Nearcasting CAPE vs Visible Imagery

The Nearcasting model is an excellent observation-based decision-support product that has been developed by UW-CIMSS.  One of the product's strengths is to forecast where instability is and is not located.  The image below is a 7-h forecast of Nearcasting CAPE (valid at 1900 UTC) with the 1900 UTC Visible imagery and previous hour's lightning observations.  Unstable air is in shades of blue and purple while stable air is in shades of green and brown with the CAPE and Visible images transparent for ease of comparison.

Some things to take away from this figure.

1. The majority of the CU field and convection is to the south of a large-scale midlevel instability gradient that extends from central AR east northeastward into the Mid-Atlantic states.

2. The deeper convection is aligned well with the deeper blues and purples (higher values of CAPE) across northern MS and most of TN, eastern Gulf of Mexico, and across central FL.

3. The most stable air extends from southern MO east northeastward into the eastern Great Lakes and into New England...notice there are very few clouds in this area.

Not only is this product good on the NWS CWA level but also for the analysis of large-scale areas of convection (and lack of convection) that are important for operations at the Storm Prediction Center and the Aviation Weather Center.  The SPC could use the Nearcasting model to assist in their forecasts of the "thunder line" and the AWC could use the output to assist their partners with highlighting which high altitude jet routes are clear for the day (e.g., the NYC to Chicago corridor today).

ThetaE Gradient

Forecaster comments from the EWP blog...

Noticed an interesting diagonal differential heating sort of boundary in across northern Florida where the next round of convection appears to be initiating.  It show up well on the nearcast as s thetaE gradient and is now showing up on Vis and CI products…if this end up being the focus of more critical convection later in the afternoon this could be a really good application the nearcast product.  There was a similar looking (though not sure if it was the same cause) feature when we were MLB two days ago that was where the bigger stuff triggered.  Waiting to see…

GOES sounder nearcasting CAPE

Forecaster comments from EWP blog...

We’ve been keeping an eye on the relationship between the GOES based CAPE field and the coverage of the developing convection over Florida today.  The synthetic satellite imagery has also done a good job on delineating the areal coverage of developing convection.  The four-panel image has actual Infrared imagery on the upper left, synthetic satellite imagery on the upper right, visible satellite and lightning data on the lower left and a layered CAPE field derived from GOES-13 sounder retrieval data.  The highest CAPE values, around 1200 joules/kg, are located just to the northwest of Lake Okeechobee, with an axis of lower CAPEs further to the north.  There is also a subtle increase in CAPE values over northern Florida, where additional convection is developing.  The region with the lower CAPEs has weaker convection at this point.  The synthetic satellite imagery is a 12 hour forecast using the 06z run of the WRF model, valid at 18z.  It also does a nice job depicting less convection in the region of lesser instability.

GOES-15 in Super Rapid Scan

For those of you interested in looking at some GOES-R-like temporal resolution imagery, GOES-15 (GOES-West) is operating in Super Rapid Scan Operations (SRSO) today over eastern CO in support of the DC3 experiment until 2359 UTC.  You can find an imagery loop at the link below (courtesy of Dan Lindsey at CSU)...

http://rammb.cira.colostate.edu/templates/loop_directory.asp?data_folder=dev/lindsey/loops/goes15&image_width=1020&image_height=720

UW-CTC in eastern CO

While the warning teams are keeping an eye on things in FL, thunderstorms have developed in eastern CO associated with a midlevel short wave.  The UW-CTC algorithm detected cloud-top cooling rates in the -25C/15min range with a lead time of approximately 45 minutes on a 50 dBZ echo with the storms.  The storms initiated in juicy air with dewpoints around 35F along the front range of the Rockies.  The images below span an hour between 1815 and 1915 UTC.

NearCasting for JAX

Forecaster comments from EWP blog...

We are focused on the JAX CWA today.  While there is convection ongoing in the extreme northern portion of the area (bordering FFC and CHS), the GOES-R NearCast CAPE (new this week) and vertical theta-e difference suggest that the southern half of the JAX CWA is more conducive for new convection.

The NearCast has been challenging to interpret this week due to interpretation confusion (are negative versus positive theta-e differences indicative of instability) and then a mid-week visualization change in AWIPS 2.  The NearCast shows promise for helping with the tremendously-difficult 1-6 hour timeframe, but the product stability has presented a problem.

EWP daily debrief 5/17

Below is some feedback we captured during today's daily debrief of the EWP forecasters...

SATCAST / UW Cloud-top Cooling
- Burlington, VT CWA... "We were seeing high CI signals to start... but when the storms started growing and pluming out, they became less useful, but it did pick up on some outflow boundaries where more development was occurring."
- Albany, NY CWA... "There were several areas where both algorithms flagged this lonely cell but it ended up falling apart... it had CTCs in the range of -30 C/15 min and it didn't do anything significant based on my expectations from what we should be comparing it to... there were some cells to the southeast, so the inflow may have been interrupted."
- "The CI started with a high signal (red), then the CTC flagged the storm and it became the strongest storm in the line."

Nearcast
- "I don't remember seeing an indication necessarily correlating to a tongue of moisture... I don't think it was really well defined... there was some indications of slightly moister airmass, but nothing significant... I don't know if the nearcast had the resolution to get down to that narrow of a feature."
- "We saw higher theta-e values as well as the new CAPE product over N. AR and that was useful which depicted and area where convection occurred."
- "It was an overall helpful tool... not necessarily picking out where severe occurred, but where convection would occur."

Simulated Satellite
- "On a 23-24 hour forecast, it nailed the location of convection over N. AR... I was impressed at how well it did over the NE as well."
- "I would like the reality and the model to look a little different... maybe a slightly different color table."

Overall / Training
- Trend line displays within AWIPS would be very useful.

Synthetic imagery – verified!

Forecaster comments from the EWP blog...

The synthetic imagery did a nice job of placing convection southward along the cold front through the Ohio and Lower Mississippi Valleys.  Here’s a screen capture from 2300UTC, with the actual IR on the upper left and the model forecast on the upper right.  Visible imagery with lightning is shown in the lower left pane, with the theta-e Nearcast product on the lower right.  All in all, the GOES products did well today depicting where convection would be in relation to the cold front.  This would be an excellent application to the CWSU environment for air traffic planning purposes.

CIMSS Nearcasting: Pushed to the limit - the lower limit!

The past three days have been challenging for the CIMSS nearcasting model since atmospheric conditions have not been favorable for the development of strong convection.  Ɵ_e lapse rates have rarely exceeded 8-9 dƟ_e.  However, forecasters have been able to use the nearcasting guidance, both Ɵ_e and the new CAPE products to correlate where convective development is expected.

 

Shown below is a 4-hour nearcast of dƟ_e  and CAPE valid 22UTC, May 16, 2012.  The nearcast model consistently placed a line of instability along the Ohio River Valley associated with a trailing weak cold front.  At this time convective cells formed in NE Arkansas which matched well with the nearcasted dƟ_e and the nearcasted CAPE.  Also shown below is the verifying GOES IR and radar images.  Although convective cells formed, they were not expected to become severe based on other available guidance.

 

 

 

Quick Glance at SimuSat for Today’s Convection

Forecaster comments from the EWP blog...

Severe thunderstorms kept the WFO ALY team busy today!  With storms in our area diminishing (and moving out), I wanted to check out the CIRA/CIMSS Simulated Satellite Imagery from the NSSL-WRF to see how it compared to the last few hours.  Overall, I’d say quite well, aside from some placement issues, and the extent of the coldest cloud tops (which was to be anticipated).