In one hour it is trying to already form storms over parts that are covered by the cirrus blanket so I wonder if this won’t happen since it hasn’t accounted for it. We shall see. ~ Vollmar
Thursday, May 22, 2014
Simulated Satellite Vs. IR Sat
In one hour it is trying to already form storms over parts that are covered by the cirrus blanket so I wonder if this won’t happen since it hasn’t accounted for it. We shall see. ~ Vollmar
Simulated Satellite to Early
Was comparing the simulated satellite to observations to attempt to get a handle on the pre-convective environment. It appears that the simulated satellite today is to quick in developing convection from southern KY into TN. It also is not handling the convection that is developing over eastern KS well. Finally it has convection developing over southwest KS that is not there in reality. This lowers the confidence in the forecast of this model for later in the day.
-JB
-JB
EWP Operations Update – Thursday 5/22 – 1:10pm
Our forecasters are set up in Baltimore/Washington D.C. (LWX) to take advantage of the DCLMA as this region currently has relatively few clouds conducive to stronger surface heating than previously anticipated. Adjusted 12z sounding with current surface temperatures show a large region in LWX with minimal CIN to hinder further convective development.
Super-Rapid Scan visible imagery from 1706z-1820z over the East/Central US.
Our second team is currently situated in Nashville, TN (OHX) as we wait for the remaining cap to erode over this area in the next few hours. We only anticipate isolated severe storms in the northern part of the CWA but would like to use this opportunity to really look at some of the GOES-R pre-CI products to see how they fare in a still capped area. On the plus side, if storms do fire here and move to the southeast, we may get some activity in the NALMA. On our final day of Super-Rapid Scan operations, we also wanted to get both of our forecaster teams in domains where they could leverage these data.
We’re watching the WTLMA and COLMA regions closely as well and are ready to jump out west if absolutely necessary.
-Darrel Kingfield
EWP Week 3 Coordinator
Super-Rapid Scan visible imagery from 1706z-1820z over the East/Central US.
Our second team is currently situated in Nashville, TN (OHX) as we wait for the remaining cap to erode over this area in the next few hours. We only anticipate isolated severe storms in the northern part of the CWA but would like to use this opportunity to really look at some of the GOES-R pre-CI products to see how they fare in a still capped area. On the plus side, if storms do fire here and move to the southeast, we may get some activity in the NALMA. On our final day of Super-Rapid Scan operations, we also wanted to get both of our forecaster teams in domains where they could leverage these data.
12z ROAB sounding from Nashville, TN.
We’re watching the WTLMA and COLMA regions closely as well and are ready to jump out west if absolutely necessary.
-Darrel Kingfield
EWP Week 3 Coordinator
Wednesday, May 21, 2014
Daily Summary: Week 3, Day 3 (May 21th, 2014)
Another productive day in the HWT! We ended up operating in three domains today, Denver/Boulder, CO (BOU), Indianapolis, IN (IND), and Lubbock, TX (LUB). For most of the day we had two forecasters in the BOU and IND CWAs but moved IND to LUB within the final two hours when most of the severe convection moved out of the CWA. A weak mid-level impulse and steep lapse rates supported initiation along the TX/NM border and provided both teams of forecasters to look at the PGLM products and utilize the tracking meteogram tool with these datasets.
Our BOU team remained in their CWA and mainly tracked what will likely be called the “Denver Metro Area Supercell of May 21, 2014″ for years to come (this title may need to be shortened…) All GOES-R, several LAPS 3D analysis products (from the 200×200 domain), MR/MS, and Super-Rapid Scan imagery were utilized in performing mesoscale and nowcasting analyses as well as warning issuance. Preliminary reports are coming in on the spatial extent of the main rain-wrapped tornado produced by this storm around 2000z.
0.5 reflectivity from KFTG on May 21, 2014.
Our IND team faced another training sequence of convective cells that came out of the northwest into the CWA. Numerous severe hail and wind reports through the central/southern parts of the CWA. The LAPS 800×800 domain was situated over this area of complex convection, providing 2D analyses and forecasts that our participants compared to MR/MS and other observational products.
Feedback (BL)
Simulated Satellite Imagery
- Indy cwa, it was pretty good with where storms would pop timing wise. Timing a bit off.
- Boulder – convection and cells in correct spots, timing a bit off. 30 min to an hour fast.
-Darrel Kingfield
EWP Week 3 Coordinator
Our BOU team remained in their CWA and mainly tracked what will likely be called the “Denver Metro Area Supercell of May 21, 2014″ for years to come (this title may need to be shortened…) All GOES-R, several LAPS 3D analysis products (from the 200×200 domain), MR/MS, and Super-Rapid Scan imagery were utilized in performing mesoscale and nowcasting analyses as well as warning issuance. Preliminary reports are coming in on the spatial extent of the main rain-wrapped tornado produced by this storm around 2000z.
0.5 reflectivity from KFTG on May 21, 2014.
Our IND team faced another training sequence of convective cells that came out of the northwest into the CWA. Numerous severe hail and wind reports through the central/southern parts of the CWA. The LAPS 800×800 domain was situated over this area of complex convection, providing 2D analyses and forecasts that our participants compared to MR/MS and other observational products.
Feedback (BL)
Simulated Satellite Imagery
- Indy cwa, it was pretty good with where storms would pop timing wise. Timing a bit off.
- Boulder – convection and cells in correct spots, timing a bit off. 30 min to an hour fast.
NearCast
- Lots of blank spots. It would be nice to blend nwp model data
- An on off toggle with model data might be neat
- Boulder – pw difference with evolution of storm. With discrete cell,m there was a narrow tounge of moisture feeding. A wall of moisture was moving into area. As wall moved in, storms formed along the boundary. I could have got change from discrete to cell.
- A little difficult to use today due to missing data from cloud cover.
GOES-R CI
- It did alright.
- In INDy One cell in particular, ci showed 90%, convection fired about hour later, it was hard to tell if convection was actually associated with it.
- In Denver it did quite well with main Denver cell it had >70%
- GOES west not doing as well as east
Prob Severe
- Did well in high plains
- In the diminishing phases is was useful in keeping me from warning on a couple storms
- Lots of pink in Indiana (>85). Many of these storms produced severe.
- I warned one based on an upward trend, then it died, lots of storms were doing this
OTD
- In NE Colorado later, the Colorado storm weaked, but still had great OTs, so still intense rainfall. Even though storm isn’t severe, updrafts could still be strong enough to produce heavy rain
- In texas, it showed up nicely, it made me focus on it, hled it for 4 or 5 scans, then it fizzled and the storm weaked.
- Perfect for overnight convection
PGLM
- Impressed, by overlaying the different products, I issused a warning off of it, it hit a sigma of 3 and I pulled the trigger. Inch hail a few minutes later.
- Flash extent and initiation. Max did not work. I thought it was great. Watching a storm merger in Lubbock. Initiation density really lit up on front end of merger. A few minutes later extent density lit up as well.
- CG’s do not tell the whole picture at all, so the total lightning was helpful
- Provided lead time over radar.
LJ
- The larger scale one would sometimes try to combine cells. So I would go to smaller one and it would be better at identifying the individual cells
- I did get the bit towards the end of day when it was nice
SRSOR
- It helped me in radar down experiment. It helped to identify boundary that was feeding into storm. It would have taken a little longer to piece in my mind what was going on in the 15 minute scan mode.
-Darrel Kingfield
EWP Week 3 Coordinator
Switching to West Texas…
Excitingly, we’ve switched over to a region that is covered by GOES-14 SRSOR, so I pulled up the visible imagery and overlaid the overshooting top detection product. Interestingly, we noticed a very slow-moving storm in south-central AMA’s CWA that erupted on vis satellite and consequently, the OTD picked up on it. Additionally, a weak but still present enhanced V was observed with this storm on SRSOR’s IR imagery.
Now to center on LUB, I overlaid SPoRT’s Flash Initiation Density product on KLBB 0.5Z and observed an increase in the initial density as a cell merger occurred in northern Lamb County, TX. This increase was evident for a few minutes and then simmered down. What’s particularly important is that the SPoRT data, since it is coming in every minute, indicated this “jump” before the next volume scan came in on radar.
In the next loop (below), I did a similar setup with 0.5 reflectivity, SPoRT’s Flash Extent Density product, and included a 1-minute CG lightning plot as well. As anticipated, flash density responded similarly to the cell merger by showing an increase.
At around the same time, WFO LUB received wind/wind damage reports in the area.
Having the 1-minute total lightning data was very helpful in being able to see the response of the cell merger and give a couple minutes more lead time to anticipate what was to come. While CGs can provide great information, having total lightning to get a better idea of what is going on is exponentially more valuable.
~Linda
Now to center on LUB, I overlaid SPoRT’s Flash Initiation Density product on KLBB 0.5Z and observed an increase in the initial density as a cell merger occurred in northern Lamb County, TX. This increase was evident for a few minutes and then simmered down. What’s particularly important is that the SPoRT data, since it is coming in every minute, indicated this “jump” before the next volume scan came in on radar.
In the next loop (below), I did a similar setup with 0.5 reflectivity, SPoRT’s Flash Extent Density product, and included a 1-minute CG lightning plot as well. As anticipated, flash density responded similarly to the cell merger by showing an increase.
At around the same time, WFO LUB received wind/wind damage reports in the area.
Having the 1-minute total lightning data was very helpful in being able to see the response of the cell merger and give a couple minutes more lead time to anticipate what was to come. While CGs can provide great information, having total lightning to get a better idea of what is going on is exponentially more valuable.
~Linda
Negating false positive/problems from larger LJDA scales.
With the image above I just wanted to show that sometime the LJDA can combine storms and trigger a larger sigma warning than seems realistic. Especially when the storms seem to barely touch on radar. In this case it was a 2-3 Sigma jump in Yellow which would prep the forecaster to think about warning.
This false positive can be fixed by overlaying a smaller scale LJDA over a larger one to potentially split the storms and break the sigma back down to a more realistic approximation that would not trigger a reason for warning (after suspecting it to be a false positive).
Grant H.
Over-Shooting Top and Heavy Rain
Been watching the convection over northeast CO, and although the general trend of the severity of the convection is downward, the over-shooting top algorithm continues to ping over this area. This highlights that although the storms may not be producing severe weather, i.e. hail and/or wind, they could be significant rain producers and should continue to be monitored.
-JB
Total Lightning to forecast strengthening storm
I have been watching the Total Lightning product over the developing storm and it shows both flash initiation density and flash extent density that have been increasing over the last 30 minutes. That coupled with the increase on the Prob Severe, would lead me to believe this is nearing severe limits and I would then issue a warning. I really like these products because it shows where the updraft is developing and also where hail would be found. ~ Vollmar
Prob Severe and Demise of Storms
Watching trends in the probability of severe over northeast CO indicates a significant downward trend in prob severe with the low-lived storm that went through DEN. This occurred as the local office allowed the warning expire and the reflectivity signatures were decreasing as well. Provided confidence in letting the warnings expire.
-JB
A CI Update
Remember that screen capture of the 90% probability of convective initiation over Lawrence County, IN at 2130Z? By 2248Z, a decent cell had popped up in the NE corner of the county. Not sure if this necessarily verifies the product or not (after talking with the product’s contact here), but it does instill some confidence on using this product in the future.
~Linda
~Linda
Subscribe to:
Posts (Atom)