2. ec2021_balwada_etal¶
(For accessing a jupyter lab environment to interact with the notebook)
(For accessing the dashboard directly)
Interactive visualization tools for ocean glider data¶
Dhruv Balwada, Scott Henderson, & Alison Gray¶
University of Washington¶
Ocean gliders are observational platforms that sample the ocean at fine spatial and temporal resolutions. The resulting datasets contain measurements of physical variables, such as temperature, salinity, velocity, as well as biogeochemical variables, such as oxygen, nutrients, chlorophyll, at scales where the variability is generated by mesoscale and submesoscale flows (1-100 km scales). The measured variables often vary over orders of magnitude as a function of scale and are correlated with each other. Thus, insights can be gained rapidly if a scientist has the ability to interact with the data, and is able to adjust the zoom level, vary colormaps, or co-plot and rapidly switch between different variables. Traditionally achieving these interactive capabilities has been a challenging software development task, which deters most time-crunched scientists. In some cases when the interactive visualization platforms are built, they have been so data- and task-specific that they can not easily be ported to new datasets or additional analysis capabilities can not be added as required.
Here we demonstrate how Python’s Holoviz visualization ecosystem (https://holoviz.org/) can be coupled with the glider data analysis toolbox GliderTools (https://glidertools.readthedocs.io) to rapidly generate interactive dashboards to explore glider datasets. The tools and code presentation in the jupyter notebook have been developed with the purpose of ease of use and portability, and they can easily be modified to new datasets and analysis/visualization needs.