Lexcube: A multi-platform "data cube" ecosystem for immersive exploration of Earth system datasets in education, outreach and research

Many widely used tools for communicating and teaching Earth observation and modeled climate data still struggle to convey spatiotemporal phenomena: visualization is often limited to 2D map views, interfaces can prove difficult for non-experts, and workflows might not be easily transferred from curated examples to large or in-progress research datasets. This creates a gap between public-facing visualization, classroom use and research workflows.

We present Lexcube, a multi-platform ecosystem for interactive exploration and visualization of Earth system "data cubes", i.e., large remote sensing and modelled data sets. Lexcube provides an immersive, interactive 3D "data cube" view where all dimensions (space and time) are treated equally, enabling users to easily reveal spatiotemporal dynamics that are not visible in 2D map-based interfaces. Over its development in the last years, Lexcube has been used in education, outreach, and research. Our goal was to emphasize intuitive navigation and low barriers of entry while being capable of visualizing large data sets with minimal hassles. Lexcube has been deployed in multiple forms: 

• (1) Lexcube.org, an interactive data cube exploration and visualization web app, with no coding or infrastructure required. It runs on desktop and mobile devices with minimal hardware requirements, and has been regularly used in teaching.
• (2) Lexcube for Jupyter, an open-source Python package aimed at scientists, that allows to visualize any 3D data set as an interactive data cube in Jupyter notebooks.
• (3) Two museum exhibits, featuring simplified versions of the Lexcube.org interface with curated data sets and explainer texts relevant to its respective exhibition.
• (4) A physical interactive data cube, a large museum-style installation that displays data cubes in physical space through five square touch screens assembled in the shape of a cube, offering the same capability and data sets as Lexcube.org, but proving even more accessible as no virtual 3D environment or software has to be navigated at all.
• (5) The option to create physical paper data cubes from templates generated by Lexcube, assembled by cutting and glueing, offering a low-cost and engaging piece of science communication.

In the future, we are looking to strengthen the education and science communication use cases for the Lexcube platform and are very interested for feedback and ideas for possible future developments. These could include a virtual reality deployment to particularly explore extreme events as 3D voxel clouds over space and time as well as offering simple data processing operators beyond pure data visualization.