Tip:
Highlight text to annotate it
X
The storm prediction problem is a problem that really focuses on a few hours in duration.
Storms modify themselves quickly; complexes of storms do the same thing. Modeling is changing
in a few areas; in the need for hourly updating, the need for radar assimilation and also the
need for high performance computing, so that we can actually resolve thunderstorms. A thunderstorm
might be around ten kilometers wide, maybe five to ten kilometers, could be down as small
as a couple of kilometers, and then to have some resolution of updrafts and downdrafts
inside that storm you really need to get down to a maximum of three kilometers, and preferably
higher resolution than that. For the Tuscaloosa event of twenty-seven April, we can contrast
the depiction of convection at ten to fifteen kilometer model, with the current RUC model,
which is run by NCEP, in which we can see these broad areas of some kind of convective
storms taking place, versus the three kilometer HRRR Model in which we can see not only are
there storms in a more accurate location in Tuscaloosa County but we can even see that
they are rotating supercells and we can actually see that in the three kilometer HRRR Model
for this particular case. We could see that with the HRRR, but we could not see that with
the thirteen kilometer RUC. I think there is a great need for severe weather and other
related applications with aviation and energy and other aspects to greatly improve a lot
further over what the HRRR can do right now. The HRRR’s been a big step ahead that we
didn’t really see taking place about six years ago, before we could actually put in
radar information into a model effectively. We didn’t see that coming; nor did we see
that the computers would advance so quickly that we could actually start to address this
problem from a computer point of view. We have a lot of shortcomings in that still right
now; we’ve taken a big step forward with the HRRR but we really have a lot of errors
in these short range forecast models and there’s going to be a lot of work to do in NOAA over
the next eight years.