Building Bridges for Indian Ocean Rim Marine Scientists Across the "Big Data Geoscience" and Cloud-computing Divide

The food security of millions depends on coastal, artisanal fisheries in the Arabian Sea and Bay of Bengal, spanning East Africa, West and East India, Bangladesh, and Myanmar, and ocean climate change is causing a myriad of impacts that affect fisheries. In a pattern seen elsewhere, climate change disproportionately affects vulnerable communities in the global south, and the impacts are exacerbated by a disparity in available resources. This disparity extends to Indian Ocean Rim scientists, who face limited access to big data geoscience resources and cloud-based computing platforms, a crucial area of advancement in earth sciences and ocean climate change research in the era of high-data-volume Earth observations. Training on and access to these new cloud-based approaches has been overwhelmingly in High Income countries, leaving the next generation of ocean and marine scientists in Low and Middle Income countries in danger of falling behind precisely when big data geoscience is needed to help countries adapt to ocean climate change. To help bridge this gap, we conducted a pilot program using a 2i2c JupyterHub provisioned for cloud-based big data geospatial research in marine and ocean science. Seventy early career scientists from Indian Ocean Rim countries (East Africa, India, Bangladesh, and Myanmar) were trained in using big data geoscience tools during a 2-week course and hackweek on machine-learning for marine species distribution modeling at the International Training Centre for Operational Oceanography (ITCOocean) in Hyderabad, India, and given access to the hub to continue their research after the course. This eliminated the need for participants to install tools on their laptops, which often have diverse and older operating systems that are not necessarily supported by the latest software packages. A JupyterHub alleviates this issue, as it requires only a browser and modest internet connectivity. However, concerns remain about the practicality of cloud platforms in regions with intermittent internet connectivity, and whether the scientists will be able to continue to use cloud-platforms effectively when they return to their home institutions. This talk will outline the insights gained from the ITCOocean Hackweek and evaluate the viability of JupyterHubs for this research community.