Agustina Pesce and Santiago Soler
| Information | |
|---|---|
| Event: | International Meeting for Applied Geoscience and Energy (IMAGE) 2023 |
| When? | 30 August, 2023 |
| Where? | Houston, Texas at the George R. Brown Convention Center |
| Session: | Mining and Mineral Exploration |
| Slides: | Download slides as PDF |
Fatiando a Terra is a community-driven project with the goal of providing open-source Python tools for geophysics that are well tested, well documented and easy to use. Their open-source license allows students, researchers and professionals to freely access the code, use it for any purpose, and modify it. This enables scientists to use it as the foundations for their research, and industry professionals to include them as part of their toolkit. During this talk we'll introduce the tools available in Fatiando a Terra, showing how they can be used to solve geophysical problems.
Fatiando a Terra is composed by a collection of Python libraries, each one with a specific set of goals and scope. Pooch allows to easily download and cache data from the web. Ensaio hosts curated open-licensed geophysical datasets useful for teaching, learning how to use our tools and probing new methodologies and code. Verde offers tools for processing and interpolating spatial data with an interface inspired in the well-known machine learning packages. Boule defines geodetic reference ellipsoids that offer coordinate conversions and normal gravity calculations. Lastly, Harmonica enables processing and modelling gravity and magnetic data. It offers tools for applying standard corrections to gravity data; FFT-based transformations to regular grids; forward modelling gravity and magnetic fields using different sources such as point sources, dipoles, rectangular prisms and tesseroids (a.k.a. spherical coordinates). It also includes tools for interpolating, gridding and upward continuing harmonic fields data with equivalent sources; and for reading geophysical data from industry formats, like Oasis Montaj® GRD files. Fatiando a Terra follows the higher standards in software development following best practices for developing, maintaining, testing and documenting our code. Our tools rely on well-known Python libraries from the Python scientific ecosystem. This allows us to write fast and efficient implementations of standard and state-of-the art geophysical methodologies.
With more than 12 years of development, Fatiando a Terra have managed to produce high quality open-source software tools, and to create a community of users and contributors from different regions of the world. Its development is driven by the needs of the community members, motivated by their research and exploration goals. Fatiando has been used in multiple PhD thesis and scientific articles, including several ones authored by researchers that aren't involved in the project.
Most of the Fatiando developers and contributors are scientific researchers who regularly use it for their research. This presents two main advantages. First, the tools are built from the user perspective, leading to high quality software design and documentation. Secondly, new methodologies developed by these researchers are usually implemented into the same libraries that enabled the research. This creates a positive feedback between research and open-source software: scientific research benefits from high quality software, and the software grows after the advances of the research.
An example of this is the development of the gradient-boosted equivalent sources, a technique that enables interpolating very large datasets of potential fields data. The research was motivated by the preexisting equivalent sources in Harmonica, and after its publication, the gradient-boosted equivalent sources were made available for its use through the same library.
Open-source scientific software stands as a clear alternative to black-box solutions, allowing their users to study their code and build own solutions on top of them. This allows the creation of foundational tools both for science and industry. Users are welcomed to contribute with new code, documentation, examples, use cases, and new ideas for extending their capabilities. Fatiando a Terra is a great example of how collaborative and community-driven projects can create high quality open-source tools that the broader geoscientific community can leverage.