1 of 21

Exploratory analysis of an early Holocene sea salt record via the Skytrain Ice Rise Core

Gillian McGinnis

Reed College ‘22

Department of Chemistry

Advised by Prof. Mackenzie M. Grieman

Presented 28 April 2022

2 of 21

Inspiration

Ongoing: “Well that’s weird.”

Questions:

  • What happened there?
  • To what extent does the mean change? And the standard deviation?
  • Is this unique to the site?
  • Can this be used to indicate WAIS collapse?

2

3 of 21

Challenges

Solutions

→

3

Subject familiarity

→

Lots of learning!

Quantity of data

→

Integration of external packages�(Thank you exaexa/scattermore ❤)

Ongoing project updates

→

Flexibility and communication

Ongoing dataset adjustments

→

Prioritizing ease of reproducibility

Many possible directions for analytics

→

Opting for preliminary exploration

4 of 21

Sea salt aerosol

Sources:

  • Open ocean
  • Sea ice
  • Frost flowers

4

Wietz, 2009

Rhodes et al., 2017

5 of 21

General process

  • Core recovery
  • Continuous Flow Analysis
  • Age/Depth scale
  • Preliminary analytics
  • Comparison to other ice cores
  • Advanced modeling

5

6 of 21

1. Core recovery: WACSWAIN - Skytrain Ice Core

6

Grieman et al., 2021

→ Come see 3D-printed topo maps in Mackenzie’s office!

Warm Climate Stability of WAIS (West Antarctic Ice Sheet) in the last Interglacial

Mulvaney et al., 2021

7 of 21

2. Continuous Flow Analysis (CFA)

  • Inductively coupled plasma mass spectrometry (ICP-MS)
  • Fast ion chromatography (FIC)

7

Grieman et al., 2021

8 of 21

3. Age-Depth Scale: first 2000 years, then the rest!

8

Hoffman et al., 2022, pers. comm.

Mulvaney et al., 2022, pers. comm.

9 of 21

4. Preliminary analytics

Analytics & visualization via R in RStudio

Sea salt component calculations

9

Rm = 0.038

mean marine Ca:Na

non-ss component = ion – (ss component)

Rt = 1.78

mean terrestrial Ca:Na

Values of interest:

Age, Ca2+, Na+, Mg2+, Cl–,

ssNa, ssMg, ssCl,

non-ssNa, non-ssMg, non-ssCl

10 of 21

Full core change over time: ramp event (7800 – 7350 yr BP)

10

11 of 21

Ion variability and change over time

11

Mean (SD)

12 of 21

5. Comparison to other Antarctic records: South Pole SPICEcore

12

Winski, 2021

13 of 21

5. Comparison to other Antarctic records: ssNa

13

Winski, 2021

14 of 21

Summary of analytically-supported results

  • Statistically significant difference before & after the ramp event
  • Ramp event not on a constant slope for non- sea salt component
  • Impact of calcium should not be underestimated
  • Non-universal event, not seen at South Pole

14

15 of 21

6. Advanced modeling: future research

  • Similar analytics of other time ranges of interest
    • Spike at ~3000 years BP
    • Compensating for gap at ~5500 years BP
  • Comparisons with other Antarctic ice cores
  • HYSPLIT (Hybrid Single-Particle Lagrangian Integrated Trajectory) back-trajectory
  • p-TOMCAT atmospheric modeling

15

16 of 21

♥ Acknowledgements

  • Mackenzie M. Grieman
  • Reed College Dep’t of Chemistry
  • Math-stats professors
  • Jay Ewing, physics fab-lab
  • WACSWAIN project members
  • Ben Salzburg & Sam Gauck (CUS)
  • RStudio online community & forums

16

CFA team

WACSWAIN

In the field!

Bruno�& Skye

>not Antarctica :’(

:))

♥

Friends, family, co-workers, animal companions, and viewers like you ;)

17 of 21

~Bonus Slides~

aka this is what happens when your thesis is comprised entirely of figures and statistics

17

18 of 21

Ramp event abnormalities

18

Ca

Slope (SE)

19 of 21

Ion variability and change over time

19

Calcium “deviations”?

(focus area: 8750 – 6250 years BP)

20 of 21

5. Comparison to other Antarctic records: SPICEcore - SPC14

21 of 21

Sea salt vs. non- sea salt fractions

21

Linear regression eqt’n & goodness-of-fit

y = mx + b (R2)