Duckweed Amino Acid Profile: The Two Limiting Ones
The duckweed amino acid profile beats the FAO pattern on most essentials and misses on methionine. What the table leaves out is digestibility.
Ask for the duckweed amino acid profile and you get a table. The tables broadly agree, which is rare in this literature and worth saying plainly at the top: duckweed protein meets or beats the FAO reference pattern on most essential amino acids, and falls short on the sulphur ones. That is the settled part of the answer.
It is also the easy half. A composition table tells you what is in the biomass. It does not tell you what reaches the animal, and the gap between those two numbers is far larger than any difference between species you will find in the tables.
The short answer, in the units it is usually printed in
The broadest survey covers five genera at once. Across Spirodela, Landoltia, Lemna, Wolffiella and Wolffia, protein ran 20 to 35 percent of dry weight and the amino acid distribution came out close to the WHO recommendations, at 4.8 percent lysine, 2.7 percent methionine plus cysteine and 7.7 percent phenylalanine plus tyrosine.
There is a structural reason the profile is so consistent. Duckweed tissue is almost entirely metabolically active, and RuBisCO, the enzyme that runs photosynthesis, is roughly 50 percent of the proteome. When half the protein is the same enzyme, the amino acid ratios do not have much room to move. That review puts protein usually at 20 to 30 percent of dry mass, above 40 percent with clone selection and optimised culture, and reports a corrected protein score near 89 percent.
Before you compare two tables, check what they are counting. Most are in grams per 100 g of protein. Some are in grams per 100 g of dry biomass. A pond study of Lemna minor reports leucine at 4.132 and methionine at 0.859 g per 100 g of biomass, alongside 39.20 percent essential, 53.64 percent non-essential and 7.13 percent non-proteinogenic amino acids, that last fraction including taurine, citrulline and hydroxyproline. Those leucine and methionine figures sit in the same numeric range as per-protein figures from other papers and mean something entirely different. This is the same discipline problem as the quantity side, where the answer to duckweed protein content depends on the conversion factor used, and where it is worth knowing why every source quotes a different protein figure.
Where duckweed beats the reference pattern
One recent study lines a single Lemna minor clone up against soybean and against the FAO 2011 recommendation in the same table, in g per 100 g of protein, which is the comparison most readers actually want.
Branched chain amino acids and threonine
Threonine came in at 4.46 against soybean’s 3.44 and the FAO’s 3.1. Valine 5.46 against 4.74 and 4.3. Isoleucine 4.29 against 3.60 and 3.2. Leucine 8.42 against 8.20 and 6.6. Aspartic acid 18.97 against 11.53.
The branched chain group is where duckweed is strongest as a category, not just as a clone. In the pond study above, leucine, isoleucine and valine together made up 48.67 percent of the essential fraction.
Lysine, and why that matters against cereals
Lysine landed at 5.90 against soybean’s 6.36 and the FAO’s 5.7. Near-equal to soymeal, above the reference. Cross-species averages from the older literature are lower, around 4.0 g lysine per 100 g protein across Lemna gibba, Spirodela polyrhiza, Landoltia punctata and Wolffia columbiana, so read 5.9 as a good clone rather than a family norm.
The comparison that matters commercially is not against soy. It is against the cereal fraction duckweed would sit next to. Corn carries 2.3 and rice 3.2 g lysine per 100 g. Duckweed complements a cereal-based ration on the amino acid that ration is short of, which is a different and better argument than replacing soymeal one for one.
Methionine is the limiting amino acid, in every source that scores it
Same table, one row that fails: methionine plus cysteine at 1.96 against soybean’s 2.70 and the FAO’s 2.7. Every other essential cleared the bar. This one did not.
Forty years of results say the same thing. Lemna paucicostata sampled from three locations at 26.3 to 45.5 percent crude protein surpassed the FAO reference pattern on every essential amino acid except methionine, which met 61.4 percent of the recommended value. The cross-species averages put methionine at 0.9 g per 100 g protein, against 3.1 in corn and 3.4 in rice. Duckweed and cereals are short of opposite things.
Then there is a study that scores the profile rather than tabulating it, and it complicates the tidy version. Against a reference pattern, raw duckweed meal scored 0.75 for methionine plus cysteine and 0.92 for lysine, so lysine came out limiting too. Processing moved both, in opposite directions: soaking lifted the sulphur score to 0.89 and dropped lysine to 0.87, while cooking pushed the sulphur score down to 0.57.
That disagreement is the honest state of the evidence. Methionine is limiting everywhere. Whether lysine is limiting alongside it depends on the reference pattern, the clone and what you did to the biomass after harvest.
The profile is close to a species constant. What arrives in the animal is not.
Here is the finding that should change how the table is read. The amino acid distribution is a characteristic of the species and is almost independent of environmental conditions, while crude protein content is not. Composition is the stable variable. That is exactly why it is the wrong one to optimise.
Digestibility swung further than composition ever does
The same broiler trial fed three duckweed batches to Ross 308 chickens: a Lemna minuta and Lemna minor mixture at 17.5 percent crude protein, Spirodela polyrhiza at 24.6, and Lemna obscura at 37.0. Standardized ileal digestibility of methionine ran from 49.9 to 90.4 percent. Crude protein digestibility ran from 40.2 to 83.7 percent, lysine from 51.1 to 93.1.
No amino acid table in this article varies by anything like that margin. Half of the methionine in one batch never made it into the bird.
Tannins, and why batch beats species
The batches carried tannin at 2943, 2890 and 303 mg per kg. Digestibility tracked that inversely, and the paper states plainly that the batches came from third parties and non-monitored production.
The best batch was genuinely good. Lemna obscura beat 45 percent crude protein soybean meal on both limiting amino acids per 100 g of crude protein, 2.0 against 1.4 g methionine and 6.2 against 6.1 g lysine. The counterweight is equally plain: feed intake fell to 109.3 and 74.9 g per day at 50 and 75 percent inclusion, so the trial does not establish an inclusion rate. That is the real state of the case for duckweed as chicken feed.
Batch-to-batch spread shows up even in a pure chemistry measurement. Across 59 batches of three species, total free amino acids ranged from 166.05 to 1899.50 micrograms per gram, with no significant difference in profile between the species themselves. Where the plant grew moved the number by more than what the plant was.
Concentrate is a different ingredient from the plant
Buyers routinely compare a protein concentrate figure to a whole-plant figure. They are not the same product.
In weanling pigs, Lemna Protein Concentrate at 67.99 percent crude protein returned a mean standardized ileal digestibility of 81.70 percent across all amino acids: lysine 82.90, methionine 81.28, tryptophan 90.53, against soybean meal at 94.84 for lysine. On percentages, the concentrate lost to soymeal on almost everything.
On concentrations it won. The concentrate carried 1.38 percent methionine and 3.50 percent lysine against soybean meal’s 0.54 and 2.34, so it delivered more digestible protein and more of every indispensable amino acid per kilogram. Fish meal still beat it on lysine at 4.81 percent. Percentage digestibility and digestible concentration answer different questions, and only the second one prices a ration. The same distinction runs through any serious look at duckweed as animal feed.
Where the high-protein samples in this literature came from
Look back at the growing conditions behind the numbers above. Ponds fertilised with organic manure and poultry droppings. Batches from non-monitored third-party production. Nitrogen-rich water is what lifts protein, and the cheapest nitrogen anywhere is a waste stream.
SERAPH does not treat that biomass as feed, and the reason is a published rule rather than a preference. Biomass grown on wastewater, manure or contaminated water never becomes food or feed here, with the two streams separated from the first sensor reading. The same page states our maturity: TRL 3, proof of concept, with outdoor field validation still ahead. The nutrient-removal case for duckweed wastewater treatment is strong and it ends somewhere else entirely.
We have no field-validated amino acid data of our own to publish, and we will not imply otherwise. What we have is a map of the evidence: 256 indexed duckweed papers across 31 research areas, each carrying a coverage weight and a maturity band. Composition is one of the thickly covered areas. Delivered amino acid under real outdoor conditions, on water whose provenance is documented, is not. Species choice sets the starting profile, which is one reason Lemna minor keeps appearing in this literature, but it is the thinner of the two problems.
Five questions to ask of any duckweed amino acid table
Which units, per 100 g of protein or per 100 g of dry matter.
Which species and clone, because the distribution is a species trait rather than a growing outcome.
Concentrate or whole plant, because those are two ingredients with one name.
Digestibility measured in which animal, or not measured at all, because that is where the variance actually lives.
What water it grew in, and whether that water permits a feed destination at all.
A profile that survives those five is a specification. One that does not is a table.
FAQ
Is duckweed a complete protein?
It contains all the essential amino acids and a 2025 review reports a corrected protein score near 89 percent. The sulphur amino acids still fall below the FAO pattern in most analyses, so completeness and adequacy are not the same claim.
Does the amino acid profile change with the water you grow it in?
Barely. The distribution is a characteristic of the individual species and is almost independent of environmental conditions. Crude protein content changes a great deal with nitrogen, and digestibility changes more still, but the ratios between amino acids stay largely put.
Which duckweed species has the best amino acid profile?
Less separates them than the tables suggest. Across 59 batches of Spirodela polyrhiza, Landoltia punctata and Lemna aequinoctialis there was no significant difference in free amino acid profile between species, while batch-to-batch totals varied more than tenfold.
Does duckweed contain taurine?
Yes, in small amounts, along with citrulline, hydroxyproline and other non-proteinogenic amino acids that together made up 7.13 percent of the amino acids measured in one pond-grown Lemna minor study. They are not counted in a standard essential amino acid table.
How much duckweed can replace soybean meal in a ration?
The published broiler work does not answer that. It measured digestibility, not a formulation limit, and feed intake fell sharply at 50 and 75 percent inclusion, which points at palatability and biomass consistency as the binding constraints rather than the amino acid profile.