_images/platecurie_logo.png

PlateCurie Documentation

Crustal magnetic anomalies carry information on the distribution of remanent magnetization in the Earth’s crust (Blakely, 1995). The Curie point corresponds to the depth at which crustal rocks become paramagnetic where they reach their Curie temperature, and is obtained by fitting the power spectral density (PSD) of magnetic anomaly data with a model where magnetic anomalies are confined within a buried layer (Bouligand et al., 2009 Audet and Gosselin, 2019 Mather and Fullea, 2019). Mapping the Curie point provides important information on geothermal gradients in the Earth; however, this is a spatio-spectral localization problem because the PSD needs to be calculated within moving windows at wavelengths long enough to capture the greatest possible depth to the bottom of the magnetic layer. The wavelet transform is particularly well suited to overcome this problem because it avoids splitting the grids into small windows and can therefore produce PSD functions at each point of the input grid (Gaudreau et al., 2019).

PlateCurie is a software that extends the package PlateFlex, which contains Python and Fortran modules to calculate the wavelet transform and scalogram of 2D gridded data, by providing new classes MagGrid and ZtGrid that inherit from the PlateFlex Grid class, and a new class Project with methods to estimate the properties of the magnetic layer (depth to top of layer zt, its thickness dz, and power-law exponent of fractal magnetization β) using either non-linear least-squares or Bayesian inference. The software is bundled with Jupyter notebooks that describe the processing steps required for the estimation of model parameters.

This software was inspired by a sister software PyCurious by Mather and Delhaye. The two softwares differ by their use of different spectral estimators (wavelet transform for platecurie and periodogram for pycurious) as well as the estimation process (either non-linear least-squares or pymc3 bayesian estimation for platecurie, and a Bayesian framework for pycurious).

Quick Links

Getting Started