Strong and weak trait-environment associations in subarctic stream diatoms

Ecological traits are functional characteristics measurable at the species level and provide valuable insights into how organisms respond to environmental constraints. Here, we investigated how diatom trait-groups and individual species respond to environmental variables, and identified indicator species that are particularly sensitive to environmental variation. Diatoms were sampled at 129 sites in the subarctic streams of Norwegian islands and mainland, and were categorised into three trait groups: high-profile species that live in an erect position, low-profile species living in low position along the surface, and motile diatoms. Data were analysed using a recently developed method known as Hierarchical Modelling of Species Communities, which is a flexible framework for joint species distribution modelling. We found that diatom trait-groups responded relatively weakly to measured environmental variables but showed positive or negative relationships with major ion levels of the water (e.g., conductivity, calcium [Ca2+], sodium [Na+] or chlorine [Cl-]). Variance partitioning showed a similar, important contribution of the chemical variables for all of the trait groups, while the contributions of physical variables and especially random (spatial) factors were notably lower for all trait groups. Our findings also highlighted considerable among-species variation in their relation to environmental variables within the trait groups. Notably, we identified a high number of indicator species within each trait group that were explained by specific environmental factors, mostly chemical variables (conductivity, pH, total nitrogen and phosphorus, Ca2+, Na+, Cl-). Our study suggests that certain diatom species can be considered as useful environmental indicators but the variability in species preferences within the trait group may in some circumstances hamper the use of ecological traits in environmental assessments. Thus, we suggest using species-level ecology combined with trait information to better track environmental change when using diatoms as indicators.