Climate models are complex codes that simulate Earth's climate system (atmosphere, ocean, land, ice). These models demand immense computing power and present significant software engineering challenges, including managing massive datasets, ensuring numerical stability, optimizing performance, and coupling diverse components. The complexity and legacy nature of existing models hinder their ability to fully utilize modern computing infrastructure (GPUs, machine learning, etc.). The Climate Modeling Alliance is developing CliMA, a new climate model built in Julia and designed from the outset to leverage GPU acceleration and modern software engineering practices to overcome the limitations of traditional climate models.
ClimaExplorer: An Interactive Visualizer of Climate Model OutputsVisualizing simulation output is crucial for both scientific understanding and outreach. This project involves developing ClimaExplorer, an interactive visualizer for the output of the CliMA Earth system model. ClimaExplorer will leverage the Makie ecosystem and web technologies, providing a user-friendly interface for exploring complex climate data. This will enable researchers to more easily analyze and interpret simulation results, accelerating scientific discovery. Furthermore, the web-based component will facilitate broader dissemination of results to a wider audience.
Desired Skills: Familiarity with front-end web development (HTML, JavaScript, and CSS), Julia, and Makie.
Difficulty: Medium
Duration 175 hours
Expected Results: ClimaExplorer, a new module for interactive visualization of simulation output (with tests and documentation).
Mentor: Gabriele Bozzola
Contact: Feel free to ask questions via email or Julia Slack (DM to Gabriele Bozzola).
Interested in other aspects of climate modeling in Julia but not this particular project? Get in touch—we have many more projects!