ECCO Hackweek 2024

Drifters

We learned the basics of git and Julia, mapped particle tracking experiments. We plotted Lagrangian particle tracking starting from 6° S across the Atlantic for 250 days.

Exploring Sensitivity Patterns and RESponses in the Southern Ocean (ESPRESSO)

Our goal was to replicate previous results using the adjoint from EMU and ensure we can attribute variability in the Drake Passage transport to recognizable patterns of primarily zonal wind stress.

Gulf of Mexico (GOM)

Each summer in the Gulf of Mexico, a hypoxic zone develops. It is imperative to gain a mechanistic understanding of the drivers of interannual variability. ECCO-Darwin provides a wonderful opportunity to study these mechanisms.

Processors-LLC

Our goal was to produce a new Python package to preprocess profiles and obsfit data. We also created Jupyter notebooks demonstrating different uses of the package, examining different routines within the package.

SPGecco

We focused on the Subpolar Gyre in the North Atlantic and ECCO, specifically the heat content. We put our effort into understanding the heat content in the subpolar gyre and its influences using the EMU tool that we have learned.

Spinners

We looked at closing the global vorticity budget using ECCO.

ToyECCO

Automatic Differentiation (AD) tools are hard to compare across different languages and algorithms because they all require different implementation. We worked on some basic examples with comparisons across languages (Python, Fortran, Julia) and open-source AD tools (Enzyme, PyTorch, JAX, Tapenade).

ECCO Representation of Water Masses

Understanding the consequences of Antarctic sea ice change, especially amidst recent extremes, is important for understanding global ocean circulation. We evaluate if ECCO accurately represents water masses and their volumes correctly. Specifically, we delve into Antarctic Bottom Water and its volume over time.