Spatial and spatiotemporal variation in metapopulation structure affects population dynamics in a passively dispersing arthropod

Abstract

The spatial and temporal variation in the availability of suitable habitat within metapopulations determines colonization-extinction events, regulates local population sizes and eventually affects local population and metapopulation stability. Insights into the impact of such a spatiotemporal variation on the local population and metapopulation dynamics are principally derived from classical metapopulation theory and have not been experimentally validated. By manipulating spatial structure in artificial metapopulations of the spider mite Tetranychus urticae, we test to which degree spatial (mainland-island metapopulations) and spatiotemporal variation (classical metapopulations) in habitat availability affects the dynamics of the metapopulations relative to systems where habitat is constantly available in time and space (patchy metapopulations). Our experiment demonstrates that (i) spatial variation in habitat availability decreases variance in metapopulation size and decreases density-dependent dispersal at the metapopulation level, while (ii) spatiotemporal variation in habitat availability increases patch extinction rates, decreases local population and metapopulation sizes and decreases density dependence in population growth rates. We found dispersal to be negatively density dependent and overall low in the spatial variable mainland-island metapopulation. This demographic variation subsequently impacts local and regional population dynamics and determines patterns of metapopulation stability. Both local and metapopulation-level variabilities are minimized in mainland-island metapopulations relative to classical and patchy ones.

Spatial Selection and Local Adaptation Jointly Shape Life-History Evolution during Range Expansion

“…We assumed density-independent aerial dispersal among grid cells, because the large spatial scale used (10 # 10-km 2 grid cells) did not allow us to sensibly incorporate density dependence (which is important at the level of a single leaf or plant). Dispersal mortality was set relatively high and reflects transience and settlement costs (90%; see De Roissart et al 2015). The probability for an individual mite to engage in aerial long-distance dispersal was modeled as an unconditional nearest-neighbor dispersal rate, determined by a single locus subject to selection/mutation.…”

The importance and adaptive value of life history evolution for metapopulation dynamics

“…We build on earlier experimental work demonstrating a profound impact of metapopulation structure on both local and metapopulation-level demography, life-history evolution and gene expression in the spider mite Tetranychus urticae [18,39]. In short, we found divergent multivariate trait evolution in metapopulations with either stable patches of variable size (mainland-island metapopulations) or with patch size varying in space and time (classical metapopulation) relative to metapopulations with equal and temporally stable patch sizes (patchy metapopulation).…”

“…The model is individual-based and simulates mite development and population dynamics as in the experiments [18,39] on a daily basis for 1200 days, after an initialization phase of 100 time steps. In the experimental evolution [18,39], mites from one genetically variable source population were either placed in a metapopulation consisting of nine equally sized patches (patchy metapopulation), nine patches with the same amount of resources randomly assigned to each patch every week until the metapopulation carrying capacity was reached (classical metapopulation), or six patches of which three were of double size (mainland-island metapopulation).…”

“…In the experimental evolution [18,39], mites from one genetically variable source population were either placed in a metapopulation consisting of nine equally sized patches (patchy metapopulation), nine patches with the same amount of resources randomly assigned to each patch every week until the metapopulation carrying capacity was reached (classical metapopulation), or six patches of which three were of double size (mainland-island metapopulation). Patches consist of leaves of 20 cm², or 40 cm² for the large patches in the mainland-island metapopulations.…”

Correction to ‘Adaptation to fragmentation: evolutionary dynamics driven by human influences’

“…the mosaic of coevolution of plant-pollinator -herbivore interactions [94] altered sexual selection divergence in sexual traits in fish due to changes in density and local habitat quality in Gambusia [95] eco-evolutionary feedbacks on other traits and correlated responses joint evolution of dispersal and other traits seed dormancy in non-dispersing morphs in Heterotheca latifolia [33] mating strategies in prairie chickens: larger clutch size and fewer nests [46]; evolution of stress resistance in Tetranychus urticae [50] genetic deterioration fitness loss herb: plants from smaller fragments had lower reproductive success in transplant experiments [96] rstb.royalsocietypublishing.org Phil. Trans.…”