Dispersal ability and niche breadth influence interspecific variation in spider abundance and occupancy

The relationship between species local abundance and their regional distribution (occupancy) is one of the most extensively recognised and investigated patterns in ecology. While exceptions exist, the generally held model is that locally abundant species also tend to be more widespread geographically. However, there is only a limited understanding of both the mechanisms driving this relationship, and their scale dependency. Here we use occupancy and abundance data for 123 species of spider from across the Canary Islands to understand how both dispersal ability and niche breadth might mediate variation among species for local abundance and occupancy. We test the predictions that (i) dispersal ability explains variation among species for both abundance and occupancy, and (ii) species with a higher degree of habitat specialisation, reflecting more limited niche breadth, will have both higher occupancy and abundance. We find no evidence within habitat patches for an effect of dispersal ability on either local abundance or site occupancy, while across all patches species with higher dispersal ability tend to occupy more sites. Species largely restricted to laurel forests have higher abundance than species with broader niche breadth, but similar occupancy. The study revealed that dispersal ability and niche breadth were significant predictors of the abundance-occupancy relationship, highlighting the importance of both factors for understating patterns of abundance and occupancy among spider species.Funding provided by: Agencia Estatal de InvestigaciónCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100011033Award Number: CGL2017-85718-PFunding provided by: Agencia Estatal de InvestigaciónCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100011033Award Number: PID2020-116788GB-I00Funding provided by: Ministerio de Ciencia e InnovaciónCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100004837Award Number: PRE2018-083230Funding provided by: "la Caixa" FoundationCrossref Funder Registry ID: http://dx.doi.org/10.13039/100010434Award Number: ID 100010434Funding provided by: Marie CurieCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100000654Award Number: 847648The 5' region (658 bp) of the mtDNA COI gene was amplified using the primers LCO1490 and HCO2198 in order to identify presumed biological species (PBS). A custom R script was used to produce an unweighted pair group method with arithmetic mean (UPGMA) tree from pairwise K2P distances.PBS were then taxonomically assigned to species or genus level based on GenBank and BOLD Systems search results.PBS (hereafter species) were categorised for dispersal ability, considering the potential for juvenile stages to be passively dispersed by air currents, while suspended from silk threads, henceforth referred to as 'ballooning', which has been broadly used as a proxy for dispersal ability.A two-step process was used to categorise species as either 'non-specialist' or 'specialist', based on their affinity to the laurel forest habitat, i.e., whether they are present in a broad range of habitats or restricted mainly to laurel forest. As a first step, the species list was assessed by five Canary Island spider specialists, ), to categorise species as either laurel forest specialists or non-specialists based on their knowledge of the biology of the species.In a second step, species association to the laurel forest was quantified using distribution records within the Biodiversity Data Bank of the Canary Islands (https://www.biodiversidadcanarias.es/biota/). The total number of 500 x 500 m cells occupied by a species in the archipelago was quantified and the percentage of those cells corresponding with laurel forest habitat was then calculated. For the subset of species that were categorised by the panel of experts as laurel forest specialists, species were ranked from highest to lowest percent laurel forest occupancy, and the percentage corresponding to the 50% quantile was used for the categorisation of the species non-characterised in the first step (i.e., above this percentage species were considered as 'specialists', whereas below it, species were categorised as 'non-specialist'). Additionally, 10%, 25% and 75% quantiles were also used for niche breadth categorisation to explore the robustness of further inferences. For each species, local abundance was calculated as i) the mean site abundance (i.e., the sum of all individuals divided by the number of occupied sites at both island and archipelago scales, so with species as the replicate unit), and ii) the individual site abundance (i.e., the sum of all individuals within each site, so with site as the replicate unit). Occupancy was calculated both across islands (i.e., presence or absence of each species within each of the islands) and across sites (i.e., presence or absence of each species within each site). The latter was again calculated both at the archipelago (across all sites) and at the island (across all sites within a given island) scales.AORs were expressed as a logistic regression between the occupancy and the log of the abundance per species (mean site abundance per species)