Demography and Environment Shapes Genetic Variation: Spatiotemporal Genetic Dynamics in Cyclic Voles at Low Latitudes

Understanding how genetic diversity persists in populations subject to extreme demographic fluctuations and recurrent bottlenecks is a central question in evolutionary and population ecology. Cyclic species provide ideal models for investigating this ecological puzzle, as they allow testing how demographic phases shape genetic dynamics. Here, we examine spatial and temporal variation in neutral genetic diversity in two populations of the common vole (Microtus arvalis) inhabiting agricultural landscapes in northern Spain. Using microsatellite data collected over 5 years and eight sampling events spanning different phases of the cycle, we show that genetic diversity remained consistently high despite > 200-fold fluctuations in abundance, although trajectories differed between sites and cycle phases. Demographic crashes caused transient reductions-particularly in allelic richness-but genetic diversity was restored within one to two generations. Our results indicate that the short duration of bottlenecks limited the opportunity for drift, while immigration was the predominant driver of post-collapse recovery. Gains of novel alleles during recovery phases, rather than vole reproduction, best explained increases in genetic diversity. Our results demonstrate that genetic structure is highly dynamic, varying predictably with cyclical temporal transitions rather than simple spatial heterogeneity. While demographic cycles drive regionally consistent patterns of genetic diversity and gene flow dynamics across sites, local spatiotemporal contingencies cause deviations during specific phases. These patterns reflect site-specific variation in genetic connectivity, movement dynamics and ecological pressures, including predation. Gene flow and genetic relatedness also varied across scales and phases: local movements between field margins and into crops peaked during population increase, reducing relatedness and enhancing genetic mixing, whereas long-distance immigration ceased during decline, allowing relatedness to recover. Together, these dynamics reflect a combination of short-distance redistribution and long-distance immigration, and point to the increase phase as the key period when voles engage in effective dispersal within agroecosystems. Finally, our findings provide partial empirical support for the Charnov-Finerty model of density-dependent dispersal and gene flow, while underscoring the limitations of both the Charnov-Finerty and Lambin-Krebs models in accounting for spatial heterogeneity and asynchronous dynamics. Overall, we identify immigration as a critical buffer against genetic erosion in cyclic populations.