Repository for filtering and comparing zircon age spectra for ID-TIMS datasets

Online usage

For trying out some of the code without installation, click the link below to run a Jupyter notebook to run some of the functions. Be patient, it can take a good few minutes for everything to compile.

Prerequisites

POT: python optimal transport sh pip install POT

Installation

1. Open terminal
2. Change the current working directory to the location where you want the cloned directory.
3. Clone the repository sh git clone https://github.com/ChetanNathwani/zircon_age_spectra.git
4. You should now see the repository has appeared in your current working directory

Functions

Systematically filter "antecrysts" (older tails) in age distributions

This can be done by defining an age distribution, here we use (Szymanowski et al. (2023): ```sh ages = [0.151, 0.284, 0.293, 0.195, 0.237, 0.21 , 0.367, 0.546, 0.941, 0.194, 0.422, 0.219, 0.29 , 0.242, 0.319, 0.269, 0.267, 0.138, 0.217, 0.327, 0.206, 0.263, 0.359, 0.303, 0.449, 0.138, 0.365, 0.261, 0.142]

unc = [0.006, 0.005, 0.005, 0.007, 0.006, 0.005, 0.004, 0.006, 0.005, 0.005, 0.005, 0.006, 0.005, 0.006, 0.005, 0.005, 0.005, 0.007, 0.006, 0.005, 0.007, 0.006, 0.01 , 0.005, 0.015, 0.01 , 0.008, 0.007, 0.006] Then calling the function: sh geochron.filterolderages(ages, unc) ``` Here are some examples of ranked age plots for three age distributions where the red bars are non-filtered ages and grey bars are filtered ages:

Compare the shape of an age distribution to published ID-TIMS age distributions

We first initialise a pd.DataFrame() containing the results of a principal component analysis of the Wasserstein dissimilairty matrix of the filtered ID-TIMS age distribution compilation:

sh geochron.generate_pca_scores() Let's take a look at what that produced:

| | PC1 | PC2 | Type | Locality | |---:|-----------:|------------:|:---------|:---------------------| | 0 | -0.465769 | -0.00700539 | Plutonic | Adamello | | 1 | -0.391456 | -0.328376 | Volcanic | Agua de Dionisio | | 2 | -0.667256 | -0.124333 | Porphyry | Bajo de la Alumbrera | | 3 | -0.464677 | -0.0586558 | Porphyry | Bajo de la Alumbrera | | 4 | -0.534635 | -0.286713 | Porphyry | Bajo de la Alumbrera | | 5 | -0.552401 | 0.225343 | Porphyry | Batu Hijau | | 6 | -0.109091 | 0.362511 | Porphyry | Batu Hijau | | 7 | -0.251048 | -0.294216 | Porphyry | Batu Hijau | | 8 | 0.173185 | -0.343716 | Plutonic | Bear Valley | | 9 | -0.38985 | -0.30109 | Plutonic | Bear Valley | | 10 | -0.343748 | -0.127883 | Plutonic | Bear Valley | | 11 | -0.304634 | -0.436756 | Plutonic | Bergell | | 12 | 0.980955 | -0.139933 | Plutonic | Bergell | | 13 | 0.203133 | -0.362156 | Plutonic | Bergell | | 14 | -0.17997 | -0.288732 | Plutonic | Bergell | | 15 | 0.349097 | -0.532901 | Plutonic | Bergell | | 16 | 0.0796051 | -0.499426 | Plutonic | Bergell | | 17 | 0.700678 | -0.325638 | Plutonic | Bergell | | 18 | -0.635634 | 0.168698 | Porphyry | Bingham Canyon | | 19 | -0.575977 | -0.204604 | Porphyry | Bingham Canyon | | 20 | 0.77727 | -0.183307 | Plutonic | Capanne | | 21 | 1.97742 | 0.916772 | Plutonic | Capanne | | 22 | -0.473491 | -0.181504 | Plutonic | Capanne | | 23 | 0.347536 | -0.394673 | Plutonic | Capanne | | 24 | -0.464833 | -0.40269 | Plutonic | Capanne | | 25 | -0.0190579 | -0.515712 | Plutonic | Capanne | | 26 | -0.227219 | -0.432933 | Volcanic | Carpathian-Pannonian | | 27 | -0.328832 | 1.25728 | Volcanic | Carpathian-Pannonian | | 28 | -0.543165 | 0.443821 | Volcanic | Carpathian-Pannonian | | 29 | -0.245996 | -0.191063 | Volcanic | Chegem | ...

Now we can cast our example Youngest Toba Tuff age distribution into the same PCA space and compare the results with the age distribution compilation:

sh geochron.calc_W_PCA(ages_fil, unc_fil) # Use the filtered age distribution which removes one older outlier

An example of the results:

We can see that the youngest Toba Tuff has a younger skew and plots where most volcanic age distributions plot

Citation