CI along sea breeze over NE Florida

Forecaster comments from EWP blog...

Westward moving sea breeze front helped to spark sporadic showers/thunderstorms during the afternoon hours. We captured the interaction of an already ongoing weak shower, as front moved in from the east.

 1932 Z; UAH-CI and Cloud top cooling

1940 Z; UAH-CI and Cloud top cooling

The southern parts of Nassau and Charlton county and Baker county were monitored. At 1932Z an increase in cloud top cooling was noticed and 8 min later, CI product also had probabilities of 70 % plus for this area. At the same time, a cooling rate of roughly -15K/15 min was detected. Rapid development occurred thereafter with 50 dBz and more seen in the JAX radar side at 1947 Z, which gave a good lead time for focusing on the potential hot-spot over NE Florida. The storm was a pulsating one with rapid weakening thereafter.

UAH-CI Signal Precedes UW-CTC Signal…Both Precede Convective Initiation

Forecaster comments from EWP blog...

The more I look at these products, the more I’m convinced they work very well in tandem with each other.  Given an environment that supports convective growth, a strong UAH-CI signal (upper image, top left panel) typically precedes a strong UW-CTC signal (middle image, top right panel).  In this case, the strong UAH-CI signal (1615 UTC) preceded the UW-CTC signal (1645 UTC) by about 30 minutes.  By 1715-1730 UTC, 30-45 minutes later, deep-moist convection did indeed form as evidenced by the numerous CG strikes (bottom image, bottom left panel).  We’ve seen this all week…UAH-CI gives us a heads up that general convection is initially building, then a short time later UW-CTC typically picks up on the stronger convection building.  Given a strong cap isn’t present and instability is sufficient, deep-moist convection typically develops 30-60 minutes thereafter.

1615 UTC. Notice the 70% UAH-CI signal in the top left panel.

1645 UTC...30 minutes later. Notice strong cloud-top-cooling signal in the UW-CTC product (top right panel).  

1715-1730 UTC...30-45 minutes later. Deep-moist convection did indeed develop in the vicinity of aforementioned CI/CTC signals, as evidenced by numerous CG lightning strikes (bottom left panel).

UW-CTC Working as Expected in MN/IA

The UW-CTC product has been capturing the initial and ongoing convective development taking place first within the MPX CWA domain, and later further south in north central to central IA.  The beauty of these events is that this showcases how the UW-CTC product still does pretty well, even though the area has a lot of apparent cloud contamination within the scenes.  The below images are shown in chronological order with the original strong convection beginning in the MSPL area at 1515 UTC (Fig. 1).

Fig. 1

Later on at 1725 UTC, the UW-CTC captured good development in the MSPL area that later is an ongoing robust thunderstorm (Fig. 2).

Fig. 2

Some further strong convective development along the frontal boundary was noted in northern and central IA.  Figs. 3 (1932 UTC) and 4 (1955 UTC) capture this development unzipping further south along the boundary and still within areas of significant amounts of cloud debris.

Fig. 3

Fig. 4

CI mainly inland across FL peninsula

Forecaster comments from EWP blog...

Latest CI image shows moderately high probabilities for CI along ahead of the westward moving sea-breeze boundary.  Best chance for thunderstorms appears to be from the interior counties towards the west coast.  Some initiation is possible across Northern Brevard County…though the prospects for lightning are looking slim at this point.

Overview of sea breeze fronts in Florida

Forecaster comments from EWP blog...

We’re currently monitoring the rapidly inland moving see breeze front over NE Florida with enhanced TCU present along the convergence zone, whereas the west coast see breeze has trouble to move inland within dominant easterly wind regime. Both convergence zones enter an area over N-C Florida where slightly lower dewpoints (lower to mid 60s) are present. With ongoing diabatic heating, cap continues to weaken or has already vanished with isolated initiation possible over N-C Florida in the following hours but ongoing large-scale weak subsidence seems to delay initiation a bit. HRRR quite aggressive with initiation mainly along the western sea breeze front with RUC also showing sporadic storms over N-C Florida. Overall set-up is not impressive and favors only the development of a few thunderstorms. Moderate CAPE and weak shear environment may result in a few strong and slow moving storms.

Neither CI nor Cld top cooling products showing robust signs for initiation … mainly over C Florida. Both products however show more vivid activity offshore over the Atlantic, where mid-levels remain a bit cooler.

1800Z VIS and station data.

1745Z; UAH-CI (upper left) and CIMSS Cld Top Cooling (upper right).

Southeast US still providing areas of CI

The pesky upper low over the Southeast US continues to move offshore. With strong surface heating and "cool" mid-level temperatures with this low, airmass thunderstorms continue to develop over the Southeast US. UAHCI has been indicating many areas with Strength of Signals over 60 from the Southeast up into the Mid-Atlantic States. Many of these storms will not become severe due to the surrounding synoptic environment, yet will produce heavy rainfall and frequent lightning with the stronger storms. This example of UAHCI tracks a slow developing storm in Washington County, GA. These two images are from 1632 UTC and 1645 UTC with Strength of Signals (SOS) of 52 and 57 respectively, suggesting only slow growth at the time.

At 1702 UTC, the SOS ramped up to 73, suggesting more rapid growth and increasing confidence that this cell will actually CI.

At 1715 UTC, the SOS increases further to 83. The UAHCI algorithm stopped tracking the object after this time period as it was deemed mature within the cloud mask, which is evident within the visible imagery at 1725 UTC.

Outlook 24 May 2012: Week 3, Day 4

From the EWP blog...

Operating today during the first mod risk of the experiment, however, that is not where we are operating today… at least not to start.

In order to try and get lightning data into the experiment this week, we’ve initialized in the Sterling, VA and Melbourne, FL domains.

For the FL domain, we’re hopeful that storms will get going as the sea breeze front sets up across the state.  Already, the UAH-CI product is flagging cumulus for development (consistently 50-60%) along the line from Sanderson to -Orlando to just northwest of Okeechobee.

UAH-CI & Visible Satelllite for central FL @ 1830 UTC on 24 May 2012.

For the DC domain, the CI alg is also flagging development generally a bit lower on the probability though. A region of SBCAPE of >1500-2000 J/kg extends from eastern NC to over much of the LWX CWA, so it appears possible that we could at least see some convective development across some of the DC LMA though likely only marginally severe.

Our observers and some PIs are keeping an eye over central WI and the associated moderate risk region for development of surface based storms (as to now all development across Minnesota appears to be elevated).  Currently instability is limited across much of the region, but we do expect additional destabilization with daytime heating and enhanced lift as the cold front moves in.

EWP daily debrief 5/24

Yesterday we had a good day in operations for the CI products especially.  Below is some feedback we gathered from the forecasters during our daily debrief...

UAH SATCAST (CI) / UW Cloud-top Cooling (CTC)
- "They were very helpful... it was environment you knew the cloud would cool quickly, but you didn't know which clouds would go up... I think it's benefit is really in that marginal environment... discriminating between that severe/non-severe borderline is where we really need help."
- "Really all the satellite based products would be useful for us in terms of aviation forecasting... if we could get an hour lead time on CI, it would be really helpful to us... especially in a place like NY where just one thunderstorm will mess up everything with regards to air traffic... just to see the CI and the CTC, you can update your TAF and contact the proper folks and give them a more confident message than if you're just using radar... with just radar we can't give them the lead time they need."
- "When we were deciding when to switch, it seemed like the CI was showing a widespread 30-50% for that whole area, but it didn't really happen." ... Developer explained that sometimes many cumulus clouds close together can become clustered together (especially at night when the visible channel isn't utilized), which will give you relatively lower probabilities over a broader area.
- "That region of Cu was a strongly forced/capped area... that broad view of the CI product gave you an idea that there is some upward motion and told you to pay attention."
- "I really loved the CI products yesterday... seeing the CI and CTC zippering down the line, we actually warning on one CTC signal when we saw that everything else ahead of it was going."

Overall / Training
- "With what you show the forecasters, simpler is better."

Nighttime UAHCI Success

As daylight begins to wane along the East Coast, the UAHCI algorithm transitions to a nighttime cloud mask that no longer depends on visible imagery. With the lack of visible imagery, the algorithm appears much more blocky in nature due to the decrease in spatial resolution (1 km visible to 4 km infrared). However, we are able to still track cloud objects that are not stratiform in nature. In this case over Clay County in Southeast South Dakota into Dixon County in Northeast Nebraska, the UAHCI Strength of Signal reached 92 on cells behind the main line of storms. Although not severe, the cell did develop with reflectivity values over 40 dBZ. It is important to note that the UAHCI product is not meant for the predictability of severe storms, but any type convection.

CI product over E-Nebraska

Forecaster comments from EWP blog...

The cold front over NC-Kansas and SE Nebraska was monitored for further development. As expected, storms began to build SW-wards towards the KS boarder and CI products were effective in highlighting the area with highest probabilities for potential development along that boundary.

The event occurred over Clay and York counties (Nebraska). A series of images will be added to show the evolution

UAHCI product upper left, CIMSS Cld top cooling upper right, lightning data lower left and surface reflectivity at -10 C isothermal level at 2232 Z

UAHCI product upper left, CIMSS Cld top cooling upper right, lightning data lower left and surface reflectivity at -10 C isothermal level at 2245 Z

UAHCI product upper left, CIMMS Cld top cooling upper right, lightning data lower left and surface reflectivity at -10 C isothermal level at 2302 Z

Roughly 30 min before surface reflectivity at -10C isothermal level revealed first signs of development, the CI product highlighted both counties with 50-60 % probabilities. The cloud top cooling product showed values of -15 to -25K/15 min at 2315Z (not shown) and the first significant reflectivity signal (56 dbZ ) became visible at 2330 Z. At 00Z onwards a 63 dBz cell evolved out of that NE of York.