Posted by: magmatist | June 5, 2013

Large debris flow in Middle Fork Nooksack River- May 31, 2013

By Dave Tucker, with scientific input from John Thompson (Whatcom County Water Resources), Seth Moran and Steve Malone (Pacific Northwest Seismic Network).

Wednesday, June 5, 2013

This post was updated on June 7 with new information. Please visit http://mbvrc.wordpress.com/2013/06/07/update-nooksack-debris-flow-initiated-by-landslide-not-outburst-flood/

Please share.

Boulders and fallen trees litter the terrace 15 feet above the river. These boulders are 5-6 feet across. Click to enlarge any image.

Boulders and fallen trees litter the terrace 15 feet above the river. These boulders are 5-6 feet across. Click to enlarge any image.

A sizable debris flow descended the Middle Fork Nooksack River early in the morning of  Friday, May 31. The source is not known with certainty, but most likely the flow began with a landslide off a glacial moraine above the terminus of the Deming Glacier. The earlier hypothesis that it began as a  glacial outburst flood (jokulhlaup is the commonly used Icelandic term) is probably incorrect. f (More on this process at the end of the post.) The flood rapidly bulked up with debris in the river channel and from glacial moraines lining the upper part of the valley. In particular, moraines are the likely source for the clay-rich matrix that dominates the deposit. The deposit was examined today by Bob Mooers and me where the Ridley Creek trail reaches the river bank, just beyond the end of the Middle Fork Road and 2.2 miles downstream from the Deming Glacier. Our observations and some photos follow.

Terminus of the Deming Glacier in 2006, 2.2 miles upstream of the terraces at Ridley Creek crossing.

Terminus of the Deming Glacier in 2006, 2.2 miles upstream of the terraces at Ridley Creek crossing.

From Seth Moran, PNSN: Twelve-hour plot from seismometer station MBW clearly shows the onset of the a very obvious debris-flow signal at 0953 UTC (0253 PDT). The main event produced appreciable above-background seismic energy for 12-13 minutes, mostly in the first 5-6.

From Seth Moran, PNSN: Twelve-hour plot from seismometer station MBW clearly shows the onset of the a very obvious debris-flow signal at 0953 UTC (0253 PDT). The main event produced appreciable above-background seismic energy for 12-13 minutes, mostly in the first 5-6. (Sorry, poor image. DT)

This would have been a dramatic but probably terrifying event to witness in person. (See a YouTube video of an event that was probably very similar to this). People commonly visit the river bank for picnics after hiking 10 minutes through the forest. Fortunately, this event occurred in pre-dawn hours. It is unlikely anyone on the river bars would survive the wall of sediment and water (more sediment than water- this was flowing concrete, not a ‘flood’) that came roaring down the narrow valley floor. To get away would have required a rapid scramble up a steep loose bank into the forest.

Bob stands at the foot of the 15' levee. Mud is plastered on the tree above him.

Bob stands at the foot of the 15′ levee. Mud is plastered on the tree above him.

The large boulders pitched 15 feet above the stream channel show that survivors would have had to move FAST for some way distance from the channel.

Bob (circled) stands beside a 10 foot boulder perched 15 feet above the stream. An even larger boulder sits at the foot of the levee.

Bob (circled) stands beside a 10 foot boulder perched 15 feet above the stream. An even larger boulder sits at the foot of the levee.

Bob and I found a new levee of boulders  embedded in sticky clay-rich mud a full 15 feet above the river channel. This levee veneers the slope of a much more extensive terrace left by a June 1927  glacial outburst flood. Mud-coated andesite boulders up to 10 feet across are pitched onto the crest of the debris flow levee. Many trees were knocked over. Surviving trees were coated

Gage recording at Cedarville. The outbust flood is shown by the spike in the center.

Gage recording at Middle Fork gage, 12.5 miles below the Ridley Creek area. The outburst flood is shown by the spike at the right edge at 5:15 AM. Another big pulse just before noon the day before, but this was a gradual build up due to melting. Please click to open this image fully.

with mud on the river-facing side for another 5-6 feet higher above the top of the levee. The channel of the Middle Fork is now a 100-foot-wide flat plain of mud and small boulders and cobbles. The opposite side of the new channel fill is banked against a Sumas-age glacial moraine. There is no terrace on that side, but a train of 5- to 10-foot boulders extends for several hundred yards along that margin of the channel. There is little to no wood in this debris. The river was very turbid, much more than

usually observed even in streams draining glaciers on a warm day (it was in the upper 60s when we visited). The river is rapidly cutting down through the sheet of sediment that fills the channel, and mud-free boulders are found immediately adjacent to the water. All else is coated with  gray mud. Some of the mud is saturated, and it is easy to sink in over your boot tops.

The 14-foot-high boulder from the top of the May 31 levee.

The 14-foot-high boulder viewed  from the top of the May 31 levee. Bob Mooers photo.

Across the river, a train of boulders lies at the base of the latest-Pleistocene moraine. These look to be 8-10 feet across. They are at least 100 feet away from the camera.

Across the river, a train of boulders lies at the base of the latest-Pleistocene moraine. These look to be 8-10 feet across. They are at least 100 feet away from the camera.

We visited the river again at the Elbow Lake trail crossing, 1.8 miles downstream. Here, the layer of fresh mud coated terraces 5 feet above the river, but logs piled higher on the terraces were not coated. We saw no newly-deposited boulders. A view of the river from three miles further down the road revealed no new mud.

This event was first detected instrumentally. The MBW seismometer only 4 miles away detected the event at 2:53 AM. Two hours and 23 minutes later, a large spike in turbidity and discharge was noted on the Middle Fork stream gage) at 5:15 AM, 12.5 miles downstream from the terminus.  Discharge (volume of water per unit time) increased from 865 cubic feet per second) at 5 AM to 976 cfs at 5:15 and back to 874 cfs by 5:30. Velocity of the big pulse of water to the Middle Fork gage was 5.3 mph. Discharge spikes were also detected on gages at Nugent’s Corner and Lynden. Ned Currence, a fish biologist with the Nooksack Tribe, went to the scene Tuesday afternoon and circulated an email with the first photos. He recognized that a glacial outburst was the most likely trigger for this large debris flow.

As the climate warms glacier termini will continue to thin. They may become more honeycombed, holding more water. If this is the case, we can expect more of these events in drainages below Mount Baker’s glaciers. Other glaciers at Baker are susceptible to glacial outburst floods. The Mazama in particularly has a long, stagnated, debris-covered tongue. It lies in the narrow, talus-walled gorge of Bar Creek which enters Wells Creek. An outburst flood from the Mazama would rapidly bulk up into a heavily laden debris flow, reaching the North Fork Nooksack just below Nooksack Falls.

About glacial outburst floods

This sketch shows water [white] ponding in cavities in rotting ice [gray]. Water can accumulate within the glacier. The lower cross section...

This sketch shows water [white] ponding in cavities in rotting ice [gray]. Water can accumulate within the glacier. This diagram will be published in Geology Underfoot in Western Washington. (C) Dave Tucker.

Stagnating glaciers can store a lot of water in cavities within their rotting termini. If the load of water gets too high, or if a hot day provides more water than the ice can store, the ice can give way and release a sudden flood of water. This surge can bulk up with sediment scoured out of the river channel, and erode steep walls of lateral moraines lining the channel. The latter provide plenty of clay and silt to the evolving flow, making it more viscous. Debris flows can grow if enough sediment is entrained. By definition, these are flows containing more sediment than water.  You may be familiar with volcanic lahars, which are simply debris flows initiated on volcanoes , sometimes, but not always by eruptions. Some might call outburst floods from Baker’s glacier lahars, but since many associate the term with eruptions, I prefer not to. An event identical to the Middle Fork debris flow of May 31 could begin on any glaciated mountain with big moraines, volcanic or not.

The view upstream at Ridley Trail crossing. A sheet of mud and rocks fills the valley bottom bank-to-bank. I glanced upstream often hoping to see or hear another one before it got to us.

The view upstream at Ridley Trail crossing. A sheet of mud and rocks fills the valley bottom bank-to-bank. I glanced upstream often hoping to see or hear another one before it got to us.

Monitoring and avoidance

If you are in a river valley on a hot day downstream from a stagnating glacier, be alert to the rumble or vibration of an approaching outburst flood, and have an escape route in mind. You can never outrun one of these things. Just charge straight up hill.

Outburst floods can not be predicted

The view downstream. Note the large boulders on the right bank, and the train of boulders against the left bank. Please click to enlarge.

The view downstream. Note the large boulders on the right bank, and the train of boulders against the left bank. Please click to enlarge.

other than to say such-and-such a glacier has the right conditions to spawn them. We saw that the May 31 Deming Glacier outburst flood was detected on seismometers, which telemeter their data directly to the Pacific Northwest Seismic Network’s center at the University of Washington. However, a received signal would do no good for anyone closely downstream. If the shaking detected at PNSN was recognized as a really large outburst flood, it is conceivable that warning could be given to people well downstream. However, it is unlikely one of these events would be of the scale of a big lahar coming off of Baker.


Responses

  1. Chilling news. Does this appear to increase the possibility of similar events ?

    • Chris,
      This particular event does not increase probablility, but continued climate warming will probably produce more of these as stagnant glacier termini trap more water.
      Dave

  2. […] http://mbvrc.wordpress.com/2013/06/05/large-debris-flow-in-middle-fork-nooksack-river-may-31-2013/ […]

  3. John Thompson, of Whatcom County Water Resources, sent some corrections, which I have made. John said: “Quick correction for your blog is that the stream gage shown in your report with the 0515 hrs spike is the Middle Fork gage located a few miles downstream at river mile 5.5 on the Middle Fork (half mile upstream from Mosquito Lake Road Bridge) not the Cedarville gage at Nugents Corner. The latter is where the turbidity readings were taken by Nooksack Tribe closer to noon on 5/31. For reference, your photos were taken about river mile 17.5. I’ll try to go through the emails in the next few days and list a chronology to share. What is really cool is the seismic data and MFK gage and the downstream turbidity data (Cedarville, Lynden, etc.) give us some pretty good info to calculate travel times (~12.5 miles in 2:22 = ~5.3 mph from source to MFK gage).”

  4. Nice summary, Dave. I hope you guys are going to write this up and publish it somewhere. It would be a neat addition to the literature. These types of debris flows occur fairly frequently on Cascade volcanoes, but few are well described and written up.

    • Tom,
      I suspect a paper will come of this event. It is nicely captured in seismometry, discharge at a number of points, and turbidity. The deposit is accessible and fresh so sedimentology can be easily described.

  5. It appears that the debris flow was caused by a large landslide at the terminus of the Deming Glacier at the headwaters of the Middle Fork Nooksack. I have posted some aerial photos here: http://www.stephabegg.com/home/photography/aerial/36

    • it was fortunate that you and John were able to fly over and get excellent images before too many days went by after the debris flow. There is nothing quite like good aerial photography to see what really happened.
      Doug (Dave is field tripping today)

  6. Dave, this is thorough information. Thanks for sharing and describing the process and calculations. The photos are fascinating. The advice is well taken. Thanks for your efforts to keep us informed.

  7. Nice article, Dave. The size of the relocated boulders is pretty impressive. BTW, how thick do you think the “muck” is below the glaciers?

    • Maybe 15′-30′ of muddy sediment carpeting the valley close to the glacier terminus. A wild hair guess. Plan to get in later this summer to look at it.
      Dave

  8. […] changes occurred in the time since the preceding visit on June 5. More on the May 31 flow is posted here on this blog. The initial debris flow activity was caused by a landslide from a glacial moraine into the river […]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

Follow

Get every new post delivered to your Inbox.

Join 362 other followers

%d bloggers like this: