Steamed, Not Pan-Fired: Why Japanese Green Tea Tastes Unlike Chinese
Put a cup of sencha next to a cup of longjing and they do not look like relatives. The Japanese cup is opaque and almost grass-green; the Chinese one is pale, clear, faintly yellow. The Japanese one smells like seaweed and cut grass. The Chinese one smells like toasted chestnuts.
It is tempting to explain that with terroir, or cultivars, or centuries of separate tradition. All of those are real, and none of them is the main reason. Both cups come from Camellia sinensis. Both were picked green and neither was allowed to oxidise. The gap between them opens in a window of a few hours after harvest, in a single step, and it is a step you can describe in one sentence: Japan stops the leaf with steam, and China stops it with a hot pan.
That step has a name in the trade — sassei (殺青), literally "kill-green" — and everything downstream follows from it. Colour, aroma, how much astringency ends up in your cup, even the shape of the dry leaf.
What makes it worth a whole article is that the neat version of the story is wrong in two directions. Japan makes pan-fired tea too. And the steamed default that now defines Japanese tea is younger than the United States.
The one step, side by side
| Steamed (Japanese default) | Pan-fired (Chinese default) | |
|---|---|---|
| How the leaf is stopped | Live steam, measured in seconds | Contact with hot dry metal |
| Heat travels through | Water vapour — fast, even, all surfaces at once | The pan surface — slower, uneven by design |
| Dry leaf | Dark green, needle-shaped | Olive to yellow-green, flat or twisted |
| Brewed colour | Green, often cloudy | Pale gold, clear |
| Aroma | Marine, vegetal, fresh-cut | Toasted, nutty, chestnut |
| Astringency | Higher — the tannins survive | Lower — measurably so |
| Typical teas | Sencha, gyokuro, tamaryokucha | Longjing, biluochun, and most Chinese green tea |
If you only take one row: steam preserves, dry heat transforms. Steaming is a brake. Pan-firing is a brake and a cooking step at the same time, which is why one tastes like the field and the other tastes like the kitchen.
Why the step matters so much
A tea leaf starts changing the moment it leaves the bush. Enzymes inside the leaf begin oxidising its own compounds, and if nobody interferes, that process runs all the way to oolong and then to black tea. Green tea is simply tea where somebody interrupted it early.
Heat is the interruption. Enzymes stop working above a certain temperature, so the whole craft of green tea comes down to getting heat into the leaf quickly enough, evenly enough, and then stopping.
Steam does this through water. Water vapour condenses on every surface of the leaf at once and dumps its heat there, so a Japanese steamer can neutralise a batch in well under a minute with almost no variation between the leaf at the top and the leaf at the bottom. Nothing scorches, because nothing gets hotter than the steam.
A pan does it through dry metal. Contact is uneven by nature — the leaf touching the surface gets far more heat than the leaf above it, and the operator keeps everything moving to spread that around. It is slower and it is a skill. It also means the leaf surface reaches temperatures where sugars and amino acids start browning, which does not happen in a steamer at all.
That last sentence is the whole flavour difference. Pan-fired tea has been cooked, lightly, in a way steamed tea never is.
What the difference measures out to
Saga Prefecture's tea research station ran sencha, guricha and Japan's own pan-fired tea through a taste sensor and a compositional analysis, which turns the usual hand-waving into numbers.
Setting sencha at 100:
| Sencha (steamed) | Guricha (steamed) | Kamairicha (pan-fired) | |
|---|---|---|---|
| Catechins — the astringent group | 100 | 104 | 72 |
| Amino acids — the savoury/sweet group | 100 | 106 | 82 |
| Taste-sensor reading | reference | astringency and umami both level with sencha | astringency clearly lower, umami level |
Pan-firing takes roughly a quarter of the catechins out of the equation while leaving umami close to where it started. That is not a matter of taste description. It is why people who find Japanese green tea harsh often get on with Chinese green tea immediately, and it has nothing to do with them having a more refined palate.
The aroma side came out just as cleanly. The pan-fired samples carried a group of low-boiling-point aroma compounds — nine of them identified, including dodecane — at more than twice the concentration found in sencha. Those compounds are the kamaka (釜香), the "pan aroma", and they are not present in steamed tea in any meaningful quantity because nothing in a steamer ever gets hot enough to make them.
Japan makes pan-fired tea, and almost nobody drinks it
Here is where the tidy version falls apart.
Kamairicha — Japanese pan-fired green tea — exists, is sold commercially, and has done so for centuries. It is concentrated in Kyushu, at the southwestern end of the country. Miyazaki produces 72% of it, Fukuoka 16%, Oita 6%.
The numbers behind those percentages are the point. Miyazaki's share is 154 tonnes. That puts national production of pan-fired tea at somewhere around 210 tonnes a year — against a national crude tea output of 75,200 tonnes. Pan-fired tea is less than three-tenths of one percent of what Japan makes.
For comparison, that is about one twentieth of the tencha crop, the leaf that gets ground into matcha. It is a rounding error in the national figures.
Which produces a situation worth stating plainly: most people in Japan have never knowingly drunk Japanese pan-fired tea. It is not on supermarket shelves outside Kyushu, it is not what arrives when you order tea in a restaurant, and it is not what a Japanese household keeps in the tin. A visitor could spend a month in Tokyo drinking tea daily and never encounter any. The "Japan steams its tea" shorthand is not a simplification of a complicated picture — for practical purposes it is just true, and the exception survives as a regional speciality rather than an alternative.
The steamed default is younger than you think
The second thing the tidy version gets wrong is the timeline.
Kyoto Prefecture's official chronology of Uji tea puts the invention of aosei sencha seihō (青製煎茶製法), the steaming-and-rolling method that produces modern Japanese green tea, at 1738, in Ujitawara, southeast of Kyoto. It spread from there, and the major producing regions had adopted it by the end of the Edo period.
Everything Western drinkers now recognise as Japanese green tea sits after that date. Gyokuro was developed in 1835. Japan's tea export trade begins with the opening of Yokohama in 1859 — meaning the tea that first reached foreign markets was already the new steamed style, which is a large part of why "Japanese green tea" and "steamed green tea" became synonyms abroad.
Before 1738 the country drank other things by other methods: powdered tea whisked in the bowl, arriving from Song China in 1191, and steeped leaf tea, arriving from Ming China in 1654. Japan has been making tea for well over a thousand years. It has been making this tea for fewer than three hundred.
That reframes kamairicha usefully. It is not a curiosity that broke away from the Japanese norm. It is closer to the older approach, still being made in the region that received Chinese methods first, in a country where a very good eighteenth-century innovation happened to win everywhere else.
Guricha: the tea that is both
There is a third category that confuses the neat split further, and it is worth knowing because you will meet it on labels.
Tamaryokucha (玉緑茶), usually sold as guricha, comes in two versions: a steamed one and a pan-fired one. The steamed version is by far the larger — Saga produces 31% of it, Nagasaki 29%, Kumamoto 20%, for a national total near 1,700 tonnes.
What makes it distinctive is not how it was stopped but what happened next. Sencha is rolled hard against a surface, which is what stretches the leaf into needles. Guricha skips that final shaping step and lets the leaf curl naturally, so it dries into small comma shapes. Saga's own summary of the difference is blunt: between sencha and steamed guricha, only the last shaping step differs.
The composition data agrees. Steamed guricha came out level with sencha on catechins and amino acids alike — 104 and 106 against 100. If sencha tastes like Japanese green tea to you, so will steamed guricha. The pan-fired version is the one that tastes different, and it tastes different for the reasons in the section above.
So the honest version of the map is: the steaming decides the flavour, and the rolling decides the shape, and Japanese tea uses both variables independently.
What this means when you are buying
Most of this is background until it changes a purchase. Where it does:
If Japanese green tea has tasted harsh to you, the problem may be the category, not the brewing. Catechins are 28% lower in pan-fired tea by measurement. Before concluding you dislike green tea, it is worth finding out whether you dislike steamed green tea — those are different findings with different fixes.
Do not go looking for kamairicha as a first purchase. At 210 tonnes a year it barely leaves Kyushu, and outside Japan the labelling is unreliable enough that you may not get what you paid for. If the roasted, toasty end of the spectrum is what appeals, Japan's widely available answer is not pan-fired tea at all — it is hojicha, green tea roasted after processing, which is stocked everywhere and costs very little.
Sugimoto Tea
Kuki-hojicha — roasted green tea stems, loose (Shizuoka, 227g)
Hojicha is green tea roasted until brown: toasty, low in caffeine, and hard to brew badly. This one is made from roasted stems rather than leaf, which gives a lighter, sweeter cup than leaf hojicha. It is a bulk size, which suits how hojicha is actually drunk in Japan — by the pot, in the evening, without ceremony. Grown in Shizuoka.
Guricha is a safe buy if you like sencha. It is the same tea with a different shape, and the shape has one practical consequence: the curled leaf packs differently, so a roomy basket infuser serves it better than a narrow one.
Chinese green tea is not a lesser version of Japanese green tea, or the reverse. They solved the same problem — stop the leaf before it oxidises — with the tools each had. One preserved the field; the other added the kitchen. Preferring one says nothing about either.
The short version
- Japanese and Chinese green tea differ mainly because of one step: Japan stops the leaf with steam, China with a hot dry pan
- Steam preserves the fresh, marine, vegetal character. Dry heat browns the leaf surface and adds toasted aroma that steaming cannot produce
- Measured against sencha at 100, Japanese pan-fired tea carries 72 catechins — noticeably less astringency, with umami intact
- Japan does make pan-fired tea, but at roughly 210 tonnes against 75,200 tonnes nationally — under 0.3%, concentrated in Kyushu, and unknown to most Japanese drinkers
- The steamed style dates to 1738. It is the default, but it is not ancient
- Guricha is steamed like sencha and only shaped differently — the composition is effectively the same
Production figures are from the Ministry of Agriculture, Forestry and Fisheries, "Current State and Issues of the Tea Industry and Tea Culture" (November 2024), citing the National Federation of Tea Producers' Associations; figures are for 2023. The catechin, amino-acid and aroma analysis is from Saga Prefecture's tea research station, "Kamairicha and Tamaryokucha: Saga's Distinctive Tea Types". Historical dates are from Kyoto Prefecture's official chronology of Uji tea.
If you want the steamed side in detail, read what steaming time does to sencha — the same variable, taken further. For the roasted alternative that is actually easy to buy, see hojicha explained, and for where each type is grown, where Japanese tea comes from.
Hiromu Wagatsuma
Based in Japan. I run cross-border e-commerce stores for Japanese brands, and I write about what I actually use and taste here.