Flood risk

Which areas are likely to flood, and how badly?

Which model should run?

Pick on the trade-off, not the name. Combined runs both and reports where they disagree.

Random ForestSteady on small or noisy samples and easy to interpret. The safe default.

XGBoostUsually the sharpest accuracy on dense feature stacks, at the cost of tuning.

Combined modelEnsembles both and reports their agreement, so disputed areas are visible.

Note.No model is trained yet. Flood risk runs a weighted overlay on the analysis service; the other three topics validate your request and show the exact payload the service will receive, so the whole flow can be used and reviewed today.

Flood riskSingle locationRandom Forest0 areas

Scope Continue