🌿 Energy-Budget Predator-Prey Agent Model

About this simulation

Instead of two smooth population curves (as in the classic Lotka-Volterra equations), this simulation tracks every plant, prey animal and predator as a separate agent with its own position and its own energy reserve. Population-level booms, busts, and extinctions all emerge from thousands of individual foraging, reproduction, and starvation events — nobody tells the total population what curve to follow.

How it works

• Every organism drains energy each second just by being alive (basal metabolism), and drifts around the terrain, bouncing off the edges.
• Prey that come within the foraging radius of a plant, while under an energy threshold, eat it — the plant disappears and the prey's energy rises.
• Predators that come within the foraging radius of prey, while under an energy threshold, catch it the same way.
• Any organism whose energy crosses above 80 has a chance each second to reproduce, paying an energy cost to spawn an offspring; any organism whose energy reaches zero dies of starvation — permanently, unlike a smooth equation that can shrink toward zero without vanishing.
• Because predation only happens within a short local radius, prey in an empty part of the terrain form a spatial refuge — exactly the effect a well-mixed, non-spatial model can't represent.

Controls

• Plant growth rate — probability per second a new plant sprouts, up to the terrain's carrying capacity.
• Foraging radius — how close a predator or prey animal must get to its food to catch it; widening this makes the ecosystem more forgiving.
• Metabolism rate — how fast every organism burns through its energy budget; raise it to see starvation-driven crashes happen faster.
• Simulation speed — a time-scale multiplier so you can slow down or speed up the whole ecosystem.
• +Predator / +Prey — manually inject an individual to test how the system responds to arrivals.
• ↺ Reset ecosystem — respawns the starting population. Drag to rotate, scroll to zoom.

Did you know?

Because every agent's fate depends on its own local encounters rather than a population average, running this exact simulation twice from identical starting numbers can produce a stable long-running ecosystem in one run and a predator die-off in the other — a kind of unpredictability called demographic stochasticity that smooth differential-equation models cannot show at all.