By the time crop stress is visible to the eye, it's often already had days or weeks to develop. Satellite imagery can see it earlier — if someone has time to look at it carefully, every time.
By the time crop stress is visible to the naked eye during a field walk, it's often already had days or weeks to develop. Early-stage disease, irrigation gaps, and nutrient deficiency all tend to show up in satellite and multispectral imagery well before they're visible on the ground — the underlying data exists earlier than the visible symptom does. The gap isn't a data problem. It's that reviewing that imagery carefully, across every field, on a consistent schedule, is a significant amount of ongoing work.
Vegetation health indices derived from satellite imagery can reveal stress zones in a field well before that stress becomes visible as discoloration or wilting a person would notice walking the rows. The signal is there. What's usually missing is the consistent, careful review needed to catch it while it's still an early warning rather than a visible problem.
The Agriculture copilot is built to do that consistent review: reading satellite imagery for a specific field, checking it against recent weather patterns and the specific crop being grown, and flagging early stress signals rather than requiring someone to manually inspect imagery on a schedule they may or may not be able to sustain.
Ask it to check a field's satellite imagery for early disease signs, and it doesn't just describe what's visible in the image. It cross-references the visual signal against weather data — recent rainfall, soil moisture trends, temperature patterns — and against what's typical for the specific crop at its current growth stage, to distinguish a genuine early warning from normal seasonal variation that doesn't need action.
A stress signal in isolation can mean several different things — disease, irrigation gap, or simply normal stress from a recent dry spell that will resolve on its own. The difference matters enormously for what action, if any, makes sense. This is why the copilot doesn't just flag anomalies in imagery; it reasons across imagery and weather together, the way an experienced agronomist would, before suggesting that something actually warrants a closer look or an irrigation adjustment.
None of this replaces walking the field. It's built to make sure that when someone does walk the field, they're checking the right section for the right reason — a week or two earlier than a visible symptom would have prompted the same visit.