follows, and not one single individual description is common to both.  That's how that goes.  I don't necessarily see that as a problem with tasting skill or a critical limitation of tasting description, it's just that any impression isn't as specific as the descriptions and associations tend to sound.  It's an interpretation, and those would always vary.  Every part of what I wrote matches in general, about complex fruit and floral range, limited spice range, freshness, transition in character across infusions, warm tones, balanced and integrated individual aspects, and thick feel.

Related to accounting for real change in tasting experience, trying a tea a year later (letting it settle due to aging), brewed with different water, using a slightly different brewing process (proportion and timing) would shift results a little.

Related to value / pricing, $14 per 50 grams is getting up there, but for tea that there is only one version of available that's hard to peg in terms of a market rate.  Related to quality level and positive experience this is probably still a good value; it's that pleasant and unique.

It goes without saying, but for people not so familiar with this scope buying "wild" origin teas is a way to ensure they are not grown using chemicals.  There is no way that a producer is out spraying down the forest to protect tea plants already thriving within a balanced ecosystem.  You have to be careful that teas aren't misrepresented, that they're not really plantation grown tea that sounds better described differently, but I'm pretty sure there is no significant risk of that from this source.  If a vendor is buying and reselling something they didn't work closely with producers to have created they might just be passing on a story they heard, not necessarily making that up themselves.  

Evaluating tea origin stories can be tricky.  Hatvala of Vietnam, Kinnari of Laos (which doesn't necessarily serve as a direct retail outlet, but are out there), and Monsoon of Thailand are other examples of vendors who work closely with local producers.  They could all be relied upon to tell accurate stories about origins, because they all played a role in both sourcing and in initiating specific forms of tea production, working directly with local residents who actually harvest and make the teas.

Review:

First infusion:  there's a touch of smoke in this; that's different.  It's so faint that I don't expect it to hang around long as infusions pass, but it does affect this first infusion quite a bit.  It's relatively positive; in the right form and balance smoke can be nice.  The tea probably came in contact with smoke, versus that being a natural taste.  Beyond that it's sweet, rich, and complex, even for the first infusion often being lighter in flavor.  

I'll do better with a flavor list next round but for now a bit of warm depth compliments lighter and sweeter fruit range nicely.  This might taste a bit like peach and vanilla, with the depth relating to mild mineral tone and aromatic wood, which is towards spice.  Feel is rich, not so unusual for white teas.  It's great how warmth, sweetness and depth all indicate some degree of oxidation and freshness and brightness show that to be limited.  A great start!

Second infusion:  I didn't try to use a longer infusion time to bump up intensity, brewing this for no longer than 10 seconds, including the pours.  Again it's interesting how this covers so much aspect scope.  I suppose going from the appearance slight inconsistency in degrees of oxidation might have led to that (more apparent in wet leaves), probably more an accident than something intentional, which worked out well in this case.  

The smoke isn't noticeable at all, replaced by heavy and complex floral tones; it would probably take at least two flowers to describe that range.  Fruit is also present but less forward and intense, now maybe more towards dried apricot.  The warmer range is pleasant, still a light warm mineral with cedar.  As if that wasn't enough there is a vegetal range of aspects that seems to extend across both, as warm, sweet dried autumn leaf tied to the warm part, and towards an actual vegetable matching the lighter tones, or let's say a lot like holy basil (tulsi, to some).

This seems a good place to guess where this is headed; I'd expect transitions to level off this round and then stay stable for most of the rest.  Even minor difference in timing would change brewed tea effect so that could seem like transitions, but not really represent that, if I shift timing a little to longer or shorter.

Third infusion:  warmth picked up; that range increased.  It's interesting how there's so much going on in this that lots of interpretations would make sense [interesting that I wrote that in notes before reading Susmit's description].  This fruit tone could be regarded as any number of things, like dried mango instead, maybe part of that heading towards pumpkin, including citrus peel (a warm version; maybe even grapefruit).  Same for the warmer range.  To me it's quite close to cedar but it's not that far from dried tamarind, or "on the other side" not completely unlike metallic range.  

The floral range is a bigger part of the story I'll probably not do justice to.  One part is rich and heavy, like lavender.  It's more complex than that though, so a hint of sweet depth might seem like rose, or the brighter range like jasmine, just a subdued and integrated version of jasmine.  On the lighter and sweeter side it's more like orchid.  That's a lot of flowers; the range really does seem to cover a lot of scope.  Still it all makes perfect sense together; it integrates.

This is interesting for being one of the more flavor complex teas I've tried in awhile.  The great versions of Darjeeling I've been trying (most from Gopaldhara) were like that but they seemed to often hit 3 or 4 notes hard, with a couple of supporting aspects, and this seems broader, maybe even to represent a range of processing outcomes.  It's not like a blend though, since they probably "got there" by mixing plant versions that may be slightly different genetically (the plants shift in genetic profile as they grow in the wild, as people change in related background over time).  The leaves could've oxidized just a bit differently, as I mentioned, but were all grown and processed together.  Those Darjeeling got the most out of one type of material processed in one way, accentuating novel, positive aspects through consistent processing, which just happened to include enough diverse supporting aspect range to give those good balance. 

Fourth infusion:  this isn't so different than last round, but it did shift a little.  The fruit picked up, evening up with the floral, or maybe moving past it.  It tastes a little like juicyfruit gum, unusually bright and sweet.  I probably did accidentally brew this a few seconds faster, causing warmer range to drop back, and it probably would've shifted a little without that.  

This is the exact opposite of white teas that suffer from either being too subtle, not tasting like much, or not spanning much range, just covering some warm cinnamon with a touch of floral (for shou mei character), or bai mu dan that's floral and sweet with a good bit of melon, and that's it.  The only way that someone wouldn't like this tea is if they are really into other aspect range and aren't flexible about that.  That's wouldn't be as negative as it might seem; someone could get really hooked on sheng pu'er and only make exceptions for some oolong or black tea range, and that would still be plenty to experience.  And I suppose not everyone loves fruit range in teas like I do.

Oddly a Thai Oriental Beauty variation I tried this year isn't so far off this.  It always had reminded me of a white tea more than an oolong, and was heavy on cinnamon and floral tones, maybe with a touch of citrus or light fruit mixed in, but not covering quite as much range as this.  You would think it would've been much more oxidized, matching that typical style, but it really wasn't.  For being moderate cost tea I thought that was great for quality level and novelty, it just didn't land in the typical OB style range.  Then this part is crazy; I liked that tea even better mixed with another white tea to stretch out the range and give it a bit more depth of feel.

Fifth infusion:  I'll leave off here; it's enough to say about one tea, and later transitions won't change an overall impression by much.  Of course this tea will have good durability and will make another 5 very pleasant rounds, using limited timing and high proportion to prepare it.  Balance is just great, and it's still evolving.  The warmth now reminds me of cinnamon, with fruit on to more like cooked peach.  The earlier floral range gave way to that fruit, and the earlier cedar aromatic aspect faded to become that spice, but hints of both give it great depth (in the sense of complexity).

I've said that the feel is rich but I can add to that; in addition to feeling full there is just a hint of dryness that gives it a pleasant edge.  Aftertaste experience is nice, with parts trailing over to make the experience seem longer, for including an extra dimension.  This probably is one of the better white teas I've ever tried.  I expected it to be novel and pleasant but not this complex and refined.  They really nailed it, which had to start with high quality material suited to this plant type.  The best version of a wild source white Monsoon Thai white tea included some similarities, just not quite the same positive result across so much scope.  I thought that was really good even for that; white tea doesn't usually compete with good Darjeeling and fruity oolong for complexity and intensity.

I've not really said where the level of sweetness stands, but explaining a lot about floral and fruit range implied it was above average in level, as I experienced it.  That makes a lot of difference in final effect.  It's hard to imagine a tea including too much natural sweetness, although I suppose that's possible, but if the level is lower that causes tea aspects to not tie together as well.  For a tea like this it could swap out some sweetness for some savory range, like sundried tomato, and that would still work, it would balance.  In saying that this kind of resembles sundried tomato a little too, but to me this last round is closer to dried persimmon, which is related but different.  In between plum and prune works to approximate that, if dried persimmon isn't familiar.

All in all really nice tea.  I expected it to be pleasant but not like this.