#22: Climate change effects on enhanced sediment delivery to streams

[Dan Isaak]

Climate Affects Everything, Even The Way a Stream Looks

Hi Everyone,
So last time out we wrapped up the hydrology module and we’ll
eventually transition into the next formal module on biological
effects, but before starting that, I’m going to work through a few
miscellaneous topics & new research on previous topics that bear on
the issue of climate and streams. Typically, when we think about
climate change effects on streams we think temperature and hydrology
first but that’s only part of the story. Drainage network structure,
stream channel morphology, and the array of fish habitats like pools,
riffles, and runs are the way they are because streams have adjusted
their form to efficiently transport sediment yielded by a landscape
downstream to the sea. Anything that shifts the historical balance
between this sediment load and a stream’s hydrograph will cause the
stream to adjust its form so that its work is continued efficiently.

A simple conceptual tool for thinking about how stream channels may
adjust to changes in sediment & flow regimes is Lane’s Balance
(graphic 1). Probably all the fluvial geomorphologists in the crowd
are cringing at this point regarding the crudeness of this tool, but
it’s about as deeply as us fish people want to think about dirt.
Basically, Lane’s Balance predicts whether a channel will either
aggrade (accumulate sediment) or degrade (export sediment) through
time as a function of the sediment load, hydrologic regime, stream
slope, and sediment size. Lots of moving parts there but the biggies
are sediment and hydrology. We’ve previously discussed climate-induced
hydrologic changes (blogs 16 – 21), with the annual high flows most
important for transporting sediment increasing in some areas and
decreasing in others (blog 17), but we haven’t yet brought sediment
into the equation. As it is the brick and mortar of fish habitat, it
is important to think about how climate change will affect the left
side of this fluvial balance.

So this week, we’re highlighting two recent papers that begin to
address this topic. The first is by Sue Cannon and colleagues and
describes the probability and volume of postfire debris flow sediment
deposition into stream channels (graphic 2). The basic story is that
if a wildfire happens in steep terrain, then a high-intensity
precipitation event like a thunderstorm drops a lot of water on there
in the next few years, a big chunk of sediment may succumb to gravity
and find its way into the channel network. If you’re a fish living in
the path of such a debris flow, your best days are behind you as the
slug of sediment scours and transports your entire world somewhere
else further downstream. Once that slug of sediment enters a larger
channel and/or where the valley slope decreases below a critical
threshold, the ride & your life are over. But if you’re a fish living
in that downstream depositional channel, you only give passing notice
to the demise of your relative and spend more time enjoying all the
great new spawning gravels and other diverse habitats that a fresh
slug of sediment often provides.

To think more broadly about how this influx of fire-related sediment
may affect the fish housing market and what the government can do to
bail them out (credit for this analogy to J. Goode), Goode et al
(2012) synthesized all available sediment delivery data from central
Idaho to assess the potential climate drivers (top figure in graphic
3) of sediment yields in fire-influenced landscapes. The potential
changes were framed in the context of implications for infrastructure
and habitat management and the question was, “what can be done to
offset the increased sediment from fires?” In a warmer and drier
climate, vegetation disturbance from wildfires, insect/pathogen
outbreaks, and drought induced die off (graphic 4) are likely to
promote to hillslope instability and lead to a general increase in
sediment delivery.

Although the magnitude of sediment delivery from fires far surpasses
that from anthropogenic sediment sources (grazing, mining, timbering,
and roading; lower figure in graphic 3), fish in fire-prone landscapes
across the northern Rockies have evolved with wildfires & debris flows
often provide fresh spawning gravels and nutrients from hillslope
soils that are needed to maintain diverse and productive habitats.
More problematic to fishes and infrastructure are the fine-sediments
that may smother fish eggs and interstitial spaces between substrates
where aquatic insects live. Wildfires provide these fine-sediments
periodically, but these background levels are greatly enhanced by
chronic delivery from anthropogenic sources & especially roads.
Moreover, it is these fine sediments that are of greatest concern for
the maintenance of downstream infrastructure (dams and levees) because
they are easily transported downstream by relatively steep mountain
rivers. A sensible approach for mitigating climate-wildfire-sediment
effects on fish populations and infrastructure, therefore, is not
through fire suppression activities, but by minimizing anthropogenic
sources of sediment.

So if the general prediction holds in semi-arid areas that there will
be fewer forests and other vegetative types with deep root systems to
keep sediment on hillslopes as climate change progresses this century
(graphic 4; blog 15), we’d expect more sediment to be finding it’s way
into streams. What that ultimately translates to for stream channel
morphology and fish habitats depends on the interaction with the
evolving hydrologic regime in that same place. In streams where peak
flows are declining and sediment rates are increasing, aggradation may
be expected over the long-term. In areas where peak flows are
increasing and sediment rates increasing, perhaps no big changes occur
(graphic 5). It all depends on where the stream fits in the relation
between sediment supply and transport capacity. So again, and as we’ve
seen several times previously, even though everything’s changing in
response to climate forcing, the details of those responses depends on
the local context.

Until next time, best regards.
Dan