On a quiet stretch of shore, you can usually tell where the beach is headed just by watching—sand shifting, dunes thinning, waterline creeping after storms. Now imagine that process happening faster, under rising seas and more intense weather. Coastal landscapes are constantly being reshaped by natural forces, and as climate change causes more frequent storms and sea level rise, that change will only intensify. Because these areas are densely populated with homes, tourist destinations, and industries, understanding how and where the coast will change is a pressing issue. Unfortunately, reliable predictions that lead to actionable knowledge are rare.
Misryoum newsroom reported that Lentz and colleagues take a hard look at what scientists do and don’t yet know about coastal evolution. The big picture is not that researchers lack models—it’s that the field’s pieces don’t always fit together cleanly. In particular, current coastal evolution predictions are often focused on too specific a location, which makes them hard to generalize. They also tend to lack detail when scientists try to analyze broader regions. And there’s another stubborn problem: linking the effects of acute events—storms—with long-term trends like sea level rise isn’t straightforward.
Improving simulations, Misryoum editorial desk noted, likely requires combining many different types of models. That means mixing physics-based numerical models, models grounded in empirical measurements, and statistical models that incorporate machine learning. It’s not just a technical tweak; the researchers argue that to fully understand potential changes, you also have to account for both coastal processes and human actions. That last part matters more than many people expect—engineering, land use, and coastal management can steer the coastline as much as waves and currents.
The researchers also argue that better forecasts depend on consistency and collaboration. Right now, the variety of tools used across different locations makes it difficult for scientists to compare results and communicate effectively. Standardizing approaches and outcomes would make it easier to produce national-scale predictions, they say. They also emphasize coordinated research approaches—stronger transdisciplinary collaboration, paired with essential training and support, would help scientists make better predictions. There’s a bit of a catch here: if teams don’t share methods or even align on what counts as a good outcome, progress can feel like parallel tracks that never quite meet.
Another route, Misryoum analysis indicates, is to compare predictions directly with real-world observations of coastal landscape change. That kind of validation could help untangle a complex challenge by revealing which models are actually performing best. But doing it properly requires datasets that adequately capture coastal landscape change across both time and space. That’s where remote sensing can help, and where the researchers suggest AI for data processing may be especially useful. You can almost picture it—someone staring at coastline change maps, then realizing the “signal” only looks clear after the right processing pipeline, not before.
There’s also a social layer to the science. Engaging end users during project planning can help, because only end users truly know what kind of information they need to adapt to landscape change. Misryoum newsroom reported that knowing how to engage end users can be difficult for physical scientists, but the researchers point out that various tools and specialized personnel exist to coordinate these interactions. It sounds simple, and yet it probably isn’t—getting the right people in the room early, and keeping them there as models evolve, is the sort of work that rarely fits neatly into a grant cycle. Still, if the goal is actionable knowledge, skipping that step would be… well, probably not the best idea.
DESI finishes biggest ever cosmic map of galaxies and quasars