SURFICIAL ICE VELOCITIES OF THE LOWER NISQUALLY GLACIER AND THEIR RELATIONSHIP TO OUTBURST FLOOD HAZARDS AT NATIONAL PARK, , UNITED STATES WALKUP, Laura C.1, BEASON, Scott R.1, KENNARD, Paul M.1 , OHLSCHLAGER, Justin G.2, and STIFTER, Anna C.1 (1) Mount Rainier National Park, 55210 238th Ave E, Ashford, WA 98304, [email protected], (2) Geology, Portland State University, 17 Cramer Hall, 1721 SW Broadway, Portland, OR 97201

I. ABSTRACT II. LOCATION III. BACKGROUND IV. HISTORIC OUTBURST FLOODS V. METHODS

592,000 593,000 594,000 595,000 596,000 597,000 598,000 Mount Rainier, Washington is the tallest of a number of glacier-clad The Nisqually Glacier has one of the longest histories of scientific examination of any glacier in the Pictured: Scott Beason Pictured: Paul Kennard and Laura German Pictured: Jeff Fortner and Paul Kennard Emmons Photo by: Paul Kennard Photo by: Laura Walkup Photo by: Laura Walkup Glacier United States, having been studied since the 1850s. Most historic studies on the Nisqually look at volcanic peaks within the Cascade Range. Standing at 4,392 m (14,411 ft) Columbia Columbia Bellingham A C Crest Crest Little Tahoma mass balance, glacial extent, or surface elevation while few have examined glacial velocity. Past and having 25 major glaciers, it contains more glacial ice than any other Glacier Peak Disappointment 5 Washington Cleaver Spokane

single peak in the continental United States. Glaciers are the headwaters of 5,189,000 Seattle 5,189,000 research indicates that the lower Nisqually glacier has displayed varied glacial velocities throughout

Point Gibraltar its measurement history. The available historic velocity data is presented in the table below. a number of rivers emanating from Mount Rainier, many of which flow to Success Rock Olympia Tacoma Ellensburg 90 Pictured: Anna Stifter Mount Rainier Nisqually Photo by: Laura Walkup Yakima the Puget Sound, approximately 60 km (37 mi) away. Glaciers on Mount National Park Glacier Nisqually Ice Cliff Maximum Velocity Minimum Summer 82 182 Survey Interval Location on Glacier Data Source Rainier have unleashed damaging outburst floods (jökulhlaups) in the (mm/day) Velocity (mm/day) Cathedral Nisqually Cowlitz Vancouver Ingraham past, often without warning. Previous observations have indicated an Cleaver Rocks 5,188,000 Cleaver 100Glacierkm 5,188,000 Near ELA 380 210 Hodge, 1974 Kautz Ice Wilson June - Sept 1968 Photo by Brockman, empiric link between changing glacial velocities and glacial outburst flood Cliff Headwall Camp Ingraham Near Terminus 146 88 Hodge, 1974 05/29/1939 hazards. Specifically, the lower portions of the glaciers were often stagnant Tahoma Muir Glacier Cleaver Near ELA 460 340 Hodge, 1974 when outburst floods occurred. Additionally, the current Nisqually glacier May - Sept 1969 Pictured: Paul Kennard Near Terminus 130 68 Hodge, 1974 Photo by: Laura Walkup Success A: Historic photo showing a “surge” of water pouring C: The Longmire maintenance area was inundated during extent is similar to glacial extent in the 1950s when an outburst flood Cleaver 5 km Cowlitz 5,187,000 Glacier 5,187,000 Muir Near ELA 740 660 Hodge, 1974 over the face of the Nisqually Glacier. Historic the 1955 glacial outburst flood. In addition to its impacted Longmire, one of the visitor destinations and work areas in the Wilson Wapowety Snowfield May 1970 Kautz Kautz Cleaver Cleaver Cleaver Moon Near Terminus 77 68 Hodge, 1974 accounts of glacial outburst floods include importance as the maintenance area for the park, this The change in position of rocks on/entrained in the park. Success Glacier Nisqually Rocks Rocks were chosen based on their location and likelihood of staying in place and survivability against rolling or Glacier Glacier 1943 – 1949 ~1,900 m 110* 70* Meier, 1968 observations that the outburst can occur above the area is adjacent to other administrative buildings and glacial ice were measured weekly and used as a Wilson falling into crevasses. Rocks chosen as survey points were labeled with a unique ID number. The location where Glacier Little proxy to categorize total velocity of the glacier. Africa 1950 – 1959 ~1,900 m 315* 137* Meier, 1968 terminus of the glacier (possibly originating from the visitor destinations such as the Longmire museum and the the survey rod was placed was also marked on the rock in order to ensure that the same location was surveyed This study explores whether the surficial velocity field of the Nisqually Pyramid

5,186,000 5,186,000 These velocities also helped determine the aerial glacier is changing and if ice velocity changes can be used to indicate glacial Glacier 1960 – 1966 ~1,900 m (Near ELA) 452* 247* Meier, 1968 interface between active and stagnant ice) and sweep Longmire Inn. each time (so as not to introduce additional measurement error). Flags were placed in the ice adjacent to the extent of the glacier that was moving rapidly outburst flood potential. Multiple sites on the lower Nisqually glacier were Paradise the glacial ice clean of debris. survey point in order to provide a point of reference in case the rocks rolled or slid. Van Trump Glacier * Meier’s data is average yearly data, and thus averages out seasonal velocity changes Kautz Glaciers D: Impacts of the 1955 outburst flood to the riverbank and versus barely moving. Surveying was conducted surveyed weekly during the summer/fall of 2011 and 2012. The surficial Glacier 2 2 Empiric evidence suggests that jökulhlaups at Mount Rainier tend to occur when large amounts of B: Photo of a debris flow initiated by a glacial outburst park road between the main park entrance and Longmire. with a Topcon GPT-3105 5-second electronic total Pictured: Bryce Limon and Laura Walkup Pictured: Justin Ohlschlager velocity field of the lower 1 km (0.4 mi ) of the glacier was calculated from The Sugarloaf stagnant ice are present in their lower reaches. Additionally, Hodge (1974) found that the summer station using standard surveying practices. Photo by: Paul Kennard Photo by: Jeff Fortner repeated measurements. Between 2011 and 2012, observed velocities in 5,185,000 STUDY 5,185,000 flood on Tahoma Creek, Mount Rainier National Park. Nisqually ice velocity of the Nisqually glacier increased significantly in the vicinity of the equilibrium line Normally if a rock falls into a Glacier the upper portions of the study area accelerated while the lower portions McClure Photo by G.G. Parker, 07/26/1988 AREA Rock (ELA), while slowing occurred toward the terminus (relative to the previous summer) just prior to crevasse that data point is lost, K of the glacier slowed or remained the same. Shortly after the on-glacier E S afet y e q u i p m e nt E R D C NE the 1970 outburst flood. B however, in this case the I LE SKYL P B H A Basalt B IG study concluded in October 2012, we observed a short-lived anomalous 1 E H R including versatile, high P Cliff Panorama A D K

I R Point S crevasse was closing and was A E

P m (3 ft) rise in river stage at Longmire during a rainstorm on the upper 5,184,000 5,184,000 He suggested that observed variations in ice velocity were correlated with the amount of liquid visibility clothing, a P G M L U A C relatively safe to venture inside. R I T E mountain. Field evidence corroborated the stage record and indicates a N R water temporarily stored within the glacier. This indicates that the presence of more englacial c l i m b i n g h e l m e t , Mildred A SKYLINE V 3 3 Point E This measurement was assigned T Golden small outburst flood (approximately 15 m /s (525 ft /s)) was released from A water leads to faster ice velocities and a higher potential for outburst floods, while the absence of harness, rope, ice axe, G Gate N TRUMP E a high measurement error since AN D -V L S O large quantities of englacial water leads to slower glacial flow and fewer outburst floods. If this is the glacier during this rain event. L crevasse rescue gear, L G FA T it had clearly changed positions E CREEK 1 M km Cushman consistently the case, annual glacial velocity measurement could be a powerful tool in the DEADHORSE A mountaineering boots, O A 5,183,000 5,183,000 C Crest T

MORAINE AL vertically. However, despite Our work contributes to the understanding of glacial outburst floods and VIST prediction of potentially devastating glacial outburst floods. crampons, etc were 592,000 593,000 594,000 595,000 596,000 597,000 598,000 having dropped several feet provides an additional line of evidence that changing ice velocities precede required in order to At least four major, damaging glacial outburst floods have been released from the Nisqually Glacier, down into the glacier, this rock outburst floods. Further study of glacial velocity fields could provide a Map showing the study area extent on the Nisqually Glacier on the southern side of Mount safely survey much of along with many smaller outburst events. The floods of 1926 and 1934 damaged the bridge below In addition to being measured and recorded by the Total Station, GPS positions of the rocks were also measured was traveling at the same rate predictive methodology for glacial outburst floods in similar terrain. These Rainier. The Nisqually Glacier is the largest south-facing glacier on Mount Rainier and it is the study area due to the glacial terminus, and the floods in 1932 and 1955 destroyed it (Hodge, 1972). In Mount Rainier weekly using a handheld Garmin GPSMap 78. The exact placement of the flag and survey rods were and in the same direction as results are critical for employee, visitor, and infrastructure protection in the source for the , which parallels much of the main access road into the steep, icy, terrain and National Park, downstream tourist areas like Longmire are susceptible to flooding associated with photographed and recorded weekly. Rocks that were determined to have moved on the ice were surveyed nearby rocks. dynamic proglacial environments experiencing the effects of climate Park. Glacial velocity surveying occurred from the terminus, at approximately 1,580 m crevasse danger. (5,180 ft), up to about the 2,190 m (7,180 ft) level (above the seasonal equilibrium line). glacial outburst floods, as evidenced in the 1955 event when the community was inundated with Pictured: Laura Walkup where they were found, and the amount and direction of roll/slide was measured using the position of the flag Pictured: Jeff Fortner and Paul Kennard change, such as that at Mount Rainier. Photo by: Paul Kennard Photo by: Laura Walkup The area surveyed is approximately 1 km2 (0.4 mi2). flood water. In addition, much of the main access road into the park parallels the Nisqually River. (which nearly always stayed in place), photos from the week prior, a measuring tape, and a compass.

VI. NISQUALLY GLACIAL VELOCITY RESULTS VII. FALL 2012 GLACIAL OUTBURST FLOOD VIII. OUTCOMES

2011 and 2012 Velocity Fields (Full Record) August 2011 and August 2012 Velocity Fields (Same time intervals) This study shows that measurable glacial velocity changes do occur between 1.5 5.0 subsequent summers. Our data supports Hodge's observation of increased PARADISE PRECIPITATION (cm/hr) A C Outburst Flood RIVER HEIGHT (m) velocity in the vicinity of the ELA, concurrent with decreased velocity near the Nisqually Glacier WATER TEMPERATURE (°C) terminus prior to the occurrence of an outburst flood. If this trend is consistent July 1970 4.5 ) r A Glacier Velocity B C D (Hodge, 1974) h Outburst Flood 2011 2012 2011 2012 2012 - 2011 August 2011 August 2012 Glacier Velocity August 2011 August 2012 August 2012 - 2011 and measurable, monitoring of glacial velocity would be a powerful tool in the Nisqually Glacier Rate (mm/day) 1.0 TION(cm/ prediction of outburst flood hazards.

600 A October 2012

600 )

Rate (mm/day) C 660 4.0 ° (

0 - 75 E UR PRECIPIT 0 - 75 T Key Findings:

480 600 480 DISE PERA 540 ? 75 - 150 RA Seasonal patterns of velocity exist EM 540 540

540 A

600 T 75 - 150 P 3.5

ER ?

T Yearly velocity surveys could allow us to monitor glacier conditions and predict (m), 420 480 0.5 A 300 W

180

420 420 GHT 360 150 - 225 240

EI outburst flood hazard 150 - 225 240 ?Real-time GPS tracking could potentially indicate glacier float and provide short- 360 360 3.0 RIVER H

360

240 term outburst flood warning 225 - 300 240 300 480 300 225 - 300 ?Stream gauge closer to terminus would monitor fluctuations in glacial discharge, and

360

420 240 0.0 2.5 300 180 180 10/26 10/26 10/27 10/27 10/28 10/28 10/29 180 OCTOBER 26 OCTOBER 27 OCTOBER 28 could allow us to detect smaller outburst floods that might be less noticeable on the 300 - 375 420 300 300 - 375 stream gauges farther downstream 240 120 120 B D 375 - 450 Continued Questions: 375 - 450 On October 27, 2012, a small storm (RI = 1.04 yr) occurred at ?What velocities are “normal” for the lower Nisqually? 180 Mount Rainier, dropping 7.98 cm (3.14 in) of rain. Between ?How has “normal” changed since Meier and Hodge did their studies?

450 - 525 60 120 450 - 525 4:00 PM – 7:00 PM, rainfall intensity was between 0.6 – 1.0 ?What is the outburst flood hazard now and what is it expected to be in the future?

60 60 120 60 ? cm/hr (0.24 – 0.40 in/hr). Starting at 8:00 PM, a stream gage in How will we know if outburst flood hazard changes?

60 60 ? 525 - 600 Longmire recorded a dramatic and anomalous rise in stream Do jökulhlaups occur when “active” ice overrides “stagnant ice?” 525 - 600 ? stage, from 0.51 m (1.66 ft) to 1.47 m (4.81 ft). The peak Can kinematic waves cause outburst floods? ? occurred at 9:45 PM, and employees living near the river Can kinematic waves reactivate stagnant ice and lower outburst flood hazard? ?How do velocities change in the winter? 600 - 675 0.25 km 0.25 km 0.25 km 600 - 675 0.25 km 0.25 km 0.25 km reported hearing boulders rolling in the river and loud sounds from the river around that time. Within 15 minutes, the Sources Cited: Hodge, S.M., 1974. Variations in the sliding of a temperate glacier: Journal of Glaciology, Vol. 13, No. 69, p. 349-369. Velocity Fields (2011 and 2012, mm) 2012 - 2011 Change (mm) Velocity Fields (2011 and 2012, mm) 2012 - 2011 Change (mm) stream stage dropped back to around 0.6 m (2 ft). This 0.96 m Meier, M.F., 1968. Calculations of slip on the Nisqually Glacier on its bed: No simple relation of sliding velocity to shear High : 800 < -120 30 - 60 High : 800 < -120 30 - 60 675 - 750 Velocity Contours (CI = 60 mm) 675 - 750 Velocity Contours (CI = 60 mm) (3.15 ft) rise in river level was not noted in other tributaries stress. U.S. Geological Survey, Tacoma, WA, p. 49-57. -120 - -60 60 - 120 Study Area Study Area -120 - -60 60 - 120 that feed into the Nisqually River. Additionally, stream Shortly after the October 2012 outburst flood was recognized in stream gage records, field reconnaissance was undertaken to find Richardson, D.T., 1968. Glacier outburst floods in the Pacific Northwest. Geological Survey Professional Paper, 600, 79. Low : 0 Low : 0 -60 - -30 >120 -60 - -30 >120 Study Area Study Area temperature dipped noticeably during the peak flood, evidence for the event. Figure A shows a dramatic trim line approximately 1 m (3 ft) above the low flow of the Nisqually River near No Change No Change Special thanks to all of our volunteers who made this project possible: 0.25 km 0.25 km 0.25 km 0.25 km potentially indicating a glacial source of the surge. Cougar Rock, approximately 5.3 km (3.3 mi) downstream from the glacier. More evidence of a trim line (Figure B) was noted in Longmire Kirsten Anderson, Jeff Antonelis-Lapp, Katie Ashe, William Baur, Kevin farther downstream and a variety of other locations on the Nisqually River. Dlugos, Peter Ellis, Corrie Floyd, Jeff Fortner, Emily Frost, Dan German, Measured data were post-processed and analyzed using ArcGIS 10.1 and Excel. A spreadsheet was The above maps represent (from L to R) the average velocity field of the Nisqually glacier in 2011, the Laura German, Andrew Hendrickson, Lucas Jones, Adam King, Bryce The exact source of the outburst flood on the Nisqually Glacier was not found; however, images of the Nisqually River shortly before designed to calculate direction of movement and positions of rocks that rolled/slid using field-measured The above maps represent (from left to right) the average velocity field of the Nisqually glacier Comparison of glacial velocity between the same time periods in both 2011 and 2012. In this velocity field during the same period in 2012, and the difference between the two years. The upper The hydrograph from the October 2012 event (shown right) Limon, Chris Magirl, Conor McDowell, Eric McPherren, Dan Miszewski, (Figure C) and after (Figure D) the event show a band of ice that appears to have been “washed” by the event. The band begins just Esteban Monreal, Donald O'Brien, Dave Oleson, Dave Olson, Patricia values. Resultant data was imported into ArcGIS and used to create maps showing direction and measured in 2011, the average velocity field measured in 2012, and the calculated difference map, the 2012 data was limited to only include the measurements taken between the end of portions of the study area showed an increase in glacial velocity in 2012, while the lower portions either resembles an outburst flood hydrograph from Hodge (1974) downstream of the confluence of the Wilson and Nisqually Glacier down to the terminus of the Nisqually. Winter weather precluded a Poulin, Shane Rathbun, Mason Reid, Eric Rejman, Heather Rogers, Amy magnitude of individual survey points on the glacier, as well as interpolated velocity fields in the survey between the two years. In general, the upper portions of the study area showed an increase in July and the beginning of September to correspond with the period of record from 2011. The showed a decrease or no change in velocity between 2012 and 2011, similar to the comparison between (shown left). Rutz, Rebecca Shaub, Brian Shermeta, Matthew Shlim, Max Stevens, comprehensive search for the source, however, it appears that the event was sourced on top of the glacier in the vicinity of the area. In the above map, each arrow represents the average velocity of one measured point on the glacier. glacial velocity in 2012, and the lower portions either showed a decrease or no change in scale is exactly the same as that in the earlier map showing the entire record. the full records (Fig VI-A). Despite slight differences between the full and limited 2012 velocity, the Anna Stifter, Carrie Tomlinson, and Jenna Zechmann boundary between active, fast moving glacial ice and stagnating ice, as determined by the 2011 and 2012 surveys. This is consistent The 2011 data points were measured weekly from August 2011 through September 2011. The 2012 data velocity between 2012 and 2011. pattern is similar enough that our interpretation remains the same. This implies that a comparison of our Additional thanks to Lead Climbing Ranger Stefan Lofgren and Climbing Rangers Sam Siemens-Luthy and Dan points were measured weekly from July 2012 through October 2012. data with previous surveys remains valid despite differences in survey timing. with historical accounts indicating that portions of the glacier were “swept clean” during outburst flood events (Richardson, 1968). Veenhuizen for providing crevasse rescue training.