How do bees develop complex multi-destination foraging routes?
Joseph L Woodgate, Joseph L Woodgate , James C Makinson , Ka S Lim , Andrew M Reynolds , Lars Chittka
Queen Mary, University of London, UK; Queen Mary, University of London, UK ; Queen Mary, University of London, UK ; Rothamsted Research, UK ; Rothamsted Research, UK ; Queen Mary, University of London, UK
Animals that visit multiple foraging sites face a problem, analogous to the Travelling Salesman Problem, of finding an efficient route between them. We continuously tracked the flights of bumblebee foragers to examine the processes by which routes develop. We used harmonic radar technology to track bees’ movements on an array of five artificial flowers in which minimising travel distances between individual feeders conflicted with minimising overall distance. We developed new techniques to compare flight paths as well as sequences of feeder visits. Most bees did not achieve optimal visit sequences, apparently because they prioritise movements to nearby feeders over efficient overall routes. Despite this, their routes show marked improvement with experience, largely due to refinement of the flight paths taken between destinations rather than improved sequences of feeder visits. As they gained experience, flights became straighter and bees reduced exploration beyond the feeder array. Both visit sequences and flight paths became more repeatable with experience. Flight paths of all legs of a flight improved at similar rates, whereas visit sequences showed a pattern whereby initial legs of the route became fixed early while bees continued to experiment with the order of later visits. Stabilising early sections of a route and prioritising travel between nearby destinations will reduce the search space, allowing rapid adoption of efficient routes. The cost of such efficient heuristics is that certain spatial configurations of foraging patches will lead to suboptimal routes, as seen in our bees.