November 7 Fault Reconnaissance
Alaska Science Center










Photographs by
Rod March and Dennison Trabant,

U.S. Geological Survey

Written by Dennis Trabant and Rod March, USGS:

On November 7, 2002 Rod March and Dennis Trabant (USGS-Fbx), and Dr. Martin Truffer (UAF Glaciologist) flew by fixed wing Cessna 206 to document the effects of the Nov. 3, 2002 magnitude 7.9 Denali Fault earthquake. We were able to fly and photograph the fault trench from near Cantwell to just NW of Mentasta Lake and additionally south of the fault trench between the Richardson Highway and Susitna Glacier to observe the areas around Eureka, Maclaren, and East Fork Glaciers. The north flank of the central and eastern Alaska Range was obscured by clouds and was not investigated. Deep shadows due to low sun angles made some areas difficult to see. However, digital enhancement of the photos overcomes this obstacle and reveals significant features not observed in the field. Features observed and photographed included numerous rock, snow, and ice avalanches, single linear faults, sub-parallel linear fault pairs, shear zones with multiple sub parallel faults, long curved faults that appear to follow glacial structures, zigzag faults, at least one significant fault perhaps 1 km south of the Denali Fault that was not parallel to the Denali Fault trench, odd looking holes on a glacier surface along the main fault that may be surrounded by blast debris, avalanche-debris dammed lakes both on and off glaciers, numerous small glacier surficial lakes that had drained, cracking in lake ice, glacier cracks that indicate drainage of subglacial lakes (these may be pre-quake features), dark rims around lakes that may have been evidence of seiches. The most dramatic features by far were the very large mostly rock avalanches onto Black Rapids and the very large glacier collapse/avalanche onto Gakona Glacier. The following is a chronological description of our flight observations.

We began near the magnitude 7.9 epicenter adjacent to the lower West Fork Glacier. Clouds north of the Alaska Range largely obscured our view during the approach. However, near the crest of the Ak Range we saw mixed snow and rock avalanches in increasing numbers and size as we approached the epicenter. The upper Yanert Glacier near Mt. Deborah was severely avalanched. The Yanert ended a surge during the summer of 2001, so its surface was very rough, but that roughness was not caused by the earthquake. We saw no linear fault trace on the surface near the epicenter or to the east across the main branch of the West Fork Glacier. There were many avalanches, the largest that we saw began of the east side of the West Fork Glacier below peak 7628 and ran out onto the surface of West Fork Glacier 100 meters of more. That avalanche, like most, was a mixture of rock and snow. In general, the snow cover in the Ak Range is very light for this time of year at lower altitudes, but well above normal at higher altitudes based on early Oct. measurements on Gulkana Glacier, just east of the Richardson Highway. Most large seracs in the icefall where the 1st tributary on the NW side of West fork Glacier enters the main stream had fallen giving the ice fall an unusual appearance in that almost no "flat-topped" seracs remained standing. Above the icefall, many snow bridges over crevasses and bergschrunds had collapsed.

The first identified surficial rift was on the hills between the West Fork and Susitna Glaciers in T16S R3E (Fbx Meridian). At this location, there was little evidence of vertical displacement accompanying the lateral displacement (the rift is on a north facing hill slope so any judgment about vertical displacement is poorly founded). There was no way to estimate the magnitude of the lateral displacement from the air. We were able to track the surficial rift almost continuously from near the east side of West Fork Glacier onto the Susitna Glacier. On the lower Susitna Glacier, in an area of post-surge deflation, some vertical differential between the two sides was noticeable due to the low-angled November light. The north side of the rift was higher than the south side. Between the West Fork Glacier and the Black Rapids divide, the observed rift is a single, ragged crack. On the Susitna Glacier, the rift was located about 20 percent of the glacier-width from the northern edge of the glacier along a medial moraine. Sometimes near the southern base of the moraine and sometimes near the crest of the moraine. The rift was continuous to about 5000 feet on Susitna Glacier.

In the saddle between Susitna and Black Rapids Glaciers and below the tributary that enters the north edge of the main glaciers west of Aurora Peak, the rift lineament was not obvious on the surface. The lineament trend in that area was crossed, almost perpendicularly, by a few freshly opened cracks (very similar in appearance to the rift cracks) that were as much as several hundred meters long and separated by several hundred meters (these values are from memory only and can be revised when the photos are available for reference/analysis). In this pass area, some of the "pot hole" lakes were still filled with water and others were empty. There was no sign that any of the lakes had had any recent changes; i.e., filling, seiching, or emptying. Those that contained water had continuous, uncracked ice covers with no signs of distortions at their edges. Those that were empty had their locally typical, light snow covers, undisturbed. The rift lineament was identifiable on the upper Black Rapids for several kilometers east of the Susitna/ Blk Rapids saddle favoring the northern 15-20% of the glacier width near medial and lateral moraine edges and a second lineament was identified in a surface drainage channel. This second lineament continued through a series of odd holes that resembled neither moulins nor entrances to crevasses; two look a little like they might be surrounded by blast debris. The, 5 or 6 holes were above the snow line and are difficult to understand. They were the only such features we observed though much of the flight may have been too high to see such features.

By far, the most dramatic mass failures we saw emanated from the highest peaks along the south walls above Black Rapids Glacier east of the Lockit tributary. Three, largely rock debris avalanches crossed almost the entire, 1-mile width of Black Rapids Glacier, overriding the monster medial moraine that is 10's of meters high and leaving a blanket of rock debris that is probably thick enough to inhibit ablation, in places. Two avalanches came from the peak between the Lokit tributary and the next tributary to the east. The third, and probably most voluminous, came from a peak in the next easterly separator between tributary entrants. The total glacier area covered with rock debris is about 13 km2 (an average of the three observers independent map drudles). The thickness of the deposit is the largest uncertainty in a volume estimate, but bounding the estimates are: a minimum average thickness of 1 meter; gives a volume of 0.01 km3, and a large average thickness of 5 meters equates to a volume of 0.07 km3 (NOTE: these are very crude values which desperately need verification). The rift lineament was not identifiable through the avalanche area.

The lineament was again identified where the Denali Fault deviates from Black Rapids trench and approaches the Augustana Creek drainage from the northwest. The rift was discernible above the snowline, but could not be identified below the snowline in the vegetated slopes or fluvial sediments of the Delta River.

The Trans Alaska Pipeline crossing and lower Cantwell Glacier rifts have previously been described. We flew up the Eel Glacier and found comparatively minor avalanching and surface disturbances. There was a large mixed rock and snow avalanche off M'Ladies Mountain on to the upper Castner Glacier. A graben, estimated to be 20 meters between bounding rifts and possibly one hundred meters long, was seen high on Gulkana Glacier south east of Cony Mountain, on the Gulkana side of the ridge crest between Gulkana and Canwell Glaciers. There were several sizeable avalanches onto Gulkana Glacier, one very near one of our long-term measurement sites below the Moore Icefall.

The next most intensively shattered ridge is the SE wall of the Gakona Glacier along the Denali Fault between the Canwell-Gakona pass and where Gakona Glacier leaves the Denali trench. There, one very large (possibly larger than those on Black Rapids) ice and snow dominated avalanche was released. Most of the deposit was deposited less than half way across the glacier width, but a considerable volume was spread across nearly the full width of the glacier. There were many smaller avalanches off that same ridge, most of which did not cross more than one half of the glacier width. Some made curiously curved run out paths. The surface rift is almost continuously visible in the parts of the Canwell, Gakona, West Fork, and Chistochina Glaciers that lie on the Denali Fault. For most of this route, the rift is confined to the northern 20-25% of the glacier widths and is a single ragged crack. However, in the ice divide between Gakona and West Fork of Chistochina the lineament is two sub parallel traces separated by a couple of hundred meters (or more) approximately centered in the glacier's width. On the West Fork of the Chistochina Glacier only a single ragged lineament was seen and it became "Z"ed (c/ 100-m arms) for a few hundred meters then straight for a while then "Z"ed again for a ways. The ice on a glacier dammed lake between the main branch of the West Fort of the Chistochina Glacier and its most easterly tributary was lightly shattered ? the rift lineament ran directly through the lake. There was no strong evidence for the lake level changing.

East of the Chistochina's east terminus, the fault lineament rises above the valley floor and parallels the valley floor varying in height above the valley floor up to 1/3 of the wall height, along the northern valley wall. The lineament was lost in the upper Slana River drainage. However, a couple of kilometers of strong lineament was seen about 1 km east of peak 6565 (near the head of Canyon Creek, a tributary to Mankomen Lake). The strike of the lineament is north-northwest ? south-southeast. Extension in either direction was not found.

A small debris-avalanche-dammed lake was seen in the Lost Creek drainage (NW of Mentasta Lk ~10 miles). There was definitely fresh debris added to the dam. However, we judged that the lake had been dammed by a previous landslide because the lake surface was frozen, therefore probably had not formed in the last few days, although the level might have changed.

Cloud cover prevented investigation of the large glaciers along the north flank of the Alaska Range. General impressions from returning along the south flank of the Ak Range (south of the southern crest flanking the Denali Fault) were that the intensity of avalanching and surface disruption diminishes quite rapidly laterally away from the rift valley to the south though this might partly be due to gentler topography. Indeed, we remarked among ourselves that flying along the south flank bare miles south of he main rift valley, that one would little expect to see the carnage we had just seen over the next ridge north. West of the epicenter and back in the Denali Fault valley system, we saw no new rifting, although the ancestral rift was evident in a few places. Just a few miles west of the terminus of the Yanert Glacier, but back in the Denali Fault trench, there were frozen lakes with unbroken ice and little or no evidence of seiching or even edge cracking. Again, it is surprising how quickly the extreme surface disturbances are reduced by distance. If, by extension, the rapid decrease in avalanching laterally away from the main rift valley applies to the large glaciers on the north flank of the Alaska Range, then there may be few large avalanches or glacier disturbances that were not documented due to the cloud obscuration on the north flank.

In no case was there any evidence of unusual glacier flow having been stimulated by the recent seismicity. However, the avalanche deposits will affect the mass balance and flow of Black Rapids, Canwell, and Gakona Glaciers for decades.