Skip to content

Energy Flow, Food Webs and the Microbial Loop

CSIR NET Ocean Sciences: trophic levels and transfer efficiency, marine pyramids, cascades, the microbial loop and viral shunt, and the biological carbon pump.

  • 17 min read
  • Updated
On this page · 13 sections
  1. Prerequisites
  2. The Hook: The Missing Half
  3. Energy Flow
  4. Food Webs: Structure and Control
  5. The Microbial Loop
  6. The Biological Carbon Pump
  7. Memory Tricks
  8. Don’t Confuse
  9. Common Mistakes
  10. Practice
  11. Quick Revision Table
  12. Key Takeaways
  13. What’s Next

Prerequisites

The Hook: The Missing Half


Energy Flow

Principles

  • Energy flows one way: from sunlight (or chemical energy at vents) through producers to consumers and is lost as heat at every step (second law of thermodynamics). Nutrients cycle.
  • Lindeman’s (1942) trophic-dynamic view treats each trophic level as a compartment receiving energy from the one below.
  • Trophic transfer efficiency (TE) is production at level \( n+1 \) divided by production at level \( n \). It is the product of how much is eaten (exploitation efficiency), how much of that is assimilated, and how much of that becomes growth. In marine food webs the average is about 10%, with a range of roughly 5–20%.

The production at trophic level \( n \) is

$$P_n = P_1 \times \text{TE}^{\,n-1}$$

Fractional trophic levels

Most marine consumers eat at more than one level. Their trophic level is

$$\text{TL}_i = 1 + \sum_j \text{DC}_{ij}\,\text{TL}_j$$

where \( \text{DC}_{ij} \) is the fraction of prey \( j \) in the diet of \( i \). Primary producers and detritus have TL = 1; herbivorous copepods about 2; sardines about 2.5–3; tunas and sharks 4 or more.

Ecological pyramids in the sea

PyramidShape in the open oceanWhy
Numbersoften inverted at the base (one diatom, many bacteria; one whale, many lice)depends on body size
Biomass (standing stock)can be inverted: zooplankton biomass may exceed phytoplanktonphytoplankton turn over in days and are grazed as fast as they grow
Energy (production per unit time)never invertedthermodynamics: each level produces less than the one below

Food-chain length and fish yield

The number of steps between producers and fish depends on the size of the producers:

SystemDominant producersTypical steps to harvested fishConsequence
Coastal upwellinglarge diatoms, chainsabout 1–2 (diatoms → anchovy)most of production reaches fish
Shelvesmixedabout 3moderate
Oligotrophic open oceanpico-cyanobacteria (Prochlorococcus)about 4–5 (pico → flagellates → ciliates → copepods → small fish → tuna)very little reaches fish

Ryther’s classic 1969 analysis used exactly this reasoning: although the open ocean makes most of the world’s primary production, its long food chains and lower efficiencies mean that upwelling zones and shelves, covering a small fraction of the area, produce most of the fish. Because \( \text{TE}^n \) is a power, a few extra steps change fish production by orders of magnitude.

Size spectra

Marine food webs are strongly size-structured: predators are usually much larger than their prey, so trophic level rises with body size. When biomass is grouped into logarithmic size classes, it is roughly constant from bacteria to whales in many ocean samples (the “Sheldon spectrum”), because the loss of energy at each step is balanced by the slower metabolism of larger organisms.


Food Webs: Structure and Control

Grazing and detrital pathways

PathwayStartImportance
Grazing food chainliving phytoplankton eaten by zooplanktondominates upwelling and blooms
Detrital food chaindead organic matter (marine snow, faecal pellets, plant litter) consumed by bacteria and detritivoresdominates estuaries, mangroves, salt marshes and the deep-sea floor
Microbial food webdissolved organic matter used by bacteria, which are eaten by protistsdominates oligotrophic open ocean

Top-down and bottom-up control

  • Bottom-up control: nutrient supply and primary production set the abundance of every level above (upwelling strength controls sardines).
  • Top-down control: predators set the abundance of the levels below.
  • Trophic cascade: a change at the top propagates down, alternating in sign. The classic marine case: sea otters eat sea urchins; where otters were hunted out, urchins multiplied and grazed kelp forests into “urchin barrens”; where otters returned, kelp recovered.
  • Keystone species: a species whose effect is far larger than its abundance suggests. Paine’s experiments on a rocky shore in Washington State, USA, in the 1960s showed that removing the predatory sea star Pisaster let mussels monopolise the space and cut the number of species sharply.
  • Wasp-waist control: in upwelling systems a few species of small pelagic fish (sardines, anchovies) sit at a narrow middle level; they control their zooplankton prey from above and their predators from below.
graph TD
    A[Sea otters] -->|eat| B[Sea urchins]
    B -->|graze| C[Kelp forest]
    C -->|shelter and food| D[Fishes and invertebrates]
    A -. otters removed so urchins boom and kelp is lost .-> C

The Microbial Loop

The idea

In 1974 Pomeroy argued that microbes, not copepods, handle much of the ocean’s energy flow. Azam and colleagues in 1983 named and described the microbial loop: phytoplankton release dissolved organic matter (DOM); heterotrophic bacteria take it up; bacteria are eaten by heterotrophic nanoflagellates; these are eaten by ciliates; and ciliates are eaten by copepods, returning some of the carbon to the classic food chain.

graph LR
    P[Phytoplankton] -->|exudation, sloppy feeding, lysis| D[Dissolved organic matter]
    P -->|grazing| Z[Copepods]
    D -->|uptake| B[Heterotrophic bacteria]
    B -->|grazing| F[Heterotrophic nanoflagellates]
    F -->|grazing| C[Ciliates]
    C -->|grazing| Z
    Z --> FI[Fish]
    B -->|respiration| CO2[CO2 and regenerated NH4+ and PO4]
    V[Viruses] -->|lysis| D
    B -.infected by.-> V

Sources of DOM

  • Exudation (leakage) of photosynthate by healthy phytoplankton.
  • Sloppy feeding by zooplankton that break cells while eating.
  • Excretion by zooplankton.
  • Viral lysis of bacteria and phytoplankton.
  • Dissolution of sinking particles.

Key numbers and terms

QuantityTypical value
Heterotrophic bacteria in surface waterabout \( 10^5 \)–\( 10^6 \) cells mL\(^{-1}\)
Virusesabout \( 10^6 \)–\( 10^7 \) mL\(^{-1}\), roughly 10 per bacterium
Share of NPP passing through bacteria as DOMlarge; often quoted as up to about half
Bacterial growth efficiency\( \text{BGE} = \dfrac{\text{BP}}{\text{BP} + \text{BR}} \), commonly about 10–30% in the open ocean
Bacterial carbon demand\( \text{BCD} = \text{BP} + \text{BR} = \text{BP}/\text{BGE} \)

BP is bacterial production and BR bacterial respiration.

Is the microbial loop a link that passes energy to larger animals, or a sink that respires it away?

  • Each step (bacteria → flagellates → ciliates → copepods) has low efficiency, so after three or four steps only a small fraction of the DOM carbon reaches copepods. In energy terms, the loop is mostly a sink.
  • But it regenerates nutrients (ammonium, phosphate) in the euphotic zone, sustaining regenerated production. In nutrient-poor gyres this recycling keeps the whole system running.
  • It also feeds small cells to protists whose own grazers are larger, so in oligotrophic water it is the main route by which any energy reaches metazoans.

The viral shunt

Viruses lyse bacteria and phytoplankton, turning living particulate matter back into DOM that is used by other bacteria. This viral shunt:

  • keeps carbon and nutrients inside the microbial loop and away from grazers;
  • raises community respiration and nutrient regeneration;
  • may make cells more prone to aggregation and sinking when lysis products are sticky;
  • sustains diversity by “killing the winner”: the most abundant strain is most likely to meet its virus, so no strain dominates for long;
  • moves genes between hosts (transduction); cyanophages carry photosynthesis genes.

Mixotrophy

Many protists are both photosynthetic and phagotrophic (many dinoflagellates and haptophytes, some ciliates). Mixotrophs blur the line between producer and consumer and can increase the transfer of energy to larger organisms in nutrient-poor water, because they gain nutrients from prey and energy from light.


The Biological Carbon Pump

The ocean moves carbon from the surface to depth by three pumps:

PumpMechanism
Solubility pumpcold, dense high-latitude water absorbs more CO\(_2\) and sinks
Organic (soft-tissue) biological pumpphytoplankton fix CO\(_2\); organic matter sinks as aggregates (marine snow), faecal pellets and dead organisms, or is carried down by vertical migrators and by mixing of DOC
Carbonate (hard-tissue) pumpcalcifiers (coccolithophores, foraminifera, pteropods) make CaCO\(_3\) that sinks; calcification itself releases CO\(_2\) at the surface (the carbonate counter-pump)

How much sinks, and how deep

  • Export production leaves the euphotic zone; the e-ratio = export / NPP, typically about 0.1–0.3 and higher in blooms.
  • Most exported carbon is remineralised in the upper few hundred metres by bacteria and zooplankton. The flux decreases with depth according to the empirical Martin curve (Martin and colleagues, 1987):
$$F(z) = F_{100}\left(\frac{z}{100}\right)^{-b}, \qquad b \approx 0.86$$

With \( b = 0.858 \), only about 14% of the flux at 100 m reaches 1,000 m. Carbon remineralised below the deepest winter mixed layer is stored for centuries.

  • Ballast: dense minerals (calcite, opal, dust) make particles sink faster and may protect organic carbon, so export is linked to the kind of plankton present.
  • Microbial carbon pump: microbes convert labile DOM into refractory DOM that resists further breakdown and persists for thousands of years in the deep ocean, a further route of carbon storage.

Linked cycles

ProcessOrganismsWhere
N\(_2\) fixationTrichodesmium, unicellular cyanobacteria, symbionts of diatomswarm, stratified, nitrate-poor water
Nitrification (NH\(_4^+\) → NO\(_2^-\) → NO\(_3^-\))ammonia-oxidising archaea and bacteria, nitrite oxidisersbelow the euphotic zone
Denitrification (NO\(_3^-\) → N\(_2\))heterotrophic bacteria using nitrateoxygen minimum zones (Arabian Sea, eastern tropical Pacific) and sediments
Anammox (NH\(_4^+\) + NO\(_2^-\) → N\(_2\))anammox bacteriaOMZs and sediments
DMS productionDMSP from coccolithophores, Phaeocystis, dinoflagellates, cleaved to dimethyl sulfidesurface ocean; DMS forms sulfate aerosol and cloud nuclei (the CLAW hypothesis, 1987, proposed a climate feedback through cloud albedo)

Memory Tricks

Don’t Confuse

ThisvsThat
Pyramid of biomass: can invert in the seavsPyramid of energy: never inverts
Transfer efficiency: production ratio between levelsvsGross growth efficiency: one population’s growth ÷ ingestion
Microbial loop: DOM → bacteria → protists → copepodsvsViral shunt: lysis returns cells to DOM, bypassing grazers
BGE = BP/(BP + BR)vsBCD = BP + BR
Bottom-up: resources controlvsTop-down: predators control
Keystone species: large effect relative to abundancevsDominant species: large effect because abundant
Soft-tissue pump: organic carbonvsCarbonate pump: CaCO\(_3\) (releases CO\(_2\) at the surface when formed)
Export production (leaves euphotic zone)vsSequestration (below the deepest winter mixed layer)

Common Mistakes


Practice

All questions below are original practice questions written in the CSIR NET style. They are not previous-year questions.

Level 1: Foundation

  1. Who named and described the microbial loop in 1983?
  2. Which pyramid can never be inverted?
  3. What is the approximate value of the exponent \( b \) in the Martin curve?
Answers
  1. Azam and colleagues (building on Pomeroy, 1974).
  2. The pyramid of energy (production).
  3. About 0.86 (0.858).

Level 2: Part B Style

Q1 (practice). A fish’s diet is 60% copepods (TL 2.2) and 40% small fish (TL 3.1). Its trophic level is:

  • (a) 2.56
  • (b) 2.65
  • (c) 3.56
  • (d) 4.30
Answer and explanation

(c). \( \text{TL} = 1 + 0.6(2.2) + 0.4(3.1) = 1 + 1.32 + 1.24 = 3.56 \). (a) forgets the “1 +”.

Q2 (practice). Bacterial production is 2 and bacterial respiration 6 µg C L\(^{-1}\) d\(^{-1}\). The bacterial growth efficiency and the bacterial carbon demand are:

  • (a) 33% and 6
  • (b) 25% and 8
  • (c) 25% and 6
  • (d) 75% and 8
Answer and explanation

(b). BGE = 2/(2 + 6) = 25%; BCD = 2 + 6 = 8 µg C L\(^{-1}\) d\(^{-1}\). (a) divides BP by BR.

Q3 (practice). The particulate organic carbon flux at 100 m is 5.0 mmol C m\(^{-2}\) d\(^{-1}\). Using the Martin curve with \( b = 0.858 \), the flux at 1,000 m is about:

  • (a) 0.50
  • (b) 0.69
  • (c) 1.25
  • (d) 4.3
Answer and explanation

(b). \( F = 5.0 \times 10^{-0.858} = 5.0 \times 0.139 = 0.69 \) mmol C m\(^{-2}\) d\(^{-1}\): about 14% survives. (a) assumes the flux falls in proportion to depth (\( b = 1 \)).

Q4 (practice). Removal of sea otters from a coast is followed by an increase in sea urchins and loss of kelp. This is best described as:

  • (a) bottom-up control
  • (b) a trophic cascade from a keystone predator
  • (c) competitive exclusion between urchins and kelp
  • (d) the microbial loop
Answer and explanation

(b). The effect alternates down the chain (predator down, herbivore up, plant down): a top-down trophic cascade. Otters act as a keystone species.

Level 3: Part C Style

Q1 (practice). Compare two regions (practice values):

RegionNPP (g C m\(^{-2}\) yr\(^{-1}\))Transfers from producers to harvested fishTransfer efficiency
Oligotrophic open ocean50410%
Coastal upwelling300220%

(i) Calculate fish production per m\(^2\) in each. (ii) By what factor does the upwelling region exceed the open ocean? (iii) If the upwelling region covers 0.1% of the ocean area and the open ocean 90%, which produces more fish in total?

Answer and worked solution
  1. Open ocean: \( 50 \times 0.1^4 = 0.005 \) g C m\(^{-2}\) yr\(^{-1}\). Upwelling: \( 300 \times 0.2^2 = 12 \) g C m\(^{-2}\) yr\(^{-1}\).
  2. \( 12/0.005 = \) 2,400 times, although primary production differs only sixfold.
  3. Totals scale with area: open ocean \( 0.005 \times 0.90 = 0.0045 \); upwelling \( 12 \times 0.001 = 0.012 \) (in the same arbitrary area units). Upwelling produces about 2.7 times more fish from 0.1% of the area. This is the logic of Ryther’s argument.

Q2 (practice). In an oligotrophic gyre, 30 units of carbon per day are released as DOM and taken up by bacteria. Assume a growth efficiency of 30% at each step: bacteria, heterotrophic nanoflagellates, ciliates, and copepods eating ciliates. Compare with 30 units of phytoplankton eaten directly by copepods (copepod growth efficiency 30%).

(i) How much copepod production results from each route? (ii) How much carbon is respired in the microbial route before reaching copepods? (iii) Which statements are correct?

(A) The microbial route delivers about 2.7% as much copepod production as the direct route. (B) Viral lysis of 50% of the bacterial production would further reduce what reaches copepods. (C) The microbial route has no ecological value because it is mostly a sink. (D) The microbial route regenerates ammonium that can support regenerated production.

  • (a) A, B and D
  • (b) A and C only
  • (c) B, C and D
  • (d) A, B, C and D
Answer and worked solution
  1. Direct: \( 30 \times 0.3 = 9 \) units of copepod production. Microbial: bacteria \( 30 \times 0.3 = 9 \); flagellates \( 9 \times 0.3 = 2.7 \); ciliates \( 2.7 \times 0.3 = 0.81 \); copepods \( 0.81 \times 0.3 = 0.243 \) units.
  2. Carbon reaching ciliate biomass is 0.81, so before reaching copepods \( 30 - 0.81 = 29.19 \) units have been respired, egested or excreted (most of it respired to CO\(_2\)).
  3. (a) A, B and D.
    • (A) \( 0.243/9 = 0.027 \), i.e. 2.7%. Correct.
    • (B) Correct: the viral shunt returns half the bacterial production to DOM, so flagellates get 4.5 instead of 9.
    • (C) Wrong: the loop recycles nutrients and is the main route to metazoans when cells are too small for copepods.
    • (D) Correct.

Q3 (practice). At a station, NPP is 600 mg C m\(^{-2}\) d\(^{-1}\) and sediment traps catch 90 mg C m\(^{-2}\) d\(^{-1}\) at 100 m. Assume the Martin curve with \( b = 0.858 \). Which statements are correct?

(A) The e-ratio is 0.15. (B) The flux at 500 m is about 23 mg C m\(^{-2}\) d\(^{-1}\). (C) About 2% of NPP reaches 1,000 m. (D) If the community shifted to heavily ballasted coccolithophore aggregates, \( b \) would likely become smaller.

  • (a) A, B, C and D
  • (b) A and B only
  • (c) A, B and C only
  • (d) C and D only
Answer and worked solution

(a) A, B, C and D.

  • (A) \( 90/600 = 0.15 \). Correct.
  • (B) \( 90 \times 5^{-0.858} = 90 \times 0.251 = 22.6 \). Correct.
  • (C) \( 90 \times 10^{-0.858} = 12.5 \) mg C m\(^{-2}\) d\(^{-1}\); \( 12.5/600 = 2.1\% \). Correct.
  • (D) Correct: faster-sinking, ballasted particles spend less time in the twilight zone, so attenuation is weaker and \( b \) smaller.

Quick Revision Table

ItemFact
Energy vs nutrientsenergy flows one way; nutrients cycle
Lindemantrophic-dynamic concept, 1942
Transfer efficiencyabout 10% (5–20%); \( P_n = P_1 \text{TE}^{n-1} \)
Fractional TL\( 1 + \sum \text{DC}_{ij}\text{TL}_j \)
Pyramidsbiomass may invert; energy never
Chain lengthupwelling 1–2 steps; open ocean 4–5
Controlbottom-up, top-down, cascades (otter–urchin–kelp), keystone (Pisaster, Paine), wasp-waist
Microbial loopPomeroy 1974; Azam et al. 1983; DOM → bacteria → HNF → ciliates → copepods
Abundancesbacteria about \( 10^6 \) mL\(^{-1}\); viruses about 10× more
BGE, BCDBP/(BP + BR); BP + BR
Viral shuntlysis to DOM, keeps C in loop, “kill the winner”
Pumpssolubility, soft-tissue, carbonate (counter-pump)
e-ratioexport/NPP, about 0.1–0.3
Martin curve\( F = F_{100}(z/100)^{-0.858} \); 14% from 100 to 1,000 m
N cycleN\(_2\) fixation (Trichodesmium); denitrification and anammox in OMZs
DMSfrom DMSP; CLAW hypothesis (1987)

Key Takeaways

What’s Next