This is all a bit much. I've been meaning to test the effect of using different water versions on tea for awhile, it was just always too much messing around to get to it. Now that I finally did results were interesting, but one part related to sample type selection could have went better.
In retrospect it was odd using teas I hadn't reviewed before, and teas of this quality level for this purpose. Cindy's Wuyi Origin Rou Gui versions are about as good as that type ever gets, and the fruitier version is a strong contender for my favorite tea among all types and versions. I guess the reasoning was that using good tea is going to tell more of the story related to what input the water has, versus using teas that are of medium quality, and trying to extrapolate results to likely outcome if the teas had been better. It was also about just using interesting versions that I had around.
The same concern and atypical theme applies for the King Tea Mall Lao Man E sample; this is a gushu version from a respected, in demand origin area. It's questionable judgement, shifting focus off the tea onto the effect water mineral content variation has on preparing it. One might argue that both can be covered at the same time, but per the actual experience a lot gets lost.
About water type / input testing in general, it had always seemed next to impossible to identify what a really good or ideal source of water would be. Really doing that process justice would involve trying lots of water types, and then probably folding in how that varies across at least a few different tea types. Online discussions of water mineral levels as an input have been interesting, and standard patterns of opinions do emerge, but nothing that works well as starting point guidance.
Avoiding high levels of calcium is generally advocated, for example, but my own interpretation of what would be low, medium, and high based on sampling lots of hearsay may or may not be remotely reliable. I mentioned some advice passed on by Peter Jones in covering some theory in this post, which was really about trying a Myanmar Kokang shu version:
We are still several months away from any conclusions. But basically you want your alkalinity to be 1/2 your total hardness, a pH in the 7s, and a balance of 2/1 of Ca to Mg. The other positive ion mineral salts also play a role in extraction...
The form I finally arrived on--definitely not a final word on a practical, functional approach--was to try two completely different teas using Volvic water (which gets mentioned as some people's personal favorite), compared to the filtered local Bangkok tap water I use at home. It was a stretch but I picked two relatively random, opposite type teas I happened to have on hand, a Rou Gui and a sheng pu'er. It turned out to be not the best selection because I didn't factor in the high level of bitterness in the sheng version (a Lao Man E); kind of an obvious mistake to make, in retrospect. Trying an already familiar, milder version would've made a lot more sense.
I can't describe what the local water is like in relation to lots of other sources, but of course I have tried the same teas prepared using it and also other spring water, RO water, and whatever is being used in other places. My friend Sasha had a favorite spring water he used, for example, and some of those teas I had tried at home, a number of versions tasted there many times. A vendor at an expo had been using RO water, and I re-tasted the same teas I tried with them again at home within a couple of days, and I try teas in shops sometimes. To really isolate differences trying the tea prepared together is all but essential; the point here is that how differences in water source work out in general, and how the local water compares, isn't completely unfamiliar.
Bangkok municipal water had been truly awful, years back. It varied so much even the color and smell could be off. We used bottled spring water for years, or at least local water sold as such, which really could be different things. As of a few years ago we switched over to using a multi-stage filtration system, and the water seems fine. Per online hearsay the tap water here is based on rainwater sources instead of natural spring origin, so it's low in mineral content, which makes sense given that it rains a lot and we're not exactly near a mountainous or even hilly area. Per other hearsay local distribution systems are a main problem for water quality, the pipes from there to houses, not so much what the processing centers put out, or even use of old pipes in your house. It probably helps to live right beside one of the main Royal palaces in Bangkok as we do; infrastructure seems pretty stable and well maintained.
One might expect the water quality and mineral contents to vary based on lots of local factors, and I'm not certain that it doesn't. I've included a number of captures from the local Bangkok real-time water quality monitoring system, from when I tried these teas, but of course those don't include a measurement of contents broken out by mineral grams / liter at any given time.
It would make the most sense to run this type of comparison multiple times, evaluating tea types one at a time. I didn't do that, I tried two types together. That I still don't consider a mistake; I've been training to do comparisons that make sense, or else those that don't instead, for about three years now. I've tried hundreds of teas over uncountable combined tasting sessions. It's familiar ground. If someone without that odd tasting habit wanted to get the clearest results I'd definitely recommend using the more intuitive approach, trying the teas one at a time, perhaps using 3 or 4 types of water to add review complexity. Selecting mild, well-known versions of teas would probably work best, regardless of type; that part was a mistake.
I'll probably try this again with a rolled oolong version to see how it works with a very mild but complex tea, which should also help to isolate feel effect even better.
Review, comparison of Wuyi Origins Rou Gui with two water types:
Rou Gui with Volvic: just amazing tea! I'm not sure if that's the water though. This didn't say if it was fruity type or cinnamon Rou Gui but judging from the heavy, warm cinnamon taste it's that. Flavor is intense, feel is thick, level of roast is perfect, slightly noticeable but only complimentary.
Rou Gui with filtered tap water: this is similar but the flavor may pop a little less. It seems to be drawing out a lightly different vegetal note, as if the warm mineral resembles celery just a little. Cinnamon is slightly flattened out. This difference is so minor that tasting across different times it wouldn't really stand out as much, but side by side it does quite a bit. Sweetness is down just a little too. It's the same great tea, almost the same flavor profile and feel, just a little different.
 |
an early round; color differences didn't seem to tell a lot of story, so I'll pass on adding more about that |
Second infusion:
Rou Gui with Volvic: this tea is just perfect as cinnamon intensive Rou Gui goes. The balance, intensity, all the profile taken together; this is it. The cinnamon part really jumps out. It's warm and rich, almost connecting a little with a very subtle roast level, or maybe it only seems that way.
Rou Gui with tap water: it's almost the same experience but it is missing some of the one high point that makes the other tea experience so special, the brightness, sharpness, and intensity of that cinnamon flavor. The deeper tones and supporting elements carry across fairly well; fullness and body isn't bad, kind of similar. The first version might have a slightly heavier, oilier feel, which is more positive. It's a quite similar tea experience but funny how those parts that seem most interesting and critical are what flattens out.
It's too early for conclusions but the results already seem clear enough; this local water doesn't work as well. I don't foresee buying bottled water to drink tea with though. I wouldn't throw out a few large plastic bottles a week to get this effect difference, or probably even one 1.5 liter bottle a week to review using a different version. It's a shame the local water isn't better but adding expense and impacting the environment to drink slightly different versions of tea isn't going to work out. Maybe a charcoal treatment effect will help; that idea comes up a lot.
Later edit: it was too early for conclusions; the opposite kind of trend turned up in the fourth and fifth infusions, a trend that was much more pronounced in the sheng version.
Third infusion:
Rou Gui with Volvic: This tea is probably evolving a bit in character but I've lost track of clear prior impression to describe that. It's still great, probably with a slightly shifted balance of aspects.
Rou Gui with tap water: it's great with this water too, with less difference standing out than had initially. It's still not quite as intense, giving up just a little sweetness, and more so dropping out just a little punch in the warm cinnamon aspect. There is a very minor difference in the mineral tone, so subtle it would only come across in direct comparison, but it doesn't work to describe it or judge how negative it is. I think it is slightly negative versus neutral, but that's a judgement call at this point, as much just a difference.
Fourth infusion:
Rou Gui with Volvic: a slightly dry underlying mineral in this might have increased this round, it just doesn't stand out as much in this as it had for the sheng at this stage, or take exactly the same form. The tap water version is just as good this round, or perhaps even slightly better. So strange; I have no idea why that would be. I would expect effects to shift a lot for increasing brewing time in later round but this isn't really onto that as much as it would be in two more infusions.
Rou Gui with tap water: still really exceptional, not losing anything for intensity or positive range. I think depth is picking up, that the one-note emphasis on cinnamon is deepening to include a lot of other warmer tone range, that it balances even better. It was great the other way too though, focused more on that one flavor aspect.
Fifth infusion:
I'll lengthen infusion time slightly but these teas are holding up; adding a lot of time for that wouldn't be positive. I'm brewing these over 10 seconds now, for what I'd take to be optimum, maybe around 15, but it goes without saying that I'm not timing this, since I just don't.
Rou Gui with Volvic: mineral is a bit much in this, a dry version of it. Looking at the mineral profile calcium and magnesium seem to be at the level I'd expect to work best, based on following online discussions about that (12 mg and 8 mg, respectively). It's not "hard" water, but that should be enough to play a standard role in extraction and setting profile. The rest, the total dissolved solids value, would be a bit much; there's lots of other minerals in this. This is still really pleasant, and a very nice citrus aspect is ramping up in this tea, a "chen pi" dried mandarin peel type effect.
Rou Gui with tap water: slightly better than the Volvic, for not including so much of that dry mineral effect. The rest of the profile is similar. The citrus part may be at a slightly lower level, so not as intense or positive, but the overall effect is better for not tasting as much like that dry version of mineral. Mineral is great in tea, in the right balance; it grounds the experience. Older plant source teas and natural origin teas seem to express a lot of that, an underlying mineral element supporting overall complexity. It's just slightly nicer in this tap water version for balancing better, in this particular infusion of this tea, per my judgement. The difference in extracting other flavor range hasn't dropped out but is has shifted in balance; it seems to matter less.
I didn't note differences or aspect range for feel or aftertaste much for these; that didn't seem to vary as much as flavor related to the water type input.
 |
a later round; adjusting infusion times would change effect of color difference |
Lao Man E sheng water source input testing:
King Tea Mall 2018 Lao Man E sheng gushu with Volvic: bitterness stands out; to be expected. One could learn to love that effect, even at this level, but I haven't. If it balances well with the rest I can appreciate it, even like it, but this is a bit much. The tea seems to obviously be good tea; all the other markers for quality work out, thick feel, clean flavors, depth and complexity, lingering aftertaste. That front end bitterness really defines the experience though, then it trails on, not dropping that much intensity for awhile. At least sweetness goes along with it, and there is plenty of other range. I'd expect it to drop off a little by round 3 or so but I'm not sure how far past that I'll get.
Lao Man E with filtered tap water: a touch of metallic taste seems to be added to the rest. That extended aftertaste is going to make this a terrible choice for combined tasting though, about as poor a choice as one could make. I might have considered that, but it's not like I'm drinking Lao Man E all the time. Rinsing your palate with plain water (not warm, not cold) works, but it's a lot to rinse in this case. Making mistakes to learn from can be a good thing, it's just more pleasant when they're less obvious, less stupid mistakes.
In general the flavor profile seems to pop a little less, again, with that mineral difference the main change. Sweetness may be slightly reduced but bitterness is dominating the flavor profile anyway, so it would stand out less.
Second infusion:
Lao Man E with Volvic: bitterness is easing up; it's balancing a little better. It won't be until the next round this approaches a more conventional sheng profile though, more evened out. The point here is the difference anyway.
Lao Man E with tap water: beyond a touch of mineral difference in this (not really a positive one; a touch of metallic flavor) and a slightly reduced fullness of feel it's similar. The other Volvic brewed version comes across as slightly sweeter. In the case of the Rou Gui it's about a shift in the best attributes being slightly dimmed, but the character isn't so different beyond that. In the case of this it comes across differently. That shift in underlying mineral, slightly less full feel, and slight reduction in sweetness seems to add up to change the tea character more. They're only small shifts, and the basic aspects are the same beyond that, the general feel, flavors range, aftertaste, feel, but the effect is more of a change, or a change at a deeper level. Again the Volvic brewed tea is just better; there was no trade-off that benefited using this local Bangkok tap water, at least not yet at this stage of infusions.
Third infusion:
Lao Man E with Volvic: this is much better balanced for transitioning over those rounds. Not in comparison with sheng that doesn't emphasize bitterness, if someone doesn't like bitterness, but I'm fine with it at this level, it's just definitely not a close match to personal preference. The thick feel is nice, how clean this tea is, how the different layers of aspects work together. One part under that is actually a little floral, drifting a little towards fruit, but deeper light mineral is really stronger.
Rich, full feel is nice. Aftertaste works a lot better for dropping off. Usually people would couple those last two descriptions with where the feel impacts your mouth or throat, with aftertaste also related to a lingering feel. I'm never so good with all that, in part for not valuing it. It probably is as much a valid quality marker as what I do pick up more, but it seems a developed preference that would be as natural to never develop.
Lao Man E with tap water: this finally works about as well as the other version. It gives up a little thickness in feel but somehow the fruit and floral range (complex, but subtle, hard to define) is standing out a little more in this. It might be working out as well or even better for not stripping out as intense a version of that bitterness. Feel losing a little thickness isn't as positive.
The floral and fruit range I can try to define, to describe, but it would be a guess: it's bright and sweet, so along the line of plumeria and dried mango. There are lots of types of dried mango, so those tend to vary a lot, but that gets complicated to wade into that. The brighter, sweeter, more citrus oriented kind. In retasting both that bitterness hitting a little harder in the Volvic version isn't quite as positive, to me. It may make for an exception that stripping out a little less intensity wouldn't always be a bad thing.
Fourth infusion:
Lao Man E with Volvic: it's really strange how this is exhibiting a stronger mineral aspect, a slight metallic range. It's almost as if I've mixed up the water sources, but then this probably does have more mineral content, so maybe that makes sense. I have no idea what that means.
Lao Man E with tap water: it's nice how that balance evened out. Bitterness is still pronounced but the proportion is more relatable now. The complexity is nice, the range of flavor, and how feel and aftertaste contribute to overall effect.
Fifth infusion:
Lao Man E with Volvic: a dry mineral input is much stronger in this tea version. It tips just a little towards how chlorine comes, that's just not it. It's much more positive to say it tastes like limestone, and to be fair it's really in between the two.
Lao Man E with tap water: there is less mineral effect in this, making it come across as sweeter in character, and better balanced. I don't know that it is actually sweeter; that may just be how it seems, without as much of the mineral input changing things. The tap water version was actually slightly better over the last two infusions, but to be fair the difference is a bit minor.
Conclusions:
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 the filtered tap water later (which probably contains a lot less mineral).
It's worth noting that different people can and do use completely different timing sequences related to brewing different rounds to different infusion strengths. It's more common to use lower proportions and longer times, and then to extend the timing more across later rounds than I tend to. Why that is becomes a lot to account for, and beyond citing personal preference some of the rest wouldn't be clear. It's hard for me to account for why people like tea prepared differently than I do; to some extent I suppose I never could.
It could be interesting to consider the water mineral content though, starting with the Volvic listing:
I'm kind of guessing, since I have been involved with plenty of discussions about water content but haven't memorized ranges, but the calcium and magnesium levels look ok, if maybe even a little low. Per discussions those two are the main factors, with a moderate amount improving results, and high levels throwing them way off. Looking back through old discussions and references, Rie Taluli did a review post comparing tasting results using of a number of types for green tea, and in a discussion comment of that post Peter Jones (who is into reviewing mineral effect themes) cited listings of standard mineral content of a number of brands:
EVIAN - Natural spring in France - 290mg/L sodium bicarbonate, 81mg/L calcium, 27mg/L magnesium, 6.7mg/L sodium.
AQUAFINA (Pepsi), DASANI (Coke) - Municipal water through RO - no sodium bicarbonate, no calcium, no magnesium, no sodium.
GLACEAU (Coke) - Municipal water through RO - no data, but unlike Dasani, they then add back in for flavor the calcium, magnesium, and bicarbonate.
FIJI WATER - Natural spring in Fiji - 153mg/L bicarbonate, 18mg/L calcium, 14mg/L magnesium, 18mg/L sodium.
NESTLE PURE LIFE - Municipal water through RO - no bicarbonate, 7.4-11mg/L calcium, 2.9-5mg/L magnesium, 4.4-9mg/L sodium.
ARROWHEAD (Nestle) - Natural spring in Arkansas - no bicarbonate, 4-66mg/L calcium, 1.4-19mg/L magnesium, 3-17mg/L sodium.
I can't imagine that using Evian would work well, and it's odd that Arrowhead water is so inconsistent, but I'll leave off saying much more about the rest. I do come back to the sodium level idea related to the local water version level.
About the local water description, some input is offered by Bangkok real time tap water quality monitoring (http://twqonline.mwa.co.th/EN/map.php?type=). We live next to the Chitralada station, in the center. These don't mean that much to me but they could to someone else:
 |
free residual chlorine levels, Nov 9, 2019, 1 PM |
 |
chloride levels (mg / L) |
 |
turbidity (NTU) |
 |
ph |
 |
salinity, listed as g / L (seems wrong, too high?) |
 |
July 2019 ph capture (to show variability) |
I don't know what was in that water, for specific mineral content, and I don't know what the filtration process would have removed. Chloride level was around 4 mg / l in this water, versus 15 mg / l in Volvic, but what of that? Salinity just seemed wrong; those readings would seem like way too much salt, in the 150 mg / l range. What if that is right; shouldn't the water taste like salt? As a baseline I checked how much is in seawater:
On average, seawater in the world's oceans has a salinity of approximately 3.5%, or 35 parts per thousand. This means that for every 1 litre (1000 mL) of seawater there are 35 grams of salts (mostly, but not entirely, sodium chloride) dissolved in it.
So 35,000 mg / l, which still doesn't answer if I could taste 150 mg, but it's a context of sorts. It occurred to me that salt is sodium chloride, so unless I'm missing something the two values don't match up, the measured amounts of chlorides and salinity. That prompted me to look up answers to both questions, how much salt can one taste, and what are normal levels for these. This WHO reference covers the first:
Taste thresholds for sodium chloride and calcium chloride in water are in the range 200–300 mg/litre (2). The taste of coffee is affected if it is made with water containing a chloride concentration of 400 mg/litre as sodium chloride or 530 mg/litre as calcium chloride (3).
I don't think we can take that second sentence as good input about coffee enthusiast perspective. More input from that source pins down range of chloride in typical fresh water:
The mean chloride concentration in several rivers in the United Kingdom was in the range 11–42 mg/litre during 1974–81..
That still doesn't resolve the monitoring inconsistency between salinity and choloride levels, but it is interesting to compare that to sodium levels in the other bottled waters.
Further conclusions:
That last section was mislabeled, wasn't it; not much for conclusions in there, or even speculation.
I suspect this tap water version ended up with very little mineral content in it, due to starting out with limited amounts and filtration removing some. It's interesting to consider that tea producing more positive infusions initially might generally relate to producing less positive infusions across later rounds. At least that did happen here, per my interpretation, and it may work out as a somewhat uniform generality (or may not; that goes without saying). This could explain why some people push teas to produce very long counts of infusions, and others tend to not like them as much after a much shorter sequence, although simple preference difference seems to be just as relevant an input.
Then again, who knows. Round after round of trying out lots of water and tea versions would tell a longer, more accurate story. In online discussion it's not uncommon for people to claim that they prefer different water (mineral content, presumably) for different tea types, and that would intuitively make sense.
I've never ran across much detailed speculation of why soaking bamboo charcoal in water is said to improve the results for making tea with it, even though people advocating that practice comes up a lot in discussion. Even when it's being sold for this use, as in the case of this Yunnan Sourcing version, there's not necessarily a reason for why it might make any difference. It probably does, it would just be interesting to hear guesses about why. Here is one that Google search turned up, Tony Gebely's input, but somehow it's not satisfying as a complete and plausible explanation:
The idea behind bamboo charcoal, and any charcoal for that matter is the fact that it is extremely porous and will absorb impurities in water. It really isn’t “filtering” water if you set a stick in the water, in order to filter the water, the water must pass through granulated charcoal or some other medium (this is how Brita filters work)...
It would seem that it could be both adding and removing minerals from the water, and that only testing, instead of speculation or traditional input, would shed light on this. I would tentatively accept that it really does improve results, although that would be clearer to me if I tried it myself.
No comments:
Post a Comment