Given that plate tectonics have only been widely accepted and studied for less than a century, there’s still a lot of interest and important knowledge to be had from examining its global geophysical phenomena– especially earthquakes, which can have a devastating impact on human life. With the historic January 15th, 2022 volcanic eruption in Tonga, we were interested in examining that trench system (and others nearby) to understand better how earthquakes relate to boundary conditions, particularly the influence depth and/or frequency of earthquakes had on magnitudes.
Using USGS data on seismic activity, we examined a few main questions around this topic:
Does the distribution of earthquake magnitudes along two faults come from the same distribution?
Are intensity of earthquake magnitudes correlated with time since last activity (theory being that more frequent earthquakes release tension), or daily frequency of earthquake activity? How about the relationship between magnitude and hypocenter origination?
As the statistical analysis was being conducted, other points of interest arose in the data and are explored later on in the study. A paper detailing our hypothesis, assumptions, and rigorous statistical methods and results/interpretation are in the repository.
Statistical Methods Include:
- Time Series Analysis, Smoothing, Autocorrelation
- Hypothesis Testing, K-S Tests for Divergence of CDFs, T-Tests, Confidence Intervals
- Regression Techniques, Deming Regression, MA/SMA Regression
- Quantifying Uncertainty, Standard Error of Sampling Distributions
- Visualization: Contour Plots, Histograms, Time Series
- Massaging Data, Filtering, Algorithms to perform calculations across observations/variables in datasets