The answer can be asked directly to the author (mihaela.caian++at++meteo.inmh.ro ) or can be found in her report (in French) or in the article "Some limits to the variable mesh solution and comparison with the nested LAM one", published in the Quarterly Journal of Royal Meteorological Society, n°539, vol.123, pp.743-766, April 1997.
From a practical point of vue, the main result of this thesis is that the best strategy for dynamical local adaptation is a coupling between a moderately stretched global model and a higher resolution LAM without too big a jump between both mesh sizes.