Salmon begin and end their lives in fresh water. After spending most of their lives in the ocean, adult salmon return to rivers to spawn mostly in the spring, summer, and fall. Chum and some populations of Coho are adapted to spawn during winter months. There are two types of Steelhead. Winter steelhead enter rivers in winter and spawn relatively quickly in the late winter and spring. Summer steelhead enter rivers in summer then reside in streams through the fall and winter where they sexually mature and spawn in the late winter and spring. The timing of river entry, spawning, juvenile emergence, and juvenile rearing are associated with seasonal patterns in water flows and food availability affected by climate change. Environmental conditions during these critical life stages are vitally important to the survival of salmon.
Figure 1. Salmon adult and juvenile freshwater life stages and associated climate impacts (King Co. 2017; adapted from Beechie et al. 2012).
Seasonal patterns in the source and delivery of water determine river hydrology. Topography (elevation and shape of the landscape) interacts with local climates and weather in different seasons to determine rain and snow patterns (figure 2), as well as delivery of meltwater. Higher elevation watersheds are dominated by snowmelt (Sauk River – figure 2), while low elevation watersheds are dominated by rain (Samish River – figure 2). Many large watersheds in the Pacific Northwest drain both mountains and lowland valleys and have seasonal hydrology that is a mix of rain and snow (Snohomish River- figure 2). Specific climate change impacts are highly dependent on the seasonal hydrologic patterns of a particular watershed.
Figure 2. Streamflow is projected to increase in winter and decrease in late spring and summer for all basin types for the A1B (moderate) greenhouse gas emission scenario. Shifts will be more severe if emissions continue to increase. (Mauger et al., 2015).
As the climate warms, freezing levels rise, causing more precipitation to fall as rain and less as snow (figure 3). Regional precipitation is also expected to increase in intensity in winter and decrease in summer (Figure 4, Mauger et al. 2015). More winter precipitation with higher freezing levels results in less snow accumulation at high elevations to melt in the summer and more water flowing through river systems in the winter.
Figure 3. Effects of rising temperatures and freezing levels on mountain snowpacks in the Skagit river watershed. source: Skagit Climate Consortium (http://www.skagitclimatescience.org/wp-content/uploads/2016/07/FINAL_CaseStudy_Glaciers-Snowpack_2.pdf)
Figure 4. Projected changes in summer (Apr – Sept) and winter (Oct – March) precipitation in the Snohomish and Stillaguamish basins. (CIG Tribal Climate Tool, 2018)
Hydrology is the fundamental river process that underlies all other processes in freshwater systems, it is also directly tied to climate and seasonal cycles. As climate change alters the nature of seasonal weather patterns, seasonal hydrology will change with them. Hydrology affects flows and floods, temperatures, sediment delivery and deposition, habitat development, stormwater systems, and even marine conditions in the Salish Sea. Hydrologic processes are the central determinant of climate change impacts to salmon during their freshwater life stages, and are necessarily a central focus of climate change adaptation in these complex terrestrial ecosystems.
Mountain snow packs are a natural reservoir of water that accumulates during the freezing winter and melts through spring into the summer. Snowmelt provides cool water to aquifers, streams, and rivers, often when precipitation is limited. As the climate warms, freezing levels will rise, changing seasonal hydrologic patterns (Figure 2 and 3). Rising freezing levels cause less snow to accumulate in the mountains, where it remains until the weather warms with the season. The Puget Sound region is projected to have 77% less snowback between 2070 and 2100, (Figure 4, Mauger et al. 2015). Weather conditions in the winter of 2015 resulted in unusually low snow packs similar to conditions we expect in the future (Figure 5). In 2015 the amount of precipitation was normal (black line), but the snow pack (blue line) was well below the normal accumulation (red line). This means precipitation fell as rain instead of snow. Water usually stored as ice until it melts in the spring snowmelt left the system in winter, which means less water is available to melt into the summer months. This also has important implications for winter hydrology.
Figure 4. Decreases in snow fall will result in less snowpack. Snow pack is measured in snow water equivalents, which is how much water a certain amount of snow contains. Under a high emissions scenario, the April 1st Snow water equivalent is expected to decrease by ~77% between 2070 and 2099. (CIG Tribal Climate Tool, 2018)
Figure 5. SNOTEL data from Alpine Meadows, north of the North Fork Tolt River. The graph shows normal precipitation volumes, but snow volumes consistently near zero (NRCS 2017).
Low snow packs negatively impact freshwater conditions. This is particularly evident during the summer. There are significant declining trends in snow pack over the last 70 years (figure 6). Snowmelt feeds streams and rivers and fills groundwater aquifers that supply stream flow throughout the summer. Climate modeling predicts less precipitation during the summer, 22% less by 2050, which makes snowmelt even more important for summer hydrology as droughts become more common (Mauger et al 2015).
Figure 6. Snow pack for Beaver Pass in the North Cascades 1944-2015 showing a statistically significant downward trend, with a decline of -1.8in per decade. (https://cig.uw.edu/resources/analysis-tools/seattle-city-light-trends/)
Low snow packs result in low stream flows and decreased aquifer supply. Low flows can reduce spawning habitat by drying gravel beds, and kill eggs if the water table drops below salmon redds. Low flows can result in fish passage barriers limiting spawning migration and access to spawning habitat.
Figure 7. Changes in low flow in the Snohomish River for historical (blue) vs. Projected (orange) flows that occur statistically every 2 and 10 years (return interval; Mauger et al. 2015).
Decreased snow pack and snow melt decrease the volume of water that flushes into the Salish Sea, especially in later months. This decreased volume concentrates pollutants like nutrients from point and non-point sources like wastewater treatment plants, residential areas, and farms. Flows are also shifting earlier in the melt season (Figure 8). Decreased river flows (especially those from the Fraiser River in British Columbia which accounts for 2/3rds of all river inflow to the Salish Sea) are responsible for driving salt water exchange between Puget Sound and the ocean (PSEMP Marine Waters Workgroup. 2015). Decreased exchange has the consequence of concentrating human derived nutrients in the Puget Sound and warming marine waters due to decreased circulation. These conditions are hypothesized to result in a shift in food web dynamics within the Puget Sound, resulting in unhealthy phyto and zooplankton communities harmful to salmon and the food web that supports them (figure 9; Krembs). This phenomenon exemplifies the complexity of the relationship between climate and salmon, and shows that even actions like decreasing nutrient pollution are actually strategies to address climate impacts.
Figure 8: An increasing proportion of the Frasier River snowmelt is melting earlier in the year. (https://www.vancouverobserver.com/blogs/climatesnapshot/fraser-river-salmon-dying-climate-change-heats-waters)
Figure 9: Effect of climate change on summer hydrology – food web shifts due to concentration of pollutants and decreases in upwelling exchange. (Source: Christopher Krembs, Washington State Dept of Ecology)
Flooding in the Snoqualmie valley extends from valley wall to valley wall even at moderate flood stages.
Snowmelt happens in the spring and summer, but the snow that melts accumulates during the winter. Climate models predict increased precipitation in the winter, which might lead one to expect a corresponding increase in snow pack. However, despite more precipitation, warmer winter temperatures will cause a larger proportion of that precipitation to fall as rain and less as snow. As precipitation shifts toward more rain, less water is stored as snow and more will flow into rivers. This increases winter flows and flooding. Climate models suggest that the historical 50 year flood will become a 10 year flood. That is to say that major floods that only happen roughly every 50 years will be much more frequent, happening every 10 years. Likewise, today’s 100 year flood will be tomorrow’s 50 year flood (Figure 10; Mauger et al. 2015). In this way, increased winter flooding is directly related to decreased summer flow, both of which are harmful to salmon.
Figure 10. Changes in peak flow in the Snohomish River for historical (blue) vs. Projected (orange) flows that statistically occur every 10, 50, and 100 years (return interval). Mauger et al. 2015.
Flooding is a natural risk to salmon in the winter. It can disrupt spawning in the late fall and early winter and bury or scour salmon eggs in redds. Flooding also strands juvenile fish on the floodplain out of river channels, and in unhealthy drainage ditches, and flushes juveniles downstream before they are physiologically prepared for estuarine or saltwater life stages.
Flooding is also a major risk to humans, who protect communities from floods in ways that degrade salmon habitat, like building levees, removing wood, and dredging channels. Levees cut off rivers from the floodplain, amplifying the negative effects of flooding by constraining flows in fast, deep, simplistic channels. Levees also limit or eliminate the natural processes that provide flood benefits like wood recruitment and channel formation.
Figure 11. Changes in projected return frequencies and volumes for 50 and 100 year floods in the Puget Sound.
As floods get worse, and outdated flood protection infrastructure becomes insufficient, there will be pressures to build new bigger levees, to dredge river channels, and remove wood. These pressures are already occurring in areas like the lower Tolt River near the City of Carnation where old levees were built without adequate understanding of geomorphic processes. Flood protection actions are complex, expensive, and fraught with political influence. When protections from floods are simplistic and shortsighted, like dredging or building levees taller around constrained channels, protections also pose major threats to salmon. The channelization of rivers (levees and dredging) and armoring of banks are harmful to salmon populations (The Plan, 2005). Furthermore, dredging or just building a levee taller are temporary fixes, which might be cheaper in the short term, but defer solutions and greater expenses to future generations. Fortunately, actions like levee setbacks offer significant opportunities for salmon restoration by improving habitat while simultaneously offering substantially more flood protection far into the future, especially as a changing climate increases flood risks. Levee setbacks are one of the best tools we have to improve both salmon habitat and flood protection for communities. However, these kinds of multi-benefit projects often require the acquisition of private land that can be expensive, politically sensitive, and limited by the cooperation of landowners.
While flooding is harmful to salmon when they are in the river, it is also a natural habitat forming process that erodes banks, recruits trees and large wood into rivers to form log jams, moves sediment, and forms new channels and spawning and juvenile rearing habitat. Salmon are well adapted to bounce back from major floods when they are infrequent and given the opportunity to create habitat. More frequent and larger floods, particularly those constrained in leveed reaches, and the tendency for humans to respond to flooding by degrading habitat, will increase risks to salmon populations.
Increased winter precipitation will increase stormwater runoff in both duration and volume. Increased stormwater runoff is likely to amplify stormwater impacts from toxic contaminants and flashy amplified hydrographs in urbanizing catchments. These impacts will intensify as population growth, development, and traffic increase independent of climate change. Stormwater systems will be challenged by increased flows. Insufficient culverts can cause erosion and road failure and limit access for salmonids. Heavier winter rain will increase erosion, soil saturation, and landslide potential, which will increased sediment loads in rivers and streams. High sediment loads can bury salmon or erode redds and harm juvenile and adult fish by smothering gills, and limiting foraging opportunity.
A salmon attempts to swim over the roadbed of 399th Street outside Gold Bar in October, 2009, after waters from the Wallace River flowed over the road during flooding. (Mark Mulligan/Herald file photo; https://www.heraldnet.com/opinion/editorial-no-avoiding-duty-to-restore-salmon-habitat/)
As the fundamental process that underlies the river system, hydrologic processes are responsible for forming and maintaining the habitats salmon depend on and actions that improve climate resiliency for salmon are typically associated with restoring and protecting hydrology. While climate change will manifest itself differently in different seasons, summer and winter hydrology are two sides of the same coin, so effective strategies for protecting and restoring natural hydrologic process benefit hydrology and address climate impacts all year long.
Diseased and dead pre-spawn Sockeye in the Columbia River basin in 2015. (Steve Ringman/Seattle Times)
As the climate changes, summers will be warmer and drier, with more intense and longer heat waves, and longer periods without rain, decreasing summer low flows. Winters will also be warmer, with fewer freezing days, and more precipitation deposited as rain instead of snow, leading to higher winter flows and smaller snow packs. Warmer air can also hold more moisture, which is the primary reason winter storms are expect to increase in severity. Increased air temperatures (2.2 to 3.3 degrees C by 2050; Mauger et al., 2015; figure 6) will cause diminished snow packs to melt quicker and increase stream water temperatures (Figures 12, 13).
Figure 12. Projected increases in average August stream temperatures in the Puget Sound Basin for the 2040s and 2080, compared to historic conditions (1970-1999). Projections are an average of ten global climate models based on a moderate greenhouse gas scenario (A1B). If emissions continue at the current rate, projections will be significantly higher (Mauger et al. 2015). Figure created by Jonathan Picchi-Wilson, Western Washington University, based on the CMIP3 projections used the IPCC 2007 report. Data source: Isaak et al. 2011.
These conditions increase the likelihood that stream temperatures will exceed the healthy tolerances of salmon more often and for longer periods (Figure 12). Generally, increased temperatures negatively affect metabolic rates, feeding requirements, prey composition and availability, disease/infection resistance, and oxygen concentration, resulting in reduced productivity, fitness, disease, and mortality – similar to effects observed in the summer of 2015 (Figure 13).
Figure 13. Temperature (7 day average daily maximum [7-DADMAX] and minimum) in the mainstem Snoqualmie River during the summer of 2015. Department of Ecology temperature standards for the period are shown as dashed lines. (King County 2016)
Figure 14: As a changing climate results in less water and warmer air temperatures, water temperatures increase risks to salmon. https://www.vancouverobserver.com/blogs/climatesnapshot/fraser-river-salmon-dying-climate-change-heats-
Thermal refugia are places where cold water is maintained in a river systems. They can be deep pools, places of groundwater discharge in a stream channel or cold tributaries. Thermal refugia will become increasingly important as the climate changes. Identifying and protecting these cold water refugia will be increasingly important as the climate changes. One important thermal refugia is the South Fork of the Skykomish River above Sunset Falls. Sunset falls is a natural barrier to migration, but a trap and haul program operated by the Department of Fish and Wildlife provides for passage above the falls. In low return years, such as 2015, the habitat above the falls is a vital thermal refuge for spawning salmon (Figure 15). The program must be maintained and operations improved to maintain and improve access. Other thermal refugia include the Sultan River in the lower Skykomish, and the Tolt River in the Snoqualmie River. Both of these systems are controlled by dams, which release cold water from thermally stratified reservoirs. In 2015 unusually high numbers of Skykomish Chinook entered and spawned in the Sultan River.
Figure 15. Sunset Falls trap and haul counts (grey), relative to total population in the Skykomish and greater Snohomish Basin. In low return years, the South Fork Skykomish supports a significant portion of Snohomish Chinook.
Sunset Falls on the South Fork of the Skykomish River. Fish cannot swim up the falls, and WDFW operates a trap and haul operation to pass fish above the falls to access the large amount of high quality habitat upstream. The river system above the falls is an important thermal refugia that supports a significant the Skykomish Chinook population in low return years like 2015.
Stream temperatures are one in a series of stresses that will challenge salmon populations as the climate changes. As with most recovery actions, focusing on restoring the natural processes that support salmon habitats, particularly hydrology, is the most effective way to restore habitat and build resilience for climate change.
Projected increases in precipitation, particularly heavy winter rainfall events, will increase the volume of stormwater runoff and stress the capacity of stormwater systems, especially older systems. Increased stormwater runoff will increase the delivery of pollutants to streams and rivers and the Puget Sound. Increased stormwater runoff will increase the potential for delivery of pollutants, including nutrients, fine sediments, fecal coliforms, pesticides, heavy metals, oils, hydrocarbons, and hydrocarbon combustion waste products. These pollutants generally degrade the freshwater and marine ecosystems salmon depend on, they also directly impact fish in the water column by increasing toxic stresses and unhealthy conditions. Coho salmon are particularly susceptible to stormwater runoff from urban areas. Coho in the Puget Sound region have been dying prematurely in urban streams at high rates (60-90% of total runs), due to undefined toxins in urban runoff. Scientists are currently studying mortalities to identify the contaminants responsible (Figure 16; Feist et al 2017). While coho are particularly sensitive to runoff, stormwater certainly has effects on other salmonid species, and changes in climate are expected to increase those impacts.
Figure 16. NOAA researchers are estimating coho morality across the Puget sound region (left) based on urban density and stormwater. The right side estimates the certainty of estimates (darker = less certain). http://www.knkx.org/post/simple-infrastructure-fixes-could-keep-stormwater-killing-puget-sound-coho
Artificial drainage systems such as stormwater systems can also degrade stream and river ecosystems by altering flows. When rainwater is not allowed to infiltrate into soils naturally, but is collected on impervious surfaces and transported directly and quickly to streams, these streams experience acute and flashy high flows that they usually would not experience. These flows increase erosion, sediment loads, bank incision, and simplify and channelize streams, and remove habitat elements like wood and leaf litter.
The landscape is like a sponge. It fills with water and stores it during and after rain events. Stormwater systems, by quickly transferring surface water to streams and rivers, reduce the capacity for the landscape to soak up water and slowly deliver it throughout the year. In watersheds with significant impervious surfaces and stormwater systems, less groundwater can decrease or eliminate summer flows, amplifying the negative effects of climate change on summer hydrology. Furthermore, stormwater systems are expanding with the human population. As the population grows and the landscape is increasingly developed, stormwater systems must be expanded to accommodate additional stormwater. This human impact will amplify the climate impacts from stormwater.
Figure 17. Green stormwater systems that mimic the natural hydrology of the landscape provide benefits to hydrologic systems, but also offer additional community benefits.
Stormwater systems are often highly engineered to accommodate the needs of the local environment. Historically these systems were designed simply to deliver water to waterways like streams. As scientists have observed the ecological impacts of stormwater, control measures have developed to increasingly rely on water retention systems like vaults and stormwater ponds. These engineered systemic elements can significantly ameliorate the impacts of stormwater. They can decrease delivery volume and allow water to infiltrate through designed wetland-like systems, which allows biological communities and plants to clean the water, store sediment, decrease high flows, and increase infiltration to groundwater, improving hydrologic impacts.
Other stormwater control measures:
Changes in winter precipitation and hydrology is expected to increase sedimentation of river systems as the climate changes. These changes will increase soil saturation and landslide potential, as well as erosion, flooding and associated sediment transport. Increased sediment loads will increase sediment deposition in low gradient reaches of streams and rivers. Suspended fine sediment in the water column is detrimental to juvenile and adult salmon. It can clog and erode gills and interfere with hunting, particularly for juveniles. Sediment transport can also bury salmon reds and eggs, suffocating them.
Sedimentation risks are associated with hydrology, particularly winter hydrology, as well as bank conditions, riparian vegetation, and hillslope vegetation. Strategies to reduce sedimentation risks including decreasing impacts from winter hydrology and restoring natural floodplain processes along banks with native riparian vegetation, and restoring upland forests along hillslopes, particularly those with high landslide risks. Restoring upland forests is also a strategy for ameliorating hydrologic impacts generally.
Beechie T, Imaki H, Greene J, Wade A, Wu H, Pess G, Roni P, Kimball J, Stanford J, Kiffney P, Mantua N. 2012. Restoring Salmon Habitat For a Changing Climate. River Research and Applications. https://doi.org/10.1002/rra.2590
Feist BE, Buhle ER, Baldwin DH, Spromberg JA, Damm SE, Davis JW, Scholz NL. 2017. Roads to ruin: conservation threats to a sentinel species across an urban gradient.
Johannessen J and Maclennan A. 2007. Beaches and Bluffs of Puget Sound. Puget Sound Nearshore Partnership Report No. 2007-04. Published by Seattle Dsitrict, U.S. Army Corps of Engineers, Seattle Washington.
Kopp RE, Horton RM, Little CM, Mitrovica JX, Oppenheimer M, Rasmussen DJ, & Tebaldi C. (2014). Probabilistic 21st and 22nd century sea-level projections at a global network of tide-gauge sites. Earth’s Future, 2(8), 383-406. https://doi.org/10.1002/2014EF000239
Mantua, N. J., S. R. Hare, Y. Zhang, J. M. Wallace, and R. C. Francis. 1997. A Pacific decadal climate oscillation with impacts on salmon. Bulletin of the American Meteorological Society 78:1069–1079.
Mauger GS, Casola JH, Morgan HA, Strauch RL, Jones B, Curry B., Busch Isaksen TM, Whitely Binder L, Krosby MB, & Snover AK. 2015. State of Knowledge: Climate Change in Puget Sound. Report prepared for the Puget Sound Partnership and the National Oceanic and Atmospheric Administration. Climate Impacts Group, University of Washington, Seattle.
Miller IM, Morgan H, Mauger G, Newton T, Weldon R, Schmidt D, Welch M, Grossman E. 2018. Projected Sea Level Rise for Washington State – A 2018 Assess¬ment. A collaboration of Washington Sea Grant, University of Washington Climate Impacts Group, Oregon State University, University of Washington, and US Geologi¬cal Survey. Prepared for the Washington Coastal Resilience Project.
Murphy DM, Solomon S, Portmann RW, Rosenlof KH, Forster PM, Wong T. 2008. An observationally based energy balance for the Earth since 1950. Journal of Geophysical Research. https://doi.org/10.1029/2009JD012105
NRC (National Research Council). 2012. Sea-Level Rise for the Coasts of California, Oregon, and Washington: Past, Present, and Future. Washington, DC: The National Academies Press. https://doi.org/10.17226/13389.
NRCS. 2017. Report Generator, Alpine Meadows (908) SNOTEL site. https://wcc.sc.egov.usda.gov/reportGenerator/view/customChartReport/daily/908:WA:SNTL%7Cid=%22%22%7Cname/2014-11-25,2015-07-04/WTEQ::value,WTEQ::median_1981,PREC::value,PREC::average_1981?fitToScreen=false
Perry, T.D. and Jones, J.A. (2016) Summer streamflow deficits from regenerating Douglas-fir forest in the Pacific Northwest, USA. Ecohydrology, n/a-n/a.
Peterson WT, Fisher JL, Peterson JO, Morgan CA, Burke BJ, and Fresh KL. 2014. Applied fisheries oceanography: Ecosystem indicators of ocean conditions inform fisheries management in the California Current. Oceanography, 27: 80-89.
Shipman, 1989. Vertical land movements in coastal Washington: Implications for relative sea level changes. Technical Report PV-11, Shorelands and Coastal Zone Management, Washington Department of Ecology, Olympia, wa. 37p.