Bee Meadow Pollination (3D)

Foragers fly from a Blender-modelled hive to four fields, pay for the trip in flight energy, bring nectar home and pollinate on the way. Colony energy balance in real time.

Presets

Playback

Weather

Temperature below 10 C, wind above 10 m/s, rain and night keep the foragers home.

Landscape

Flower richness and the distance of the four fields from the hive.

Colony

Drawn bees are a sample; the active foragers number scales every delivery.

Forager behaviour

theta = 1 is Charnov's optimal leaving rule. Constancy and dancing shape pollination and patch choice.

Time

One real second can be many simulated minutes.

Field shown in the Charnov chart (graph 3)

Camera

Layers

Quality

Model checks

The science

The question

A honey bee colony is a small energy business. Foragers fly out to flower patches, pay for the trip in flight metabolism, collect nectar and return the sugar to the hive, where it is stored as honey and spent on brood, heating and the foragers themselves. The same visits move pollen between flowers, so the meadow's seed set depends on how the bees choose their routes. This lab ties the three together: foraging flights, nectar economics and the colony energy balance.

Nectar as fuel

A nectar of sugar mass fraction c has density rho(c) = 0.998 + 0.3856 c + 0.1 c^2 mg/uL (1.145 mg/uL at 35 %), so one microlitre carries c rho(c) mg of sucrose. Sucrose releases 16.5 J/mg, hence E(c) = 16.5 c rho(c) J/uL: about 6.6 J/uL for 35 % nectar and 8.2 J/uL for 42 % clover nectar. A full honey stomach holds 40 uL, roughly 330 J.

Cost of flying

An unloaded worker (100 mg) burns about P0 = 42 mW in flight and cruises at 6.5 m/s. Metabolic power scales with total mass, P = P0 (1 + m_load / m_bee), and loaded bees fly a little slower, v = 6.5 (1 - 0.12 m_load/m_bee). Wind w multiplies both time and power by 1 + 0.015 w^2. Working on flowers (walking, hovering, lapping) costs 0.5 P. The net energy of a trip is gain - cost and the energy return on investment is EROI = gain / cost; for a 600 m trip to a rich patch this comes out near 10:1, a figure in the range reported for real foragers.

When to leave a patch: Charnov's marginal value theorem

Each patch gives diminishing returns, g(t) = G (1 - e^(-t/tau)) - c_F t (the honey stomach fills and the flowers are depleted; c_F is the foraging cost rate). The long-run net rate for a bee that spends t in the patch and T travelling is R(t) = g(t) / (T + t). It is maximal where dR/dt = 0, i.e. g'(t*) = g(t*) / (T + t*): leave when the instantaneous rate falls to the average rate that the whole trip achieves (the tangent from the point -T in graph 3). In the simulation each forager leaves when its expected rate at the next flower, estimated from its recent takings, drops below theta R, where R is the colony's running mean net rate; theta = 1 is Charnov's rule. Longer travel time T moves t* to the right: bees should stay longer in far patches.

Nectar supply

Every flower head refills toward its maximum, dq/dt = (q_max - q) / tau_r, so a visited head returns to q_max (1 - e^(-t/tau_r)) after time t. q_max is scaled by flower richness, soil moisture and temperature (nectar flows poorly below about 10 C). Bees avoid heads that were visited in the last 15 minutes with a probability set by a scent-mark rule.

Colony energy balance

The hive store changes by dG/dt = I - D. Income I is the net energy delivered by all active foragers (the drawn bees stand in for the colony: each return counts N_active / N_drawn times). Demand is D = D0 + 1.5 W/C (18 C - T) below 18 C: brood nest warming and the colony's own metabolism. Foragers stay home when it is colder than 10 C, windier than 10 m/s, raining, or outside 06:00 to 20:30. The check box "energy ledger" verifies that G = G_start + integral I - integral D holds to numerical precision in the live run.

Recruitment and flower constancy

A bee that returns with a rate above the colony mean "dances" for its patch; the dance strength decays with a 15 minute time constant and adds to the score of that patch for other bees, which choose with a softmax of score, plus 8 % random scouting. Flower constancy is the tendency to stay with the species visited last: with probability k the next target is the nearest head of the same species even if another is nearer.

Pollination

Bees pick up pollen from anthers (35 % of the stock per visit, refilling over a couple of hours), lose 12 % to grooming per visit and deposit 1.2 % of what they carry on the stigmas. The mean seed set of a head is S = (1 - e^(-D_c/30)) e^(-D_h/80), where D_c is conspecific and D_h heterospecific pollen per head. In the mixed daisy and clover meadow low constancy fills stigmas with the wrong pollen and seed set falls; high constancy fixes it.

Numerical checks

(1) The Charnov optimum computed by bisection of the marginal-value condition is compared with a brute-force scan of g/(T + t). (2) A single simulated trip to a uniform patch, with no randomness, is compared with the closed-form sum of flight, working and return energy. (3) The live colony ledger closes to numerical precision.

Honest simplifications

This is a teaching model, not a calibrated hive. Each drawn plant stands for 30 to 60 real flower heads, visited one after another in a time K * 2.5 s; hive-to-field distances (180 to 900 m) are shown on the map at 10:1, while all energies use real metres. Parameters (42 mW, 6.5 m/s, 40 uL, nectar volumes, refill times, pollen fractions) are typical round numbers from the literature, not fits to one study. Real colonies use scouts, odour and visual learning, the sun compass, and many more patches; honey ripening, comb building, pollen foraging and water collection are left out. Bees and flowers are drawn larger than life so they can be seen.

What to try

  • Load Midsummer nectar flow and watch the hive store (graph 1) rise by day and stay flat at night; income and demand are on graph 2.
  • Switch to Fields far away (x2.2): EROI drops along the curves of graph 4 and foragers shift to the nearer borage strip.
  • Set the leave threshold theta to 0 (never give up) and then to 2: both lower the colony's net rate compared with theta = 1.
  • Pick Drought and see the nectar bars (graph 5) stay low, trips lengthen and the store fall.
  • In No flower constancy, compare the mixed meadow seed set on graph 6 with the 60 % constancy default.
  • Choose each field in the Charnov chart selector: the farther the field, the later the optimal departure t*.
Loading model...
Loading Blender model...
Hive energy store (MJ)
Colony income and demand (W)
Charnov chart: net energy vs time in the field
Energy return (EROI) vs distance, four fields
Nectar left in each field
Mean seed set per field
Made with 🟠 Blender · Three.js