The fate of the Arctic seaweed Fucus distichus under climate change: an ecological niche modeling approach

Abstract

Rising temperatures are predicted to melt all perennial ice cover in the Arctic by the end of this century, thus opening up suitable habitat for temperate and subarctic species. Canopy-forming seaweeds provide an ideal system to predict the potential impact of climate-change on rocky-shore ecosystems, given their direct dependence on temperature and their key role in the ecological system. Our primary objective was to predict the climate-change induced range-shift of Fucus distichus, the dominant canopy-forming macroalga in the Arctic and subarctic rocky intertidal. More specifically, we asked: which Arctic/subarctic and cold-temperate shores of the northern hemisphere will display the greatest distributional change of F. distichus and how will this affect niche overlap with seaweeds from temperate regions? We used the program MAXENT to develop correlative ecological niche models with dominant range-limiting factors and 169 occurrence records. Using three climate-change scenarios, we projected habitat suitability of F. distichus -and its niche overlap with three dominant temperate macroalgae -until year 2200. Maximum sea surface temperature was identified as the most important factor in limiting the fundamental niche of F. distichus. Rising temperatures were predicted to have low impact on the species' southern distribution limits, but to shift its northern distribution limits poleward into the high Arctic. In cold-temperate to subarctic regions, new areas of niche overlap were predicted between F. distichus and intertidal macroalgae immigrating from the south. While climate-change threatens intertidal seaweeds in warm-temperate regions, seaweed meadows will likely flourish in the Arctic intertidal. Although this enriches biodiversity and opens up new seaweed-harvesting grounds, it will also trigger unpredictable changes in the structure and functioning of the Arctic intertidal ecosystem.

Projected 21st‐century distribution of canopy‐forming seaweeds in the Northwest Atlantic with climate change

“…It will also depend on the availability of suitable substrate, where rocky coasts are typical along western Greenland and the Canadian Archipelagos but can be limiting elsewhere in the Arctic (Filbee‐Dexter et al, ). Similar trends are expected for seaweed populations across the Northern Hemisphere in Europe (Assis, Araújo et al, ; Franco et al, ; Jueterbock et al, ; Raybaud et al, ), the Arctic (Jueterbock, Smolina, Coyer, & Hoarau, ), and in Japan (Takao, Kumagai, Yamano, Fujii, & Yamanaka, ). Yet future studies are needed for the west coast of North America as well as Africa and South America.…”

Takeoff temperatures in Melitaea cinxia butterflies from latitudinal and elevational range limits: a potential adaptation to solar irradiance

“…Other models derive predictions by combining experimental measurements of physiological responses with climate envelopes of current distributions (Kearney & Porter, 2009;Araujo et al, 2013;Sunday et al, 2014). Jueterbock et al, 2016). Jueterbock et al, 2016).…”

Machine learning of large‐scale spatial distributions of wild turkeys with high‐dimensional environmental data

“…Our findings indicated that regularization with the MaxEnt default setting was sufficient to account for multicollinearity of the original data set of 27 landscape variables, of which 14 variables exhibited multicollinearity (Appendix Table A1). Francis et al (2017) determined the optimal complexity of MaxEnt models for American beaver by selecting variables with Akaike's information criterion and relative contribution to model fit, tuning the β parameters for regularization, and removing correlative variables following Jueterbock, Smolina, Coyer, and Hoarau (2016). Francis et al (2017) determined the optimal complexity of MaxEnt models for American beaver by selecting variables with Akaike's information criterion and relative contribution to model fit, tuning the β parameters for regularization, and removing correlative variables following Jueterbock, Smolina, Coyer, and Hoarau (2016).…”