EMC's Model Evaluation Group

 The Model Evaluation Group (MEG) community is established to interact with the user community on issues related to the forecast systems of NCEP's Environmental Modeling Center (EMC).    It is used to discuss overall model performance and provide feedback on operational and parallel versions of the models which comprise the production suite.



Navigation Menu

What's New

angle-left RE: Better handling of high-level Hurricane moisture outflow?
Mark Klein, modified 2 Months ago.

RE: Better handling of high-level Hurricane moisture outflow?

Youngling Posts: 4 Join Date: 4/15/15
Hi Tim,

Thanks for posting this case!  It's good to see that the FV3 performance was better overall here, and given it was consistently drier than the GFS for several runs, it seems unlikely it's a one-off. Also, it's interesting in that our (WPC) forecasters might pay less attention to these light amounts (our focus being on excessive rainfall), but it shows we need to be cognizant of any QPF during fire weather season.


On Wed, Jul 11, 2018 at 6:25 PM, Timothy Barker <VLab.Notifications@noaa.gov> wrote:

Here is something completely different.  For Fire Weather in the Intermountain West, it is often very subtle model differences that are VERY important. 


When Fabio was decaying off the Pacific Coast last Friday (July 6th), there was a stream of moisture getting sucked up into the Pacific Northwest.  It is exceedingly dry up here at this time of year.  Moisture streaming off of Hurricanes like this is typically very "high level", and yields only mid and high clouds - with maybe a sprinkle at low levels.  Especially, since hurricane moisture is typically also accompanied by relatively warm air aloft - which keeps instability at bay.  But, sometimes, if there is enough existing moisture in the low levels, it can generate some precip.  That has HUGE implications for fire weather.  Most surface fuels, especially the grasslands at the lower elevations, are starting to "cure out", and are now, or will soon be, "ready to burn".  A little bit of moisture the generates some high-based thunderstorms with lightning, but no precipitation, would be very bad.  That is unlikely, but possible, in this scenario - if the warming aloft is not quite enough to offset the instability increase from increased moisture at mid levels.  But, at the same time - even if there is no convection, a rain event of 0.10" or 0.25" would be HUGE as well (I know, the rest of you, with your 2" rainfall events are laughing at us now...).  That would set back the curing time for light fuels like grasses - and would reduce the chance of fires for the next couple of weeks.


Here is a comparison of the 132-hour forecasts of 24-hr precip ending at 12Z July 7th from GFS/FV3.  Note how the GFS is generate a large area of 0.10" of precip across Eastern Oregon - with some small areas of even 0.25".  While, at the same time, the FV3 barely generates any area with 0.01" of precip:

Here is the Precipitable Water forecast at the same time - so that you can see the plume of moisture coming off of Fabio in the Pacific and into the Northwest:

An interesting thing to note here is that, at least in this crude look at Preciptable Water values, GFS/FV3 are pretty comparable with the AMOUNT of moisture available aloft - though the FV3 has the plume slightly further to the East.

This is often something we have seen in the past with the GFS, that it can generate too much threat of precip from high-level moisture, especially when low-levels start out relatively dry.  In this case, it appears to me that the FV3 is handling it better.


So, here are the same 24-hr precip forecasts for each of the comparable runs as the potential event got closer:


Now, in most of these, the GFS precip amounts were less than in that 132-hour forecast, which is good.  But, what stood out to me was that in virtually every case, the FV3 had less precip than the GFS (with the possible exception of the 60 hour forecast made from 00Z July 5).  In my mind, this is a really good "win" for the FV3.


I looked at surface dewpoint differences between the GFS/FV3 in this case (but I won't show them) - and the surprising thing was that the FV3 was actually slightly WETTER than the GFS in most runs.  But, to me, that shows how irrelevant the surface moisture can be in such as case.  The real story is the handling of the dry layers between, say, about 850-500mb - and any sub-cloud evaporation of precip.  Is there a difference in sub-cloud evaporation between the models?  If so - it seems to have worked out well in this case.


As far as verification, there was virtually no precipitation at the surface in this case.  Here is the RFC analysis at 12Z on the 7th:


And, just to show that the models were correct that the plume of upper-level moisture was there, here is a satellite loop from Friday the 6th:


Maybe it is silly, but this is a good case of the FV3 performing better in a "subtle", "marginal", and "seemingly unimportant" case.


Tim Barker - SOO - Boise, ID

Timothy Barker EMC Model Evaluation Group Virtual Lab Forum https://vlab.ncep.noaa.gov/web/emc-model-evaluation-group/home/-/message_boards/view_message/4417973VLab.Notifications@noaa.gov

Mark Klein
Science and Operations Officer
Weather Prediction Center
(301) 683-1490