Utilising resources: optimality (foraging focus)

Question 1

Natural selection

Answer 1

favourable heritable traits become more common in successive generations

Question 2

Optimisation by selection

Answer 2

–traits leading to greater efficiency become more common over evolutionary time
–new mutations that increase efficiency will be favoured

But how can animals behave optimally?
–optimality is mathematically demanding

Optimality in animal behaviour:
–objections
–controversy
–Usefulness

In this lecture we focus on foraging but this can apply to other behaviours e.g. mate acquisition

Question 3

optimality and behaviours - objections

Answer 3

Optimal behaviour is predicted with calculus and algebra

• but foragers do not determine their actions using higher mathematics

Question 4

early explanations: rule of thumb

Answer 4

-Janetos & Cole (1981) – e.g. move to a new site if you have a bad day’s foraging

-simple rules often come close to payoffs predicted by optimality

Rules of thumb oversimplifies
Foragers use intricate and sophisticated mechanisms involving sensory, neural, endocrine and cognitive structures, as well as interactions with genes

Question 5

Behavioural selection acts on many genes for many systems simultaneously

Answer 5

behaviour no less likely to approach optimality than other complex, multi-system outcomes

-crypsis in morphology and colouration

-design of eye / optical senses for visual predators

Question 6

Optimality is all about trade-offs

Answer 6

• foraging in current location vs travelling to a new (potentially better!) location

–feed & risk predation vs hide & risk starvation

Question 7

Optimality: a controversial gambit

Answer 7

Gould & Lewontin (1979): not everything is adapted to purpose to which it is put. Alleles can be randomly fixed; behaviours may be necessary by-products of morphological necessities

Pierce & Ollason (1987): how can optima evolve in changing environment? Optimality theory is untestable (we don’t know what animals seek to optimise, nor the available strategies)

Stearns & Schmid-Hempel (1987) “optimality” refers to the best solution within given boundary conditions; important to recognise constraints changing environments may prevent optimisation but:

• much theory is devoted to optimising under uncertainty
• micro-environments may remain relatively stable over meaningful evolutionary time-frames
• optimality theory is not, in itself, testable; however, we can test - hypotheses about currencies
  - hypotheses about constraints

Question 8

uses of the optimality approach

Answer 8

allows us to make testable predictions -What, where, when to eat

gives better insights into: currencies & constraints on adaptation

exposes the logic of our understanding of behaviour - assumptions are made clear

emphasises generality of simple problems that animals face - unites understanding of different taxa

Question 9

Diet is a fundamental driver of diversity
But why specialise?

Answer 9

why aren’t all animals omnivores, eating whatever they can find/catch?

why not always eat the one potential prey that yields most energy?

How can we answer these questions without thinking about what represents the optimal strategy?

Question 10

MacArthur & Pianka (1966) & Charnov (1976)

Answer 10

(see notes for equations)

But consider:
*prey have different handling times, hi
*prey are encountered at different rates, li
*prey have different energy content, Ei

Consider the situation where there are 2 prey types that are always attacked when located.

*“Profitability” defined as Ei / hi
*Prey 1 more profitable than prey 2, i.e. E1 / h1 > E2 / h2

The MacArthur & Pianka (1966) & Charnov (1976) model allows us to predict a switch point at which a 2nd food item should be included in the diet

*IF the predator is aiming to maximise net energy gain and
*IF it is unconstrained in its ability to take prey types 1 & 2 (i.e. it doesn’t lack the morphological features required)

Question 11

Models make specific predictions, + allow general insights

Answer 11

–higher frequency of encounter with preferred prey → more likely to specialise
–less productive environments (with lower frequencies of encounter with prey) → more generalists

Question 12

Test of the MacArthur & Pianka (1966) and Charnov (1976) model: Elner & Hughes (1978)

Answer 12

*mussel size discrimination by predatory crabs – the larger the mussel the harder it is to access the food inside requiring larger claw muscles so only larger crabs can access them
*preferred intermediate-sized mussels, as predicted
*peak of optimality observed
*choices as predicted in environments of low and intermediate productivity –
*chose large mussels in high productivity environments, counter to predictions

Question 13

Another test of the MacArthur & Pianka (1966) & Charnov (1978) with damselfly

Answer 13

Pyrrhosoma type fits predictions

Whereas Enallagma type doesn’t perhaps due to different foraging strategies e.g. ambush hunting rather than feeding on debris. If prediction is far from the observed then further research is required.

Question 14

where to eat?

Answer 14

–ideal free foragers
–marginal value theory
–central place foragers

Question 15

constraints

Answer 15

e.g. returning to the same location repetitively to feed young in a den/burrow/nest

Question 16

Ideal free distribution

Answer 16

Ideal free distribution (Fretwell & Lucas)

e.g. 10 foragers: how should they distribute themselves?

-> according to productivity (ripe fruits / day) :

                                  proportion of productivity  no. foragers if tree 1 produces 8                        0.2                               2    tree 2 produces 20                      0.5                               5    tree 3 produces 12                      0.3                               3   

At simplest, individuals compete equally for the productivity of patch so intensity of foraging (e.g. animal days) in each patch should be proportional to the relative quality of the patch.
Based on this model it is possible to predict how foragers should distribute their foraging to maximise food uptake as above the expected distribution of monkeys between fruit trees

Question 17

Diminishing returns during foraging

Answer 17

Diminishing returns during foraging
For prey choice model:
E/T = Ep/ Ts + Th

E/T = Energy / Time. i.e. the rate Of energy intake
Ep= Energy content Of prey (constant)
Ts = average search time - variable - depending on abundance increasing as prey are depleted
Th = Average handling time - usually constant

Question 18

Diminishing returns during foraging further notes

Answer 18

Intake rate declines as foraging progresses this is known as “diminishing returns”

–foraging removes food from an area
–as prey get less abundant, searching for remaining prey takes longer
–intake rate diminishes as foraging progresses

→ Sets scene for a series of models known as “patch” models

Many foragers feed on patchy resources – how long should each individual spend at each patch before moving on? And after that how far/long should it travel before stopping in another patch?

We can calculate travel time between patches and constraints of patch usage e.g. areas where predators are found.

Question 19

Patch models: graphical

Answer 19

Optimal food benefit compared to time spent travelling

For shorter travel times optimum time spent in the patch is also shorter

(see notes for graphs)

Question 20

Charnov’s Marginal Value Theorem (1976):

Answer 20

Patch models
* Formally: — Should leave patch when Intake rate in the current patch E(Tp) is equal to =
Intake rate for the environment as a Whole (including travel
time between patches)

This is the marginal value theorem’ (Charnov 1976)

• Economics term for (loosely) the gain from pursuing the
  current course of action, in preference to another option

Question 21

A test of the patch model (Cook and Cockrell 1978)

Answer 21

a test of waterboatmen:
Dry mass extracted follows law of diminishing return – in this case maps the feeding time between catch and discarding of prey animal remains

Question 22

In some cases handling time may be constant but some interference effect is associated with a prey item e.g. prey depletion and decreasing predator efficiency

Answer 22

e.g. a starling can only collect food in its beak and therefore must return to feed its offspring more regularly than say a penguin which can regurgitate food for chicks. Starling foraging methods also require it to forage with an open beak* again diminishing potential food return for the chicks.

Both penguins and Starlings are central place foragers when raising young**

*starlings foraging for leatherjackets (cranefly larvae) to take to their young probe the turf with closed bill, open to part turf and extract leatherjackets –this process becomes less efficient as number of leatherjackets in bill increases hence diminishing returns
**Starlings are classic “central place foragers” in breeding season. They must leave the nest, forage and return, up to 400 times per day in breeding season causing high selection for efficiency

Question 23

Central place foraging

Answer 23

Kacelnik (1984)

–devised test of optimal central place foraging
–trained parent starlings to collect mealworms from a feeder
–mealworms dropped in to feeder through a tube to control the shape of the gain curve
–feeder placed at different distances from the nest

starlings face decision:
–when to stop foraging and return to the nest?
–collect too few – majority of time spent flying for limited reward
–collect too many – too much time spent waiting for next prey

(see graph in notes)

Question 24

Patch models
allow us to predict optimal individual decisions regarding the amount of foraging before moving on

Answer 24

IF the predator is aiming to maximise rate of energy gain

and

IF it is constrained by the travel time between patches, or the required travel distance, as well as by a given diminishing gain curve

Question 25

Currencies
“fitness” only true measure

Answer 25

*number of descendants left far into the future
*probability that a trait is perpetuated

–intake rate – often a poor proxy

*plenty of evidence that animals don’t eat as much as they can – with good reason:
–exposure to predation
–costs of being too fat (reduced manoeuverability to avoid predators e.g. tit vs hawk)
–other activities may be important (grooming, fighting, courting, mating, resting … etc.)

Question 26

Constraints emerge from trade-offs

Answer 26

*e.g. birds can’t forage actively without causing feather-wear
and they can’t moult (to replace worn feathers) and forage actively at same time

– so it is not just imposed by animals’ abilities

*environment also important – e.g. predators & weather

Question 27

constraints compared to tradeoffs

Answer 27

Constraints:

–unalterable (by the organism)
–no decision required
–e.g. gut volume, day-length (for diurnal foragers)

Trade-offs:

–required because of constraints
–necessitate decision making (in some sense)
–e.g. how to trade-off dietary components but remain within gut volume limit; whether to forage at night (given risks and lower efficiency)

Question 28

Constraint example: diet

Answer 28

Constraints can affect currencies (foraging objectives may be more than pure energy)

e.g. moose (Belovsky 1978) feed on shores of Lake Superior and have two feeding habitats

–forest – deciduous leaves – relatively high energy, low sodium
–lakes – aquatic plants – relatively low energy but high sodium

Question 29

Optimal foraging and science
- describe
- explain
- predict

Answer 29

Describe:
*anecdotal description of foraging habits is natural history
*rigorous, methodical, quantitative description of foraging habits is science

Explain:
*optimality theory leads us towards explanations for the foraging habits of animals

Predict
*more importantly, optimality theory allows us to make predictions about how animals will behave