How Dealate's Forecast Works
A reason to look up, not a guarantee. Here's honestly what goes into that number, and what it doesn't claim.
What goes into a forecast
Dealate reads live weather for your location — temperature, wind, humidity, and recent rainfall — from Open-Meteo, then compares it against what's known about when each ant species tends to fly.
Species-specific thresholds, grounded in published research
For species with real published research on their flight triggers, Dealate uses thresholds specific to that species — for example Boomsma & Leusink's 1981 study of weather during nuptial flights in four European ant species (Oecologia 50(2): 236–241) and Hart et al.'s 2018 citizen-science analysis of UK flight triggers (Ecography 41(6): 877–888). For species without that level of published detail, Dealate falls back to a general model built from the same body of ecological literature at the genus level. See which species have dedicated thresholds →
From weather to odds
Each day's conditions are weighed against a species' known preferences and turned into a plain-language tier — very low, low, moderate, high, or very high — rather than a raw model score. The exact weighting and thresholds are the engineering behind Dealate and stay unpublished, the same way most scoring systems explain their approach without handing over the tuned parameters.
What the confidence range means (and doesn't)
The range shown alongside a forecast is a heuristic margin that widens the further out a day is — not a statistically fitted interval. Nuptial flights depend on local triggers no model fully captures, so treat a high score as a strong reason to look up, never a promise.
Frequently asked questions
Is Dealate's forecast a black box?
No. Dealate scores live weather against each species' known flight-triggering conditions, drawn from peer-reviewed research where it exists and a general ecological model otherwise. We publish how the method works and the science it's grounded in — we don't publish the exact tuned numbers, the same way most scoring systems keep their tuned parameters private while explaining the approach.
Why doesn't Dealate publish the exact thresholds and formula?
The underlying science is public — you can read the same papers we do. The specific tuned thresholds and how we combine them into a score are the engineering behind Dealate, so we keep those private rather than handing over a ready-to-copy spec.