Apply shading model
When ticked, forecasts use your layout’s fitted shading model: at every timestep the calculator checks whether the sun is above or behind your real horizon — the trees, fences and buildings detected from your inverter’s history — and only counts direct sunlight when the sun is actually visible from your roof.
What changes in the forecast
Sunlight reaches your panels two ways: the direct beam from the sun’s disc, and diffuse light scattered across the whole sky. The shading model only gates the direct beam — when the sun is behind your eastern tree line at 7am, the beam contribution is zeroed, but the diffuse light (which arrives from the whole sky, not just where the sun is) still counts. In practice you’ll see slightly lower morning and evening shoulders on the forecast curve, exactly where your real production dips. Midday output is usually untouched.
On a shaded site this makes day forecasts noticeably more honest — the predicted curve tracks your measured production near sunrise and sunset instead of optimistically assuming a flat, open horizon.
When to untick it
Untick to compare against the bare no-shading prediction — useful for seeing how much energy your obstructions cost you, or for “what if I removed that tree” questions. The toggle is per-calculation and doesn’t change anything saved on the layout.
Why it might be greyed out
The toggle needs a fitted shading model on the loaded layout. If there isn’t one yet, connect an inverter to the layout and run Auto-detect on the saved-layouts bar — it infers your horizon from production history in one click. The model is specific to each layout (it depends on your panels’ orientation as well as the obstructions), so each saved layout gets its own fit.
Closes this tab and returns to the calculator.