Conservation of freshwater thermal habitats for Pacific salmon in a changing climate

Thesis (Ph.D.)--University of Washington, 2016-06

Climate adaptation strategies for freshwater biota have focused on how water temperature and hydrology will change over time, but understanding spatial patterns in water temperature will also be essential for evaluating vulnerability of biota to future climate and for identifying and protecting diverse thermal habitats. I used high-resolution remotely sensed water temperature data for over 16,000 km of 2nd to 7th-order rivers throughout the Pacific Northwest and California to evaluate spatial patterns of summertime water temperature at multiple spatial scales. I found a diverse and geographically distributed suite of whole-river patterns. About half of rivers warmed asymptotically in a downstream direction, as expected, whereas the rest exhibited complex and unique spatial patterns. Patterns were associated with both broad-scale hydroclimatic variables as well as characteristics unique to each basin. Within-river thermal heterogeneity patterns were highly river-specific, but median size and spacing of cool patches <15 °C were both around 250 m. Patches of this size are large enough for juvenile rearing and for resting during migration, and the distance between patches is well within the movement capabilities of both juvenile and adult salmon. The density, size, and spacing of patches were nonlinearly related to the resolution of water temperature; there was a lot of heterogeneity at very fine scales that may be important to fish that would be missed if data were analyzed at coarser scales. Climate change will cause warmer temperatures overall, but thermal heterogeneity patterns may remain similar in the future for many rivers. Maintaining this diverse portfolio of habitats will promote resiliency of salmon to natural and anthropogenic disturbance. I also developed an individual-based model to evaluate whether influences of climate change on growth and phenology of juvenile salmon could be mediated by the shape of stream networks. I used three network shapes of increasing spatial complexity: long, typical, and compact. Growth and movement of fish were based on water temperature and conspecific density. Under current-day climate conditions, salmon grew best and were large enough to smolt earliest in the long network. However, salmon grew best and outmigrated earliest in the compact network under future climate scenarios, suggesting that areas of high productivity may shift in the future. Increases in summer maximum temperature had a greater effect on fish responses than did increases in the rate of spring warming or day-to-day variability. Results from my research can be used to inform restoration and conservation strategies that minimize vulnerability of Pacific salmon to climate change and other stressors.