Food webs, indirect effects and trophic cascades

Read the relationship carried by the arrow

In this chapter's trophic panel, arrows run from resource to consumer: producer P to grazer H, grazer H to consumer C, and producer P directly to C. A trophic link records a feeding relationship. It does not, by itself, specify how much energy flows or how the populations change.

The influence panel uses a different convention. An arrow runs from an influencing actor to the affected node and carries a signed response coefficient. Consumer C negatively affects grazer H; grazer H negatively affects producer P; C also has a direct negative effect on P. These are stipulated relationships in a small model, not effects inferred from co-occurrence.

Indirect paths can oppose a direct effect

An increase in C can reduce H, and a reduction in H can release P from some grazing. That two-step path has a positive sign because it combines two negative effects. A trophic cascade is a sequence of indirect responses across feeding levels. Its sign and strength depend on the links and conditions, including alternative resources and additional pathways.

In our one-pass linear model, a unit increase in C changes H by −2. Each unit change in H changes P by −3, so the indirect contribution to P is (−2) × (−3) = +6. The direct C-to-P contribution is −7. Adding the two contributions gives −1. The positive indirect path is present even though the net assigned response of P is negative. This simple addition assumes compatible response scales and ignores feedback and nonlinear changes.

Energy flow is a conserved account with changing form

Food can carry chemical energy between organisms. Some of that energy is transferred onward in food or detritus, while metabolic processes dissipate energy as heat. Heat remains energy, but it is not recycled through the pictured feeding relationships into the original chemical-energy input. Continued living activity requires a supply of usable energy.

The energy ledger begins with 1,000 J of captured chemical energy. The producer's assigned outputs are 600 J as heat, 250 J to detritus and 150 J to the grazer. Further outputs also balance. These are integrated flows over the chosen interval with no net change in the represented energy stocks, not a biomass pyramid or a universal transfer-efficiency rule.

Nutrient recycling uses a different ledger

A separate material ledger follows units of one conserved nutrient element. A mineral pool supplies forty units to producers; feeding and detritus routes carry those units through the represented community; decomposition returns forty units to the mineral pool. Its starting and ending amount is one hundred in this closed example.

The two ledgers must not be added together. Joules measure energy and the assigned nutrient units measure an amount of material. Real communities exchange materials across boundaries and can retain them in stocks for different lengths of time. The closed, balanced example illustrates how material can return to an available pool while energy is dissipated as heat. The next chapter develops ecosystem and biogeochemical accounts in more detail.

Impact needs a response and an abundance denominator

A keystone effect is large relative to a species' abundance as well as important for a defined community response. A numerically or physically dominant species can have a large collective effect simply because it is abundant. Relative impact asks a different question. Abundance may mean biomass or another justified measure; switching that denominator can change the comparison.

Our counterfactual teaching comparison assigns A 5% of included biomass and a 30% change in a defined community response when its contribution is absent. The response-change fraction divided by biomass fraction is six. B has 50% of biomass and a 20% response change, giving 0.4. C is rarer than A but has only a 1% response change, giving 0.5. These stipulated contrasts do not identify actual keystones or supply a universal cutoff.

A web hypothesis is not a demonstrated effect

A rare species is not automatically a keystone, and neither is every top predator. Important effects can involve several trophic levels and mechanisms beyond consumption. A highly connected node in a selected diagram need not have the largest causal effect. The inference requires evidence about the response, baseline, context and time horizon.

Co-occurrence can suggest a relationship to investigate, while a before-and-after change can have several explanations. Environment, movement, detection or simultaneous changes can contribute. Our signed model assumes its causal links; it does not discover them. A defensible real account must establish what the evidence supports without treating a clean-looking arrow or a suggestive sequence as proof.

Worked example

Check your understanding

1. Why do the trophic and influence arrows between P and H point in opposite directions?

   Show reasoning

   P-to-H in the trophic panel means that P supplies a resource to H. H-to-P in the influence panel means that H affects P. The conventions describe different relations, and only the influence link carries the stipulated negative response.

2. Does nutrient recycling mean that the same usable energy is recycled in the food web?

   Show reasoning

   No. Nutrient material can return to an available pool. The energy ledger conserves energy while dissipating it as heat, which does not feed back into its captured chemical-energy input. Their units and routes are separate.

3. Species C is the rarest in the impact model. Why does this not make it the strongest keystone candidate?

   Show reasoning

   Its assigned response change is only 1%, and dividing that by its 2% biomass share gives 0.5. A's comparison gives six. Rarity alone does not establish disproportionate or important impact, and no actual causal evidence is supplied by the toy model.

What to carry forward

State what each arrow means, balance the relevant flow, and connect an impact claim to a defined response and abundance measure.