Stepping-stone populations in linear landscape elements increase pollen dispersal between urban forest fragments

Background and aims - Understanding how to ensure connectivity by gene flow between insect-pollinated plant populations is essential for designing functional ecological networks in a human-dominated matrix such as in urban settings. Linear landscape elements (LLEs) might contribute to such connectivity by increasing pollen transfer. Methods - Pollen dispersal was investigated for the insect-pollinated herb Primula elatior in urban forest remnants, using fluorescent dye as a pollen analogue, in three study sites comprising population pairs located in fragments connected or not by a wooded LLE. We examined whether the presence of a small patch of plants of P elatior placed within the LLE (stepping-stone population) might increase pollen dispersal compared to an LLE where P. elatior is absent. Key results - Dye dispersal showed a leptokurtic decay distribution, with most dye transfer events at short distances (80% at less than 15.4-114.8 m) and a maximal distance of 647.5 m. Populations not connected by an LLE, separated by distances of 249 m or more, did not exchange dye, whereas populations that were connected by LLEs (up to 602 m) did show interpopulation dye transfer, suggesting that LLEs function as biological corridors facilitating dye dispersal. The presence of a stepping-stone population of P. elatior within the LLE increased between-population dye transfers (between 1.8 and 3.3 times the proportion of visited individuals) and deposition distances, compared to an LLE with no stepping stone, at least when pollen flow already existed. Conclusions - These results provide evidence that LLEs may be more effective for pollen dispersal when serving as both movement conduit and habitat. In urban settings, developing a network of core populations connected by LLEs comprising small, even temporary, stepping-stone populations should be encouraged.