Processing a new generation of hyperspectral data on the Cloud using Pangeo

Science Center Objects

We aim to migrate our research workflow from a closed system to an open framework, increasing flexibility and transparency in our science and accessibility of our data. Our hyperspectral data of agricultural crops are crucial for training/ validating machine learning algorithms to study food security, land use, etc. Generating such data is resource-intensive and requires expertise, proprieta...

We aim to migrate our research workflow from a closed system to an open framework, increasing flexibility and transparency in our science and accessibility of our data.   Our hyperspectral data of agricultural crops are crucial for training/ validating machine learning algorithms to study food security, land use, etc.  Generating such data is resource-intensive and requires expertise, proprietary software, and specific hardware.  We will use CHS resources on their Pangeo JupyterHub to recast our data and workflows to a cloud agnostic open-source framework.  Lessons learned will be shared at workshops, in reports, and on our website so others can increase the openness and accessibility of their data and workflows. This project explores the capabilities of Pangeo to support workflows, training others in best practices for open science. It also makes high profile data more accessible, which can be used by others to increase knowledge about the Earth and its processes.