Ecosystem stoichiometry: the Redfield ratio and nutrient limitation

Intuition Beginner

Living cells are built from carbon, nitrogen, and phosphorus in a remarkably steady proportion. In marine phytoplankton that proportion is about 106 carbon atoms to 16 nitrogen atoms to 1 phosphorus atom, written C

= 106:16:1. This proportion is the Redfield ratio, named for Alfred Redfield, who noticed in 1934 that the chemistry of the deep ocean mirrors the chemistry of plankton. As plankton grow they strip nutrients from surface water in this proportion; as they die and sink they return those same nutrients to the deep ocean, the total amount of each recycled in step.

Think of a recipe that calls for flour, eggs, and sugar in a ratio of 106:16:1. If you have plenty of flour and sugar but only a few eggs, the eggs set the limit on how many cakes you can bake, no matter how much flour sits unused. Algae obey the same rule. If a lake holds nitrogen and phosphorus in a ratio above 16:1, phosphorus is the scarcest relative to demand, so phosphorus limits growth. Adding phosphorus spurs a bloom; adding nitrogen alone does nothing. This is Liebig's law of the minimum.

These ratios matter for the whole planet. Ocean algae pull carbon dioxide out of the air as they grow, and when they die they carry some of that carbon to the deep ocean in a process called the biological pump. How much carbon gets exported depends on the carbon-to-nutrient ratio inside algal cells. Stoichiometry, the study of these elemental proportions, links the chemistry of a single cell to the climate of the globe.

Visual Beginner

The Redfield ratio as a bar diagram, with carbon, nitrogen, and phosphorus drawn to scale:

   C : N : P  =  106 : 16 : 1

   Carbon  [████████████████████████████████████████████████]  106
   Nitrogen[████████]                                           16
   Phosph. [█]                                                   1

   Ocean surface:  algae take up N and P in this ratio,
                   then sink and release them back to deep water.

Ratios vary across organisms, but the marine average clusters near 106:16:1 ^{[Redfield 1958]}:

Organism / compartment	Typical C (molar)	Notes
Marine phytoplankton (Redfield)	106:16:1	Canonical ocean average
Diatoms (silicified)	~106:16:1, +Si	Need silica for frustules
Freshwater cyanobacteria	~140:18:1	Higher C per P when P-scarce
Land plants (wood-rich)	~1000:28:1	Carbon-dense, P-poor tissue
Vertebrate animals	~40:10:1	Protein- and P-rich muscle
Soil organic matter	~186:13:1	Slowly cycling pool

Worked example Beginner

A coastal water sample holds 0.4 micromoles of phosphate and 3.2 micromoles of nitrate per litre. The algae growing in this water follow the Redfield demand ratio N

= 16:1.

Step 1. Compute the supply ratio. The water supplies nitrogen to phosphorus in the ratio 3.2 / 0.4 = 8:1.

Step 2. Compare supply to demand. The algae need 16 nitrogen for every phosphorus, but the water offers only 8.

Step 3. Identify the limiting nutrient. Because the supply ratio (8) falls below the demand ratio (16), nitrogen is in shorter relative supply, so nitrogen limits growth.

What this tells us: the limiting nutrient is the one whose supply-to-demand ratio is smallest. Add enough nitrogen and the ratio climbs toward 16; once it passes 16, phosphorus becomes limiting instead. In much of the open ocean the scarcest nutrient is iron; in most freshwater lakes it is phosphorus.

Check your understanding Beginner

Formal definition Intermediate+

Ecological stoichiometry is the study of the balance of multiple chemical elements in ecological interactions, and of how the elemental composition of organisms shapes, and is shaped by, their environment ^{[Sterner-Elser 2002]}. The central object is an organism's elemental profile, the molar ratios of the elements in its biomass, reported relative to one reference element (phosphorus for the canonical Redfield triple, or carbon per unit nutrient in general).

The Redfield ratio is the average molar composition of marine particulate organic matter,

$$\mathrm{C}:\mathrm{N}:\mathrm{P}=106:16:1,$$

established by Alfred Redfield from the observation that the dissolved nitrate-to-phosphate ratio in the deep ocean is close to 16:1 by atoms, matching the average composition of phytoplankton detritus sinking from the surface ^{[Redfield 1958]}.

Liebig's law of the minimum states that the growth of an organism is constrained by the resource present in the smallest amount relative to its demand. For an autotroph building biomass with fixed elemental ratios $R_i$ (atoms of nutrient $i$ per atom of carbon), the relevant quantity for each nutrient $i$ is the ratio of supply $S_i$ to demand $R_i$. The nutrient minimizing $S_i/R_i$ is the limiting nutrient.

Homeostatic versus plastic stoichiometry. Organisms differ in how rigidly they hold their elemental composition when the composition of their food or environment changes. The stoichiometric homeostasis coefficient $H$ is defined on a log-log scale by

$$\frac{1}{H}=\frac{d\ln (\text{organism C}:X)}{d\ln (\text{resource C}:X)},$$

so that $H\to \infty$ describes strict homeostasis (the organism holds a constant composition, as in many metazoans) and $H=1$ describes a fully plastic organism whose composition tracks the resource (typical of many phytoplankton and vascular plants).

The growth rate hypothesis (Elser and colleagues) attributes the phosphorus content of fast-growing organisms to their investment in P-rich ribosomal RNA: rapidly growing cells build more ribosomes, which are rich in phosphorus, and so fast growth correlates with low biomass C

and N

ratios.

The threshold elemental ratio (TER) is the food C ratio at which a consumer switches from carbon (energy) limitation to limitation by that nutrient; it sets the boundary between food quantity and food quality.

Counterexamples to common slips

• Treating the Redfield ratio as universal. It describes marine particulate matter; land plants run far more carbon-rich (C
  near 1000:1 in wood) and vertebrates run more phosphorus-rich. Applying 106:16:1 to a forest soil budget gives nonsense.
• Confusing proximate and ultimate limitation. Nitrogen often limits ocean production on the timescale of a bloom, but on geological timescales nitrogen can be replenished by biological fixation while phosphorus cannot; the ultimate limiting nutrient then differs from the proximate one.
• Equating "most abundant" with "limiting." Limitation follows the supply-to-demand ratio, not the absolute concentration. A nutrient present in large amounts can still be limiting if demand is larger.

Core model: nutrient limitation and the Redfield ratio Intermediate+

The quantitative core of ecological stoichiometry couples two ingredients: a fixed biomass composition and the Droop model of nutrient-limited growth. The Droop model ^{[Droop 1968]} relates the specific growth rate $\mu$ of a cell to its internal nutrient quota $Q$ (nutrient per cell, or per unit carbon):

$$\mu (Q)=\hat{\mu }\left(1-\frac{k_Q}{Q}\right),Q\ge k_Q,$$

where $\hat{\mu }$ is the maximal growth rate and $k_Q$ is the subsistence quota at which growth ceases. Growth tracks the internal nutrient pool, not the external concentration directly; uptake kinetics first fill the quota, and growth then reads off the quota. For multiple nutrients, Liebig's law composes the Droop curves by a minimum:

$$\mu =\underset{i\in \{\mathrm{C},\mathrm{N},\mathrm{P},\mathrm{Fe}\}}{\min }{\mu }_i(Q_i).$$

Combined with a fixed biomass composition, this yields the central stoichiometric result.

Result (Liebig limitation from stoichiometric demand). Suppose an autotroph builds biomass with fixed elemental ratios $R_i$ (atoms of nutrient $i$ per atom of carbon), with carbon freely available from carbon dioxide. Given available nutrient pools $S_i$, the maximum biomass carbon achievable is

$$B^{\ast }=\underset{i}{\min }\frac{S_i}{R_i},$$

and the limiting nutrient is the one attaining the minimum, that is, the nutrient minimizing the supply-to-demand ratio $S_i/R_i$.

Derivation. Let $B$ denote the carbon biomass produced. Producing $B$ atoms of carbon biomass consumes $B{R}_i$ atoms of nutrient $i$, which cannot exceed the available pool: $B{R}_i\le S_i$ for each $i$. Hence $B\le S_i/R_i$ for every $i$, and the binding (smallest) upper bound sets the achievable biomass $B^{\ast }={\min }_i{S}_i/R_i$. The nutrient attaining that minimum is exhausted first and so limits further growth. Substituting the Redfield ratios $R_{\mathrm{N}}=16$ and $R_{\mathrm{P}}=1$ recovers the rule applied in the worked example: nitrogen limits when $S_{\mathrm{N}}/16<S_{\mathrm{P}}/1$, equivalently when the ambient N

ratio is below 16:1. $\square$

This gives a clean operational test for limitation: compute the supply-to-demand ratio for every required nutrient, and read off the argmin. The same test, generalized by the Droop quota, predicts the transient response when a limiting nutrient is added, namely a bloom whose size is set by how far the added supply lifts the binding ratio above the next-smallest one.

Bridge. This stoichiometric limitation model builds toward 19.11.01 ecosystem ecology, where it constrains the nutrient-cycling budget of an entire catchment, and appears again in 19.11.02 pending energy flow, where the carbon fixed per nutrient exported sets the trophic ceiling above it. The foundational reason is that biomass carries a characteristic elemental composition, and this is exactly why adding the scarcest nutrient rather than the most abundant one relieves limitation, a constraint that generalises from a single algal cell up to whole-basin carbon export and to the global biological pump.

Exercises Intermediate+

Advanced results Master

1. The biological control of ocean chemistry (Redfield 1958). Redfield's 1958 synthesis ^{[Redfield 1958]} did more than report a ratio; it argued that the chemistry of the ocean is biologically controlled. The deep-ocean nitrate-to-phosphate ratio of about 16:1 by atoms is not a geochemical accident but the signature of the average composition of the plankton that stripped those nutrients from surface water and returned them at depth through sinking detritus. The constancy of this ratio across ocean basins is the evidence: an abiotic ocean would show no such regularity. The finding couples organismal biochemistry to basin-scale chemistry through a single stoichiometric proportion.

2. Nitrogen versus phosphorus as the ultimate limiting nutrient (Tyrrell 1999). Tyrrell's box-model analysis ^{[Tyrrell 1999]} separated proximate from ultimate limitation. Proximately, nitrogen limits surface productivity in much of the ocean because fixed nitrogen is scarce relative to phosphorus; but nitrogen fixers replenish it over millennial timescales, so any nitrogen deficit self-corrects. Phosphorus, supplied only by rock weathering, sets the long-term ceiling. The model predicts that on timescales of $10^4$ years and longer the total inventory of oceanic fixed nitrogen tracks the phosphorus inventory through Redfield stoichiometry, an instance where the ratio serves as a closure constraint on a global biogeochemical model.

3. The growth rate hypothesis. Elser and colleagues found that biomass phosphorus content scales positively with growth rate across bacteria, phytoplankton, crustaceans, and insects. The mechanistic basis is ribosomal investment: rapidly growing organisms build more ribosomes to sustain high rates of protein synthesis, and ribosomal RNA is phosphorus-dense. This yields a quantitative prediction relating the specific growth rate $\mu$ to biomass C

, and it explains why fast-growing blooms are P-demanding and why P-enrichment preferentially stimulates rapid-growing taxa.

4. The threshold elemental ratio (TER) and food quality. Sterner, Frost, and Urabe showed that herbivore growth depends not only on the quantity of food but on its stoichiometric match to the consumer. The TER is the food C ratio at which a consumer switches from carbon (energy) limitation to nutrient limitation; when food C

exceeds the TER, the consumer is phosphorus-limited despite abundant food. This reframes "food quality" as a stoichiometric quantity and explains why P-poor algae support poor zooplankton growth.

5. Stoichiometric homeostasis and the producer-consumer mismatch. Autotrophs are stoichiometrically plastic: their C

tracks the ambient nutrient supply, so under nutrient scarcity they become carbon-rich (high C

). Metazoan consumers are homeostatic, holding a near-constant body composition. The mismatch between plastic food and homeostatic consumer generates the food-quality effects of result 4 and is a major axis on which nutrient supply propagates up the food web.

6. The biological pump and carbon export stoichiometry. The biological pump exports carbon to the deep ocean as sinking particulate matter, and the efficiency of this pump depends on the C ratio of the export. Canonical Redfield stoichiometry gives about 106 moles of carbon exported per mole of phosphorus, but the ratio in sinking material varies regionally. Carbon-rich export strengthens the pump's draw-down of atmospheric carbon dioxide. The Southern Ocean HNLC paradox, where added iron triggers blooms that shift community composition toward diatoms, illustrates how a single micronutrient alters both the magnitude and stoichiometry of export.

7. Co-limitation typology (Harpole and colleagues). Modern nutrient-enrichment experiments showed that single-nutrient limitation is the exception rather than the rule. A three-way classification distinguishes Liebig (serial) co-limitation, where one nutrient is strictly primary and a second becomes limiting only after the first is supplied; independent co-limitation, where two nutrients are equally limiting and both must be added to elicit a response; and biochemical (serial) co-limitation, where the nutrient required in the smallest amount relative to supply limits an internal metabolic bottleneck. Syntheses of enrichment experiments find that co-limitation is widespread in both marine and freshwater systems.

8. Variability around the Redfield ratio. The canonical 106:16:1 is a mean; regional and taxonomic deviations carry information. Diatoms cluster near Redfield, while nitrogen-fixing cyanobacteria run richer in phosphorus demand, and carbon-over-rich cells appear under severe nutrient stress. Revised estimates (Anderson 1995, near 117:16:1 for recycled production) refine the canonical figure but do not displace it: the Redfield ratio is best read as the central tendency of a distribution whose width reflects physiological and taxonomic variation.

Synthesis. The Redfield ratio is the foundational reason ecosystem ecology acquired a quantitative stoichiometric backbone: a single C

proportion ties organismal biochemistry to ocean carbon export. The central insight of Sterner and Elser is that organisms differ in how rigidly they hold their composition, and this is exactly the lever that produces nutrient recycling, food-quality mismatch, and the growth rate hypothesis. Putting these together with Tyrrell's nitrogen-versus-phosphorus analysis identifies geological-scale nutrient limitation with the phosphorus supply, while the bridge is the biological pump, which couples every atom of limiting nutrient to roughly 106 atoms of exported carbon. The pattern generalises from a single phytoplankton cell to forests and to whole-ocean carbon budgets, and it appears again in 19.11.01 and 19.11.02 pending wherever energy flow and material cycling are tracked together.

Full proof set Master

Proposition 1 (Liebig limitation, rigorous form). Let an autotroph build biomass carbon in the amount $B$, requiring $R_i>0$ units of nutrient $i$ per unit biomass carbon, for nutrients in a finite set $I$. Suppose the available pool of nutrient $i$ is $S_i>0$, and carbon is freely available. Then the maximum achievable biomass is $B^{\ast }={\min }_{i\in I}{S}_i/R_i$, and a nutrient $j$ is limiting if and only if $S_j/R_j={\min }_{i\in I}{S}_i/R_i$.

Proof. For each nutrient $i$, building biomass $B$ consumes $B{R}_i$ units of nutrient $i$, which must not exceed $S_i$: hence $B{R}_i\le S_i$, equivalently $B\le S_i/R_i$. This holds for every $i$, so $B\le {\inf }_{i\in I}{S}_i/R_i={\min }_i{S}_i/R_i$ (the infimum is attained because $I$ is finite). Call this bound $B^{\ast }$. To see that $B^{\ast }$ is achievable, take a nutrient $j$ attaining the minimum. At $B=B^{\ast }=S_j/R_j$, nutrient $j$ is fully consumed (consumption $=B^{\ast }{R}_j=S_j$), and for every $i\ne j$ the consumption is $B^{\ast }{R}_i\le (S_j/R_j)R_i\le S_i$ since $S_j/R_j\le S_i/R_i$. Thus no constraint other than $j$'s is violated, and biomass $B^{\ast }$ is realized. The limiting nutrients are precisely those attaining the minimum. $\square$

Proposition 2 (threshold elemental ratio). Let a consumer with assimilation efficiencies ${\mathrm{AE}}_C$ for carbon and ${\mathrm{AE}}_X$ for nutrient $X$ build body biomass with elemental ratio $A_C:A_X$ (carbon to nutrient $X$, by atoms). Feeding on food of ratio $F_C:F_X$, the consumer is carbon-limited below a unique threshold food ratio and $X$-limited above it, with

$${\left(\frac{F_C}{F_X}\right)}_{\mathrm{TER}}=\frac{{\mathrm{AE}}_X\cdot A_C}{{\mathrm{AE}}_C\cdot A_X}.$$

Proof. Ingestion at rate $I$ delivers assimilated carbon ${\mathrm{AE}}_C\cdot I\cdot F_C/(F_C+F_X)$ and assimilated nutrient ${\mathrm{AE}}_X\cdot I\cdot F_X/(F_C+F_X)$. The carbon that can be laid down as new biomass is limited by assimilated carbon, and the nutrient laid down is limited by assimilated $X$. The two potential growth ceilings are equal when the assimilated inputs, normalized by the body requirements $A_C$ and $A_X$, coincide:

$$\frac{{\mathrm{AE}}_C\cdot F_C}{A_C}=\frac{{\mathrm{AE}}_X\cdot F_X}{A_X}.$$

Rearranging gives the threshold ratio $(F_C/F_X{)}_{\mathrm{TER}}=({\mathrm{AE}}_X\cdot A_C)/({\mathrm{AE}}_C\cdot A_X)$. If $F_C/F_X$ is below this value, the carbon-normalized assimilation is the smaller quotient and carbon limits growth; if above, nutrient $X$ limits. Substituting typical Daphnia values (${\mathrm{AE}}_C\approx 0.6$, ${\mathrm{AE}}_P\approx 0.8$, body C

$A_C/A_X\approx 80$) gives a TER near $107$, in line with the empirical threshold above which Daphnia become phosphorus-limited. $\square$

Connections Master

• Ecosystem ecology 19.11.01. This unit deepens the nutrient-cycling half of 19.11.01, where carbon, nitrogen, and phosphorus cycles were introduced qualitatively. Ecological stoichiometry supplies the quantitative constraint that closes those cycle budgets: the Redfield ratio fixes how many atoms of each element cycle together through the producer-decomposer loop, and the homeostasis coefficient controls how tightly an organism holds its composition against environmental variation.

• Ecosystem energy flow 19.11.02 pending. The carbon exported per unit of limiting nutrient, a stoichiometric quantity, sets the trophic ceiling analysed in 19.11.02 pending. Energy flow and material flow are two descriptions of the same biomass: the C

  ratio of export fixes how much energy (as carbon) a given nutrient supply can carry into the food web and down through the biological pump.

• Photosynthesis 17.04.03. The carbon side of the Redfield ratio is set by the photosynthetic machinery described in 17.04.03: the Calvin cycle, Rubisco kinetics, and C3/C4/CAM pathways determine how fast carbon is fixed, while nutrient uptake kinetics determine how fast nitrogen and phosphorus can accompany it. Stoichiometry is the bookkeeping that keeps the two in balance.

• Cellular respiration 17.04.01. When consumers face stoichiometric mismatch (Proposition 2), the excess carbon from P-poor food is respired away as carbon dioxide. The respiratory machinery of 17.04.01 is the physiological route by which a homeostatic consumer disposes of carbon it cannot use, linking cellular bioenergetics to ecosystem-scale nutrient recycling.

• Population ecology 19.09.01. Stoichiometric food quality propagates into population dynamics: a P-limited Daphnia population has lower birth rates, altering the predator-prey cycles of 19.09.01. The carrying capacity $K$ of a consumer population is set not only by food quantity but by food C ratio relative to the TER, coupling elemental ratios to Lotka-Volterra dynamics.

Historical & philosophical context Master

Alfred Clarence Redfield, a physiologist at the Woods Hole Oceanographic Institution, observed in 1934 that the ratio of nitrate to phosphate in the deep Atlantic was close to 20:1 by atoms and proposed that this ratio reflected the average composition of the plankton sinking from the surface. His 1958 paper, "The biological control of chemical factors in the environment" ^{[Redfield 1958]}, generalized the claim: the chemistry of the ocean is not imposed by geology alone but is regulated by the collective metabolism of marine organisms. The canonical ratio C

= 106:16:1 crystallized from these data, and the broader proposition that life sets the chemical environment of its own habitat became a organizing principle of biogeochemistry.

The field of ecological stoichiometry was codified by Robert Sterner and James Elser in their 2002 monograph Ecological Stoichiometry: The Biology of Elements from Molecules to the Biosphere ^{[Sterner-Elser 2002]}, which generalized the Redfield insight from the ocean to all ecosystems and to all trophic interactions. The growth rate hypothesis, the homeostasis coefficient, and the threshold elemental ratio were consolidated there into a single framework linking the elemental composition of a ribosome to the nutrient budget of a lake. Toby Tyrrell's 1999 analysis ^{[Tyrrell 1999]} sharpened the long-standing debate over whether nitrogen or phosphorus ultimately limits ocean production, concluding that phosphorus sets the geological-scale ceiling.

Two philosophical dimensions attend the subject. First, the Redfield ratio raises a question about the level at which biological explanation operates: is the ratio a property of individual cells, or an emergent statistic of a community? The answer is that it is both simultaneously, and the near-constancy of the deep-ocean ratio is evidence that an emergent community-level regularity can be grounded in organismal biochemistry. Second, ecological stoichiometry is a rare ecological theory in which a small set of conserved quantities (the atoms of carbon, nitrogen, and phosphorus) closes the books on a system, importing the mass-balance rigor of physical chemistry into a discipline that usually lacks it. Liebig's 1840 law of the minimum, originally a statement about crop nutrition, is the lineage from which the modern stoichiometric closure descends.

Bibliography Master

1. Redfield, A. C. "The biological control of chemical factors in the environment." American Scientist 46 (1958) 205-221.

2. Sterner, R. W. & Elser, J. J. Ecological Stoichiometry: The Biology of Elements from Molecules to the Biosphere (Princeton University Press, 2002).

3. Tyrrell, T. "The relative influences of nitrogen and phosphorus on oceanic primary production." Nature 400 (1999) 525-531.

4. Droop, M. R. "Vitamin B12 and marine ecology. IV. The kinetics of uptake, growth and inhibition in Monochrysis lutheri." Journal of the Marine Biological Association of the United Kingdom 48 (1968) 689-733.

5. Elser, J. J., Dobberfuhl, D. R., MacKay, N. A. & Schampel, J. H. "Organism size, life history, and N

   stoichiometry: towards a unified view of cellular and ecosystem processes." BioScience 46 (1996) 674-684.

6. Harpole, W. S. et al. "Nutrient co-limitation of primary producer communities." Ecology Letters 14 (2011) 852-862.

7. Anderson, L. A. "On the hydrogen and oxygen content of marine phytoplankton." Deep-Sea Research I 42 (1995) 1675-1680.

8. Falkowski, P. G. "Rationalizing elemental ratios in unicellular algae." Journal of Phycology 36 (2000) 3-6.

9. Martin, J. H. & Fitzwater, S. E. "Iron deficiency limits phytoplankton growth in the north-east Pacific subarctic." Nature 331 (1988) 341-343.

10. Urabe, J., Clasen, J. & Sterner, R. W. "Phosphorus limitation of Daphnia growth: is it real?" Archiv fur Hydrobiologie 138 (1997) 143-154.

11. Campbell, N. A. & Reece, J. B. Biology, 12th ed. (Pearson, 2020). Ch. 55.