Hunga Tonga Volcanic Eruption & Earth's Ionosphere

This research project was undertaken to fulfill requirements for 'Project 1' of City College of San Francisco's MATH 108 Fundamentals of Data Science course.

Introduction to Earth's Ionosphere

Earth's atmosphere is composed of several layers, each with different physical properties. The ionosphere is region of the atmosphere containing several sub-layers of electrically charged (ionized) gas particles. Depending on the density and energy of these gas particles, radio frequency (RF) traveling through these layers can be absorbed, reflected back to earth, or lost to space.

Ionosphere diagram

Sub-layers tend to follow diurnal (day/night) patterns, with E & F being persistent, and D, F1 & F2 emerging only during the day.

Ionosphere layers

Measuring the Ionosphere

Methods of measuring the properties (sounding) the ionosphere range from active radar-like radio frequency probes to passive GPS satellite signal observation. Ionosonde is the term used for active radio frequency probes.

Ionosonde schematic

Ionosonde data can be represented visually in the form of an ionogram. Ionograms typically include data such as the Frequency of F2, Frequency of E, Total Electron Content, et al.

Example ionogram

An international consortium of research institutions runs several ionospheric sounding stations around the planet. These stations probe the ionosphere periodically through the day, year round. This data is used to predict ionospheric conditions that may effect long range radio transmission, satellite signal reception, and global navigation satellite positioning.

Ionosonde Station Map

Heliophysics has the strongest influence on ionospheric conditions, including day/night, solar wind, sun cycles, sun spots, coronal mass ejections (CMEs), et al. Earth weather & geophysics also influence the ionosphere, including weather & storms, tsunamis & earthquakes.

Project Description

This project aims to use the global ionospheric surveillance network dataset to determine if the Hunga Tonga volcanic eruption in January 2022 had a measurable effect on the ionosphere.

Variables

Variables & measurements we intend to measure include:

Limitations

Potential limitations to the data used in this project include:

  1. The number of stations within range of the volcano. More stations would allow more granular readings.
  2. Periodicity is not guaranteed. Ionosonde readings do not appear to be syncronized, and some periods are missing completely.

Future Considerations

Errata / N.B.

This project originally used raw MIDS data from NOAA, but switched over to use FastChar data from GIRO. Consider all MIDS-related functions DEPRECATED in favor of FastChar.

Sources

The data used in this project comes from these sources:

Ionosonde MIDS & FastChar:

Other Notes

"A large eruption commenced on 15 January 2022 at 04:20 local time, sending clouds of ash 20 km (12 mi) into the atmosphere."

https://volcano.si.edu/ShowReport.cfm?doi=10.5479/si.GVP.WVAR20220112-243040

https://cimss.ssec.wisc.edu/satellite-blog/archives/44214

https://en.wikipedia.org/wiki/2022_Hunga_Tonga_eruption_and_tsunami

https://appliedsciences.nasa.gov/our-impact/news/tonga-eruption-sent-ripples-through-earths-ionosphere

Look at Guam barometric pressure data to see when pressure wave hit:

https://www.ndbc.noaa.gov/station_history.php?station=aprp7

The 'Code'

Boiler-plate

Imports

Constants

Project-specific function definitions

Data Science!

Create Ionosonde DataTables

Data Ingest & Pre-processing

Focus Data

We're going to focus our investigation on the station at Guam.

Visualize Ionosonde Data

Zoomed-out view

For our zoomed-out view we'll look at data for the entire month of January. Red dot represents the eruption.

Scatter: foF2 & TEC

We'll look at two values from the ionosonde, the foF2 and the total electron content (TEC). There is an association between these two values - that is: As the TEC increases so does the foF2.

Line: TEC

The plot below shows TEC measurements for all of January, which seem to follow some diurnal patterns, climaxing during the day and dropping at night.

Zoomed-in View

To 'zoom-in' we're going to take a snapshot of our data within the day of the eruption.

Line: TEC

This plot shows the typical diurnal pattern of daytime peak and nightime low, but with a considerable spike shortly after the eruption.

Line: foF2

Similar to the TEC pattern plotted above, the foF2 plot shows usual diurnal pattern, with a notable peak-trough-peak shortly after the eruption, followed by another peak later in the day.

Histogram: foF2

This histogram of foF2 shows us the distribution of foF2 readings throughout our snapshot period, with two prominent areas around 2.5 and 10.75 MHz.

Mo' Graphs

Day-over-Day Comparisons: foF2 & TEC