The Habituation Bound: How Long a Static Bird Scarer Actually Protects a Ripening Crop, and Why No Rotation of Static Scarers Holds Deterrence for a Full Ripening Window

A common piece of garden advice is that a scarecrow, a line of reflective tape, or a hanging disc will keep birds off a ripening crop. This paper tests that advice with a stated habituation model. Deterrence effectiveness starts at 0.85 for a newly installed scarer and decays exponentially, with a time constant of 3 days, towards a residual floor of 0.05; the crop is sweetcorn in a 21-day ripening window, and the unprotected damage rate is 2.5 percent of the crop per day (52.5 percent over the window). Under these assumptions a static scarer stays above a protection threshold of 0.50 — preventing at least half the unprotected damage — for only 1.73 days, which is 8.2 percent of the window. The damage rate is back to 75 percent of the unprotected rate by day 4.16 and to 90 percent by day 8.32, and the crop still loses 43.84 percent against 52.5 percent unprotected: the scarer prevents just 8.66 percentage points of loss. Moving the same scarer helps modestly at a 3-day interval (32.58 percent loss) and gets worse as moves become more frequent, reaching 43.7 percent loss when it is moved twice a day, because each move spends part of the device's remaining novelty. Changing to a different scarer type every 2 days is the best simple schedule found, at 22.88 percent loss, and the best schedule in a fine scan — a type change every 1.25 days — still loses 22.35 percent. No rotation of static scarers, at any interval tested, holds deterrence at or above the 0.50 threshold for the whole 21-day window. Resting a scarer type also recovers little: after 14 days of absence its peak effectiveness recovers only to 0.742, which buys 1.29 further days above threshold. The claim that putting up a scarecrow makes a crop safe is disproved under the stated assumptions; what the arithmetic supports is a rotation of genuinely different scarer types, changed roughly every one to two days, as damage reduction — not as protection. This study is simulated, not measured.