Computers & Geosciences

Computers & Geosciences publishes original, high-impact research at the interface between computing/informatics and the geosciences. Submissions must make a substantive contribution in both dimensions: (1) an innovative computational, informatics, or software element, and (2) a clear geoscientific contribution addressing a scientific problem of broad interest to the geoscience community.

Manuscripts are suitable for the journal only when the computational/informatics component is central to the work and advances the state of the art beyond routine implementation, straightforward application, or incremental adaptation of existing methods. It is expected that code is provided for all scientific research articles.

Relevant computational/informatics contributions include, but are not limited to: novel algorithms; computational methods; scientific software; data models; scientific databases and retrieval systems; machine learning and artificial intelligence; inversion and data assimilation frameworks; uncertainty quantification workflows; visualization and human-computer interaction for geoscientific analysis; high-performance, parallel, cloud, or distributed computing; reproducible research infrastructures; and interoperable digital frameworks for geoscientific data and modelling.

Relevant geoscientific domains include: geology; geophysics; geochemistry; geomorphology; sedimentology; stratigraphy; palaeontology; tectonics and structural geology; seismology; volcanology; hydrology; hydrogeology; and Earth-system science. Manuscripts centered primarily on generic environmental engineering, routine GIS applications, signal processing, agriculture, ecology, or other topics outside the core geosciences are normally outside the journal’s scope unless the geoscientific contribution is explicit, substantial, and of clear interest to the readership.

Code, software, data, and reproducibility requirements

For any manuscript presenting software, code, workflows, trained models, computational pipelines, or implementation details essential to the results, public access to the corresponding repository is mandatory at submission and required for acceptance. A manuscript will not be accepted for review without a GitHub repository or equivalent public repository that is properly documented and enables evaluation, reuse, and long-term utility to the community.

there needs to be a link to the repository in the manuscript and at a minimum, the repository must include:

• a clear license;

• a ReadMe file in English with basic usage and installation instructions;

• documentation of dependencies and computational requirements;

• sufficient material to reproduce the main results in the paper, or a clear explanation of any justified limitations if data cannot be shared, a dummy model or a synthetic dataset for test cases should be provided

• how-to files or tutorials for typical use cases;

• a user guide describing inputs, outputs, options, and expected behaviour;

• and for security reasons, a single compacted file is not accepted (e.g. .zip, .rar, .7z)

• any comments in the code must be in English

Repositories that are incomplete, poorly documented, inaccessible, private at the review or acceptance stage, or lacking reproducible examples are grounds for rejection. Manuscripts describing software that is not open source are normally desk rejected.

Computers & Geosciences will not consider manuscripts that fall into any of the following categories:

• geoscience studies with no genuine computational or informatics innovation;

• computer science or informatics papers with no clear and substantive geoscientific contribution;

• routine application of established methods, software, machine learning models, GIS workflows, or visualization tools to a case study;

• incremental code implementations of already published methods without a clear methodological or software advance;

• manuscripts focused mainly on engineering design, operational management, consultancy-style case studies, or site-specific technical problem solving rather than geoscientific insight of broader relevance;

• papers in which the computational element is peripheral, cosmetic, or limited to software packaging or interface development;

• purely methodological developments (e.g. geophysics, hydrology)

• purely analytical developments unless they have significant implications on computational geoscientific problems

• pure case studies and application examples less they have significant implications on computational geoscientific problems

• GUI papers unless the interface itself solves a significant, non-trivial scientific-computing problem;

• benchmark or comparison papers lacking novelty, reproducibility, or transferable insight;

• papers whose results cannot be independently evaluated because code, workflow details, trained models, key parameters, or supporting data are unavailable; and

• manuscripts generated around black-box tools or proprietary workflows with insufficient transparency for scientific scrutiny.

Submissions should make clear, from the title, abstract, and introduction onward, what is new computationally, what is new geoscientifically, and why both matter.

Manuscripts better aligned with applied implementation, software deployment, or lower-novelty contributions may be more suitable for the companion journal Applied Computing & Geosciences.