by Laura Risley, June 2024

After a 6-year hiatus, 2024 saw the return of the ‘Workshop on Meteorological Sensitivity Analysis and Data Assimilation’ also known as the Adjoint Workshop!

The workshop, first held in 1992, was the creation of Ronald Errico from National Center for Atmospheriic Research (NCAR). He envisioned a workshop with a unique style. Each Adjoint Workshop is held in a remote location where attendees stay, eat all three meals, and spend their free time mostly together. This encourages participants to get to know each other well. The talks at the workshop are longer than at most conferences. Each speaker has 25 minutes to present and 5 minutes for questions. This format allows presenters to take their time and go into detail about their work, with plenty of time for questions and discussions. Overall, the workshop facilitates more in-depth presentations, encourages discussion between participants, and allows people of all experience levels to expand their data assimilation (DA) networks. And this year was no exception!

Adjoint workshop 2024

People from all over the world gathered at the Fort William Henry Hotel in Lake George, New York. The workshop started on Sunday, 19th May, with a day of DA tutorials, mostly aimed at those early in their DA careers. This also gave everyone an opportunity to meet and chat before the workshop officially commenced.  From Monday through Friday there were talks on observation errors, DA theory and applications, the Joint Effort for Data assimilation Integration (JEDI) and Observing System Simulation Experiments (OSSEs).  Among the speakers were four members of DARC: Sarah Dance, Alison Fowler, Visweshwaran Ramesh, and myself. On the Tuesday afternoon we had a poster session, including OSSE jeopardy (see Fig. 1)! Wednesday evening saw a panel discussion about the future of data assimilation, where three panellists answered questions such as ‘how does machine learning fit into the future of DA?’.

Figure 1: Participants of OSSE Jeopardy – a game devised by Nikki Privé, where the contestants answer questions about OSSEs and win money (fake money).

Lake George was a stunning location, and we were very lucky with the weather – 30 degrees for most of the week, with only one thunderstorm. On the Wednesday afternoon we had free time, and to make the most of the weather and location, many of us hiked up Prospect Mountain (see Fig. 2). The view of the lake made it well worth the effort!

Figure 2: Adjoint Workshop attendees at the top of Prospect Mountain after a 2-hour hike.

The Adjoint workshop was an incredible opportunity to meet so many different people in the data assimilation field and to build connections. I am so grateful to have had the opportunity to attend, to share my research, and engage in interesting conversations about my research and that of others. I have made connections and friends that I hope to see again in the future. Thank you to Nikki Privé (Morgan State University / NASA) for organising such a fantastic workshop!

The next Adjoint workshop will hopefully take place in the coming years, somewhere beautiful in Europe! To read more about the workshop please see the website https://www.adjoint-workshop.org/home.

Talks by DARC members

Laura Risley  “On the choice of velocity variables for variational ocean data assimilation”

Sarah Dance: “Assessing the influence of observations in convection-permitting numerical weather prediction”

Alison Fowler: “The importance of anchor observations in data assimilation”

Visweshwaran Ramesh: “Improving Soil Moisture Estimates from the JULES Land Surface Model through 4D-EnVar Hybrid Assimilation of COSMOS-UK soil moisture Observations“