Data for: Thermal vulnerability in a mountain stream network: Temporal, spatial, and biological data

As climate change continues to increase air temperature in high-altitude
ecosystems, it has become critical to understand the controls and scales
of aquatic habitat vulnerability to warming. Here we used a nested array
of high-frequency sensors, and advances in time-series models, to examine
spatiotemporal variation in thermal vulnerability in a model Sierra Nevada
watershed. Stream thermal sensitivity to atmospheric warming fluctuated
strongly over the year and peaked in spring and summer—when hot days
threaten invertebrate communities most. The reach scale (~50 m) best
captured variation in summer thermal regimes. Elevation, discharge, and
conductivity were important correlates of summer water temperature across
reaches, but upstream water temperature was the paramount
driver—supporting that cascading warming occurs downstream in the network.
Finally, we used our estimated summer thermal sensitivity and downscaled
projections of summer air temperature to forecast end-of-the-century
stream warming, when extreme drought years like 2020-2021 become the norm.
We found that 25.5% of cold-water habitat may be lost under
business-as-usual RCP 8.5 (or 7.9% under mitigated RCP 4.5). This
estimated reduction suggests that 27.2% of stream macroinvertebrate
biodiversity (11.9% under the mitigated scenario) will be stressed or
threatened in what was previously cold‑water habitat. Our quantitative
approach is transferrable to other watersheds with spatially‑replicated
time series and illustrates the importance of considering variation in the
vulnerability of mountain streams to warming over both space and time.
This approach may inform watershed conservation efforts by helping
identify, and potentially mitigate, sites and time windows of peak
vulnerability.