The effort of this group appears quite extensive. They performed experiments using 1x and 10x vertical diffusion and experimented with setting the boundary conditons of the different nests as a function of its mother (default) or the GFDL GCM that the group uses. A series of plots was submitted by Anthony, and I present these in list form for completeness.
The following is an image that I generated to specifically compare the 1x vertical diffusion case for the 27 km nest output against the VL2 data. The black lines are the basline case (GCM drives the mother domain). The blue line is for a test case where the 27 km boundary conditions are taken directly from the GCM. The red line shows the output when the the mother domain of the 27 km nest (81 km domain) was driven by the GCM and boundary conditons for the 27 km nest are taken from the 81 km nest. As GCM boundary conditons get "closer" to the 27 km nest the pressure cycle in that nest tends to follow the GCM much closer. We believe that this will spark some interesting discussion!