Harvora™ prediction brief
Disease risk windows for New Zealand crops
Harvora turns published disease science into block-level infection-risk windows — twelve crop-disease models, each naming the method it runs, with the machine-learning blend held to ROC-AUC 0.90–0.99 against that science across 18,064 held-out rows.
Direct answer
What this supports
Disease surfaces report infection-window probability from wetness and temperature using published disease methods — the contributing model is named per crop rather than blended into one anonymous number.
Intended for New Zealand growers, technical partners, lenders and insurers.
Evidence and method
How the signal is formed
Published infection models are reproduced as formula windows, blended with guarded machine-learning window classifiers where a deployable artifact exists.
Evidence grade: SOURCED
Published scientific methods, faithfully implemented.
Boundary
How to read this evidence
The benchmark measures how faithfully the implementation reproduces the published science, not infection outcomes in orchards — no independent infection-outcome validation is claimed. Each formula's assumptions stay visible on the surface.
Sources and related evidence
Published 25 August 2026 · Last reviewed 25 August 2026 · Harvora technical review
Apply it to your operation