An exciting project that utilises unique, highly accurate spatial and temporal movement data for butterflies to understand mating, foraging and egg tending behaviours and predict gender differences. We are seeing significant gender differences and different modes of female behaviour that may relate to their different responsibilities. Working with colleagues at the University of York, University of Tennessee (US) and University of Nevada (US).

Using state-of-the-art GPS systems, highly spatial (1cm) and temporal (1s) accurate, to follow butterflies and create detailed movement data. Combined with mapping local geographic features we can analyse properties of the movement, such as distances between direction changes, mean squared displacement, time spent in patches of plants, speed, distributions of angles turned, etc. From the data we can fit (using AIC, etc) probability models to analyse underlying behaviour and use these properties to make predictions on gender to unmarked track data. In the first instance, we are using one reference site to analyse the data and generate analysis algorithms that are applied to data from other two sites to predict gender based on movement patterns.

Our analysis of data from site one is encouraging. There are significant gender differences in the movement and even modal data in the female movement - possibly indicating multiple activities. See the image to the right for 6 example movement tracks and the associated patches of vegetation.