Yield trajectory
Plan crop trajectories with crop, canopy and weather context.
Method · Agronomic prior + canopy history + gated ML
Crop-informed planning estimate
Evidence pathwayHarvora™ / Product
Explore Harvora's prediction areas, model coverage, operating tools, reasoning layer and calibration evidence — with each claim kept within its evidence boundaries.
Explore field hardware ↗Prioritise blocks and nights for frost protection.
Evidence pathwaySequence irrigation around likely dry-state transitions.
Evidence pathwaySequence scouting and disease-window decisions.
Evidence pathwayPlan bloom-day labour around bee-flyable weather.
Evidence pathwayAct early on damaging gust exposure.
Evidence pathwayPlan maturity windows from heat-unit accumulation.
Evidence pathwayPlan crop trajectories with crop, canopy and weather context.
Evidence pathwaySequence scouting and disease-window decisions.
Evidence pathwayTime scouting around named pest lifecycle windows.
Evidence pathwayPrioritise protection when convective conditions build.
Evidence pathwaySequence canopy, shade and harvest decisions before extreme heat.
Evidence pathwayPlan ripe-fruit work around calibrated rain exposure.
Evidence pathwaySchedule work and escalate preparedness as fire weather builds.
Evidence pathwayPlan crew timing, hydration and rest around regional heat conditions.
Evidence pathwayFrame seasonal water risk with long-run regional context.
Evidence pathwayScreen the coming work window for machinery access.
Evidence pathwayFind the most reliable rain-free operating windows.
Evidence pathwayBring block observations into the evidence lineage.
Evidence pathway44 registered prediction variables span 16 app surfaces, 12 crop families and 13 representative NZ regions. Group views may overlap when one variable supports more than one decision surface; the same serving contract travels across the platform, while evidence status remains surface-specific.
Plan crop trajectories with crop, canopy and weather context.
Method · Agronomic prior + canopy history + gated ML
Crop-informed planning estimate
Evidence pathwayPrioritise Psa scouting windows with published disease science.
Method · Published infection-pressure formula
Published infection-pressure method
Evidence pathwayPlan maturity windows from heat-unit accumulation.
Method · Published thermal-time and chill formulas
Heat-unit maturity timing
Evidence pathwaySequence irrigation around likely dry-state transitions.
Method · FAO-56 physics + proxy calibration
Connected moisture-state indicator
Evidence pathwaySequence scouting and disease-window decisions.
Method · Published formulas + gated ML classifiers
Formula-based disease pressure
Evidence pathwayTime scouting around named pest lifecycle windows.
Method · Degree-day formulas + gated ML classifiers
Degree-day lifecycle timing
Evidence pathwayMove from weather conditions to prioritised operating decisions.
Method · Mixed ML, physical proxies, thresholds, heuristics and climatology
Mixed weather and hazard signals
Evidence pathwayFocus the crop stage and use the left and right arrow keys, previous and next buttons, or a horizontal swipe to browse crops.
01 / 12
Kiwifruit, crop 1 of 12
Product status is explicit: operational tools are live, provisional tools remain assumption-driven, gated tools require a connection, and planned tools are not represented as live.
Named methods, applicability, limits, and prohibited claims are versioned.
Revenue, costs, margin, and risk are scoped to attributable block evidence.
Cash windows preserve scenario and source provenance.
Model multi-season P&L scenarios with clear assumptions carried through every year.
Track tax and covenant views against the accounting and facility assumptions you configure.
Bring authorised, freshly synced Xero ledger numbers into the app's financial views.
AI-generated reports persist their evidence snapshot and generation state.
Cross-block and year-over-year views use archived predictions, block-attributed actuals, and labelled forward scenarios.
Threshold alerts are evaluated independently from LLM narration.
AI-generated calendar events and grower actions remain attributable and reviewable.
Hardware integration is a later-phase capability and is not represented as live.
Harvora's prediction engine turns atmospheric and field data into block-level probabilities, with each surface's evidence status shown beside it. It forecasts frost, disease, water stress and harvest windows as honest probabilities, not vague icons. An AI reasoning layer then turns every prediction into dollars and a decision: what's likely, what it costs, and what's worth doing about it.
Every AI feature exists because of a Harvora prediction. The discipline is grounding-before-generation: deterministic engines find the signal, the AI puts it into words, and a structured fallback ships if it can't. Agentic tool-calling, retrieval-augmented answers, cross-checked responses on high-stakes questions, all sitting on an audited reliability layer, with you making every call.
Explore the intelligence layer →A prediction-driven calendar. Spray, irrigation, frost, harvest and more, planned ahead, and kept current as each new forecast sharpens the picture.
Every night, Harvora reads each block's trajectory and gathers what builds too slowly to trip an alarm: drifts, accumulating pressure, quiet anomalies, waiting in-app by morning.
The urgent calls, pushed to your phone. The reasoning layer decides what's critical and sends it by SMS and email, AI-driven out of the box, with manual thresholds still there if you want them. Deduped, intelligently bundled and paced around quiet hours and action windows.
An agentic assistant grounded in your blocks, your history, and your numbers, including actual costs when Xero is connected through its read-only, authorised and freshly synced connection. A bounded tool-calling loop that retrieves, reasons and runs what-ifs, powerful enough to act, disciplined enough to never act without you.