Impact of habitat fragmentation on the behaviour of a Belgian butterfly, Leptidea sinapis

Details

Supervisors

Nieberding, Caroline M. ; Marcantonio, Matteo

Faculty

Faculté des sciences

Degree label

Master [120] en biologie des organismes et écologie, à finalité approfondie

Abstract

Habitat reduction and fragmentation are a major threat to biodiversity and have been linked with a decline in butterfly populations in numerous countries. To this day, we have a good understanding of the drivers of this decline, but not of the way butterflies could adapt to them. Learning is a form of reversible phenotypic plasticity that could play an important role in this case, allowing butterflies to adapt to habitat changes faster than through genetic adaptation. Insects are capable of different types of learning and many of them also display social learning. Learning can be adaptive and its impact on fitness depends on environmental conditions: its benefits should be maximum in heterogeneous habitats with a high predictability on the scale of the lifespan of the studied organism, which is generally the case in fragmented landscapes. Social learning is likely to evolve when learning is beneficial but associated with high costs or when resources distribution are very heterogeneous, and could therefore also be favoured by fragmentation. Fragmentation could as well impact other behaviours such as dispersal abilities. This study was therefore an attempt to assess the effects of fragmentation on the behaviour of the butterfly Leptidea sinapis. To do so, trials were conducted in large outdoor cages divided in quadrants with or without host plants, using butterflies caught in areas with varying levels of habitat fragmentation in Wallonia, Belgium. The trials were divided into a training and a memory phase. Butterflies were distributed into a social cohort, for which template females were placed in the arenas during the training trials, and a non-social cohort, for which they were absent. In the memory phase, which started after oviposition occurred in the training phase, host plants were covered except for two of them for the non-social cohort and three for the social cohort. During each trial, the type of behaviour displayed by the butterfly as well as the quadrant in which the behaviour occurred were recorded. In the training phase, butterflies spent less time exploring quadrants with host plants while in the memory phase they spent more time exploring them than the ones without host plants and the time spent in these quadrants decreased with trial numbers. This suggests that after several trials butterflies were able to remember host plants’ locations, thus displaying a similar exploration pattern around host plants than in the training phase, while butterflies that underwent fewer trials could not and thus displayed more curiosity towards covered host plants. Butterflies from the social cohort spent less time dispersing than those of the non-social cohort in the training phase, which could indicate the presence of conspecific attraction, and the time spent in this type of behaviour decreased with trial number for both experimental phases. This was interpreted as being the result of habitat familiarisation, as unfamiliar habitats are likely to trigger dispersal. The increase in dispersal with trial numbers for the social cohort in the memory phase could indicate that butterflies gradually learned to associate the habitat of the arena with the presence of conspecifics and that their “departure” triggered dispersal. This could indicate the presence of social learning in L. sinapis. Butterflies from the most fragmented area tried to disperse less, indicating a negative selection on dispersal, which is common in butterflies. The effect of fragmentation was different for the two cohorts: social cues were associated with a decrease in exploration and an increase in resting in fragmented habitat while the contrary was observed in more continuous habitats. This may have been due to the fact that copying conspecifics is beneficial in fragmented habitats with clustered resources while in continuous ones competition avoidance could be a more important factor influencing butterflies’ behaviours. This difference could have been the result of social learning, as this study showed evidence for the possibility of social and non-social learning in L. sinapis. Although further experiments are required to confirm the findings of this study, this is a promising start for the understanding of the effects of habitat fragmentation on insect behaviours.