Collection-window means removed heat-exposure rank differences in four insect samples in Sweden and Madagascar in 2019-2020

We tested whether a temperature base applied after a collection-window mean preserves heat-exposure ranks. We used 5836 measured insect collections in two periods each in Sweden and Madagascar. At a 10 C base, averaging changed distinct daily exposure rates to ties in a mean 16.05% of within-trap pairs in early Sweden and 5.06% in late Sweden. Strict reversals were below 0.5%. Both Madagascar samples had no changes at 10 C. At 20 C, ties occurred in 4.96% and 8.78% of their pairs. All created ties had zero window exposure. Checks with nearby dates and 5-10 day collections retained losses. Trap selection changed the Swedish period contrast. These are arithmetic audits with fixed weather, not biological effects. One survey year per country, repeated traps, selected sites and diagnostic bases limit wider use. The package gives code, aggregate results, source hashes, validation and a figure. External source records stay outside the bundle.

thermal exposure · insect seasonality · observation bias · temperature averaging · rank preservation · Sweden · Madagascar