💡 Climate-drift adaptive management calendar with automatic milestone shifting

[github-actions[bot]]

Summary

Build a living seasonal management calendar that automatically adjusts ALL management milestones — spring buildup start, swarm season window, nectar flow timing, fall preparation deadlines, winter cluster formation — based on real-time climate data vs historical norms for the beekeeper's specific location. "Spring buildup is running 2 weeks early in your area this year — your feeding schedule and swarm prevention checks have been shifted forward." Uses freely available satellite NDVI data (Sentinel-2), local weather station data, and historical bloom records to detect climate drift and proactively adjust management timing.

Market Signal

Research published in 2026 in Ambio (source) confirms that climate change is making traditional beekeeping calendars obsolete, with flowering patterns shifting unpredictably across regions. The BeeGuards consortium spanning 11 countries is studying exactly this problem with digitally-instrumented hives. Beezum offers static weather integration, and the BEEKIND platform provides hyperlocal climate data, but neither adjusts the management calendar itself. No beekeeping app currently auto-adjusts management timing based on climate drift — all rely on static seasonal calendars or manual adjustment by experienced beekeepers.

User Signal

The PRD includes localized guidance (FR3, FR4) and seasonal planning (FR15, FR16) as MVP features. Discussion #365 (climate-adaptive bloom forecasting) and #476 (software-only environmental intelligence) validate strong interest in climate-adaptive features but focus on data provision and bloom prediction — not automatic management timeline adjustment. The critical gap: showing beekeepers that blooms are shifting is useful, but automatically adjusting WHEN they should feed, treat, and inspect is transformative. This changes the "when" of every management decision.

Technical Opportunity

The architecture already specifies weather API and flora integration as recommendation inputs. Sentinel-2 satellite data is freely available via Google Earth Engine API — Broodly is already on GCP, making this a natural integration. Historical bloom records are available from USDA, the National Phenology Network, and equivalent agencies globally. The priority engine and seasonal planning calendar (Epic 6 in epics.md) provide the adjustment surface. The recommendation contract (action + rationale + confidence + fallback) ensures shifted timelines always come with confidence levels and the traditional calendar as an explicit baseline fallback.

Assessment

Dimension	Score	Rationale
Feasibility	high	Satellite data (Sentinel-2) and weather APIs are freely available; GCP provides Earth Engine integration
Impact	high	Directly addresses the climate change challenge documented in 2026 research; changes timing of every management decision
Urgency	high	Climate shifts are already causing losses NOW; traditional calendars are demonstrably failing; first-mover advantage

Adversarial Review

Strongest objection: Climate data granularity may not be fine enough for accurate hyperlocal predictions. Incorrect calendar shifts could cause beekeepers to manage too early or too late, potentially causing colony harm (e.g., removing winter insulation too early based on a false spring prediction).

Rebuttal: The system always presents adjustments with explicit confidence levels and the traditional calendar as the stated fallback — never removing the baseline. Multiple data sources (satellite NDVI + local weather station + historical bloom records) provide cross-validation that reduces single-source errors. The app's existing explainability contract (rationale + confidence + fallback) makes this safe: "Based on NDVI greening trends and 10-day forecast, we suggest starting spring feeding 12 days early (confidence: moderate). Traditional calendar date: April 1. Tap to learn more." The user can always override. The alternative — following a fixed calendar in a demonstrably changing climate — is already causing losses.

Suggested Next Step

Identify the top 3 freely available data sources for each signal type (satellite vegetation index, weather station networks, historical bloom records) and build a proof-of-concept for one US region showing the delta between traditional calendar timing and actual observed phenological timing for the current season.