By Ed Berg
In the June 13 Homer Tribune reporter Naomi Klouda described a visit that she and I made to the bluff collapse that is encroaching on the Sterling Highway four miles north of Anchor Point at Mile 153.3. On that visit we measured the collapse hole as 49 feet from the edge of the highway, indicating that four more feet were lost since my measurement of 53 feet last September. Numerous readers of Naomi’s article expressed to me long-standing concerns about the danger to the highway of further collapse of this hole. I have spent a number of days this summer checking out a variety of bluff collapses along the coast from the Fox River to Kasilof and now have a much better idea about the mechanisms of these collapses.
The simplest collapses occur as rock falls from steep cliffs; these are very common between Homer and Diamond Creek, where Tertiary-age sandstone, shale and coal are strong enough to support vertical cliffs. These rock falls typically produce a rubble heap at the base of the cliff, which runs out into the intertidal zone tens of meters, and is vigorously attacked and whittled back within a few years by high tides and winter storms. Long shore currents carry the eroded material south along the coast and ultimately deposit it on the Homer Spit and Archimandritoff shoals west of the Spit, as shown by time-lapse photography with the Argus cameras at Munson Point.
The second type of collapse might be called simple slumping, where the bluff simply slides downhill, usually in several slices or steps. These slumps can be small or as much as a half mile wide, like the one below Seaside Farm five miles out East End Road. They are often thickly vegetated with alders and spruce, and appear to be old and well stabilized.
A third type of collapse involves a slumping block that rotates as it slides downward on a concave fault plane. The fault plane is steep at the headwall and then flattens out lower down. The huge 2.8-mile wide slide at Bluff Point (just west of Baycrest Overlook) is a beautiful example of a rotated slump block. As you walk past it along the beach you can see the bedrock layers dipping north into the slump because this lower part of the slump block was rotated upward. This is a very old slump that probably extended more than 500 yards out in the intertidal zone. The upper part has been eroded off by wave action, leaving a rocky residue of offshore boulders. On a good minus tide you can see the rotated coal layers sticking out of the intertidal mud and dipping north as much as 50 degrees.
The fourth type of collapse – that of MP 153.3 – is less instantaneous and is more a matter of slow whittling by erosion of many small slides. This type of collapse is called “spring sapping” and is common in the US Southwest. This situation occurs with groundwater percolating downward through sand to an impermeable “basement” layer, like coal or dense glacial till. When the groundwater reaches the basement, it weakens the overlying sediment, flows out horizontally and exits the bluff as springs. This constant flow of spring water erodes the base of the sandy bluff. Periodically the steep sand face collapses and sends a pulse of sand, mud and water down the stream channel and out onto the beach.
Spring sapping typically produces a clean semi-circular hole with little slump material at the bottom. The steady flow of water from the springs, aided by occasional mass flows, seems to keep the holes pretty well flushed out. A look at GoogleEarth will show a number of these rounded holes north of Diamond Creek to Ninilchik, but none south of Diamond Creek. They are north of Diamond Creek because of the thick glacial sands that sit on top of glacial till and Tertiary bedrock with impermeable coal layers. These sands were deposited when Cook Inlet was full of glacial ice in lakes that formed alongside the melting ice.
South of Diamond Creek the glacial sands are much thinner and intermittent, with most of the bluff formed from Tertiary bedrock of sandstone, shale and coal. This bedrock can certainly slide, as at Bluff Point, but it doesn’t get nibbled away by spring sapping.
Recently I checked out the bottom of the mile 153.3 slide. From a geological point of view it was a scary scene. The slide bowl is floored with a massive layer of hardened mud – a totally impermeable glacial till layer at least ten feet thick. Spring water flows out of the overlying sand, staining the till layer orange, probably from the same iron-precipitating bacteria found in many of Homer’s wells. The banks of the narrow stream channel were covered with sandy mud, suggesting pulses of mud-and-sand five feet deep coursing down to the beach.
I would encourage interested readers to visit this site, but not when it’s raining! You can drive to the beach at Whiskey Gulch Road five miles north of Anchor Point and hike south two-thirds of a mile to the stream. Four-wheeler trails punch through the alders along the stream into the bowl of the slide. As you look up the bowl, keep in mind that Sterling Highway is only 49 feet back from the rim; you can hear the traffic zooming by.
Spring sapping at this site probably goes on all year long, even in the winter, because the groundwater is relatively warm. There is however the potential here for a much more dangerous kind of instantaneous collapse, where water-saturated sand layers are liquefied in a strong earthquake. This is called “liquefaction” and there were many examples of it in the 1964 Great Alaska Earthquake, as well in the 2011 earthquakes in Japan and Christchurch, New Zealand. It is a major earthquake hazard in West Coast cities like Seattle, Portland and San Francisco and liquefaction susceptibility maps for these areas can be viewed on the Internet.
When a water-saturated sand layer is at rest, the sand grains are touching and support each other. In an earthquake, however, the upward shaking lifts the sand grains and they briefly become free-floating, and the sand layer behaves like a liquid. As soon as the shaking stops, the sand grains settle down, lock together and stop floating. But a brief period of free-floating can be enough to get a landslide started because momentum and gravity now come into play.
Many of the simple slumps I have examined, as well as the rotated slump blocks like Bluff Point have impermeable coal seams or glacial till at their base and it is quite likely that some degree of earthquake-generated liquefaction facilitated their collapse.
If liquefaction were to take charge at the mile 153.3 slide, hundreds of yards of the Sterling Highway could be shifted into Cook Inlet. This would make the present spring sapping at this site look like peanuts.
In her June article Naomi Klouda quoted AK DOTPF spokesman Rick Feller as saying (about the mile153.3 slide) that, “It was determined that the best course of action would be to shore-up the bluff from below, as opposed to relocating the highway alignment inland.”
It may well be possible to temporarily “shore up the bluff from below” but this will ultimately be defeated by the relentless erosion of the bluff base by emerging groundwater. There is a major wetland 650 yards to the east of the slide, so it is unlikely that the groundwater source will dry up anytime soon. My real concern, however, is that the AK DOTPF analysts have completely missed the possibility of massive failure by earthquake-triggered liquefaction.
It would be my recommendation that this site be evaluated as soon as possible by engineers with expertise in geology, perhaps brought up from the West Coast, who can fully assess the risk of trying to simply repair the problem with steel and concrete. This is not rocket science, but it does require engineers who are familiar with building on this kind of geology.
To regular drivers of the Sterling Highway, such as myself, it would appear that the only reasonable course of action is to move the road well away from the bluff. This will cost some money, but the Sterling Highway has already been rerouted in many places since its construction began in 1947. It’s better to bite the bullet now before the road connection to Anchor Point and Homer is severed and someone goes airborne into a slide- expanded hole in the dark of night.
Ed Berg, PhD, has taught geology at the Kenai Peninsula College since 1983. He recently retired from the US Fish & Wildlife Service, where he was the ecologist at the Kenai National Wildlife Refuge since 1993. He lives with his wife Sara in Homer and is presently working on a book about the geology of the Kenai Peninsula. In September he will be teaching his one-credit Geology of Kachemak Bay course at KPC-Homer, which will include a fieldtrip to the MP 153.3 collapse site.
Comments are closed