FACTORS AFFECTING THE SIZE OF LEATHERJACKET (DIPTERA, TIPULIDAE) POPULATIONS IN PASTURES IN THE WEST OF SCOTLAND

Leatherjacket (Tipula spp.) populations were surveyed in 77 pastures in the west of Scotland in the winter of 1991/92 and again in 1992/93. General soil characteristics and a number of other environmental variables were also recorded for each pasture in 1992/93. Leatherjacket numbers and associated environmental variables were analysed using Generalised Linear Interactive Modelling (GLIM) to produce a generalised linear model relating 1992/93 leatherjacket numbers to field characteristics over 68 of the pastures surveyed. The model was constructed using four categorical variables (aspect, silage use, tendency to waterlogging, organic fertiliser use) and three continuous variables (1991/92 leatherjacket numbers, sward height at time of sampling, distance from the Atlantic). All seven variables had a significant effect in the model and together accounted for 50% of the total variability in 1992/93 leatherjacket numbers, Possible modes of action are suggested for each of these variables, The predictive efficiency of the model was tested using the data from nine pastures excluded from the original data set. The results of the analyses confirmed the positive effects of waterlogging on leathejacket numbers, but also indicated the importance of other previously undocumented factors such as aspect and the influence of prevailing winds. In addition, none of the standard soil characteristics recorded (such as pH, bulk density or percentage loss on ignition) was found to have any effect on leathejacket numbers. The relevance of these findings to future research on leatherjackets in pastures is discussed.