Using traits to explain interspecific variation in diatom occupancy and abundance across lakes and streams

Abstract

Aim To discover how biological traits, ecological preferences and taxonomic relatedness are associated with occupancy and abundance of diatom species across lakes and streams. Location Finland. Taxon Diatoms. Methods We studied 288 diatom species from 492 stream sites and 230 diatom species from 290 lake sites. For each species, we calculated logit‐transformed regional occupancy and log‐transformed mean local abundance, and further determined biological traits, ecological preferences and taxonomic levels for each species. Boosted regression tree (BRT) analysis was used to reveal the linear and nonlinear associations of biological, ecological and taxonomic predictors with occupancy or abundance of lake and stream diatoms. Results There were strong and positive interspecific occupancy–abundance relationships across both lakes and streams. The BRT models explained more deviances in variation in occupancy and abundance and their relationship for lakes than streams. Biological traits, especially cell size, but also life‐form and guild, were the strongest predictors of diatom occupancy and abundance in lakes and streams when controlling for ecological preferences and taxonomic relatedness. Main conclusions In general, biological traits were the strongest predictors of occupancy and abundance in both freshwater systems. Species with similar biological traits thus tended to show similar occupancies and abundances. As indicated by lower explained deviances, occupancy and abundance in streams seemed to be more complexly structured than in lakes, suggesting that these two freshwater system types differ in the formation of biodiversity patterns. This difference may be related to the differences in hydrological connectedness between lakes and streams. Understanding how variations in species’ occupancy and abundance are formed across various waterbodies is important for meaningful biodiversity conservation.

Thermal flexibility and a generalist life history promote urban affinity in butterflies

“…First, we extracted the relative influence for each predictor variable, which shows the effect of each predictor variable on the response variable normalized to sum to 100 (Elith et al, 2008; Friedman, 2001). Second, for any variable that explained >5% of the total relative influence, we produced partial dependency plots that illustrate the influence of a given predictor variable accounting for the average effects of other predictor variables (e.g., Vilmi et al, 2019). The boosted regression tree analysis was performed using the dismo package in R (Hijmans et al, 2020).…”

“…The boosted regression tree analysis was performed using the dismo package in R (Hijmans et al, 2020). We used a tree complexity of 5, a learning rate of 0.001, and a bag fraction of 0.5 (e.g., Buston & Elith, 2011; Elith et al, 2008; Vilmi et al, 2019). Exploratory analyses varying the level of tree complexity, learning rate, and bag fraction showed no difference in the quantitative or qualitative results.…”

Rescue effect drives local persistence of fish species in the Upper Paraná River floodplain

“…Traits may also explain variation in occupancy and abundance through their effects on species environmental niches (Bergerot et al., 2015; Marino et al., 2020). Several studies have discussed the relative importance of niche breadth and niche position in explaining occupancy and abundance (Osorio‐Olvera et al., 2019; Rocha et al., 2018; Siqueira et al., 2009; Vilmi, Karjalainen, et al., 2019; Vilmi et al., 2019). In our study, the contribution of these niche measures to temporal occupancy was more contingent on the site studied.…”

“…In addition to local abundance, several not mutually exclusive mechanisms have been proposed to explain variation in spatial distribution (Borregaard & Rahbek, 2010; Gaston et al., 1997), which could also account for among‐species variation in temporal occupancy (Taylor et al., 2006). A growing body of studies has suggested that this variation can be explained by species traits (Buckley & Freckleton, 2010; Vilmi et al., 2019) and metapopulation dynamics (Freckleton et al., 2005; Verberk et al., 2010). For example, species occupancy is hypothesised to increase with dispersal ability (Gaston et al., 2000; Rocha et al., 2018; Verberk et al., 2010), for high movement among local populations would decrease the chance of local extinction (rescue effect, Hanski, 1982) and possibly increase colonisation rates (Verberk et al., 2010).…”

“…Niche characteristics have also been proposed as drivers of interspecific variation in occupancy (Siqueira et al., 2009; Rocha et al., 2018; Vilmi et al 2019). For example, the niche breadth hypothesis (Brown, 1984) considers that species that utilise a greater variety of resources would be more widespread, persistent, and abundant than specialist species.…”