#43: Part 2, Mechanisms of change in fish populations: Floods and streambed scour during incubation & emergence
Dan Isaak
So picking up again with Part 2 of the new module on “Mechanisms of change…” wherein we delve more into the various means by which climate change may affect local habitats and population dynamics…Recall that earlier we put these local considerations within the context of the BIDE processes (Birth/Immigration/Death/Emigration; Blog #41) and discussed how the temporal variability of habitat conditions at a site needs to mesh with species biology & life cycle requirements for things to work. One of the big things determining that temporal variation is the annual flow regime as manifest by the amount & timing of water moving through a stream channel. As Leroy Poff and his collaborators have amply demonstrated in previous decades, populations of fish and other aquatic critters are often strongly adapted to what are the historical norms regarding flow conditions (graphic 1; this hyperlink takes you to the paper: http://rydberg.biology.colostate.edu/~poff/Public/poffpubs/Lytle%26Poff2004(TREE_natflow).pdf). Fish breed, their eggs incubate, and juveniles hatch and grow at certain times of year that are synched up with the flow regime. In a similar sense, the structure of the stream channel itself (i.e., sediment size distribution, distribution of pools/riffles/runs) is also adapted to the flow regime and represents a quasi-equilibrium among the flux of water, sediment, and wood through a river network (Blog #22). It’s reasonable to assume, then, that as climate change alters temperature, precipitation, and runoff regimes (blogs 13, 16, and 17), conditions within channels have to adjust & will cause concomitant adjustments to the habitats that streams offer their biological denizens.
Sometimes these adjustments will be gradual, occur over many decades, and not necessarily be that disruptive. But other parts of the adjustment process will be relatively dramatic & rapid (from a fish’s perspective) and could negatively affect some of the BIDE processes. One thing often discussed for streams draining watersheds with substantial seasonal snowpacks is increased risk of channel scour from rain-on-snow events. When rain falls on snow, heat is transferred to the snow, the snow melts rapidly and dramatic runoff events sometimes occur (graphic 2). As temperatures slowly warm & isotherms creep upwards (Blog #36), snowpacks at higher elevations in mountains that were previously too cold to experience rain will slowly begin to do so. Stream channels and fish populations that were unaccustomed to big floods at certain times of year will have to adjust and if you’re an egg incubating in the streambed or newly emerged fry developing along the stream’s margin, adjustment may not be an option. For these little guys, Martin Short’s quote, “I’m not a real strong swimmer” from his classic Saturday Night Live skit on men’s Olympic synchronized swimming comes to mind (the YouTube video helps if this reference is too obscure http://www.youtube.com/watch?v=A2AU2xu3CeQ). In these situations, the D in BIDE may spike for small fish as their calm marginal habitats become chaos and previously protective streambed substrates turn into fish blenders.
Fortunately, not all stream channels, even in areas with snowmelt dominated hydrologies, are going to be susceptible to these types of disruptive flow events. And our papers this time focus on sorting out the factors that govern spatial differences in susceptibility to channel scour. Goode and colleagues linked global climate model outputs to a hydrologic model to predict changes in the frequency of channel forming (flows with approximately a 2-year return interval) & larger floods across a mountain river network. Flood levels were then linked to reach-averaged shear stress equations to predict the depth & likelihood of bed scour for individual stream reaches and this information was cross-referenced to those times and places when the eggs of different salmon & trout species would be incubating in the gravel (graphic 3; this hyperlink takes you to the paper: http://www.treesearch.fs.fed.us/pubs/43353 ; as an aside, this paper appearred in a recent special issue in Hydrological Processes (Volume 27, Issue 5) entitled “Catchments in the future North: interdisciplinary science for sustainable management in the 21st Century” that has lots of related studies on similar topics). Goode and colleagues found that susceptibility to scour events varies depending on position within the river network (high vs low elevation), reach morphology (pool-riffle, planebed, step-pool), species life history (spring spawners vs fall spawners, the latter of which are generally more vulnerable because their eggs incubate in the substrate all fall and winter), and body size (larger species bury eggs deeper, which makes them less vulnerable to scour). For some species in some parts of the network, the future looks less promising as scour frequencies & depths are predicted to increase to harmful levels. But for other species in that same part of the network or nearby, it may not matter.
The second study by Shellberg and colleagues addresses this topic in more detail through an empirical approach (this hyperlink takes you to the paper: http://gis.ess.washington.edu/grg/publications/pdfs/Shellberg_etal_Scour_CharSpawning_Washington.pdf). Scour chains were used to measure the depth of streambed scour along elevational transects during several years of monitoring (graphic 4). These depths were compared to depths that fall-spawning bull trout are reported to bury their eggs when building a nest. Shellberg and colleagues found that scour depth at a site varied across years with the magnitude of the flood, by channel type (pool-riffle, planebed, step-pool), and microhabitat location within a channel. Notably, some microhabitats in diverse channels experienced relatively little scour (sometimes even aggradation) even when in close proximity to those that did. One of the paper’s conclusions, therefore, was that structural channel diversity might provide population resilience by ensuring that some spawning fish would consistently find areas where eggs incubated successfully.
These studies make it apparent that there are many complexities involved with understanding vulnerability of even a single life stage to a single type of habitat alteration that could occur from climate change. As such, we need to continue validating and refining our models and thinking by comparison of empirical data against model predictions and one can see how studies of this sort could be used in complimentary fashion to do so. Also important is doing the biological validation to confirm that the predicted effects on fish populations are a reality. One possibility might be using a modeling approach like Goode and colleagues to reconstruct the temporal variation in scour events (or other relevant flow characteristics) at sites with long-term biological monitoring (Blog #41) and to look at the biological variation relative to flow variation. It’s a sort of sensitivity analysis, but a post-hoc version that could be done relatively quickly to leverage information from existing biological datasets. That’s it for now; next time we’ll think about the other side of the flow spectrum and how too little (rather than too much) water might affect fish populations.
Until then, best regards
Dan
David Peterson
It seems like there are at least two overarching stresses that have reduced resilience of fish populations. First, most populations of most species are much lower now than they were prior to extensive human disturbance (and commercial and sport fishing continue). Second, habitat has been degraded. This means that large disturbances, climate induced or otherwise, will have a relatively larger effect than they would have had 200 years ago.
I know nothing about fish genetics, but it seems like a topic that would be a logical follow-up to this post. I hear some fish biologists say that a warmer climate will rapidly reduce fish habitat quality, causing widespread extirpations of species and stocks. Other biologists say that sufficient genetic diversity and flexibility in life history traits exist to allow long-term survival. I'm guessing this might differ by species and geographic location.
Much of ecology attempts to simplify complexity to facilitate systematic study and inferences. In this case, understanding the many different aspects of complexity in fish biology and habitat is critical for projecting climate change effects and for developing effective conservation strategies.
Dave P.