Geology topics

Michael G. Loso

Publications and source records attributed to Michael G. Loso.

4 recordsLinked to original sources

USGS2017

Varve formation during the past three centuries in three large proglacial lakes in south-central Alaska

The sediments stored in the large, deep proglacial lakes of south-central Alaska are largely unstudied. We analyzed sediments in 20 cores, up to 160 cm long, from Eklutna, Kenai, and Skilak Lakes, using a combination of repeated lamination counting, radionuclide dating, event stratigraphy, and tephrochronology. We show that the characteristically rhythmic layers were deposited annually. Most of these glacial varves consist of one coarse-grained base and a fine-grained top, but varves composed of multiple coarse-grained turbidite pulses are common too. They are likely related to successive episodes of high sediment discharge during flooding, and they become more frequent in all three lakes, along with increased sedimentation rates, during the nineteenth century late phase of the Little Ice Age. These flood turbidites were generated by rain events and intense melting of snow and ice. Other (mega) turbidites are a result of earthquake-triggered slope collapses (e.g., A.D. 1964). Some event layers are present in all three lakes. In addition, the annual time series of varve thickness (normalized annual sedimentation rate) are significantly correlated among the three lakes (ρ > 0.27; p < 0.001). Differences between the varve thickness records can be attributed partly to the dam construction at Eklutna Lake and outbursts from an ice-dammed lake at Skilak Lake. Geomorphologic differences among the catchments result in further differences in sedimentation patterns in the three lakes.

Alaska↗

USGS2017

Geometry, mass balance and thinning at Eklutna Glacier, Alaska: an altitude-mass-balance feedback with implications for water resources

We analyzed glacier surface elevations (1957, 2010 and 2015) and surface mass-balance measurements (2008–2015) on the 30 km 2 Eklutna Glacier, in the Chugach Mountains of southcentral Alaska. The geodetic mass balances from 1957 to 2010 and 2010 to 2015 are −0.52 ± 0.46 and −0.74 ± 0.10 m w.e. a −1 , respectively. The glaciological mass balance of −0.73 m w.e. a −1 from 2010 to 2015 is indistinguishable from the geodetic value. Even after accounting for loss of firn in the accumulation zone, we found most of the mass loss over both time periods was from a broad, low-slope basin that includes much of the accumulation zone of the main branch. Ice-equivalent surface elevation changes in the basin were −1.0 ± 0.8 m a −1 from 1957 to 2010, and −0.6 ± 0.1 m a −1 from 2010 to 2015, shifting the glacier hypsometry downward and resulting in more negative mass balances: an altitude-mass-balance feedback. Net mass loss from Eklutna Glacier accounts for 7 ± 1% of the average inflow to Eklutna Reservoir, which is entirely used for water and power by Anchorage, Alaska's largest city. If the altitude-mass-balance feedback continues, this ‘deglaciation discharge dividend’ is likely to increase over the short-term before it eventually decreases due to diminishing glacier area.

Alaska↗

USGS2017

Glacial conditioning of stream position and flooding in the braid plain of the Exit Glacier foreland, Alaska

Flow spilling out of an active braid plain often signals the onset of channel migration or avulsion to previously occupied areas. In a recently deglaciated environment, distinguishing between shifts in active braid plain location, considered reversible by fluvial processes at short timescales, and more permanent glacier-conditioned changes in stream position can be critical to understanding flood hazards. Between 2009 and 2014, increased spilling from the Exit Creek braid plain in Kenai Fjords National Park, Alaska, repeatedly overtopped the only access road to the popular Exit Glacier visitor facilities and trails. To understand the likely cause of road flooding, we consider recent processes and the interplay between glacier and fluvial system dynamics since the maximum advance of the Little Ice Age, around 1815. Patterns of temperature and precipitation, the variables that drive high streamflow via snowmelt, glacier meltwater runoff, and rainfall, could not fully explain the timing of road floods. Comparison of high-resolution topographic data between 2008 and 2012 showed a strong pattern of braid plain aggradation along 3 km of glacier foreland, not unexpected at the base of mountainous glaciers and likely an impetus for channel migration. Historically, a dynamic zone follows the retreating glacier in which channel positions shift rapidly in response to changes in the glacier margin and fresh morainal deposits. This period of paraglacial adjustment lasts one to several decades at Exit Glacier. Subsequently, as moraine breaches consolidate and lock the channel into position, and as the stream regains the lower-elevation valley center, upper-elevation surfaces are abandoned as terraces inaccessible by fluvial processes for timescales of decades to centuries. Where not constrained by these terraces and moraines, the channel is free to migrate, which in this aggradational setting generates an alluvial fan at the breach of the final prominent moraine. The position of this fan is glacially conditioned but the process of migration of the braided channels across it is not. This broad perspective on channel controls identifies incipient avulsion into the roadside forest as part of a long-term fan-building process independent from changes in streamflow or sediment load.

Alaska↗

USGS2017

Paleoseismic potential of sublacustrine landslide records in a high-seismicity setting (south-central Alaska)

Sublacustrine landslide stratigraphy is considered useful for quantitative paleoseismology in low-seismicity settings. However, as the recharging of underwater slopes with sediments is one of the factors that governs the recurrence of slope failures, it is not clear if landslide deposits can provide continuous paleoseismic records in settings of frequent strong shaking. To test this, we selected three lakes in south-central Alaska that experienced a strong historical megathrust earthquake (the 1964 M w 9.2 Great Alaska Earthquake) and exhibit high sedimentation rates in their main basins (0.2 cm yr − 1 –1.0 cm yr − 1 ). We present high-resolution reflection seismic data (3.5 kHz) and radionuclide data from sediment cores in order to investigate factors that control the establishment of a reliable landslide record. Seismic stratigraphy analysis reveals the presence of several landslide deposits in the lacustrine sedimentary infill. Most of these landslide deposits can be attributed to specific landslide events, as multiple landslide deposits sourced from different lacustrine slopes occur on a single stratigraphic horizon. We identify numerous events in the lakes: Eklutna Lake proximal basin (14 events), Eklutna Lake distal basin (8 events), Skilak Lake (7 events) and Kenai Lake (7 events). The most recent event in each basin corresponds to the historic 1964 megathrust earthquake. All events are characterized by multiple landslide deposits, which hints at a regional trigger mechanism, such as an earthquake (the synchronicity criterion). This means that the landslide record in each basin represents a record of past seismic events. Based on extrapolation of sedimentation rates derived from radionuclide dating, we roughly estimate a mean recurrence interval in the Eklutna Lake proximal basin, Eklutna Lake distal basin, Skilak Lake and Kenai Lake, at ~ 250 yrs., ~ 450 yrs., ~ 900 yrs. and ~ 450 yrs., respectively. This distinct difference in recording can be explained by variations in preconditioning factors like slope angle, slope recharging (sedimentation rate) and the sediment source area: faster slope recharging and a predominance of delta and alluvial fan failures, increase the sensitivity and lower the intensity threshold for slope instability. Also, the seismotectonic setting of the lakes has to be taken into account. This study demonstrates that sublacustrine landslides in several Alaskan lakes can be used as reliable recorders of strong earthquake shaking, when a multi-lake approach is used, and can enhance the temporal and spatial resolution of the paleoseismic record of south-central Alaska.

Alaska↗