NaaVRE

NaaVRE, short for Notebook as a Virtual Research Environment, is a platform for turning code written in Jupyter notebooks into workflows that a team can share and run in the cloud.¹ A Jupyter notebook is a page in the browser that mixes code, text and results. In NaaVRE a researcher writes Python or R in a notebook, turns cells into containerised workflow components, chains them into a workflow, and runs it on a workflow engine, now or on a schedule. Notebooks, components, workflows and results can be shared with a team or a wider community.^1,2 Work is organised in virtual labs, shared workspaces for one purpose or community, each with its own software libraries, interfaces and workflow building blocks.¹ LifeWatch ERIC offers it as a hosted platform, and its development is partly supported by the University of Amsterdam and LifeWatch ERIC.¹

It was described in 2022 by Zhiming Zhao and ten co-authors from the University of Amsterdam and LifeWatch ERIC.^6,7 The paper starts from a gap: notebooks are good for trying out scientific code, but they do not help run heavy computations on remote machines or find code others have written, so the authors embed a virtual research environment inside Jupyter.⁶ Their demonstration took an existing workflow that derives ecosystem structure from airborne laser scanning (LiDAR) point clouds and scaled it out in the cloud, so it could handle multi-terabyte country-wide surveys.⁶ The maps above, of ecosystem height and structural complexity across the Netherlands from the third national survey (AHN3), come from that workflow.⁷

To use it, open the platform at beta.naavre.net and sign in with Google, through the EOSC federation (which offers GitHub, ORCID and other logins), or with a registered account. Then pick a virtual lab, such as the NaaVRE Open Lab for the tutorial, and launch your own instance, which opens in JupyterLab.³ Groups can propose a virtual lab of their own.¹ Behind it, the system is a set of microservices with JupyterLab extensions on top, deployed on Kubernetes.⁴ The code is on GitHub, where most component repositories are Apache-2.0 and several had commits in October 2026.⁵

LTER-LIFE, the Dutch infrastructure for digital twins of ecosystems, lists NaaVRE among its services and is developing it further for building twins.⁸ Its GitHub holds NaaVRE virtual labs for a forest landscape model of the Veluwe and for its NatureDataCube data service.⁹ Digital twins as a tool for ecosystem research cites the NaaVRE paper when it describes virtual labs as a place to assemble an ecosystem's data and models into a twin.¹⁰

For a group modelling a place, NaaVRE is a way to move analyses kept in individual notebooks into shared workflows. Because they can run on a schedule, a model can be refreshed as new data arrive without someone rerunning it by hand.