Research
Research Overview
My research interests focus on understanding the response of glaciers and ice sheets to past, present, and future climate change. My work is motivated by the recent, rapid increase in ice discharge (i.e., dynamic change) from the marine-terminating outlet glaciers draining the Greenland Ice Sheet. Spatial and temporal variations in glacier dynamics are likely the result of differences in external triggering mechanisms (i.e., environmental controls) and internal controls of ice flow; however, our understanding of dynamic controls is limited by the sparseness of glacier-ocean observations. As described below, I aim to improve the current understanding of glacier dynamics through collaborative research efforts that combine remotely sensed and in situ glacier and ocean observations with numerical ice flow modeling.
I look forward to strengthening my current collaborations with oceanographers and fellow glaciologists, developing new interdisciplinary collaborations, and incorporating student research into my projects. I believe that all research projects benefit from the outside perspectives and insights provided through collaborative efforts and that incorporating students within supervised research projects promotes the development of scientific skills that are not received from classroom instruction. Students interested in remote sensing, climate change, and/or glaciology will have the opportunity to conduct interdisciplinary research projects focused on glacier behavior in a changing climate and ice-ocean interactions. My students will be encouraged to submit their research findings to peer-reviewed journals and to present their research at relevant meetings (e.g., GSA, AGU) in order to develop their scientific writing and oral presentation skills. I am particularly excited to provide underrepresented students with the same access to mentoring and research opportunities that I was granted as a student, which inspired me to pursue my passion for glaciology.
Research Topics
Understanding Glacier Dynamics in a Changing Climate
The spatial and temporal variability of dynamic change observed since the mid-1990s revealed a lack of understanding of changes in glacier behavior in response to environmental forcing. In order to improve our understanding of glacier behavior, we must examine variability in glacier behavior and environmental forcing across a spectrum of spatial and temporal scales.
My research utilizes a variety of remotely sensed datasets to quantify dynamic change and investigate potential triggering mechanisms. Using satellite and aircraft observations, I detected the onset of dynamic change at the majority of the west Greenland outlet glaciers in the early 2000s, earlier than previously inferred from geophysical data. The dynamic acceleration and thinning of these glaciers lead to a gradual increase in ice discharge from this region since 2000. In contrast, discharge from southeastern Greenland peaked in ~2005, then decreased slightly before stabilizing. When combined with surface mass balance observations, these discharge observations suggest that 21st century mass loss from the Greenland Ice Sheet will primarily be due to surface meltwater runoff; however, variability in the ice sheet’s mass balance components makes mass loss predictions highly speculative. Annual discharge estimates for 178 of Greenland's marine-terminating glaciers from 2000-2012 can be downloaded here.
I am currently funded to construct time series of changes in iceberg discharge from the >500 marine-terminating glaciers in Greenland that are not part of the Greenland Ice Sheet. This project will address a critical gap in Arctic glacier mass change estimates and will result in the develop of an automated approach to extract time series of glacier length change from the Landsat image archive. In the future, I plan to expand upon my analysis of the controls glacier dynamics by combining remote sensing and in situ observations of ice velocity, thickness, and terminus position and surface and submarine melting estimates for a suite of glaciers that span the full spectrum of observed seasonal velocity patterns and geographic settings.
Glacier-Ocean Interactions
An increase in submarine melting of glacier termini has been called upon to explain the recent thinning of several west Antarctic ice shelves and widespread retreat of Greenland glaciers, yet few in situ hydrographic observations are available to test the hypothesized dynamic triggering mechanism. Where logistical changes prevent in situ observations, remote sensing techniques can be applied to improve the current understanding of submarine melting.
Using remotely sensed ice flow speed and thickness observations, I found that 5-85% of the volume lost from the floating termini of 13 Greenland glaciers during 2000-2010 was due to submarine melting, with the residual due to calving of icebergs. Although melt rates cannot be estimated for the majority of Greenland’s glaciers with the applied mass-conservation method because they lack discernable floating tongues, I have found that iceberg submarine melt rates derived from differencing repeat high-resolution digital elevation models can be used as a proxy for winter melt rates along glacier termini. My analysis of iceberg melting has been restricted to two fjords so far, but the preliminary data indicate that the iceberg meltwater flux is largely controlled by an iceberg’s submerged surface area, potentially providing a link between changes in the volume and/or shape of discharged icebergs and meltwater fluxes in glacial fjords.
In order to improve the understanding of submarine melting and, more generally, glacier-ocean interactions, I am expanding my iceberg analysis to Antarctica. I currently have a Masters student and an undergraduate honors student working on the generation of iceberg melt rates around the entire Antarctica periphery and analysis of these data with respect to ocean observation and model outputs as well as changes in glacier dynamics. My future research projects will focus on using remotely sensed iceberg freshwater fluxes and surface meltwater estimates to improve fjord circulation models and to investigate the use hyperspectral imagery acquired by unmanned aerial vehicles and satellites for analyses of submarine meltwater plume dynamics.
Predicting Glacier Behavior
In order to provide estimates for sea level rise due to changes in glacier dynamics, both the external and internal controls of glacier behavior must be fully understood. Although some external perturbations can be directly observed, interactions between the various components of the glacier force balance are complex and the interplay between internal and external controls only be fully examined using numerical ice flow modeling.
Thus far, I have investigated whether uncertainty in parameterizations of glacier shape, ice rheology, and basal sliding influence the predicted response to an external perturbation at the glacier terminus using the one-dimensional numerical ice flow model that I developed as part of my dissertation (download the basic Matlab modeling package here). My modeling results suggest that differences in bed elevations and ice rheology within observational uncertainty control the timing and nature of the response to a perturbation and that the basal sliding parameterization exerts a strong control on the simulated ice discharge. Models are likely equally as sensitive to deviations in the observed and simulated terminus position.
I have an ongoing project funded with Dr. Tim Bartholomaus (UIdaho) to use observational data to test the parameterizations currently used to simulate terminus position change in numerical models. We are using a variety of observational datasets from Greenland, including NASA Operation IceBridge lidar, WorldView digital elevation models, Landsat imagery, MEaSURES ice velocities, and the BedMachine bed elevation map, to test seven frontal ablation parameterizations (i.e., calving laws). In the future I will continue to use a wide variety of remote sensing, in situ, and historical glacier observations to investigate the controls of past changes in dynamics and predict mass loss under continued climate warming.
PhD Research
Greenland Iceberg Discharge
In order to develop a better understanding of recent mass loss from the Greenland Ice Sheet, I constructed time series of ice discharge change for 178 marine-terminating outlet glaciers from 2000-2012 with help from my colleagues at The Ohio State University. These discharge time series were then compiled to create a record of ice sheet discharge change. A time series of annual surface mass balance change was obtained from the Regional Atmospheric Climate Model for the Greenland Ice Sheet (RACMO/GR). The surface mass balance and discharge time series were combined to create record of ice sheet mass change since 2000. We found that changes in ice sheet discharge were driven by increased discharge from a small number of large and fast-flowing outlet glaciers (50% of cumulative discharge change from 4 glaciers). Despite the observed increase in ice sheet discharge from ~462 Gt/yr in 2000 to ~547 Gt/yr in 2012, ice sheet mass loss since 2000 was primarily due to increased meltwater runoff. Please email me if you would like to use the discharge data. I am happy to format the data to meet your needs.
Glacier Modeling
I have also developed a depth-integrated, width-averaged numerical ice flow model that can be used to efficiently model ice flow for the channelized tidewater glaciers draining much of the Greenland Ice Sheet. As part of my dissertation research projects, I used the model to examine the influence of uncertainty in glacier shape (article), ice rheology and basal sliding parameterizations (article) on predictions of future glacier change. I found that if a glacier is initially grounded across a basal over-deepening, relatively small variations in outlet width will control both the magnitude and timing of grounding line retreat into the over-deepening following a reduction in resistive stress at the terminus. Additionally, changes in the maximum over-deepening depth within the uncertainty of current ice-penetrating radar observations can result in glaciers switching from stable (i.e., ice remains grounded on a marine shoal) to unstable (i.e., rapid retreat into the over-deepening) behavior.
My numerical modeling simulations also indicate that a non-unique combination of ice rheology and basal sliding parameterizations can be used to reproduce similar steady-state thickness and speed profiles but that following the onset of a reduction in resistive stress, the dynamic behavior of a simulated glacier will vary with the parameter choice. Based on these results, I suggest that prognostic ice flow models are either accompanied by sensitivity studies that investigate the influence of geometry and parameter uncertainty on predictions of dynamic behavior or that these models are not used as precise predictors of future glacier change. The model that I developed to perform my sensitivity studies is open access (download model demo). The model is set-up to run a demonstration and is accompanied by a user guide that explains the model basics, the demonstration, and how to modify the model for application to real glacier systems. A more complex version of the model can also be obtained by emailing me.
Glacier Submarine Melting
I also calculated submarine melt rates beneath floating termini for 13 marine-terminating outlet glaciers in Greenland from 2000-2010 in order to develop a better understanding of variability in ocean forcing that may have triggered the observed changes in dynamics. Annual melt season submarine melt rates were calculated by differencing grounding line and front discharge obtained from satellite remotely-sensed ice flow observations, assuming that mass loss between the grounding line and front is due to submarine melting after accounting for mass loss from surface melt. Although my analysis was spatially and temporally limited, I found no correlation between changes in dynamics and submarine melt rates during the study period. Melt rates varied widely between glaciers, but, on average, mass loss from submarine melting was approximately equal to mass loss from the iceberg calving (article).