Data assimilation combines recent model forecasts and observations to estimate current atmospheric conditions for use as initial conditions for numerical weather prediction. We analyse the results of a series of data assimilation experiments using a simplified and inexpensive mathematical model of the atmosphere. We closely compare key properties of the models used by weather centres with our idealised setup, proving that it can help support operational data assimilation research.

Wetropolis is a table-top demonstration model with extreme rainfall and flooding, including random rainfall, river flow, flood plains, an upland reservoir, a porous moor, and a city which can flood. It lets the viewer experience extreme rainfall and flood events in a physical model on reduced spatial and temporal scales with an event return period of 6.06 min rather than, say, 200 years. We disseminate its mathematical design and how it has been shown most prominently to over 500 flood victims.

Extreme water waves impacting ships and offshore structures can not only cause severe structural damage, but also threaten the safety of passengers and crew. Accordingly, the motivation for the present work is to better understand the dynamics of extreme waves in two cases: the case of "green water" and the case of "freak waves". Our methodology can simulate those two events in order to estimate the forces of such extreme waves and thus aid engineers in the design of safer maritime structures.