2021 Virtual Undergraduate Research Symposium

2021 Virtual Undergraduate Research Symposium

CONSTRAINING WEST ANTARCTIC SNOW VARIABILITY WITH GNSS REFLECTOMETRY

CONSTRAINING WEST ANTARCTIC SNOW VARIABILITY WITH GNSS REFLECTOMETRY

PROJECT NUMBER: 59 | AUTHOR: Stephanie Holzschuh​, Applied Mathematics and Statistics

MENTOR: Matthew Siegfried, Geophysics

ABSTRACT

The surface mass balance of the West Antarctic Ice Sheet and the Ross Ice Shelf are highly variable and its change as a result of global climate change will impact Antarctica’s sea-level contribution. With few in situ observations of snow accumulation, distribution, and melt available, estimations of ice sheet mass balance are often imprecise. This hinders the ability to accurately constrain physical models or validate atmospheric model predictions of the ice sheets. In order to improve these projections, here we use global navigation satellite system interferometric reflectometry (GNSS-IR) to quantify snow accumulation over time in various locations across this terrain. We find that this method captures, with low variability, snow accumulation, and compaction seasonally. These measurements can be used to improve the parameters of physical models and to validate existing models commonly used for ice sheet behavior. Furthermore, with better constraints on the behavior of surface mass balance of these ice sheets, sea-level rise projections can be more accurately portrayed to scientists and global policymakers alike.

PRESENTATION

AUTHOR BIOGRAPHY

Stephanie is a senior in Applied Mathematics and Statistics. She is a member of the Mines Glaciology Laboratory. Currently, she is working on modeling snow accumulation in West Antarctica with her mentor, Dr. Matthew Siegfried in the Geophysics Department.

10 Comments

  1. Nice Job Stephanie! How are the stations secured to the ground, i.e., could melting cause the stations to sink themselves biasing the data?

    • Hi Michael,

      For the station setup, the antenna is mounted on top of a metal pole. The metal pole is placed into the ground roughly 1-2 m below the surface. Then, the near-surface firn will freeze the pole into the ground. While I don’t know the extent that melting affects it, the GNSS-IR method was proven to be accurate to 0.02 m and precise to 0.06 m for surface snow height estimates in a previous investigation of the Whillans and Mercer ice streams when compared to in situ observations. If you’re curious to read about that research, the paper that discusses this can be found here: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2017GL074039.

      So, I would say that GNSS-IR can be considered a reliable method for surface height estimates given that the daily uncertainty is 0.06 m.

  2. Great talk, Stephanie! I’m curious if you can distinguish snow compaction from snow removal by wind? Or would both present the same signal.

    • Hi Wilson!

      Thank you for listening to my talk. It would be interesting to consider how snow compaction versus snow removal would play into the wind analysis. In both snow compaction and snow removal by wind, the estimated reflector height should increase (indicating a “loss” to the surface height). So, in this way, both would present the same signal. However, I haven’t considered alternative methods that could be applied in this situation to be able to differentiate between the two scenarios.

      I will definitely have to think more about this, so thank you for your insight.

  3. Very cool talk Stephanie! I noticed that the 95% confidence intervals on the time series are not always constant with time. Do you think this non-constant variability of the estimated reflector height could be related to any kind of physical processes associated with snow accumulation, compaction, etc?

    • Hi Roger!

      If I understand the non-constant variability that you are referring to correctly, I would say there are significant seasonality responses in the data due to the physical setup of the GNSS stations. Due to the stations’ sunlight dependency, there are higher variances expressed when stations are turning on and off from a lack of power. This can contribute to the confidence interval being wider or showing sporadic dips/spikes in the summer months. As for non-seasonal trends in variability, it would be interesting to investigate other physical processes such as compaction or wind erosion that may affect the estimated reflector height.

  4. Nice work, Stephanie!

    • Thank you Dr. Behn!

  5. Stephanie very interesting study. I am curious about the sampling rate of GNSS for the recorded measurements.

    • Hi Ana,

      The GNSS-IR software can be adjusted for parameters that determine if a specific satellite pass was “successful” over a particular station. For this application, the stations had roughly 70-80 arcs that were deemed successful on a daily resolution. This number is affected by the consistency of the data. It is consistent in the winter months since there is more sunlight to keep the stations turned on and less consistent in the summer months when the stations may not have had enough power to record as many of the satellite passes over the station.

Share This