Evolutionary history determines population spread rate in a stochastic, rather than in a deterministic way

Fragmentation of natural landscapes results in habitat and connectedness loss, making it one of the most impactful avenues of anthropogenic environmental degradation. Populations living in a fragmented landscape can adapt to this context, as witnessed in changing dispersal strategies, levels of local adaptation and changing life-history traits. This evolution, however, can have ecological consequences beyond a fragmented range. Since invasive dynamics are driven by the same traits affected by fragmentation, the question arises whether fragmented populations evolve to be successful invaders. In this study we assess population spread during three generations of two-spotted spider mite (Tetranychus urticae) population in a replicated experiment. Experimental populations evolved independently in replicated experimental metapopulations differing only in the level of habitat connectedness as determined by the inter-patch distance. We find that habitat connectedness did not meaningfully explain variation in population spread rate. Rather, variation within experimental populations that shared the same level of connectedness during evolution was larger than the one across these levels. Therefore, we conclude that experimental populations evolved different population spread capacities as a result of their specific evolutionary background independent but of the connectedness of the landscape. While population spread capacities may be strongly affected by aspects of a populations evolutionary history, predicting it from identifiable aspects of the evolutionary history may be hard to achieve.