Research Scope and Modeling Assumptions
PHIDS models plant–herbivore interactions on a discrete spatial grid with deterministic, tick-based state transitions. It is designed as a computational ecology instrument that privileges reproducibility, inspectability, and architectural rigor over biological maximalism.
Represented Phenomena
The current simulator explicitly represents:
- flora energy accumulation and bounded growth,
- predator swarm feeding, movement, reproduction, and starvation,
- signal and toxin mediated defensive behavior,
- mycorrhizal connectivity and root-network relay,
- airflow-influenced diffusion in environmental layers,
- telemetry suitable for comparative scenario analysis.
Deliberate Simplifications
The current implementation deliberately abstracts away:
- individual organism physiology,
- continuous-space body mechanics,
- stochastic weather fields beyond configured wind vectors,
- unrestricted species cardinality,
- unconstrained biochemical networks.
Methodological Commitments
PHIDS treats the following engineering rules as scientific commitments because they shape the space of possible simulations:
- double-buffered environment updates,
- fixed-size matrix design under the Rule of 16,
- global flow-field navigation instead of per-agent pathfinding,
- O(1) spatial-locality queries,
- strict schema validation at the API boundary.
Interpretation Guidance
Readers should interpret PHIDS as:
- a deterministic ecological simulator,
- a controlled environment for comparative scenario studies,
- an executable architectural model of data-oriented simulation design.
Readers should not interpret PHIDS as:
- a full biophysical ecosystem model,
- a continuous-time differential-equation solver,
- a free-form agent sandbox unconstrained by memory and performance invariants.
Implementation Anchors
src/phids/api/schemas.pysrc/phids/engine/loop.pysrc/phids/engine/core/biotope.pysrc/phids/engine/core/ecs.pysrc/phids/shared/constants.py