Ended operations for today  for us and V2 crew with storms weakening and moving out of WY. (Note: During debrief below, storm reintensified and produced the biggest hail of the evening – such was the night). Also exhausted the PAR archive list as well as a couple more CASA events.

Some general thoughts for discussion tomorrow on today’s events:

• Good cone of silence case for MR/MS
• Beneficial to compare base data with MR/MS output to get a sense of the values
• MESH did very well with hail size estimates
• Trends very useful
• Google map display very useful
• Lots of new products to check out in addition to looking at base data

Will touch on these and more points tomorrow.

Liz Quoetone and Paul Schlatter (EWP Weekly Coordinators, 1-5 June 2009)

At the start of the 1 PM meeting on 6/4/2009, the group summarized yesterdays activities.  We did an LMA IOP for LWX/AKQ, then PAR and CASA archived events.

LMA IOP

• We re-examined the 3 storms around 1950 that had varying LMA attributes vs. CG attributes.  In particular one storm that looked impressive on radar had very high amounts of CGs, but LMA didn’t show much.
• Need to be able to explain why the storm didn’t show up in LMA yet had high CG rates and good reflectivity characteristics aloft, because this particular storm would be an example of when the LMA doesn’t not perform as expected, and in a way detrimental to storm interrogation.
• MESH didn’t agree with the VIL, LSRs, Z at -20C…seemed to overestimate the hail in the LWX CWA
• LMA helped get a warning out quicker than normally would b/c they were able to monitor trends in the LMA, looking for a rapid increase in flash rates
• For warning decisions, Bill/Veronica used VIL, Z at -20 C, MESH, and was looking for things to agree with each other and with the LSRs they were getting
• For warning decisions, Pete/Jenni focused more on LMA rather than warning decision making.
• Low values in LMA b/c of temporal smoothing ocurred which made things hard to interpret
• LMA didn’t respond the same way that CG rates do as the storm intensifies and grows in vertical depth.

PAR

• Pete very excited about using PAR for tropical supercells
• TS Erin case had very clear signatures, liked the time resolution of the data and watching the evolution of the convergence in low levels in the hook echo region.  Issued a warning when convergence increased, knowing that an increase in rotation was likely to follow.  Admittedly the convergence signatures in tropical environments have a high FAR.
• See the same things with the 88D given similar resolutions, you just see them a few minutes sooner
• Yes, warnings could come out a few minutes earlier…but also can see a situation arise when they wait for another scan…60 sec…then wait another…then another since it’s only 60 sec.  Would be interesting to do a study with experienced forecasters using PAR data to see how they react to the vast amounts of data rapidly updating in a warning environment.

CASA

• showed reflectivity do-nut which the 88D did not…also was able to see the rotation intensification better than with the 88D. (May 2007 event)
• CASA had far better location of the tornado relative to the 88D.  1500 feet (CASA) vs 2200 feet (88D)
• Good: see things faster and better low level coverage, could see things you wouldn’t normally see with the 88D, especially with the varying viewing angles
• Lots of things we don’t understand in the CASA data since haven’t seen before. Big learning curve. Change in methodology will be warrented.

Liz Quoetone and Paul Schlatter (EWP Weekly Coordinators, 1-5 June 2009)

At the 1 PM briefing on June 3, group got together to discuss yesterdays activities.

MFR IOP: MESH was useful and seemed to be accurate in size relative to the reports given. -20C Z was helpful because many storms across the domain had high reflectivities but not much into the hail growth zone, but the deepest storms that led to the warnings showed up nicely on the -20C Z products.  MESH showed 1 inch at one point, leading to the issuance of the warning, which verified. Enough shear present for MESH to do a good job at size estimation….where normally weak sheared environments MESH used to overforecast (when it was cell based).  Would be interesting to find out if gridded MESH continues to overestimate hail size.  All tilts reflectivity was used to draw the warning polygon, MESH used to zero in on threat area.  One team also used VIL to find the deepest/most threatening storms, and for this event they found 35 VIL sufficient for 0.50 inch hail, greater values for potentially severe hail.

Interesting rotational signature with greater than 50 kts velocity difference at 12 kft MSL on the western end of the storm complex.  Signature was nearly stationary for 25 minutes, and corresponded to a peak in the MESH of just over an inch before dissipating.  Reflectivity deformed with the circulation as well.

LMA Thoughts from the IOP:

• Good event for training…to show what LMA can and can’t do given isolated, not widespread coverage of storms.
• Was able to track updraft cores in the LMA data.
• DC network had some problems yesterday, missing some storms.
• Impressed with how the LMA cores followed the reflectivity cores.
• Next step for LMA data is to get trends into SCAN
• LMA would be very valuable in radar-sparse areas or in blockage areas
• GLM for warning input was sufficient to find the significant storms. Also agreed that the most benefits would likely be in the WR
• ICahead of CC, CC before IC, or same time in growing storms: Good research topic to see which occurrence might lead to useful intensification information.
• Algorithms based on LMA data could be useful, like a TVS type thing, but big questions about how this might differ from SCAN type alerts and the algorithms it uses

Discussion moved towards data overload if/when 1 min volume scans become operational, in addition to new radar products and algorithms and other sources of data.  Will be a VERY big challenge trying to figure out which data are most important and how to effectively manage all the data.  We don’t really know what the impact of all that data will have on operations.

Liz Quoetone and Paul Schlatter (EWP Weekly Coordinators, 1-5 June 2009)

Pete worked entire IOP on storms over Northern Kansas/Southern Nebraska. This was mainly a hail event with lots of quarters and a few golfballs in storms which were propagating swwd. Mesh and traditional data sets were consistant. Some of the trend products were useful.

Bill and Veronica worked marginal storms in CASA network. Some hail but not much wind. Followed this with the CASA Archive wind event. This group didn’t have D2D and was somewhat hampered with WDSSII display efforts. Looking forward to live events tomorrow, hopefully for all 3 OK network sensors (CASA, PAR, LMA).

-30-

Liz Quoetone (EWP Weekly Coordinator, 1-5 June 2009)

The LMA archive took the group up through 6 PM.  We then reconvened with pizza at 6:30pm, and began work on three tasks: PAR archive cases, a CASA surface wind prediction exercise, and a free-form exploration of CASA data from earlier this spring.  We will work these tasks right up to 9 PM, and then hope to gather a wealth of feedback during our Friday end-of-week debrief.

Patrick Burke (EWP Weekly Coordinator, 18-22 May 2009)

The day began with an archived LMA event.  We then had a successful MRMS IOP in the Cheyenne, WY, and North Platte, NE, CWAs.  We certainly made the most of what was a marginal severe weather setup in the most marginal of weeks!  Forecasters spent some time learning the MRMS products, and then began to issue warnings.  Severe warnings were scattered initially.  Eventually, one multicell formed a cold pool and propagated forward, maintaining a severe weather threat all along its path as it approached the KLNX radar. There was plenty of discussion of MRMS products in relation to wind and hail threats, and updraft strength. Operations are winding down as of 8:25 pm CDT, and we will hold an end of day discussion to gather forecaster feedback.

All of this week’s forecasters advocate intense use of base data in warning decision making.  Mike says “We are overloaded with derived radar products.  Just give me the base data, and a better understanding of the environment (to improve warning decision making).”  Others agree that environmental data is key to warning decision making, but they also point out that there may be some tasks that algorithms can perform very well, thus freeing the human forecaster to tackle other problems.  The group also discussed certain scenarios in which algorithms can aide the forecaster during low staffing or broad geographical outbreaks.  In such cases, Matthew would use algorithms to rank storms that need further investigation in the base data.  Les and Rob particularly like the LMA and lightning trend data, as it is “Base data of a different kind.”  The forecasters, as a whole, liked the trend graphs for all types of data.  That was the most useful function of the data in Google Earth.  The group noted several areas, however, where the Google Earth data needs improving.  There is no time stamp, images are smoothed, and there is no cursor readout.  They would like the color tables to be synced to the respective products with which they were intended to be used, rather than requiring the forecaster to click on a product and then click on an appropriate color curve.  The color curves also need to be consistent across platforms so that the data looks the same on Google Earth as it does on D2D/AWIPS.  In its current state, forecasters feel the images in Google Earth are better suited to verification efforts by mapping MESH and rotation tracks to GIS data, than they are suited to warning decision making.  Within AWIPS, forecasters looked at MESH swaths and current MESH, as well as height of the 50 dbZ reflectivity.  These things were used most often to assess hail and updraft strength.  As noted in another post, forecasters would like to see height products for 60 dbZ, and they did not seem to find much use for 18 and 30 dbZ.

Patrick Burke (EWP Weekly Coordinator, 18-22 May 2009)

Early on, forecasters commented that they enjoyed viewing LMA data from an early spring case in the southeast U.S.  This was their first chance to navigate the data in a hands-on fashion.  Technical problems delayed our intended MRMS IOP.  Forecasters eventually viewed some MRMS data from radars near Salt Lake City and Glasgow.  This was a good exercise in familiarization, but there were still some questions about the product labels in Google Earth, and what exactly the radar images represent.  These questions will be answered on Wednesday.  Technical problems were solved by the time we moved our IOP into the Glasgow area, but the weather did not cooperate.  Storms were not continuous in time, and were not very threatening when they did exist.  We decided to cut our losses, and spend the last two hours of the evening swapping yesterday’s teams between the CASA and PAR archive cases.  I have heard a lot of good discussion as the teams viewed these cases, and we will discuss this further to begin the day on Wednesday.

Patrick Burke (EWP Weekly Coordinator, 18-22 May 2009)

Our Week 4 forecasters were not deterred by bleak prospects for organized severe weather.  Instead, they arrived ready to tackle archive events and whatever pockets of real-time opportunity present themselves.  With four projects in the HWT this year, the task of training is quite lengthy, and it was an accomplishment to push through all of the training on the first day.  We have a very inquisitive and well-read group this week; they asked many questions during the training, often citing references.  They seemed to take in all of this material eagerly.  Although I heard no negative comments, I wonder as a coordinator, if it might be information overload to push through all 4 programs in one day.  I might have preferred to have a Day 1 IOP, and complete the training necessary for that IOP, while saving half of the training for Day 2.  In our case, though, we are now well versed, and the team of Mike and Rob have completed a PAR archive case; Les and Matthew are paused in the middle of a CASA archive case, ready to resume Tuesday afternoon.  We also hope to run an IOP for Multi-Radar/Multi-Sensor research in the northern Rockies.

Patrick Burke (EWP Weekly Coordinator, 18-22 May 2009)

The evening was used for archive exercises with PAR and CASA.

Jim LaDue (EWP Weekly Coordinator, 11 – 15 May 2009)

End of IOP.  Storms in the RAH CWA are rapidly trending down and leaving the CWA.  Since we started before 4 this afternoon, we’ll call it a day.  Discussion follows…

Used the MESH and RotationTracks for intensity trends more than location.  Storms today didn’t really deviate much.  Tracks helped to narrow warning areas.  There was some question regarding the usability of the Lightning forecast products in their current form.  Color table may need to be adjusted.

Scott – liked trends a lot.  Found them to be quite useful and informative.

Gino :: trends seem “black boxish” right now, so confidence is low.  Relies more on based data.  Worried about false peeks and whether to trust the intensity diagnostic.  Good SA tool and for “post analysis”.   Likes Reflectivity at -20C

Suggestion :: VIL (or a Reflectivity sum product) above -20.

Slider capability for product values, similar to how we adjust isosurfaces.

Products in GoogleEarth were very popular.  Most agreed that they would be useful in SA.

Trends were helpful in discriminating between “similar” looking storms.  Trends were also useful in warning decision making.

Kevin Manross (EWP IT Coordinator)