Introduction

Although the browzarr.io user interface is pretty intuitive, there are plenty of features that deserve some dedicated space for a longer explanation. But first,

What is Browzarr?

Browzarr is a powerful, browser-native framework for visualizing, exploring, and analyzing Zarr and NetCDF data stores. Load multi-dimensional datasets directly in the browser, interactively slice and inspect chunks, and gain insights without any backend or server setup.

Because everything runs client-side, your data flows directly from the store (local disk or S3/HTTP bucket) to your browser — there is no intermediate server that processes, stores, or re-serves your data.

Key Features

• Data cubes — explore spatio-temporal (x-y-t) variables as interactive 3D cubes
• Globe — render data on a 3D sphere for spatial context
• Flat maps — classic rectangular 2D projections
• Point clouds — high-density spatial data as interactive 3D points
• Time series — select any pixel in any direction and plot its values over time
• Colorbar control — interactive bounds, colormaps, univariate and bivariate mapping
• Analytics — WebGPU-accelerated computation, including custom WGSL shaders
• Export — static images and animation/video export
• No install required — runs online at browzarr.io, or fully offline via npm or Julia

Who is it for?

• Earth system scientists and climate researchers who want to eyeball a data cube before writing analysis code
• Data engineers validating Zarr stores, chunking layouts, and Icechunk repositories
• Educators and students who need an approachable, zero-setup way to explore geospatial datasets

How these docs are organized

• Getting Started — installation and a first tour of the interface
• Essentials — the core concepts: datasets, variables, colorbar, plot settings, analytics
• Guides — task-oriented recipes (remote data, local data, animations, bivariate maps…)
• Reference — plot types, supported formats, browser support, troubleshooting, FAQ
• Community — contributing and project background