A settlement expansion can make an ecosystem simulation feel connected when the land changes along with the town. Start with one village edge, a meadow tile, a stream corridor, and a small refuge. One expansion converts the meadow tile to homes and relocates the species on it.
Define one expansion step
Represent the settlement footprint as a set of tiles. For this test, expand from one ring to two and change exactly one meadow tile into a home tile. Keep the stream and refuge unchanged so the effect is easy to attribute.

Move affected species to a refuge
Place two tracked species on the meadow tile before expansion. When the tile changes, move both along the marked stream corridor to the nearest refuge. Keep their total count constant; this experiment tests location changes, not birth or death.

Compare before and after states
Start with settlementRing 1, meadowTiles 3, refugeCapacity 2, and two species on the affected meadow. After one expansion, settlementRing becomes 2, meadowTiles becomes 2, and both species occupy the refuge. A replay should reject expansion if the refuge has no room.
| Measure | Before | After |
|---|---|---|
| --- | ---: | ---: |
| Settlement ring | 1 | 2 |
| Meadow tiles | 3 | 2 |
| Species at edge | 2 | 0 |
| Species in refuge | 0 | 2 |

Prompt for an AI game maker
Ask for a compact ecosystem simulation with a village, meadow, stream corridor, and refuge. One expansion converts one meadow tile to homes and moves the two species on it to refuge. Show the habitat and location counts before and after.
Keep climate, reproduction, food chains, and species health outside this first rule. Add those systems after one expansion resolves predictably and the habitat change is visible.
Start an ecosystem simulation prototype in SEELE
Frequently asked questions
Does this model real ecosystem dynamics?
No. It checks one tile change and deterministic relocation, not population ecology or long-term viability. Keep the starting values and the single expansion step fixed during the local replay so you can attribute each change to settlement growth. This applies to this prototype.
What if the refuge is full?
Reject the expansion in this prototype and keep the original tile state. A later version can test alternate destinations or capacity rules. Keep the starting values and the single expansion step fixed during the local replay so you can attribute each change to settlement growth. This applies to this prototype.
Should the species count decrease when habitat changes?
Not in this test. Keep total count constant and track location only, so the effect of expansion is isolated. Keep the starting values and the single expansion step fixed during the local replay so you can attribute each change to settlement growth. This applies to this prototype.


