The Tree
Keeps Score
One invented season, not a forecast. Each small intervention changes a shared state: water changes roots; roots decide what the canopy can carry; the canopy decides what can be saved for later.
Day 1. The tree is ordinary, not neutral: it begins with a thin reserve from last year.
What you can change
Set the day, then add rain, shade, or a light prune. The tree is replayed from day one: late rain cannot erase early leaf loss; pruning trades canopy now for water demand later. There is no random weather after a restart.
Season ledger
No interventions yet. Move to a dry spell and make one.
What this is, and is not
This is an authored causal toy, not an ecological forecast or a calibrated tree physiology model. Its point is path dependence: every state is derived from earlier water, reserve, roots, and canopy. The numbers are internally consistent toy units, and the ledger exports those assumptions with your chosen events.