Model Intercomparison Project: Shoreshop3
Brad Murray
To all who research and model shoreline-change-relevant dynamics:
We hope you will participate in the Shoreshop3 Model Intercomparison Project and workshop (Oct. 20-22, 2026; https://shoreshop3.netlify.app/)!
Shoreshop3 builds on the previous two workshops ( Montaño et al., 2020; Mao et al., 2025) in three ways:
First, we will focus on the contrasting and complex morphodynamic setting of the North Carolina coastline (with part of the effort centered on the vicinity of Duck NC).
Second, we will investigate shoreline and coastline change over expanded space and time scales: from local to 100s of km, and from daily to century hindcasts and forecasts.
Third, treating models as representations of processes, we will focus on increasing understanding of which processes play the most important roles in shoreline change, and how the answers depend on the scales of interest. For example, how do the processes (i.e. models) that are important over daily timescales compare to those that are important over decades and centuries—and at what scales do the process dominances (i.e. model effectiveness) overlap and transition? We will also evaluate how modeled morphodynamic behaviors, and model successes, depend on the different ways of representing processes (e.g. hydrodynamics based and detailed vs more synthesized). We are hoping that participants will bring a wide variety of model types and approaches into this project, including Machine-Learning models.
Consistent with these process-exploration emphases (and a ‘Blind’ evaluation period including the year following the workshop as well as the year before), the metrics for comparing model results with each other and with observations will be primarily statistical. We will evaluate how much of the variance in the shoreline-change signal can be explained by each model (i.e. process sets), across a range of time scales (e.g. as measured by wavelet power spectra for local models or how those spectra vary alongshore for alongshore-extended models). So, please calibrate models to best match the distribution of variance (either in time or in time and space).
The MIP will involve two broad objectives—arenas of inquiry—with different forcing and calibration data sets, and different geographical and temporal ranges (see https://shoreshop3.netlify.app/about). Forcing and shoreline-change data for model calibration, testing, and hindcasting for relatively short timescales (1982 - 2024) will be available in early January 2026, with data for longer timescales (century hindcast and forecast) following by March. Model results will be due in August, and we will think together about the insights model intercomparison can produce in October.
We will be able to support at least some travel expenses for the Workshop (including for PhD students), with details depending on the level of interest.
Please join our community effort, with whatever shoreline-change-relevant modeling approach you would like!
If you might be interested, you can subscribe to the Shoreshop3 email list by emailing sy...@duke.edu with the subject “sub shoreshop3” (no quotes).
Looking forward to interacting with you,
Brad Murray, Fernando Mendez, Mark Piper, Jennifer Montano, Dylan McDougall, Aidan Dealy, and the Advisory council of Giovanni Coco, Dano Roelvink, and Kristen Splinter.