Ecology and Evolution · Unit 5: Ecosystem processes

Energy Flow and Productivity

How much energy does an ecosystem capture, and why does so little of it reach the animals at the top? This lesson defines gross, net primary and net ecosystem production as rates, shows how each is estimated with bottles, harvests and flux towers, and follows energy through consumers as an annual budget. Afterwards you can compute consumption, assimilation, net production and trophic transfer efficiencies from rates in matched units, keep detritus in the budget, and say why the "10% rule" is a rough average rather than a law.

Watch first Where All the Energy Goes · 3:03

Plants often respire roughly half of the energy they fix. Of the half it keeps, only a small share becomes herbivore, and a smaller share of that becomes predator. Where does the rest go, and how would you measure it?

This lesson treats an ecosystem as a set of accounts. Energy comes in as sunlight fixed by photosynthesis, moves between accounts when something is eaten or dies, and leaves as heat. The mechanisms of photosynthesis and respiration belong to Biochemistry: the light reactions make ATP and NADPH, the Calvin cycle uses them to fix CO2, and oxidative phosphorylation recovers energy from food through a proton-motive force. For the shared vocabulary of energy and free energy, see basic concepts of metabolism. Here you keep the totals.

Production is a rate: GPP, NPP and NEP

Production is the rate at which new organic matter, or the energy in it, is formed per unit area or volume per unit time, for example kJ m^-2 yr^-1 or g C m^-2 yr^-1. It is a flow. Biomass (standing crop) is a stock: how much is there at one moment, in g m^-2 with no time unit. Think of a bank account. Biomass is the balance; production is the deposits per year. A small balance can sit on a large income if the money is spent quickly, which is exactly what fast-dividing phytoplankton do.

Three production terms recur.

NPP = GPP - Ra NEP = GPP - Reco

Gross primary production (GPP) is the total rate of photosynthetic fixation by the producers (the first trophic level, a step in feeding defined by what an organism eats). Autotrophic respiration (Ra) is what the producers themselves respire. Net primary production (NPP) is what is left: the energy available to everything else. A common slip is to call GPP “what is left after plants respire”; that is NPP. Net ecosystem production (NEP) subtracts the respiration of the whole ecosystem, Reco, which adds the respiration of animals, fungi and bacteria to Ra. NEP is the energy or carbon the ecosystem stores over the period.

Tower studies usually report net ecosystem exchange (NEE) instead. With release to the atmosphere counted as positive, NEE = Reco - GPP, so an ecosystem that takes up carbon has a negative NEE. Check the sign convention before you read any flux graph.

Measuring it: bottles, harvests and towers

No method observes GPP directly at ecosystem scale. Each method measures something nearby and infers the rest.

Light and dark bottles. Fill paired bottles with lake or sea water, seal them, keep one dark, and incubate both. With initial oxygen I, final light-bottle oxygen L, final dark-bottle oxygen D and time t:

net community production = (L - I) / t respiration = (I - D) / t GPP = (L - D) / t

The calculation assumes respiration runs at the same rate in light and dark and that no gas escapes. The bottle also holds bacteria and zooplankton, so the light bottle’s net change is net community production, not phytoplankton NPP. Results stay in oxygen units until a stated conversion to carbon is justified.

Harvests and component sums. In grassland, clip and weigh the plants at the peak of the season. In forest, add the separately measured parts. Ji Young An and colleagues summed woody growth, litterfall and fine-root growth in nine stands in the Gwangneung Experimental Forest, Korea, and fine roots made up 18 to 44% of the estimated NPP in those stands. Leave out the roots and the total is badly short. Where NPP is high in the first place depends largely on temperature and water, the subject of climate and biomes.

Eddy-covariance towers. Fast sensors above the canopy record how vertical gusts of wind carry CO2 up and down, which gives the net exchange, NEE. GPP and Reco are then separated by a partitioning model. A common approach fits respiration to night-time data, when there is no photosynthesis, and extrapolates it into the day; a 2020 study applied a neural-network method to 36 FLUXNET2015 sites. At one Scots pine forest in Finland, over 2013 to 2017, daily GPP estimates from two carbonyl sulfide methods averaged 7% and 23% above the CO2-based estimate. Treat GPP as an estimate with a method attached.

Following energy through a consumer

Feeding links, which predation and mutualism, parasitism and coevolution describe, are the pipes energy moves through. An animal’s budget has the same logic as a plant’s. Of what it ingests (I), part leaves as feces (egestion). The rest is assimilated (A). Of that, part is respired (R) and lost as heat, and the remainder is production (P), new tissue and offspring.

A = I - egestion P = A - R

Three efficiencies follow, and their product is the trophic transfer efficiency (TTE): production at one level divided by production at the level below, in matched units and time.

TTE = (I / P below) x (A / I) x (P / A) = P / P below

Worked problem. Here is a made-up closed budget for one square metre over one year. Every number is in kJ m^-2 yr^-1, there are no imports or exports, decomposers show no net growth, and producer production that is not grazed enters detritus. GPP is 12,000 and Ra is 6,000. Herbivores ingest 1,800, egest 1,080 and respire 360. Predators ingest 200, egest 100 and respire 60. Decomposers respire 3,240.

  1. Producers. NPP = 12,000 - 6,000 = 6,000.
  2. Herbivores. A = 1,800 - 1,080 = 720. P = 720 - 360 = 360. Consumption efficiency = 1,800 / 6,000 = 30%. Assimilation efficiency = 720 / 1,800 = 40%. Net production efficiency = 360 / 720 = 50%.
  3. First transfer. TTE = 360 / 6,000 = 6%. Check: 0.30 x 0.40 x 0.50 = 0.06. The ingestion and assimilation terms cancel, and so do the units.
  4. Predators. A = 200 - 100 = 100. P = 100 - 60 = 40. TTE = 40 / 360 = 11.1%.
  5. Detritus. Inputs are ungrazed plant production 6,000 - 1,800 = 4,200, herbivore feces 1,080, herbivore production that predators did not eat 360 - 200 = 160, and predator feces 100: total 5,540. Decomposers respire 3,240, leaving 2,300 as detritus.
  6. Whole system. Total respiration = 6,000 + 360 + 60 + 3,240 = 9,660. NEP = 12,000 - 9,660 = 2,340, which is the 2,300 of detritus plus the 40 of predator production. Check: 9,660 + 2,300 + 40 = 12,000.

Notice two things. The first transfer here is 6% and the second 11%, so no single number describes both links. And the decomposers respired more than every animal combined: in this budget most NPP never passed through an animal at all.

Notice also what is not an efficiency. Gross growth efficiency, P/I, is a property of one consumer, not a transfer between levels. A ratio of biomasses is not an efficiency either, because stocks carry no time unit.

Energy budget ladder — one invented square metre, one year⚙ original · interactive
Herbivores: A = I − egestion | P = A − R
—
Herbivore TTE = P / NPP
—
Predator TTE = P / P below
—
Detritus inputs | left after decomposers
—
Total respiration | NEP = GPP − Reco
—
Balance check
—

In this sketch: a closed system with no imports or exports and no decomposer growth; GPP 12,000 and autotrophic respiration 6,000 are fixed, so NPP is 6,000; predators ingest 200 for every 360 of herbivore production, with assimilation 50% and net production 40%, as in the lesson; decomposer respiration is held at 3,240; all numbers are kJ m⁻² yr⁻¹ and the budget is made up, not measured.

Keep the rule: the transfer efficiency is the product of the three efficiencies, and in this sketch the ledger always closes, because everything fixed is respired, held as detritus, or made into predator production.

Transfer efficiency, detritus and the 10% rule

Each step loses most of what it receives, to feces and to heat. Assimilation efficiency tends to be higher for carnivores than for herbivores, because animal tissue is easier to digest than plant cell walls. Net production efficiency tends to be low in birds and mammals, which spend most of their assimilated energy keeping warm, and higher in many invertebrates and fish.

Multiplying the three efficiencies explains why food chains are short. If every step passed on 10%, a top predator at the fourth trophic level would receive 0.1% of NPP. In an analysis of 48 published aquatic food-web models, Daniel Pauly and Villy Christensen derived in 1995 a mean transfer efficiency of about 10%, which they used to estimate the primary production needed to sustain world fisheries. That is the source of the 10% rule: a rough average, not a law, and not Lindeman’s law. Real links can be far lower or higher, and food quality matters as well as quantity: producers short of nitrogen or phosphorus may limit transfer to herbivores even when carbon is plentiful. Energy is also not the only proposed limit on chain length. A 2000 study of 25 northern temperate lakes (Post, Pace and Hairston, Nature) found that food-chain length increased with lake size but not with productivity.

Detritus is the other half of the story. Dead plants, feces and carcasses feed bacteria, fungi and detritus-feeding animals, which are themselves eaten. In John Teal’s Georgia salt marsh most grass production went this way rather than to grazers. Energy does not come back from this route. Each organism’s respiration releases it as heat, and only matter, the carbon and nutrients, cycles back to producers, which is the subject of nutrient cycling.

Design problem

Students incubate light and dark bottles of lake water for 6 hours. Initial dissolved oxygen is 8.0 mg/L; at the end the light bottle reads 9.2 mg/L and the dark bottle 7.4 mg/L. These are invented numbers. Calculate net community production, respiration and GPP per hour. Then say what these numbers can and cannot tell you about the lake’s phytoplankton, and how you would improve the study.

One way to do it

Net community production = (9.2 - 8.0) / 6 = 0.20 mg O2 L^-1 h^-1. Respiration = (8.0 - 7.4) / 6 = 0.10 mg O2 L^-1 h^-1. GPP = (9.2 - 7.4) / 6 = 0.30 mg O2 L^-1 h^-1, which equals 0.20 + 0.10, a useful check.

What the numbers rest on: respiration is assumed equal in light and dark, though respiration in the light can differ, and the bottle is assumed sealed. The dark-bottle loss includes bacteria and zooplankton, so this is community respiration and the light-bottle gain is community production, not phytoplankton NPP. The figures are oxygen per litre per hour at one depth on one day. Converting them to carbon needs a stated photosynthetic quotient (moles of O2 released per mole of CO2 fixed), and converting to an annual areal rate needs measurements through the water column and the seasons.

To improve it, run replicate bottles at several depths, repeat the incubations through the year, keep incubations short enough that oxygen does not build up unrealistically, and report each result with its spread.

How we measure it

Light and dark bottles

Paired sealed bottles of water are incubated, one transparent and one dark, and dissolved oxygen is measured at the start and end. The light bottle gives net community production, the dark bottle community respiration, and their difference an estimate of gross production, in oxygen units, assuming respiration is the same in light and dark.

Harvest and component summation

Net primary production of a forest is estimated by adding separately measured parts: woody growth from repeated tree measurements, litterfall from traps, and fine-root growth from ingrowth cores of initially root-free soil. Any part left out is missing from the total.

Eddy covariance with flux partitioning

Instruments on a tower above the canopy record rapid fluctuations in vertical wind and CO2 concentration, giving the net CO2 exchange between ecosystem and atmosphere. Gross primary production and ecosystem respiration are then separated from that net flux by a partitioning model, so GPP is an estimate that depends on the model.

Mass-balance food-web modelling (Ecopath)

Measured biomasses and feeding rates, combined with parameters taken from other studies where measurements are missing, are balanced into estimated flows between groups. The flows and efficiencies it reports are model outputs that inherit those assumptions.

Finish line

Test yourself

8 questions

Practice and MCAT-style, with explained answers.

Next · Lesson 22

Nutrient Cycling

Cut a forest and its stream starts carrying away nitrogen that never left before. Why?

Continue →

← Lesson 20 · Island Biogeography

Raymond Lindeman

Raymond Lindeman

Recast a lake as a series of trophic levels, each with a productivity measured as a rate of energy, and compared successive levels as ratios, the efficiencies this lesson computes.

Lindeman did his doctoral work at the University of Minnesota on Cedar Bog Lake, a small, shallow lake at Cedar Creek, north of Minneapolis. According to biographical accounts, he began collecting data there in 1936 and completed his doctoral thesis, dated 1941. Ecologists before him often described communities by who eats whom, or by counting the animals at each step of a food chain. Lindeman's 1942 paper in Ecology, The trophic-dynamic aspect of ecology, treated the lake instead as levels through which energy passes: producers, herbivores, carnivores, and the decomposing organic matter on the lake bottom, which he kept inside the system rather than leaving it out. He expressed the productivity of each level as a rate of energy and compared one level with the next as a ratio. His numbers were rough estimates from one lake. What the paper made possible was a common accounting language: once every level is a rate in the same currency, a whole ecosystem can be audited and compared with another.

✓ verbatim Lindeman, The trophic-dynamic aspect of ecology, Ecology 23:399 (1942); Raymond Laurel Lindeman and the Trophic Dynamic Viewpoint, ASLO Bulletin 21(2), 2012.
Howard T. Odum

Howard T. Odum

Built an energy budget for a whole spring-fed stream, Silver Springs in Florida, and drew it as the boxes-and-pipes energy-flow diagram that textbooks still use.

Silver Springs, in central Florida, is a large spring whose water emerges at a nearly constant temperature and chemistry and flows away as a clear river. For an ecologist such springs were a ready-made experimental system for comparing whole communities. Odum used daily changes in dissolved oxygen to estimate the production and respiration of the community, and he estimated the standing crops of plants, herbivores, carnivores and decomposers. In his 1957 paper in Ecological Monographs he set the results out as a budget for one square metre over a year, with energy entering as sunlight, passing from level to level, and leaving each level as heat from respiration and as dead matter to the decomposers. The diagram made a whole ecosystem readable at a glance, and it set the pattern for comparing ecosystems by the size of their flows rather than by lists of species.

✓ verbatim Odum, Trophic structure and productivity of Silver Springs, Florida, Ecological Monographs 27:55 (1957).
John Teal

John Teal

Built an energy budget for a Georgia salt marsh and showed that most of the grass's production was not eaten while alive but passed into detritus.

Salt marshes on the coast of Georgia are dominated by one grass, smooth cordgrass, Spartina alterniflora, which grows in the tidal mud and produces a great deal of plant matter each year. A grazing-chain picture of ecology predicts that this production should feed herbivores, which feed predators. Teal assembled an energy budget for the marsh from measurements of the grass, the algae on the mud, and the animals and bacteria that live there, and published it in Ecology in 1962. In his budget, insects and other animals grazing on living grass took only a small part of its production. Much of the grass died, broke into fragments and was used by bacteria and by animals that feed on detritus, and a large part of the production, about 45% in one later review, was available to leave the marsh with the tides. The study moved detritus from a footnote to the centre of coastal energy budgets, and it raised a question later work took up: how much marsh production supports food webs in the estuary beyond the marsh.

✓ verbatim Teal, Energy flow in the salt marsh ecosystem of Georgia, Ecology 43:614 (1962).
Test yourself
Question 1 of 8 Score 0 / 0

Which quantity is gross primary production minus the respiration of the producers themselves?
Sources & Integrity

Every structure, slide, simulation, and clip on this page is listed below with the source it came from. None of the 3D structures is drawn by hand or by AI: each one loads by accession ID from RCSB PDB or PubChem. This panel is generated from the lesson's own data file, so what it lists is what the page loads.

Simulations

  • An energy budget that always closes · Original interactive — Biochemistrypedia (Deppmann Lab).

Scientist profiles & quotes

  • Raymond Lindeman — Lindeman, The trophic-dynamic aspect of ecology, Ecology 23:399 (1942); Raymond Laurel Lindeman and the Trophic Dynamic Viewpoint, ASLO Bulletin 21(2), 2012.
  • Howard T. Odum — Odum, Trophic structure and productivity of Silver Springs, Florida, Ecological Monographs 27:55 (1957).
  • John Teal — Teal, Energy flow in the salt marsh ecosystem of Georgia, Ecology 43:614 (1962).

Quotes are verbatim from the cited source. Profile text is original to this site and was checked against the sources cited.

Animated explainer

Where All the Energy Goes · narration: an AI narrator voice (Gemini, Charon) reading narration written from this lesson's text · animation: AI-generated in code (Manim) by Claude, a schematic with no drawn molecular structures, plots computed from their equations · captions from the narration script.

Images & figures

This lesson carries no raster figures. Every structure is rendered live from RCSB PDB or PubChem, and every plot is drawn from its equation in the browser.

integrity Machines check the mechanics. The build confirms that every local file exists, that every question's answer key points at a real choice, and that any AI-generated scientific figure names the person who checked it; a weekly job re-resolves every accession ID against its database. Questions are original; MCAT-style items are written for this site and are not real AAMC questions. Banned sources (the Tymoczko textbook, publisher decks, copyrighted figures) are not used.

Like I'm five

Plants catch sunlight and use it to build themselves. They often burn about half of it just to stay alive. Animals that eat plants get only a little of what is left, and they burn most of that too. So there is less and less for each animal up the line, which is why there are few lions and lots of grass.
High school

Plants capture energy from sunlight by photosynthesis, which ecologists call gross primary production. Plants use some of that energy for their own respiration, and what remains, net primary production, is the food available to everything else. When an animal eats, some of the food passes out as feces, some is burned in respiration and lost as heat, and only a small part becomes new animal tissue. On average only about a tenth of the production at one level becomes production at the next, but the real figure varies a lot. Much of what plants make is never eaten alive; it dies and feeds decomposers.
College

Gross primary production (GPP) is the rate of photosynthetic carbon or energy fixation. Subtracting autotrophic respiration gives net primary production (NPP = GPP - Ra); subtracting total ecosystem respiration gives net ecosystem production (NEP = GPP - Reco). With release to the atmosphere positive, net ecosystem exchange is NEE = Reco - GPP, so a sink has negative NEE. For consumers, assimilation A = ingestion I - egestion, and production P = A - respiration. Consumption efficiency (I over prey production), assimilation efficiency (A/I) and net production efficiency (P/A) multiply to give trophic transfer efficiency, consumer production over prey production, a ratio of rates in matched units. Gross growth efficiency P/I is not a transfer efficiency, and a biomass ratio is not an efficiency at all. Unconsumed production, feces and dead bodies enter detritus, where decomposers respire much of it. Transfer efficiencies average roughly 10% with wide spread, so few levels can be supported.
Graduate

GPP is never observed directly at ecosystem scale. Eddy covariance measures net CO2 exchange, and GPP and ecosystem respiration are separated by partitioning models, for example by extrapolating night-time respiration to the day, by light-response fits, or by neural networks trained on environmental drivers, as applied to 36 FLUXNET2015 sites, where agreement among methods is not independent validation. At the Scots pine stand at Hyytiälä, Finland, carbonyl sulfide based estimates of daily GPP averaged 7% or 23% above the CO2-based reference over 2013 to 2017, depending on the uptake-ratio model. Transfer efficiency also depends on food quality: nutrient-depleted autotrophs may limit herbivore transfer through nitrogen or phosphorus supply even when carbon is plentiful, so energy is not the only currency that matters.

An open question is how far the detrital pathway, rather than the grazing chain, supports production at higher trophic levels in a given system, and whether food-chain length is set mainly by energy supply or by other factors such as ecosystem size and disturbance.
Four estimates of one forest's photosynthesis

At the Hyytiälä station in Finland, in a forest dominated by Scots pine, Kukka-Maaria Kohonen and colleagues compared four ways of estimating ecosystem GPP over 2013 to 2017, two based on CO2 fluxes and two on carbonyl sulfide, a gas leaves take up along with CO2. Daily estimates from the carbonyl sulfide methods averaged 7% or 23% above the CO2-based reference, depending on the model of the uptake ratio. The study shows how much GPP depends on method at one site. It does not say which estimate is true.

— Kohonen et al., Biogeosciences 19:4067 (2022).
The production hidden underground

Ji Young An and colleagues estimated net primary production in five deciduous and four coniferous forest stands in the Gwangneung Experimental Forest, Korea, by adding woody growth, litterfall caught in twelve traps per plot, and fine-root growth measured with ingrowth cores. Fine roots made up 18 to 44% of the estimated NPP in those stands. A budget built only from wood and leaves would have missed a large share. The authors limited their inference to these nine stands and called for tests in other forests.

— An et al., PLOS ONE 12:e0180126 (2017).
More algae, less for the grazers

Hadayet Ullah and colleagues ran twelve 1,800-litre mesocosms resembling temperate coastal sea floor in South Australia under control, warming, acidification and combined conditions. The combined treatment raised primary-producer biomass, yet a food-web model built from the measurements estimated less energy flowing to herbivores. Warming increased cyanobacteria and detritus accumulated. The flows were modelled in wet mass with some borrowed parameters, so the result describes this preparation rather than the ocean.

— Ullah et al., PLOS Biology 16:e2003446 (2018).

 ↑ Top