Ecosystems & Food Webs

Nature's web - everything eats everything!

Junior level — plain language, no maths

Picture a summer meadow. Grass drinks in sunlight and grows; a grasshopper eats the grass; a frog eats the grasshopper; a snake eats the frog; a hawk eats the snake. Each link leans on the ones below it for food and energy. That chain of eating and being eaten is a food chain - and a real ecosystem stitches hundreds of them together into a tangled food web.

Every scrap of that energy traces back to the Sun. Plants and algae catch sunlight by photosynthesis and bank it as sugar - the producers. Animals that eat plants are primary consumers, the ones that eat them are secondary consumers, and so on up the chain. But here's the rule that shapes it all: at each step, roughly 90% of the energy leaks away as heat. That's why there are always far more plants than grazers, and far more grazers than hunters. You simply can't have more lions than wildebeest.

Pull one species out and the whole web can lurch in ways nobody sees coming. When wolves were wiped out of Yellowstone, the elk multiplied and stripped the riverbanks bare; the rivers eroded and wandered off course, and the fish crashed - an entire landscape unravelling because one predator was gone. Bring the wolves back, as rangers did in 1995, and the system slowly knits itself together again. Ecologists call such animals keystone species: their grip on the ecosystem is wildly out of proportion to their numbers.

Energy flow, trophic dynamics, and the Lotka-Volterra equations

Student level — the core equations

An ecosystem is an open thermodynamic system: energy pours in as sunlight, climbs through the trophic levels, and leaks out as heat. Lindeman's 10% rule (1942) names the leak - only about a tenth of the energy at one level makes it into biomass at the next, the rest spent on respiration, waste and decay. That brutal tax is why food chains rarely reach past four or five links: there's simply nothing left to feed a sixth. All of it rides on net primary production, the roughly 120 petagrams of carbon that producers fix worldwide each year.

Boil predator and prey down to essentials and you get the Lotka–Volterra equations (1925–26): prey \(N\) breed at rate \(r\) but are eaten in proportion to encounters \(aNP\), while predators \(P\) grow on what they catch and starve at rate \(d\) - \(\dfrac{dN}{dt} = rN - aNP\) and \(\dfrac{dP}{dt} = eaNP - dP\). The two circle a shared equilibrium \(N^{*} = d/ea\), \(P^{*} = r/a\) in endless out-of-phase oscillation - exactly the ~10-year lynx–hare cycle that Hudson's Bay fur records preserved for ninety years.

Zoom out to the whole web and network maths takes over. Its connectance \(C = L/S^2\) - realised links out of all possible ones - governs stability, and May's 1972 result was counterintuitive: a randomly wired community holds together only if \(\sqrt{SC}\,\sigma < 1\), so bigger, more connected webs are harder to stabilise, not easier. Real ecosystems escape this diversity–stability paradox with mostly weak interactions and a modular structure that keeps a local collapse from cascading. Keystone species are the glaring exception - few in number but strong in influence, which is why losing one lands so heavily.

Metabolic theory of ecology, biodiversity, and tipping points

Scholar level — full mathematical depth

01Ecology from the metabolism up

The Metabolic Theory of Ecology (Brown et al., 2004) is an audacious bet: that much of ecology follows from the physics of metabolism. An individual's metabolic rate obeys \(B = b_0\, M^{3/4}\, e^{-E/kT}\), tying it to body mass \(M\) and temperature \(T\), with \(E \approx 0.65\ \text{eV}\) the typical activation energy of biochemical reactions. From this one expression the theory predicts how population density, lifespan, growth rate and even mutation rate scale with size and warmth - and it holds across an astonishing 27 orders of magnitude of body mass.

02Kleiber's three-quarter law

The strange exponent at its heart is Kleiber's law, \(B \propto M^{3/4}\) - not the \(2/3\) you'd naively expect from surface-area cooling. West, Brown and Enquist (1997) traced it to geometry: life distributes resources through branching, space-filling, fractal-like networks - blood vessels, plant xylem, tracheae - and optimising flow through such a network forces the \(3/4\) power. It's why a mouse's heart races and an elephant's crawls, and why, gram for gram, big animals are so much more frugal than small ones.

03How many species an island can hold

Biodiversity has its own quantitative laws. The species–area relationship \(S = cA^z\) (with \(z \approx 0.25\text{–}0.35\)) says richness rises as a fixed power of area, and MacArthur and Wilson's island biogeography (1967) explains why: species number settles at a dynamic balance where the immigration rate (falling with distance from the mainland) crosses the extinction rate (falling with island size). Richness isn't a static tally but a running equilibrium between arrival and loss.

04Fragmentation and the design of reserves

That theory carries a hard practical edge. Carving continuous habitat into fragments turns a mainland into an archipelago of small "islands", each doomed to shed species down its own species–area curve - which is exactly why fragmentation erodes biodiversity so predictably. It also framed the long-running SLOSS debate - a Single Large Or Several Small reserve? - and gives conservation planning a quantitative backbone rather than a gut feeling.

05Alternative stable states and tipping points

Ecosystems don't always change smoothly. Many have alternative stable states, and a slow push can shove them past a bifurcation into an abruptly different regime that resists going back. A clear lake flips to turbid green as phosphorus mounts; a coral reef flips to algae; the Amazon, past a deforestation threshold, may stop making its own rain and slide toward savanna. These are genuine tipping points - cheap to cross, brutally expensive to reverse.

06Hearing the alarm before the crash

The hopeful twist is that systems nearing a tipping point may broadcast a warning. As a bifurcation approaches, a system recovers ever more sluggishly from small knocks - critical slowing down - showing up as rising variance and autocorrelation in its fluctuations. In principle, watching those statistics lets you detect an impending collapse before it happens, a fast-moving research frontier with urgent stakes for lakes, fisheries, reefs and the climate itself.