Spatial connectedness imposes local- and metapopulation-level selection on life history through feedbacks on demography

Dispersal evolution impacts the fluxes of individuals and hence, connectivity in metapopulations. Connectivity is therefore decoupled from the structural connectedness of the patches within the spatial network. Because of demographic feedbacks, local selection also drives the evolution of other life history traits. We investigated how different levels of connectedness affect trait evolution in experimental metapopulations of the two-spotted spider mite. We separated local- and metapopulation-level selection and linked trait divergence to population dynamics.Longterm_metapops: local population density over the long-term evolution experiment. Variables: TREAT (factor; code of metapopulation) PATCH (factor; code of local population) DATE (date; day of recording) AFEMA (numeric; n. of adult females recorded) AFEMA_After (numeric; n. of adult females moved back to the experimental setup) WEEK (numeric; progressive n. of weeks since the beginning of the experiment) AFEMA_tminus1 (numeric; n. of adult females recorded at time T-1) DeltaT (numeric; n. of days since the last record) LENGTH (factor; metapopulation connectedness) NO_R (factor; was randomization performed?) REPLICATE (numeric; n. of replicate metapopulation for a given treatment (connectedness*randomization)) Dispersal_Disps: n. of daily dispersers during the dispersal propensity trial. Variables: Metapop (factor; code of metapopulation of origin) Length (factor; metapopulation connectedness) was.random (binary; was randomization performed?) Density (factor; starting density of the trial) Time (days) Bridge (numeric; n. individuals recorded on the bridge) Cotton (numeric; n. individuals recorded drowned in the cotton around the trial setup) Dispersers (numeric; n. individuals who succesfully dispersed) Disps_cumsum (numeric; cumulative sum of dispersers) IndividualGrowth: n. of eggs, larvae and juveniles recorded for each individual during the individual growth rate trial. Variables: crate (factor; code of metapopulation of origin) patch (factor; code of local population of origin) position (factor; position of the patch: corner, center of edge) replicate (numeric; n. of replicate for each local population) time (numeric; day of recording) eggs (numeric; n. of eggs spotted) larvae (numeric; n. of larvae spotted) juveniles (numeric; n. of juveniles spotted) adults_F (numeric; n. of adult females spotted) adults_M (numeric; n. of adult males spotted) LENGTH (factor; metapopulation connectedness) was.random (binary; was randomization performed?) dailyEgg (numeric; n. of eggs laid that day) lifeEgg (numeric; n. of eggs laid during the whole lifetime till that day) Starvation: n. of surviving individuals at each time point. Variables: T (numeric; time point number) Time (numeric; n. of hours from the beginning of the trials) Alive (numeric; n. of alive individuals) Dead (numeric; n. of dead individuals) Missing (numeric; n. of missing individuals) Metapop (factor; code of metapopulation of origin) Lenght (factor; metapopulation connectedness) was.random (binary; was randomization performed?) Funding provided by: Fonds Wetenschappelijk OnderzoekCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100003130Award Number:Metapopulations of T. urticae were established on a 3-by-3 grid of patches connected with bridges of different length and let to evolve for ~12 generations. A subset of the replicates were weekly randomized to remove local kin and kind structure. Local population densities were recorded once a week. After the experimental evolution period, different evolved traits were tested: starvation resistence, individual growth rate and dispersal propensity