Monday, March 7, 2016

Monitoring Flooding & Severe Weather Potential for Later in the Week

The First Alert Storm Team is continuing to monitor the potential for flooding rains and severe weather later in the week. A slow-moving, potent storm system already producing flooding rains and severe weather in parts of California is the one that could be problematic for us in the coming days.

Here's what we know: parts of the Southern Plains and northern Gulf Coast are all but certain to receive excessive rainfall in the coming days. The devil lies in the details. It's difficult, if not almost impossible to pinpoint which areas will be hardest hit but we're gradually getting a better idea of where the greatest threats may lie.

Locally, we can expect scattered showers and a few t-storms to arrive by Tuesday afternoon, but neither flooding rains nor severe weather is anticipated in this initial round of wet weather. Those threats should remain to our west on Tuesday.


Our latest model runs have even started to indicate a somewhat slower arrival of the most active showers and t-storms, perhaps holding off until late Wednesday or even Thursday. It's during this timeframe though that severe weather appears to be possible along with our first real threat of locally heavy rains. The Storm Prediction Center does have areas from near Baton Rouge westward under a 'Slight Risk' of severe weather during this time, with a lower 'Marginal Risk' east of Baton Rouge.


The biggest threat to the greatest number of people during the remainder of the week will be heavy, potentially flooding rainfall. In a modestly encouraging trend, our last couple of rounds of computer model guidance have trimmed back forecast rain numbers. However, we still think widespread totals of 3" to 6" will be common in the WAFB viewing area through Saturday, with the potential for much higher amounts in localized areas. As it stands right now, a greater threat appears to be for areas near the Louisiana/Texas line extending northward to the ArkLaTex.


While the greatest threat does appear to be to our west, the weather pattern expected later in the week will favor the development of 'training' bands of t-storms -- clusters of storms moving over the same areas like train cars moving along a track. Anyone unfortunate enough to caught underneath one of these bands could easily pick up rains in excess of the 3" to 6" we're currently forecasting on average across the area.


With the expected heavy rains, we'll also have to monitor rising river levels later in the week. Additionally, persistent southerly winds mean that coastal flooding could be an issue in spots. Those southerly winds would also result in water backing up in the lower reaches of the Amite and Tickfaw river basins, slowing the ability of rain water to drain into Lake Maurepas.



Tuesday, March 1, 2016

Commentary: Is a gap in south Louisiana radar coverage putting lives at risk?

The final numbers are in -- 12 tornadoes touched down on what was a terrible Tuesday, February 23rd for south Louisiana. Included in that count were some monster twisters like the EF-3 from northern Assumption Parish that went on to devastate the Sugar Hill RV Park in Convent as an EF-2, leaving 2 dead and as many as 75 injured. Let's not forget the EF-2 that injured 17 in LaPlace or the EF-2 that plowed through Livingston with one woman hanging on to her bathroom door for dear life as the entire top of her home was sheared off.

Leveled home near Willow Street in Livingston. (Source: WAFB)

It appears to be the worst tornado outbreak on record for south Louisiana in records that date back to 1950.

Map of the 12 confirmed tornadoes in south Louisiana on Feb. 23, 2016.

And it could have been worse, much worse, if not for the heroic efforts of forecasters at the National Weather Service New Orleans/Baton Rouge office in Slidell issuing rapid-fire warnings for the tornado-producing storms on Tuesday.

But I noticed something on that day that most of you probably didn't. There were a few tornadoes that got very little warning and at least one that got none at all. That particular storm actually fell under the warning responsibility of the National Weather Service Lake Charles office.

But let me be clear: the lack of warning is by no fault of forecasters with the National Weather Service.

The issue, I believe, stems mostly from poor, if not unacceptable radar coverage for parts of south Louisiana.


The Problem: Poor Radar Coverage


The map below shows National Weather Service NEXRAD (Next-generation Radar) Doppler radar coverage at 4,000 feet and below (green), and 6,000 feet and below (yellow). Earth's curvature and the fact that radar is tilted a minimum of 0.5° above horizontal while scanning mean that the radar beam gradually gets higher and higher the farther it gets away from the radar site. What stands out is that because of the curvature and tilt issues, areas from Baton Rouge to Lafayette southward to the coast and northward toward Natchez, Mississippi have no radar coverage in the key atmospheric levels needed for tornado detection. The nearest NEXRAD radars to Baton Rouge are 80 miles to the east in Slidell and 120 miles to the west in Lake Charles.

Radar coverage at ≤4,000 feet (green) and at ≤6,000 feet (yellow). (Source: NOAA/NCEI)

A wider view of the same map shows that Baton Rouge appears to be the second largest city anywhere in the southern U.S. without reliable NEXRAD radar coverage, with only Charlotte, NC having a larger population of people with such poor coverage. Later in this article, you'll see how meteorologists and leaders in Charlotte have taken action to address that problem.

Radar coverage at ≤4,000 feet (green) and at ≤6,000 feet (yellow). (Source: NOAA/NCEI)

If we look at areas with radar coverage at 6,000 - 10,000 feet (blue in map below), metro Baton Rouge finally gets into the mix, but there are still portions of south Louisiana left out. Much of St. Mary Parish remains without good radar coverage even at this level, including areas around Franklin and Baldwin. And as it turns out, that's one of the areas where a tornado was missed on Tuesday. No warning was ever issued, but an EF-1 touched down in that area according to the National Weather Service Lake Charles. Two businesses and 28 homes were damaged, along with 20 telephone poles being snapped.

And, oh by the way, it was this same severe thunderstorm that went on to produce an EF-3 tornado near Paincourtville and EF-2 damage and casualties in Convent. I don't think it was coincidence that the St. Mary Parish tornado came and went without a warning.

Same as above with coverage of ≤10,000 feet added in blue. The location of a confirmed EF-1 tornado in St. Mary Parish is shown, clearly in an area with poor radar coverage. (Source: NOAA/NCEI & Steve Caparotta)

Secondary Problem: Little Reliable Backup


In another illustration of our radar coverage vulnerability in south Louisiana, a lightning strike took down the National Weather Service Doppler radar in Slidell just before 5 p.m. on Tuesday as severe weather was ongoing. The map below shows that there were 2 active Tornado Warnings in southeast Louisiana at that time in addition to ongoing Tornado Watches. The radar wasn't restored to service until late the next day, meaning the National Weather Service in Slidell had to rely primarily on the radar in Mobile, Alabama to issue warnings for southeast Louisiana and south Mississippi.

There is also a Terminal Doppler Weather Radar (TDWR) located in St. Charles Parish that can supplement coverage as needed, but it's much lower power than the NWS NEXRAD radars and has a much shorter range. It does NOT provide reliable coverage of Baton Rouge or much of the area in south Louisiana's radar gap. The TDWR located in St. Charles Parish was installed primarily to aid in wind shear and heavy rain detection for Louis Armstrong International Airport in Kenner.

Radar from 4:50 p.m. on Tuesday, Feb. 23, 2016 with current watches and warnings overlaid. (Credit: Iowa Environmental Mesonet)

Now imagine this nightmare scenario. That same lightning strike happens 2 hours sooner and meteorologists are left 'flying blind' (or close to it) as the powerful EF-3 tornado strikes Assumption Parish and continues as an EF-2 into St. James killing 2 and injuring dozens more. Several people in those areas have credited the warnings for saving their lives. Had the radar in Slidell gone down just a little earlier, those warnings may not have been issued and the casualty numbers could have been much higher.

The Solution?


Chances are if you've made it this far into the article, you now understand and agree that south Louisiana needs better radar coverage. So why don't we have it? As is the case with so many things in life, much of it boils down to dollars and cents.

The installation cost for a new Doppler radar runs into the millions of dollars and some estimates indicate annual maintenance and upkeep can run into the hundreds of thousands of dollars.

Additionally, you need a team of qualified personnel both to operate and maintain a radar at any given location. That's why our nation's Doppler radars are most often collocated with National Weather Service offices.

In the wake of an undetected tornado near Charlotte, NC in 2012, a bill was introduced in Congress that would require a radar within 55 miles of cities with a population of at least 700,000. If passed, that solves Charlotte's problem, but certainly not ours in south Louisiana.

I do think our Louisiana congressional delegation should lobby for a new NEXRAD installation near or around Baton Rouge, but if that doesn't occur or fails to gain traction, there are other possible solutions.

The University of Louisiana at Monroe (ULM) recently completed installation of a Doppler radar that was funded by a $3 million grant through the Louisiana Governor's Office of Homeland Security and Emergency Preparedness (GOHSEP). That radar is set to go online any day now.

New Doppler Radar at ULM. (Creidt: ULM)
ULM's radar is a great step forward and one that I believe was needed in northeast Louisiana. It also gives an added boost to ULM's Atmospheric Sciences program which has produced a number of very talented and successful meteorologists through the years.

But what a shame that an area centered around our state's capital has some of the poorest radar coverage in our state and is also among the poorest in the U.S. What a shame that an area that houses so much of the state's industry and its 'chemical corridor' -- an area that one could argue is very vulnerable to severe weather -- has very little reliable coverage.

The map below shows the number of lightning flashes per square mile from 2005-2014. If lightning is assumed to be a rough proxy for 'active' or potentially severe weather, it becomes clear that one would be hard pressed to find an area anywhere in the U.S. more in need of better radar coverage or where there would be better bang for the buck on a new installation.

Number of lightning flashes per square mile, 2005-2014. (Credit: Vaisala)

The way I see it, there's a real opportunity for our local and state leaders to lobby hard for an installation near or at LSU. The state's flagship university is home to a long-standing, well-respected Geography Department that houses a number of faculty with weather and climate backgrounds. I am currently a graduate student in that department pursuing my Ph.D., having already completed my master's degree at LSU in 2008.

LSU is also home to both the Louisiana State Office of Climatology and the Southern Regional Climate Center, an agency that collects and disseminates weather and climate information for multiple states in the South.

I believe a joint effort by the Dept. of Geography, the climate centers housed at LSU, LSU President F. King Alexander, the National Weather Service, and our congressional delegation could go a long way in trying to secure a badly-needed radar for Baton Rouge and surrounding areas. Not only could it be a life-saving device, but also an invaluable recruiting and research tool for multiple programs at LSU.

I know money is tight, but let's get moving on this. Lives could depend on it.