Hello,
Water going into the sand and back is not considered so far in Telemac. The reason of faster propagation on wet bed is that the celerity of the shallow water waves is added to the velocity of the flow. Then you should be aware that such flows, a front advancing on dry land, is far from the assumptions of the shallow waters. Another thing is that in such computations the advection scheme is very important. You can see in the Malpasset dam break test case the difference between the method of characteristics and the NERD scheme (case small, changing TYPE OF ADVECTION = 1;5 into TYPE OF ADVECTION = 14;5). The wave front goes faster.
With best regards,
Jean-Michel Hervouet