Spotted very interesting set of photos (1,2,3) from Villa Epequén in our flickr group. Reading about the background on the Wikipedia page made me even more curious: rare weather patterns causing a standing wave in a nearby lake that breaks a dam, drowning the flourishing tourist resort for 25 years.

The article didn't get into too much detail about the specific weather phenomena that causes the dam and dike to burst, although the page related to "Seiche" phenomena was an interesting read. However, based on a quick search it seems numerical modelling of the phenomenon is not necessarily a mainstream activity (although the first hit was on Deltares's OSS forum, so perhaps there are some recent developments...?).

PS. Thank you Nowhere land for sharing the photos!

Hi Matti,

Not really my specialty, but what I understand is that seiches may be triggered when advection speed of convective cells matches local speed of free surface waves on the water body. This triggers a resonance in the system.

You may be interested in the PhD research of my colleague Martijn de Jong.
His dissertation on the origin and prediction of seiches in Rotterdam harbour basins is available for download from:
repository.tudelft.nl/view/ir/uuid%3Ad7c...7-bc14-e01ce5e6856b/
An old ppt by him on this topic can be found on the website of the Dutch meteorological organization (e-mail address and phone number no longer correct):
www.knmi.nl/research/colloquia/abstracts...jong_20060209001.pdf

Similar to flash floods in mountain catchments, predicting the convective cells with sufficient resolution and reliability seems to be the main challenge in predicting these seiche events.

Best regards,

Bert

Hi Bert,

Thank you for the links, interesting read (or browse, to be honest)!

So if I understood correctly, at the moment it is possible (routinely) to predict conditions that might give raise to seiche waves, but the models tend to be specific to individual body of water?

As an aside, after living for more than a decade on the shores of Lake Geneva, it was somewhat amusing to finally find a name for some of the waves that seem to come out of nowhere!

Matti

Hi Matti,


Again with the disclaimer that it's not my core field of expertise, I think that in many cases these seiches could indeed be routinely predicted IF you have an accurate high-resolution wind model, and IF you have a sufficiently detailed hydrodynamic model (possibly including waves). The meteorological and hydrodynamic software (e.g. WRF and Delft3D) would be generic, but for the actual model instance you will need local information (outline of the water body, bathymetry/topography, hydraulic structures, ...). The local situation will have a strong influence on the exact atmospheric conditions that trigger the event.

Bert

Hi Bert,

Thanks again!

I guess the next question would be how feasible it is to map the topography under water? I'd imagine cheap drones are making sufficiently accurate 3D modelling of land areas (mountains and man-made structures) more and more affordable, but what is the situation if you wanted to map the topography underwater?

Matti

Hi Matti,

Indeed the amount of data on land is rapidly increasing due to the use of Lidar and related 3D Laser Scanning techniques. These data sets are now increasingly used for high resolution flood modeling (see for instance FloodVis and 3Di.

While we use light on land, we fall back to sound in water. High resolution bathymetry data is usually obtained using multibeam echosounders. Collecting such data for a large area still takes a lot of time, but in the nearshore it can be almost fun if you connect the device to a jetski.

Cheers, Bert