NSW Climate Data Portal
Took the NSW Climate Data Portal from beta to public production, serving NARCliM2.0 regional climate projections.
The problem
The NSW Climate Data Portal needed to move from beta to production so NARCliM2.0 regional climate projections could be released publicly on an ongoing basis.
That meant migrating the Portal's backend services onto new Kubernetes clusters inside a DIO-managed AWS environment, while meeting the security and governance requirements of multiple government departments.
What I built
Working within the Intersect engineering team alongside DCCEEW Climate and Atmospheric Science, Science Data Compute (SDC) and DPHI Digital Information Office staff, I reviewed, adjusted and migrated the Portal's backend services — including the THREDDS Data Server and the SDC API — onto the new clusters.
I validated the architecture design throughout the migration, raising critical changes and bugs with stakeholders before implementing technical adjustments rather than after the fact.
I maintained a stable SDC API specification and a consistent codebase so dataset subsetting and upload workflows stayed reliable and performant across all NARCliM data requests, and tested and refined the Kubernetes deployment templates against the newly established SDC clusters.
Outcome
The Portal was publicly released in September 2025 and now serves climate projection data to researchers, government and the public across NSW and Australia.
Stack
Kubernetes · Python · AWS · THREDDS Data Server · REST API