Meteorological determinants of spider ballooning: the roles of thermals vs. the vertical windspeed gradient in becoming airborne.

Spiders disperse by ballooning, a form of aeronautic behavior which they initiate by launching themselves into thermals. An attempt was made to define meteorological variables related to production and maintenance of thermals and use them as predictors of the number of aeronauts. Ballooning spiders were collected throughout a full growing season at an agricultural site and a native tall grass prairie 25 km distant, and numbers of ballooners were regressed against variables derived from meteorological data taken at locations near each site. The variables were the proportions of cloud cover and of possible sunshine, differences between maximum and minimum daily temperature (DT), wind speed, and a modification of the aeronautic index of Vugts and van Wingerden (1976). Ballooner numbers and meteorological variables used in the regressions were all weekly means. Significant one-step models were derived for both sites, but the addition of a second variable did not significantly increase the proportion of variation explained in either model. The modified aeronautic index explained 23% of the variation in ballooner numbers at the prairie site, while the proportion of possible sunshine explained 82% of the variation at the agricultural site. However the signs of the partial regression coefficients were contrary to expected. This may be due to the masking of short term meteorological and behavioral events by the averaging of meteorological variables and aeronaut numbers over a week. Alternatively it may indicate that the source of updrafts used by aeronauts may not always be thermals, but may sometimes be the vertical gradient in windspeed, a model which is consistent with the contrary signs of the regression coefficients.