Japan measured how much miyeok swells, and got three different numbers

A 25-gram bag of dried miyeok costs almost nothing and makes an alarming amount of soup. Every English page that sells it or cooks with it says the same thing: it expands about ten times. None of them says ten times what, and none of them cites anything.

Japan’s government measured it. The answer is not ten. It is three numbers, and the gap between them is the whole story.

The figure everyone repeats, and the four versions of it

Here is what the English-language web actually says, in the pages I opened.

Science Insights states that dried wakame “will expand to roughly ten times its dried volume.” The page cites nothing for this, and nothing for its other claims either — it refers to unnamed “research” and “studies” without a single link.

Taste Korean Food says it “expands up to 10 times its dried size,” which leaves open whether size means weight or volume. The same page gives the soaking time three times and gets a different answer each time: “about 20 mins,” then “15–20 minutes is usually enough,” then “20–30 minutes.”

Porto Muiños, a Galician seaweed producer, is the only source I found that commits to weight: “Wet, they increase their dry weight by a factor of 6.” No testing laboratory or method is named.

And one retailer manages to say both on a single page. SushiSushi’s listing for Tokon cut wakame claims in its bullet points that the product “Expands 10–12× its dry volume,” then in the specification table directly below gives “Rehydration Ratio — Approx. 10–12× dry weight.” Volume and weight are not the same quantity, and the page uses the same number for both.

Now the sources that know the most, and say the least. The Tsuji Culinary Institute in Osaka has a page devoted specifically to the technique of rehydrating wakame. It gives no multiple at all — it says “Soak in plenty of water” and “After soaking for 20-30 minutes it should look like this,” and shows a photograph instead. Just One Cookbook, which describes itself as the largest English-language site for Japanese recipes, also gives no number: “Dried wakame expands a lot, so start with a pinch or two.”

The pattern is consistent. The pages that quantify it cite nothing. The pages that would know decline to quantify it. It is the same failure that produces the “87% perilla ketone” figure repeated about perilla leaf — a number detached from the method that produced it, passed along until the condition is gone and only the digit remains.

Japan published the number, and published the conditions with it

The Standard Tables of Food Composition in Japan include a table almost nobody outside dietetics opens: Table 12, the table of cooking methods and weight-change ratios. For each prepared food it records what was done to the sample, how much water was used, and what the mass became afterwards. The ratio is defined in the same document as the mass of the sample after cooking divided by its mass before, times 100.

Three wakame entries carry a measured ratio (Table 12, p. 52):

• 09041, dried wakame, sun-dried, water-rehydrated — steeped 8 minutes in 100 times its weight of water, then drained: 590%
• 09058, cut wakame, simmered — cooked in 100 times its weight of water and drained: 1173%
• 09057, blanched-salted wakame, desalted then boiled — desalted in 20 times its weight of water, then boiled in 3 times the desalted weight: 250%

Two further wakame entries, 09043 (ash-dried, rehydrated) and 09046 (stem wakame, desalted), appear in the same table with the ratio column left blank.

So plain dried wakame, soaked the way most recipes describe, gains 5.9 times its weight — not ten. The figure closest to “ten times” is 1173%, which is 11.7 times, and it belongs to cut wakame that was simmered in boiling water rather than soaked.

That makes the retailer’s “10–12× dry weight” claim for cut wakame an oddly good number. The same document explains why the product is a different food: cut wakame, it notes, is made by washing blanched-salted wakame in brine, machine-drying it, cutting it to size and sealing it in bags. It begins from a different raw material than sun-dried wakame, and it has already been heated once. “Ten times” looks like a real measurement of one product that has been copied onto all of them.

The one measured figure has an unstated variable: water temperature

Table 12 classifies water-rehydration as a non-heated preparation, alongside soaking in water, salting and bran-pickling. It never states the temperature of the water.

That matters, because the English pages disagree about it loudly. Science Insights is emphatic: “Use cold water, not hot,” and “Hot water makes wakame mushy and slimy.” Asia Market instructs the opposite — “Replace the water with hot water at around 80 degree.” The Tsuji Institute, the one actually teaching the technique, specifies neither.

The single official figure measured with its conditions attached is missing the one condition the cooking sites argue about most.

Japan lists nine wakame foods; the English web lists one

In the Japanese tables, “wakame” is not one entry. It is nine separate foods with nine separate codes: sun-dried (09040) and its rehydrated form (09041), sheet wakame (09042), ash-dried rehydrated (09043), cut wakame dry (09044), blanched-salted desalted raw (09045) and blanched-salted (09048), stem wakame (09046), and mekabu, the sporophyll (09047).

Korea’s national table splits it differently and just as finely — raw, dried, powdered, salted, salted-and-blanched, stem, farmed raw, wild raw, and deep-fried. Its 2006 edition records the moisture content of dried miyeok (12 035) at 16.0 g per 100 g and of salted miyeok (12 039) at 62.4 g. Their sodium figures are 6,100 and 10,081 mg per 100 g — and the table prints the first in parentheses and the second without, which in its own notation means the dried-miyeok sodium value is an estimate and the salted one is a measurement.

These are not one ingredient under different labels. A figure measured on one of them is not a figure about the others, and a page that says “wakame expands ten times” has not said which of the nine it means.

Soaking throws away most of the potassium

The most direct measurement I found of what soaking removes is Korean. Han and colleagues compared four treatments — soaking, boiling, sautéing then boiling, and dressing with vinegar — across four samples of wild and farmed sea mustard (Korean J. Food Cookery Sci. 18(1), 2002). Their abstract states: “Sodium and potassium contents decreased significantly more at 68% and 85% of total content on basic soaking treatment.” The Korean abstract puts it as a loss of more than half from soaking alone, the most basic preparation step there is.

Two of their other findings run against intuition. Calcium, iron, magnesium and phosphorus were “remarkably decreased by seasoning with vinegar” — the vinegared side dish, not the soup, was hardest on those minerals. And sautéing before boiling lost more than plain boiling did.

Their methods section gives the conditions the abstract leaves out. The soaking treatment was “건미역 10g을 100배의 증류수(20℃)에 10분간 수침” — 10 g of dried miyeok in 100 times its weight of distilled water at 20°C for 10 minutes, then drained for 10 minutes. The samples were four dried products bought in Daegu: three farmed, from Wando, Boseong and Geoje, and one wild from Ulleungdo. Minerals were measured by atomic absorption spectrometry after ashing at 550°C for 20 hours, with sodium and potassium by flame photometry, in triplicate.

That water ratio is worth pausing on. It is the same one Japan’s table used for its 590% measurement: one hundred times the dry weight. Two groups, in two countries, a decade apart, reached for the same proportion. And the Korean paper supplies the variable Japan’s table omits — the water was at 20°C.

A 2021 study of four seaweeds reached a compatible conclusion by a different route (Food Chem. Toxicol. 155:112385). Ranking the species by how much they leached during rehydration and boiling, it placed Undaria pinnatifida last for sodium and first for potassium — of the four, wakame gave up the least sodium and the most potassium. The per-element percentages are behind a paywall I could not open.

Korea has no published iodine value for the dried miyeok people buy

Korea’s national food composition table does have an iodine column. In the 9th revision, the seaweed section runs to 53 foods, and the table’s own summary of how many foods carry an analytical value for each nutrient gives iodine as 5. I counted the rows independently and found the same five.

Dried miyeok is not among them. Entry 2513, miyeok, dried, publishes calcium 959 mg, iron 9.1 mg, phosphorus 307 mg, potassium 5,500 mg and sodium 6,100 mg per 100 g — and a dash in the iodine column. So do raw miyeok (2512), miyeok powder (2514), salted miyeok (2515) and miyeok stem (2517).

The two miyeok entries that do carry a value are strange. Wild raw miyeok (2520) is listed at 1.59 µg per 100 g. Deep-fried miyeok twigak (2521) is listed at 3,139.4 µg.

Set that against the measurement. In 2025, Lee and colleagues analysed 61 sea mustard samples — 40 wild, 21 farmed — by microwave digestion with tetramethylammonium hydroxide followed by ICP-MS, and reported iodine at 12.82 ± 7.42 mg/kg wet weight and 176.89 ± 88.91 mg/kg dry (Foods 14(16):2865). A wet-weight mean of 12.82 mg/kg is 1,282 µg per 100 g. The national table’s figure for wild raw miyeok is 1.59 µg per 100 g — roughly eight hundred times lower.

The same pattern shows up in the table’s other raw seaweeds: raw sea lettuce at 1.26 µg and maesaengi at 0. Its processed entries, by contrast, land close to a believable range — seasoned laver works out to about 17 mg/kg, which sits just under the 23.45–412.19 mg/kg that the 2025 study measured for laver, and fried miyeok twigak to about 31 mg/kg. Both are within a few-fold of what was measured. It is specifically the raw-seaweed entries that are off by three orders of magnitude, and nothing in the published table explains why.

A single number for miyeok’s iodine cannot be right anyway

The 2025 paper’s range matters more than its mean. Across 61 samples the values ran from 39.06 to 455.62 mg/kg dry — an 11.7-fold spread within one species, in one country, sampled across twelve coastal counties and cities.

Germany’s Federal Institute for Risk Assessment arrives at a similar picture from the European side, giving Undaria pinnatifida a range of 50 to 1,500 mg/kg of dry matter, and proposing a binding maximum of 1 gram of iodine per kilogram of dry matter for seaweed sold as food together with mandatory iodine labelling (Opinion No. 044/2026). A labelling rule is an admission that the content varies too much to be looked up in a table.

Which is the useful thing to take from all of this. Miyeok is not a stable quantity. It is a farmed crop above all — Korea’s marine aquaculture produced 631,236 tonnes of it in 2025, up 10.0% on the year, in the Ministry of Data and Statistics’ provisional count — and it is sold in at least nine processed forms.

That production figure carries the same trap as everything else here. The survey’s own glossary defines what it is counting — 어업생산량 — as “해수면 및 내수면에서 수산동식물을 포획·채취하거나 양식한 수산동식물을 생산한 생중량(kg)”, the live weight of what was caught, gathered or farmed, with a footnote specifying it is measured unprocessed. So 631,236 tonnes is wet miyeok straight out of the water, not the dried product anyone buys, and the conversion between the two is exactly the ratio this article has been chasing. The published figures for how much miyeok swells, how much it loses in the sink and how much iodine it carries all vary by product, method and sample more than any one number can carry — and so does the figure for how much of it exists.

What I could not verify

• The water temperature for the 590% figure. Japan’s Table 12 gives the water quantity and the 8-minute steep but never states the temperature, and I found no supplementary document that does.
• Whether the 68% and 85% losses are means or ranges. The paper’s abstract gives one figure for each, across four products of differing origin. It does not say whether those are averages or the extremes, and the sample-by-sample mineral figures are in tables I have not transcribed.
• Per-element leaching percentages from the 2021 study. Its abstract gives only the species ordering. The full text is paywalled.
• USDA’s wakame record. FoodData Central has one non-branded wakame entry, FDC ID 170496, “Seaweed, wakame, raw.” Its detail pages render only in JavaScript and return 404 to a direct request, and the public API rate-limited me before I could read the record. I have not confirmed its contents.
• Why the raw-seaweed iodine entries sit ~800× below the measured values. The published table carries no unit footnote or note for those rows. Either the values are wrong or they are in a different unit than the column header states, and the table does not say which.
• A third official moisture figure for dried miyeok. A Rural Development Administration portal lists dried miyeok at 6.3% moisture and 7,535 mg sodium, against 16.0% and 6,100 mg in the composition table from the same agency. The portal’s figures come from a form submission with no stable address, so I cannot give you a link to check them and have not used them above.
• Korea’s 10th-revision composition table. I verified the iodine column in the 9th revision (2016). I could not obtain the 10th-revision text to confirm whether dried miyeok has since been given a value.
• How much iodine boiling removes from miyeok specifically. The 2025 paper states that boiling and blanching can reduce iodine by 30–90%, but attributes that figure to another work and gives it for seaweed in general. I found no measurement of the loss for miyeok on its own.
• How much miyeok is wild rather than farmed. The 631,236-tonne figure comes from the marine aquaculture table. The coastal-fishery table in the same release has no row for miyeok at all, so there is no official split between farmed and wild-harvested production to give you.
• Where “ten times” came from. None of the four pages that give an expansion figure cites a source, and I could not trace the number to an original measurement.