Butterfly migration: are synoptic-scale wind systems important?

Abstract. 1. If synoptic-scale wind systems are important in determining long-distance movements of butterflies, a portion of the variation in daily counts of migrants at a site should be explainable by prior winds.
2. Using special flight traps near Gainesville, Florida, U.S.A., from 29 August until 12 November 1978, we made replicated, continuous counts of four migrant species: Phoebis sennae, Urbanus proteus, Agraulis vanillae and Precis coenia .
3. Significant SSE-ward flights occurred for one or more of the four species on 47 days between 5 September and 6 November (Fig. 1).
4. Seasonal changes in numbers of migrants were similar for the four species (Fig. 1). Median fall migrants were trapped between 22 September ( A.vanillae ) and 1 October ( U.proteus and P.sennae ).
5. Daily fluctuations in total numbers of migrants were largely attributable to local weather, viz temperature, wind speed and cloud cover (Fig. 2).
6. Neither local wind direction (Fig. 3) nor back-tracking the positions of air parcels (Fig. 4) helped explain the daily fluctuations.
7. The characteristic autumn weather patterns of south-eastern U.S.A. and the day-to-day steadiness of the numbers of migrants are incompatible with the hypothesis that synoptic-scale wind systems are important in determining butterfly migrations through Gainesville, Florida, in the autumn.