IMPACTS OF FLOOD SIZE ON SALMON SPAWNING HABITAT IN SOUTHEAST ALASKA

November 29, 2017

Sloat MR, GH Reeves, and KR Christiansen. 2017. Stream network geomorphology mediates predicted vulnerability of anadromous fish habitat to hydrologic change in southeast Alaska. Global Change Biology 23:604–620

In a nutshell
  • Changes in global climate systems are expected to result in increased rates of precipitation in many areas; in Alaska, precipitation rates are projected to rise by as much as 30% by century’s end
  • Higher future rates of precipitation will intensify seasonal floods in many of Southeast Alaska’s watersheds, potentially disrupting prevailing hydrologic regimes
  • Spawning habitat loss due to increased flooding will be largely restricted to lower quality habitat, specifically high-gradient and confined reaches and streams, whereas higher quality habitat, such as floodplains, will be relatively unaffected
  • Models project that increasing flood size in Southeast Alaska will have only minor effects on the spawning habitat of pink salmon and chum salmon, but coho salmon, which use high-gradient headwater streams to a greater extent than other salmon, may lose up to 16% of its present spawning range
  • This study underscores the need to incorporate watershed geomorphology into model projections of climate change on rivers, as well as the importance of preserving and restoring floodplain connectivity
The annual average temperature in Alaska has increased at a rate nearly twice that of the continental US as a result of global warming, with summers in Alaska now approximately 3.5°F and winters nearly 6°F warmer than they were just 50 years ago. Higher average seasonal temperatures are already driving extensive changes in the patterns of important geophysical processes throughout the state, but projections of future warming indicate that the annual average temperature in Alaska will rise much higher still, by an additional 2–4°F by 2050 and possibly 8–10°F by the end of the century.

Not surprisingly, a considerable amount of research has focused on trying to predict how such extreme future warming will affect Alaska’s unique ecosystems and the organisms they support. While many of the direct impacts are fairly obvious, and attract significant scientific and public attention, many secondary impacts associated with rising temperatures are less apparent. In addition to a warmer climate, for instance, models also project that many high-latitude regions will receive considerably more precipitation, in Alaska by as much as 15–30% annually. More precipitation, in tandem with higher volumes of springtime ice and snow melt, will significantly enhance the intensity and frequency of flood events, with models projecting ~17% and ~28% increases in the average annual flood size in Alaska’s coastal river systems by mid-century and the end of the century, respectively. Alterations to the hydrologic patterns of these watersheds of such magnitude would transform the rivers and streams that support myriad species of cultural, economic, and recreational importance, not least of which are the freshwater and anadromous salmonids that inhabit or frequent these waterways for reproductive purposes. “Increased average flood size is a potential concern for salmon because they spawn in the fall and their eggs incubate in the streambed during winter floods,” explains Matthew Sloat, a fish ecologist and Director of Science at the Wild Salmon Center in Portland, Oregon. “Salmon typically bury their eggs deep enough in the gravel that they are protected from being washed away during floods, even if some of the surface gravel is scoured away. However, if floods are larger in the future, they will have more power to scour deeper into the streambed and could wash away salmon eggs, potentially reducing the number of salmon that support fisheries and ecosystems.”

Using a combination of available field data and model simulations, Sloat and his colleagues assessed how projected changes in flood regimes would affect the spawning habitat of three salmon species – coho, chum, and pink – in more than 800 Southeast Alaskan watersheds by 2040 and 2080 under a mid-level greenhouse-gas emissions scenario (IPCC A1B). They also incorporated two states of channel response into their models, the first assuming a static morphology, whereby channel dimensions remained largely unchanged and future flows were restricted to current stream channels, and the second assuming a dynamic response, in which channel dimensions changed in response to future flow volume.

The resulting models indicated that spawning habitat response would vary greatly among both watersheds and salmon species, but somewhat surprisingly most of this variation was explained by local topography and the degree of geomorphic complexity, factors that mediate stream-channel responses, and less so by increasing mean annual flood size itself. Under the dynamic channel-response assumption, for instance, more spawning habitat was projected to be lost in high-gradient, narrow channels than in flatter, unconfined streams and reaches. “Channel confinement had the biggest effect on salmon habitat response to flooding in a changing climate, as our model simulations showed that floodplains help limit potential damage to salmon spawning habitat from floods. Basically, in unconfined streams, large floods spread out over the floodplain and dissipate much of the energy that would otherwise be acting to scour the streambed where salmon eggs are buried,” says Sloat.

Of the three salmon species included in this study, coho are known to use the widest range of habitats for spawning and thus were projected to suffer the greatest losses in spawning range, by up to 9–10% and 13–16% by 2040 and 2080, respectively; much of this loss will be restricted to moderate-quality habitat, however, specifically confined reaches with high gradients, whereas high-quality spawning sites, typically consisting of low-gradient, unconfined habitat and open floodplains, will be less impacted. “Coho are more likely than the other species to spawn in high-gradient, confined stream channels, and these channels are more vulnerable to streambed scour than low-gradient, unconfined channels,” explains Sloat. “That said, most coho spawning occurs in low-gradient, unconfined streams and the higher gradient, confined habitat is less preferred. So these bigger floods are more likely to affect lower quality spawning habitat for coho.” In contrast, pink and chum salmon, which spawn almost exclusively in unconfined reaches and streams with connected floodplains, will experience only minor losses in spawning habitat.

Although this study suggests that the overall impacts of future flood conditions in Alaska will be more mild than previously believed, the results of the model highlight the importance of incorporating reach- and valley-scale geomorphic features into projections of the effects that climate change will have on the state’s watersheds, as well as the necessity of restoring and improving floodplain connectivity to preserve high-quality salmon spawning habitat as Alaska’s climate becomes both warmer and wetter. “We really need to recognize the importance of places like intact floodplains that naturally buffer salmon habitat from climate-driven disturbances like floods,” says Sloat. “If we keep these types of places intact, they will help salmon populations maintain resilience in the face of climate change. In addition, predicting the effects of climate change on salmon and watersheds is only partly about climate per se; we need to understand how climate interacts with the river valleys and stream channels of salmon watersheds to predict the effects of floods and other climate-driven disturbances on salmon populations.”

Science Spotlight by Ken Ferguson