Jump to content
  • Member Statistics

    17,508
    Total Members
    7,904
    Most Online
    joxey
    Newest Member
    joxey
    Joined

June 1st-6th Severe Events


Quincy

Recommended Posts

  • Replies 188
  • Created
  • Last Reply

HRRR and other hi-res models struggle with subtle set-ups. Hey it's job security for operational mets! Seriously those models are optimized for traditional or classic severe weather sequences. If sensitivity were turned up for subtle events, every traditional outbreak would look like April 27, 2011. Fortunately those are only once a generation in real life. It is not really sensitivity, but the equations and physics required to do well with traditional sequences. Usually it is a trade-off in NWP, unless we have a whole other model for subtle events. With scarce resources, the community favors models that do well in traditional sequences. I'm not complaining about budgets; I mean the scarce resources from Econ 101. The most lives are saved and commerce protected with models that do well in traditional sequences or classic outbreaks.

 

Even the HRRR did not totally blow it in Kansas IMHO. Despite off the charts TOR parameters, it never showed spectacular mesos in Kansas or Nebraska. 2-5km updraft helicity (weather.cod.edu) was not spectacular. You could get at that on the hi-res NAM indirectly with 925 mb, and the NAM showed uninspiring storm inflow (and some bad outflow NE). While both had gorgeous reflectivity simulations, neither had impressive mesos. They actually did OK in Kansas. On the other hand both missed the incredible tornado show in Colorado. Apparently some human forecasters decided 2000 CAPE, vs 1000 day prior, would do the trick.

 

Congratulations to all who documented the Colorado storm! Appears it will go down as a Rozel 13 or Haper 04 type career event for you all. Event had 4+ cycles, all photogenic, including a double tornado. Tough to beat that. James Smart shot may become a classic anti-cyclonic tornado photo. Nice work to all who got those!

Link to comment
Share on other sites

What a show in eastern Colorado for sure yesterday from the tornadoes to familiar images of deep, deep hail.

Can we do it again today? A similar setup unfolding and winds in the mid and upper levels look at least slightly stronger too. Seeing some large looping hodographs via NAM for east-central to northeast Colorado this afternoon.

Link to comment
Share on other sites

Yesterday will be painful for a long time. Of course we chased CO/WY/NM on all the days that ended up lame.

We skipped the far NW KS cells because they looked elevated. Second was a beauty tho. Got the best lightning pics I've ever taken so it wasn't a total loss.

Link to comment
Share on other sites

Yesterday will be painful for a long time. Of course we chased CO/WY/NM on all the days that ended up lame.

 

That's the worst thing about yesterday... I probably shouldn't even complain since I at least hadn't chased since the Canadian day last week. I saw LOTS who were all over CO/WY for 3-4 days prior and then bailed to KS yesterday, though. Again, brutal. I haven't seen any evidence that anyone out there can forecast these CO events with real skill; e.g., seeing that yesterday would be way better than the day before (Wed). Chasing out there is basically the lotto.

Link to comment
Share on other sites

That's the worst thing about yesterday... I probably shouldn't even complain since I at least hadn't chased since the Canadian day last week. I saw LOTS who were all over CO/WY for 3-4 days prior and then bailed to KS yesterday, though. Again, brutal. I haven't seen any evidence that anyone out there can forecast these CO events with real skill; e.g., seeing that yesterday would be way better than the day before (Wed). Chasing out there is basically the lotto.

Even right before it started going to town I saw plenty of people poo pooing the day. "Weak, splitting, etc". Know the cell north of Dencer also produced but that prolific cell dove almost straight south while the others meandered. Not sure anyone honestly saw that coming.. Mostly luck. This whole chasecation we really didn't really see any cells turn into the good sfc flow. We did chat about CO the night before but being burned the day prior made it hard to ignore the huge parameters setting up over KS/NE even with cap bust potential lurking.
Link to comment
Share on other sites

HRRR and other hi-res models struggle with subtle set-ups. Hey it's job security for operational mets! Seriously those models are optimized for traditional or classic severe weather sequences. If sensitivity were turned up for subtle events, every traditional outbreak would look like April 27, 2011. Fortunately those are only once a generation in real life. It is not really sensitivity, but the equations and physics required to do well with traditional sequences. Usually it is a trade-off in NWP, unless we have a whole other model for subtle events. With scarce resources, the community favors models that do well in traditional sequences. I'm not complaining about budgets; I mean the scarce resources from Econ 101. The most lives are saved and commerce protected with models that do well in traditional sequences or classic outbreaks.

 

Even the HRRR did not totally blow it in Kansas IMHO. Despite off the charts TOR parameters, it never showed spectacular mesos in Kansas or Nebraska. 2-5km updraft helicity (weather.cod.edu) was not spectacular. You could get at that on the hi-res NAM indirectly with 925 mb, and the NAM showed uninspiring storm inflow (and some bad outflow NE). While both had gorgeous reflectivity simulations, neither had impressive mesos. They actually did OK in Kansas. On the other hand both missed the incredible tornado show in Colorado. Apparently some human forecasters decided 2000 CAPE, vs 1000 day prior, would do the trick.

 

Congratulations to all who documented the Colorado storm! Appears it will go down as a Rozel 13 or Haper 04 type career event for you all. Event had 4+ cycles, all photogenic, including a double tornado. Tough to beat that. James Smart shot may become a classic anti-cyclonic tornado photo. Nice work to all who got those!

 

It's not that the models are "tuned" incorrectly for these marginal events.  The same dynamics and physics that govern marginal events govern outbreak events.  Rather, the events are frequently sensitive to the the upscale influence of prior convection.  For instance, the location of cold pools and their boundaries - meso-delta scale atmospheric circulations that contribute to convective initiation.  These area all processes that are poorly emulated by, and initialized in models.

 

Outbreak events, on the other hand, are more synoptically driven, and models are better at getting a good handle on the synoptic scale environment. 

 

The point being - there isn't an easy way to just "tune" the models to better understand these upscale convective feedbacks and nuances of convective initiation.  These processes are inherently chaotic.

Link to comment
Share on other sites

In terms of NAM/HRRR, a bias to over-convect seems apparent in most of these setups. Yesterday, they pushed surface temps well into the 90s across much of Kansas. As a result, the cap was erroneously breached several hours early. Just one of many factors.

Link to comment
Share on other sites

Homes damaged in Berthoud, Longmont possibly

---

BERTHOUD, Colo. — Residents in northern Colorado were waking up Friday to assess damage after two tornadoes, strong thunderstorms and hail ripped through the area Thursday night. At least 25 homes between Longmont and Berthoud were damaged, with three of them destroyed, after the tornadoes touched down about 7 p.m. near the 2500 block of Blue Mountain Avenue. No injuries have been reported as the winds tore apart homes and rolled vehicles.

---

http://kdvr.com/2015/06/05/tornadoes-thunderstorms-hail-leave-damage-along-front-range/

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...