Lou Uccellini promises big US model upgrades

[dtk]

The idea of upgrading models is a good one, but how will the data be affected by the new ASOS policy and letting the UA network goto hell.. seems counter-productive to me...

To be honest, our observing system is quite impressive with many redundancies.  The UA network is not "going to hell", despite repeated attempts at various "radiosonde replacement programs and studies".  We are getting much better at using satellite (and satellite-derived) data, aircraft-based observations, GPS, and on and on and on.  Both NCEP and ECMWF have done interesting studies within the context of reanalysis that shows a majority of the improvements we have seen over the past few decades (at least in terms of large scale, medium range skill) has come from modeling and data assimilation, and not necessarily the observing system itself.

 

The ASOS point is an interesting one, and could have big implications in terms of convective scale and very short term modeling, as well as validation.  For the global models, most centers are still investigating if/how to use land temperature and wind surface data (NCEP GFS and ECMWF included)....as it is becoming more important as we increase horizontal and vertical resolution.

[dtk]

An interesting thread to read. I, for one, love having 4 cycles of the GFS/NAM. For dprog/dt within day 4, especially for major storm events, it is crucial to have those extra runs. With respect to data assimilation, dtk is spot on (as I would expect him to be) with respect to 4d-var not being the be all-end all that some think it is. The Canadian global is a great example of that. Statistics aside, from a usabilty standpoint, the GFS is substantially better than the Canadian. Finally, with respect the ECMWF, the differences are not as dramatically different than as some believe. I am routinely impressed by the overall performance of the GFS compared to the ECMWF with substantially less resources available to them (NCEP). 

To be fair....do I wish that NCEP had 4DVAR (and following hybrid variants thereof)?  Of course I (we) do.  That ship has sailed.  Instead of playing catch up, we are thinkging long term...taking into consideration what other centers are doing, the direction that supercomputing is going, and addressing customer needs.  We have made a conscious decision to invest in the hybrid EnVar algorithm (with 4D extensions), and in my humble opinion, this decision will pay huge dividends sooner rather than later.

[Amped]

One thing I would like to see on the GFS is a resolution upgrade.The NAM can be difficult to use for details, you usually have to lower the precip by about 30%

[dtk]

One thing I would like to see on the GFS is a resolution upgrade.The NAM can be difficult to use for details, you usually have to lower the precip by about 30%

We our working toward a significant horizontal resolution increase for the GFS (targetted ~1 year from now...as we wait for the new supercomputer to go operational, the backlog of software upgrades to get completed, and then finish our development and testing). 

[isohume]

Looks like the day-6 GFS verified the dry wedge, llvl theta/e pattern, and weak coastal trof...while the ECMWF was left shaking it's moist/cool/dynamic head. 

 

I thought this was impossible? 

[am19psu]

That's what DT has been saying for the last X years, so it had to be true, right?

[isohume]

That's what DT has been saying for the last X years, so it had to be true, right?

 

I'm afraid he brainwashed a whole legion of ECMWF drones, tho.