Login / Signup

Biodiversity patterns diverge along geographic temperature gradients.

Charlie J G LoewenDonald A JacksonBenjamin Gilbert
Published in: Global change biology (2022)
Models applying space-for-time substitution, including those projecting ecological responses to climate change, generally assume an elevational and latitudinal equivalence that is rarely tested. However, a mismatch may lead to different capacities for providing climatic refuge to dispersing species. We compiled community data on zooplankton, ectothermic animals that form the consumer basis of most aquatic food webs, from over 1200 mountain lakes and ponds across western North America to assess biodiversity along geographic temperature gradients spanning nearly 3750 m elevation and 30° latitude. Species richness, phylogenetic relationships, and functional diversity all showed contrasting responses across gradients, with richness metrics plateauing at low elevations but exhibiting intermediate latitudinal maxima. The nonmonotonic/hump-shaped diversity trends with latitude emerged from geographic interactions, including weaker latitudinal relationships at higher elevations (i.e. in alpine lakes) linked to different underlying drivers. Here, divergent patterns of phylogenetic and functional trait dispersion indicate shifting roles of environmental filters and limiting similarity in the assembly of communities with increasing elevation and latitude. We further tested whether gradients showed common responses to warmer temperatures and found that mean annual (but not seasonal) temperatures predicted elevational richness patterns but failed to capture consistent trends with latitude, meaning that predictions of how climate change will influence diversity also differ between gradients. Contrasting responses to elevation- and latitude-driven warming suggest different limits on climatic refugia and likely greater barriers to northward range expansion.
Keyphrases
  • climate change
  • human health
  • healthcare
  • mental health
  • genome wide
  • south africa
  • electronic health record
  • machine learning
  • palliative care