A large fraction of global biodiversity resides within biogenic habitats that ameliorate physical stresses. In most cases, details of how physical conditions within facilitative habitats respond to external climate forcing remain unknown, hampering climate change predictions for many of the world’s species. Using intertidal mussel beds as a model system, we characterize relationships among external climate conditions and within-microhabitat heat and desiccation conditions. We use these data, along with physiological tolerances of two common inhabitant taxa (the isopod Cirolana harfordi and the porcelain crab Petrolisthes cinctipes), to examine the magnitude of climate risk inside and outside biogenic habitat, applying an empirically derived model of evaporation to simulate mortality risk under a high-emissions climate-warming scenario. We found that biogenic microhabitat conditions responded so weakly to external climate parameters that mortality risk was largely unaffected by climate warming. We also found that desiccation outside the biogenic habitat drove substantial mortality in both species at temperatures 4.4 to 8.6 ºC below their hydrated thermal tolerances. This finding emphasizes the importance of warming-exacerbated desiccation to climate-change risk and the role of biogenic habitats in buffering this less-appreciated stressor. Our results suggests that, when biogenic habitats remain intact, climate warming may have weak direct effects on organisms within them. Instead, risk to such taxa is likely to be indirect and tightly coupled with the fate of habitat-forming populations. Our findings emphasize that conserving and/or restoring biogenic habitats that offer climate refugia could support biodiversity conservation in the face of climate warming.

Full description of Methods is included in the associated publication.

Metadata details with definitions of field (column) headers are given in the ReadMe file titled READMe_Jurgens_FacilitationClimate.txt and in the file titled File_Fields_Metadata.csv. Files include field trial data for 2 species in the file Animal_trials.csv and agar experiment data in the file Agar_data.csv. Data included in Figure 4 of the paper, which were used to calculate present day (2012) and future (2099) distributions of high-end evaporation stress for each species and which include environmental data with resampling for 2012 (measured) and 2099 (simulated; see Methods for details) are given by focal species C. harfordi and P. cinctipes and microhabitat. In each file, data were measured or calculated for intertidal microhabitats and species given in the file name using equations and methods described in Methods. These data are contained in the files: C.harfordi_MusselBed_EvapCalc_Fig4.csv, C.harfordi_Rock_EvapCalc_Fig4.csv, P.cinctipes_MusselBed_EvapCalc_Fig4.csv, P.cinctipes_Rock_EvapCalc_Fig4.csv