Much of the uncertainty in fate of the Greenland and Antarctic ice sheets results from a lack of constraints on ice sheet sensitivity and response to a range of forcings. The geological record of deglaciated land- and seascapes allows for the reconstruction of ice flow and retreat in the past and provides insight on processes and feedbacks that influence ice behavior.
During the Last Glacial Maximum (~20,000 years ago), the Antarctic Ice Sheet was much larger and thicker than today. In many places, the ice sheet extended to the continental shelf break, and glacial landforms and sediments associated with ice advance and retreat are preserved on the seafloor. Within the Ross Sea Embayment, the largest drainage basin for the Antarctic Ice Sheet, variable styles of retreat were influenced by regional topography and seafloor geology (Halberstadt et al., 2016, The Cryosphere; Simkins et al., 2016, GSL Memoirs). The integration of sediment facies analyses and geomorphology reveal clear distinctions between subglacial and ice-proximal facies, proximity to paleo-grounding lines based on the presence of ice shelf basal debris in sediment cores, and changing marine influence (Prothro et al., 2017, Marine Geo). During retreat of the ice sheet, an abrupt increase in atmospherically produced beryllium-10 and diatom concentrations at the facies transition to open marine conditions supports a major ice shelf collapse event in the late Holocene (Yokoyama et al., 2016, PNAS).
Ongoing projects focus on (i) assessing ice sheet (in)stability from a large dataset (n=6,275) of landforms deposited at paleo-grounding lines, the downstream most points ice was in contact with underlying substrate and (ii) addressing questions about the retreat of the southwestern Cordilleran Ice Sheet in Washington state (a relatively local field site for me!) from the stratigraphic record (Simkins et al., 2017, GSA field guide; Demet et al., in review) .
At the base of the Antarctic Ice Sheet are hundreds of identified subglacial lakes, some of which periodically drain water downstream of the lakes and even to the grounding line. Geological observations are complementary to observations of contemporary subglacial hydrology and offer unique perspectives on the nature and influence of meltwater drainage on broad spatial and temporal scales. Although paleo-subglacial channels have been identified on the continental shelf around Antarctica, none have yet been temporally associated with a single deglacial phase and linked to ice sheet grounding lines.
We present the first evidence of a subglacial hydrological system that was active during the post-Last Glacial Maximum deglaciation and connected explicitly to grounding line positions of a former ice stream in the western Ross Sea (Simkins et al., 2017, Nature Geoscience). Episodic channelized meltwater drainage locally restricted grounding line landform growth, limiting the degree to which landforms could have provided stability feedbacks to the grounding line. The drainage system was fed on a frequency of decades to centuries by upstream subglacial lakes in an area of elevated geothermal heat. Meltwater drainage configuration persisted through numerous grounding line retreat events, shifts in ice flow direction, and a circuitous retreat pattern, suggesting that the stable location of source lakes and ample production of basal melting influenced the retreating ice stream.
Locations that are proximal to ice sheets (i.e. in the 'near-field') experience relative sea-level fall during deglaciations due to glacial isostatic adjustment, causing the solid Earth to rebound in response to ice unloading. The effects of glacial isostatic adjustment extend far beyond the near-field and create spatially variable sea levels across the globe.
Coastal landforms preserved above current sea level by post-glacial rebound allow for near-field relative sea-level reconstructions. Based on a series of raised beaches in Antarctic Peninsula, relative sea level fell at a rate of 3 mm/a between 7,000 and 2,500 years ago and 1.5 mm/a during the last 2,500 years (Simkins et al., 2013, Quaternary Science Reviews). Our results indicate half as much sea-level fall during the Holocene than suggested by previous reconstructions, indicating that models have overestimated of ice loss and/or do not sufficiently characterize Earth rheology. Antarctic raised beaches also reveal information about interactions between wave energy, sea ice and coastal evolution. We found that distinct periods of heightened beach formation around Antarctica correspond to minimum sea-ice conditions (Simkins et al., 2015, Journal of Quaternary Research). Increased El Niño frequency is associated with reduced sea ice in the South Pacific and enhanced beach formation in the Antarctic Peninsula, while the South Atlantic experienced more expansive sea ice causing limited beach formation along the East Antarctic margin.
Dating geologic materials is often difficult, despite the necessity of establishing the timing of past events. Optically stimulated luminescence dating determines how long materials have been shielded from sunlight. Naturally decaying elements within minerals lead to the build up of free electrons that are trapped in crystal defects or impurities and the number of free electrons is directly related to the rate at which elements decay and the burial age. This dating method is traditionally applied to sedimentary quartz and feldspar grains; however, I am involved in improving the application of optically stimulated luminescence dating to minerals extracted from crystalline rock surfaces (Simkins et al., 2016, Quaternary Geochronology), providing a dating method that is useful for environments where materials for radiocarbon dating are sparse and other dating methods are not suitable.