Welcome to Aurora Documentation

Aurora is an open-source package that robustly estimates single station and remote reference electromagnetic transfer functions (TFs) from magnetotelluric (MT) time series. Aurora is part of an open-source processing workflow that leverages the self-describing data container MTH5, which in turn leverages the general mt-metadata framework to manage metadata. These pre-existing packages simplify the processing by providing managed data structures, transfer functions to be generated with only a few lines of code. The processing depends on two inputs – a table defining the data to use for TF estimation, and a JSON file specifying the processing parameters, both of which are generated automatically, and can be modified if desired. Output TFs are returned as mt-metadata objects, and can be exported to a variety of common formats for plotting, modeling and inversion.

Key Features

• Tabular data indexing and management (Pandas dataframes),

• Dictionary-like processing parameters configuration

• Programmatic or manual editing of inputs

• Largely automated workflow

Documentation for the Aurora project can be found at http://simpeg.xyz/aurora/

Installation

Suggest using PyPi as the default repository to install from

pip install aurora

Can use Conda but that is not updated as often

conda -c conda-forge install aurora

General Work Flow

1. Convert raw time series data to MTH5 format, see MTH5 Documentation and Examples.

2. Understand the time series data and which runs to process for local station RunSummary.

3. Choose remote reference station KernelDataset.

4. Create a recipe for how the data will be processed Config.

5. Estimate transfer function process_mth5 and out put as a mt_metadata.transfer_function.core.TF object which can output [ EMTFXML | EDI | ZMM | ZSS | ZRR ] files.

User Documentation

• Installing
  • PyPi
  • Conda Forge
  • From source
  • Issues

Examples

• Build an MTH5 and Operate the Aurora Pipeline
  • Flow of this notebook
• 1. Build an MTH5 file from Earthscope archives
  • 1A: Specify the data to access from IRIS
  • 1B: Examine and Update the MTH5 object
  • 1C: Optionally Update Metdata:
• 2: Process Data
  • Generate an Aurora Configuration file using MTH5 as an input
  • Channel Summary
  • Run Summary
  • Kernel Dataset
• Here is one way to select a single run:
• To discard runs that are not very long
  • Select only runs “b” & “d”
  • The same result can be obtained by excluding runs a & c
• To process only a segment of data
• Exercise:
  • What if I have unconventional channel names?
• Exercise:
• Dataset Definition DataFrame
  • Create the Stations container
• Excercise:

Tutorials

• Processing Configuration
  • Using ConfigCreator
• Example of making a KernelDataset from an mth5
  • Create Config from KernelDataset
  • Default Parameters
• EMTF Band Setup File
• Accessing Magnetic Field Data from EarthScope Using MTH5
• 0. Getting started
  • Confirm installation
  • Import required packages and modules
• 1. Exploring the data at EarthScope
  • Figure 1: The EarthScope Geographic Interface
  • Figure 2: Spatial distribution of the EarthScope magnetic dataset
  • Station Service
  • Figure 3: Screengrab of station list associated with magnetic field data.
  • Availability Service
  • Figure 4: Screengrab of availability URL for Network=”4P”
  • Metadata Aggregator (MDA)
  • Figure 5: Earthscope Metadata Aggregator (MDA) website landing page.
• 2. Case Study:
  • Comparison with Published Observation (Heyns et al 2020)
  • Figure 6: Title and Abstract from Heyns et al. 2020
  • Figure 7: Screengrab of Figure 1 from Heyns et al. 2020
  • Figure 8 Interactive map showing query results.
  • Select a single station
  • Figure 9 MDA info about station ALW49
  • Figure 10: 4P.ALW49 Data Availability
• 3. Define request dataframe
• 4. Build MTH5
  • (metadata only)
  • Tabulated Metadata
• 5. Build MTH5 (with data)
• 6. Calibrating time series
• 7. Comparison of time series
• TODO: 9. Access USGS data directly
• Process CAS04 with Remote Reference
• Make MTH5 from IRIS Data Managment Center v0.2.0
  • Initialize a MakeMTH5 object
  • Make the data inquiry as a DataFrame
  • Fix Survey Name
• To compare with the archived file, we need to set the coordinate system to geographic
• Part II: Logic to save FCs
• Now that the FCs are saved we can access them:|
• Absolute Minimal Example
• Synthetic Data Tutorial
  • Contents:
  • Process Synthetic Data with Aurora
  • Fourier coefficient storage in MTH5
• Feature Storage Experimental work in progress

Reference documentation

• aurora
  • aurora package
    • Subpackages
      • aurora.config package
      • aurora.pipelines package
      • aurora.sandbox package
      • aurora.test_utils package
      • aurora.time_series package
      • aurora.transfer_function package
    • Submodules
    • aurora.general_helper_functions module
      • DotDict
      • count_lines()
      • execute_subprocess()
      • get_test_path()
      • save_to_mat()
    • Module contents

Contributing to Aurora

If you’d like to contribute, start by searching through the issues to see whether someone else has raised a similar idea or question. If you don’t see your idea listed, open an issue, and if there are code changes, also a pull request.

There are many ways to contribute to the aurora, such as:

• Report/fix bugs or issues encountered when using the software

• Suggest additional features or functionalities

• Fix editorial inconsistencies or inaccuracies

Aurora is hosted by simpeg, please refer to their contributing guidelines page. for more details.