How Did It Happen: Record breaking Louisiana Flood August 2016

A record flood in Louisiana. Hundreds, thousands of water rescues. August. People are dying, houses are flooding. People flock to shelters to escape the rising water. The shelters begin to flood.

For most, the name “Katrina” probably creeps into mind. But no, this story isn’t about Hurricane Katrina. In fact, it’s not about a tropical storm of any kind. Instead, this is a story that went relatively untold on the national stage and happened literally this past week into the weekend with ongoing flooding even today (August 11th-present 2016). As storm chaser Brandon Clement puts it “For a lot of people, and I mean a LOT of people, this will end up being worse than Katrina.”

So, what happened? How did this event occur meteorologically and why did it seem to fly under the radar in the eyes of national media?

Well, a few things combined to create a perfect storm of media absence and meteorological catastrophe. The weekend of Aug. 5-7th featured an ominous storm on our computer guidance … we knew flooding was coming the next week … for Florida. The long range computer guidance was insistent on the storm drifting westward underneath a stout ridge, but at the time the flooding along the northern Gulf Coast wasn’t seen as being an upcoming disaster. Between politics and the Olympics dominating headlines, this storm having no “name”, and heading into a weekend, national media was notably quiet. As the early week came around, we knew differently. As Mr. Clement said, “The local news media did great. But the national media just seemed absent for the most part”.  Model guidance painted an ugly picture of a large area receiving around 20 inches of rain and a disaster was becoming more apparent. Below, you will see the GFS forecast earlier in the week, with the low pressure area moving away from Louisiana by early weekend. By midweek, the model shows that instead the storm would stick around a further 2 days before exiting. This resulted in a flash flooding event turning into a record shattering catastrophe.

early loop later loop

As the Weather Prediction Center (WPC) showed in these two Mesoscale Precipitation Discussions, extreme moisture content and convergence associated with the area of low pressure was evolving into extreme flash flooding, by Thursday evening the 11th.

MD 1 MD2

As the forecast became reality, the 20” forecast mark that originally concerned forecasters was realized. And eclipsed. In fact, the NWS in New Orleans reports that one station received 21.29” in approximately 1 day. And keep in mind it rained heavily (though perhaps not 21” a day, each day) for 3 days across this area.

A video from the Stu Ostro shows the radar presentation through this time period, as wave after wave of rain pummeled the area.

Tropical system / Louisiana flood

56-hour radar of relentless torrential rain from the tropical system which caused the Louisiana flood, from Thursday morning when the downpours shifted inland from the Gulf to when they finally diminished Saturday afternoon

Posted by Stu Ostro on Tuesday, August 16, 2016

So how much rain fell? Below is the radar estimated rainfall total. Areas in white saw over 20″. It should be noted that rainfall estimates by radar are typically underdone in heavy, tropical rains like this event, so totals likely exceed what is shown.
total radar rain

As this water vapor image shows, the storm had the upper level characteristics of a tropical cyclone. Had the storm been over water, in this very light shear regime with extreme upper level outflow, it very likely would have become a tropical storm. While this storm will have no name in the history books, it will be burned into the memories of those in Louisiana for generations.

water vapor

Meteorologist Logan Poole

Hurricane Hunters Tour Hurricane Earl

As Earl strengthened into a hurricane yesterday, meteorologist Sandy Delgado was given the opportunity to board the NOAA P3 hurricane hunter plane and give a birds eye view! Here are some pictures of that experience, courtesy of Mr. Delgado! I think this counts as storm chasing, right??

Sandy in plane

Sandy in the plane!


Live infrared and visible satellite


Radar image from the flight with observations overlayed

view of eye

View of the eye of Earl!

And here’s Sandy’s public tropical weather group on Facebook where you can see excellent near-daily updates of Atlantic Hurricanes past and present!

Meteorologist Logan Poole

Short Tropical Update: Aug 2 2016

A vigorous tropical wave has been on the cusp of  being designated a tropical storm for about 36 hours now. A hurricane hunter aircraft was scheduled to enter the storm yesterday, but due to mechanical failures the flight was scrubbed. Today’s mission is so far successful, with a possible center showing up in the latest observations. If this is confirmed by the experts at NHC, the storm will receive the designation ” Earl” and watches and warnings for a tropical storm impact in central America and the Yucatan Peninsula of Mexico will likely be needed.

center vis0-lalo

A tweet from the NHC confirms advisories are being issued on Tropical Storm Earl:

NHC Atlantic Ops (@NHC_Atlantic):

NHC will initiate advisories on Tropical Storm Earl with a Special Advisory to be issued before noon EDT.


Wxchasing Hurricane Season 2016 Update

Monday, August 1st, marks the beginning of a typically 3-month long period of heightened tropical activity in the Atlantic basin. The August-September-October (ASO) period accounts for on average 80% of the tropical storms in a given year and 86% of its hurricanes.

Credit: Landsea et al.

After a mostly quiet July, and indeed a mostly quiet decade regarding landfalls, understandably some folks are beginning to ask “Is it over already?”. While we don’t know for certain what the year will hold, we can continue to look at the various large-scale patterns that may drive future development this year.

To start, let’s “look out the window”. A strong tropical wave has emerged off the coast of Africa this past week, and looks to potentially develop into a tropical depression by this weekend. The National Hurricane Center gives this system a 40% chance of developing into a depression and that seems reasonable. As the system looks to transit the Atlantic, conditions will become less favorable and should the system organize beforehand, weakening or dissipation would be the most likely result. Recent modeling is a little more bullish on retaining a wave of low pressure through the Windward Islands, so should any remnant low make it to the western Caribbean, trends would have to be monitored for regeneration at that time as conditions for development may begin to improve.

A somewhat weaker wave further west will face similar conditions and will also bear watching as it enters the Caribbean or Bahamas later this week. The NHC gives this disturbance a 30% chance of development.  (Valid Thursday evening, 7/28)

Credit: NHC

One of the primary concerns for limiting tropical development is the intrusion of dry air from the Saharan Desert in Africa westward into the Atlantic. This Saharan Air Layer (SAL) can stifle the growth of robust thunderstorms that grow into tropical storms.

Credit: Cimms

This dry air coupled with the Intertropical Convergence Zone (ITCZ) and associated large scale lift being seasonally far south (around 5N latitude) has helped prevent tropical storm formation through July across the central and eastern Atlantic, though this is not at all unusual. That portion of the Atlantic Basin doesn’t often see tropical storm formation in July of any year.

By August, the central Atlantic becomes increasingly active and by September, storms are somewhat common as far east as the Cape Verde islands. By October, the action becomes confined well West for the most part once again. july




Credit: NOAA

The current forecast from remains unchanged:


We still anticipate an additional 11 named storms, 7 hurricanes, 4 major hurricanes,  2 hurricane landfalls, and 1 major hurricane landfall.

Dr. Jeff Masters points out in his blog that the western Atlantic is currently experiencing near record oceanic heat content (OHC), exceeding that seen in 2005. While OHC is not the only important aspect to an active season, it represents focused warmth in the tropics as well as increases atmospheric instability. This coupled with a departing El Nino and decaying warm-neutral ENSO pattern should combine with seasonal influences to increase activity across the basin in coming weeks.

Rest assured, more storms will develop during this period. And the threat to the United States mainland is still likely at some point during this season.

Check back next week when we’ll have another guest blogger to discuss storm photography!

Meteorologist Logan Poole



Guest Blog: A tour of the Pacific Crest Trail (PCT)

“What do you do if it storms?” is one of the most common questions you get asked while hiking. If there’s one thing to be learned about backpacking on The Pacific Crest Trail (PCT), it’s that mountain weather does not care what the weatherman said on the TV this morning or what is placed as the 5 day forecast. The tall mountain range that looms over central California known as the “Sierra Nevada’s” are 200 miles long and extend from Independence, California all the way to Lake Tahoe and have a mind of their own when it comes to weather. One day it can be bright and sunny outside. The next, a snowstorm is threatening your camp and you have to brace yourself and check food inventory to make sure you have enough food to be stranded for a couple days while you wait on snow to clear. Prime hiking time through this mountain range varies from June-August to allow sufficient time for snow to melt and make the trail passable.

Now, when talking about June one wouldn’t think you’d be talking about snow. But, when troughs make their way onto the coast of California it brings with it beneficial rain to California. But, with the colder air in the higher elevations that rain turns into snow and as hikers make their way through elevations ranging from 10, 11, 12, and 13,000 feet the temperature quickly drops and precipitation quickly turns into snow. Sometimes it wouldn’t just be snow either. Sometimes it would fall in the form of small hail. Which if caught away from the trees really hurts as it pelts against your skin as you try and hike faster to find some sort of cover. There’s no bracing yourself for it either. There is zero cell service with all carriers through the mountains, unless you get into town where cell service returns again. Occasionally there are high spots with a small ounce of cell service and you are able to check the weather. But, 9 times out of 10 you won’t know a storm is coming until you see clouds start filling in, or hear distant rumbles of thunder.

Here’s a crazy story about the drastic changes in weather: Word was spreading among hikers of an incoming snowstorm as we neared the bottom of Mount Whitney (the tallest point in the continental US). Elevations above 8,000 feet could see snow ranging from 6-12 inches! This blew my mind as I have never dealt with snow in June. Let alone a significant amount. As a hiker, this makes your heart drop. We knew we were beginning to run low on food, and simply didn’t have enough to stay on trail for an additional day, much less two or three. While out there every scenario goes through your mind. What if we get a foot of snow and are stuck here for a week? What if we try and get to the next camp but get caught out in the snowstorm unable to setup a camp? All scenarios that could very well happen. There must have been about an hour long discussion among a circle of hikers (including us) who were trying to figure out what the best option was. There were many other hikers also running low on food, as we were only a couple days away from town and were going on day 5 of hiking. All of the hikers came to the conclusion that we’d camp at the bottom of Whitney, and get out early in the morning before the storm was forecast to hit. Even lying in bed that night my stomach turned as I didn’t know how this was going to play out. Then, at about 2am I awoke to a loud rumble outside. “Is that thunder?” I thought to myself. There’s no way, today was supposed to be clear. The system wasn’t supposed to arrive until tomorrow afternoon.

A few minutes  later I saw a bright flash and a crash of thunder followed right after it. Talks of a foot of snow later that afternoon and it was thunderstorming. Not knowing how bad the storm was and where it was going because of no cell service is like walking across a highway blind folded, wondering if/when you’d get hit by a car. Now we had to ask ourselves “Well does this mean the system arrived early?” “Does this mean no snow later today?” “Is it going to storm all day leaving us stranded?” There came a point where I had to force myself asleep and just see what I awoke to that morning. Which shockingly, when I woke up was blue, sunny skies.



My hiking partner Brennan Jontz (above) and I, after collecting our thoughts, wondered if the snow would verify at all now. It stormed, and now its clear. It didn’t take us 10 miles as we began climbing up our next peak to camp to realize we were wrong.



Just like the flip of a switch, clouds rolled in and after crossing a river the mountains began to fade away behind a thick plume of clouds and snow. We made it to camp right as it began to snow on us after averaging about 3 miles per hour, an almost unmaintainable pace of progress in the mountains.

It almost makes you respect weather more being out in the wilderness. Because one can’t simply jump in their car and ride out the storm. It takes careful planning and searching. You can’t rely on a forecast either. Many days on trail that were forecast to be sunny all day, turned into stormy weather later in the day. It’s all about being situationally aware. Always have a plan even if nothing is supposed to happen. If we knew rough weather was coming, we’d have a set time to make a peak. If we didn’t think we were going to make it, we’d have no choice but to setup camp for our safety.

Before the Sierras though on this long hike is the desert portion. This is the first 700 miles that you have to hike through and while the weather wasn’t as extreme as the mountains it was very unique. During the day temperatures would climb to the mid 90s and some days even into the lower 100s. But then at night, temperatures would drop into the lower 40s. I had never been to a place with this extreme of a temperature drop at night. Normally around the midwest (where I am from) at night, temperatures will drop 10 degrees (maybe 15 at most). But out in the desert, there would be a 50, nearly 60 degree drop. Which made for great sleeping weather. However, as soon as that sun came up temperatures rose as fast as they dropped and you’d quickly find yourself sleeping in an oven. There was also a time crunch in the desert. You still had to get your 20-25 miles in but due to the heat you’d be force to hike all night and attempt to sleep a couple hours during the day (Which was hard to do in the heat) and do it all over again. Or the option we preferred was start hiking around 4-5am and stop at noon (hopefully under a tree if you could find one) for shade and ride out the heat until about 4pm and keep hiking. Sounds nice just sitting for a while after hiking all morning right? Nope. While you’re in the shade, it hardly takes away any of the heat as you roast.

While the PCT was beautiful and the adventure of a lifetime it all came with a price. It wasn’t all relaxation, fun, and games. There was so much we had to power through and each section held its own set of challenges in the 910 miles that I hiked of it. I will say, I’m much happier to be chasing in a vehicle again as opposed to on foot.

Aaron Rigsby,
Hiker, Storm Chaser

What causes heat waves and what is the “ring of fire” ?

It’s hot and getting hotter across much of the United States. With the National Weather Service forecasting high temperatures to reach the highest so far this year across much of the country, understandably folks are wondering just what is driving these unbearable temperatures. This article will discuss that as well as touch on other weather that comes as a direct result of this.

“It’s not the heat that gets you, it’s the humidity!” Most folks have heard that phrase in some for or fashion. Some shake their heads disapprovingly (looking at you, Desert Southwest!) as others nod in agreement. The fact is, it’s both. If it is 120 degrees Fahrenheit outside, no matter the humidity, you will feel it. It’s hot! On the other hand, if it’s very dry but “only” 95 degrees Fahrenheit, your perspiration will readily evaporate, cooling you. It won’t feel so bad for most folks. If, however, it is 95 and there is a lot of water vapor already in the air (high humidity), the sweat on your skin doesn’t evaporate so well and you will feel very hot indeed! (And wet, sticky, and unhappy!)

To relate these two, most agencies rely on the “Heat Index”. That is, an Index that accounts for both heat and humidity. With large swaths of the country expecting heat index values over 110 degrees Fahrenheit, it should come as no surprise that the weather community is becoming concerned. After all, heat and cold related injuries and deaths account for more weather deaths than any other weather phenomenon in the country! It’s no joking matter! Heat kills, and we are getting a serious dose of it this week. Further, it’s important to remember that Heat Index values (and temperature in general) are values IN THE SHADE. And often 2 METERS ABOVE GROUND. In the summer in particular, the actual skin temperatures of the earth can be much higher than what your local meteorologist is telling you the temperature is. Air cools very rapidly with height in the lowest few meters of the atmosphere. In the chart below (in Celsius) the data from a weather station in Yuma, Arizona shows that the reported temperature at 2 meters above ground level (AGL) is 95 degrees Fahrenheit. Contrast that to the skin temperature at ground level of 102 degrees Fahrenheit!

YUMA surface tempYUMA skin temp

Since most (or all) of your body is below 2 meters when you are walking around at ground level, it’s important to make these distinctions. And also note that these effects are made worse when in direct sunlight and/or during peak daytime heating. Notably, I selected a site in a desert to illustrate my point better, and certainly your “experience may vary” depending on your location and what materials the ground is made up of as well as how much moisture it contains. Regardless, these effects will still result in noticeable effects even in the best of scenarios. To limit this added heat, it’s useful to avoid dark, dry surfaces (like pavement) as these absorb and radiate heat the most efficiently. As most of you are aware, pavement in the summertime is not nice to bare feet!

Now, this may be on a tangent but it needs to be stated. Do you know what else typically has bare feet and is even closer than you are to the ground where it’s VERY hot ? Your pets! Heat waves are notorious for causing foot damage to cats and dogs and other animals so please do be mindful of the added stress on your animal as you implore them to go on walks with you. Early morning is your best bet as the ground is the coolest and in most places the ground will be cool to the touch and safe for walking animals. As the ground heats up in the day this changes very quickly and it’s not unheard of to have animals severely injured. Be careful!

Okay, so it’s going to be hot. It’s going to be hottest near the ground and in the sun. Take precautions. Here’s a link to OSHA’s guidelines for working in heat, and do note these heat index values are IN THE SHADE. If you are in the sun, tack on 15 degrees!

If you are content with the “what” you can stop reading here. If you want to know the “how”, here it goes!

Heat waves are caused when large domes of high pressure force air to sink across a large area. These domes of high pressure result in weak winds which prevents the atmosphere from transporting cooler air into the warm areas and results in sinking air that removes clouds and rain as well as causes air to warm as it sinks. Below, you see how on Thursday a large ridge of high pressure will dominate much of the country.


This comes at a time where there is significant agricultural efforts ongoing particularly across the Cornbelt, but elsewhere too. These large farms with mature crops release a lot of water vapor into the air through evapotranspiration. While this is a cooling effect, as water is evaporated into the atmosphere, this will result in dew points perhaps into the lower 80s in spots particularly across Iowa and into Illinois, resulting in heat index values over 100 degrees Fahrenheit over a large area and peaking in excess of 120 degrees across portions of Iowa (in the shade!).

heat index Thursday

Around the periphery of this high, across the High Plains, Great Lakes, and into New England, significant heat and moisture with weaker ridging will promote robust thunderstorms. This is known as the “ring of fire” as storms are suppressed beneath the oppressive ridge of high pressure, but form on the edges and move clockwise around it. This can result in significant growth of storms as they feed on the energy produced by such intense heating, resulting in damaging wind storms and potentially Derechos. While that isn’t currently forecast to happen, it IS Derecho Season so that will need to be watched in coming days. ( )

Next week, will feature a guest blog, followed by a discussion of the impending “ASO” (August-September-October) period of the hurricane season, the most active period in almost every hurricane season!

Thanks for reading,

Meteorologist Logan Poole

Hillary Clinton Stole My Video

“I`m running for president because everyday Americans and their families need a champion, and I want to be that champion. I want to make the words ‘middle class’ mean something again. “

-Hillary Rodham Clinton, Democratic Nominee for President.


This article is hard to write. I never wanted to have to discuss this. There was a time in my life when I had faith in my country, in its government and leadership. I still have hope that that faith can be restored. For now, my faith is broken. Allow me to explain:

On December 23rd, 2015, a devastating EF-3 tornado moved from near Clarksdale, Mississippi to near Sardis, Mississippi. As readers of this blog ( will know, I have made a career out of chasing storms. This particular storm was scary. Certainly all tornadoes are dangerous, but when you can see a violent tornado moving very quickly toward populated areas it is an entirely different fear than seeing a tornado in the middle of a field. I’ve seen that happen time and again, and each time I can feel my soul crushed a little bit more. In that regard the “Sardis Tornado” was no different.

What makes this day stand out in particular is two-fold: I witnessed firsthand as the State government put people in front of a deadly tornado, then I watched as the Federal government used that same situation to push an agenda through deceit and theft.

On that day in December, I thought I was about to see people die because of the carelessness of the Mississippi Highway Patrol (MHP). In the video below, a line of traffic is struck by a violent tornado. What isn’t shown is why there is a line of traffic in the first place. This traffic jam, barely moving in the video, had moments before been stopped by MHP officers. I rushed to an officer as soon as I was on scene. I showed him using a radar display where the tornado must have been: it was on a collision course with the traffic. I asked him to please let the traffic move. I was told that it was not their decision to make. I pleaded with him. I told him that the tornado was headed straight for them. I begged him to contact those who had the authority to let traffic move. I was told it was “not his problem”. I was at a loss. I shouted some expletives, dared him to come chase me as I peeled across the median. It was a last ditch effort: I figured if he stopped blocking traffic to pull me over, traffic could move. No such luck, he didn’t follow me. When the officers realized they themselves were in danger, they left the scene and the result is what you see in the video: traffic crawling in a futile effort to flee. As all of this was occurring, I was sending a live stream to national networks and thousands of viewers online. This encounter with the MHP was no secret.

A friend of mine, Mr. Bryan Wilson, took these photos of a house after the tornado had cleared the area. The debris you see in my video is from that house.

12416007_1662036680716371_1243237140_o 12432555_1662051077381598_2083387668_o 12435663_1662040214049351_795207099_n

Tornadoes are fickle. Sparing one house while leveling the neighbor’s sometimes. Why this happens is mostly known, but predicting which houses will be destroyed is impossible. The tornado did in fact move through that line of traffic that day, but to my knowledge no one was killed. Vehicles including a tractor trailer were overturned on the highway. Call it a miracle, call it luck, but what you can’t call it is a good job by the MHP.

Okay, I get it, “But Brandon, that happened over 6 months ago, why bring it up now?”

That’s a fair criticism, but there’s been some recent developments.

Look familiar? It should. The tornado portion of that political advertisement is from the same video I shared above. I was alerted to this apparent theft by a friend of mine (the same Bryan Wilson who provided the imagery above) in March. At the time the only thing we had to go on was that there was an ad somewhere that had my video in it from Mrs. Clinton’s campaign. Through the rest of March and April, I investigated that tip. Another close friend and chase partner, Mr. Logan Poole, was able to find both the Youtube video listed above and information on how many times and in what markets the video had been played on television. It was a popular ad piece, with thousands of online views and hundreds of airings on television.


Once I had this material, I reached out to the Clinton campaign. To their credit, they didn’t deny it was my clip. They stated they had received the piece from a particular national news outlet. I don’t claim to know everything about video rights and exclusivity, but my video broker, Mr. Brett Adair, at (LSM) does. I contacted Brett who assured me that all sales of that video were limited licenses with no redistribution rights. Further, when Brett reached out to that news outlet that the Clinton campaign claimed to have sourced the video from, they denied all involvement and were quite upset that it was said they did such a thing.

That being the case, I can only assume the video was stripped from the LSM Youtube page, the same video I posted above. With all water marks cropped out. When confronted with this information, initially it seemed the campaign would be willing to settle the theft. Perhaps it wasn’t intentional, I thought. Perhaps the misdeed was done by someone without authority, and now corrective action was being made. That would be a somewhat typical evolution of my dealings with copyright infringement. I produce a lot of content and this sort of thing occasionally happens. People can work it out when mistakes are made.

Then things changed. Mrs. Clinton secured the Democratic Nomination, and looked to be on the inside lane to the White House. Suddenly, stealing from a middle class white man in Mississippi wasn’t that big of a deal to them anymore. They told me to just release the story. They had no further interest in compensating me for the stolen work.

The voiceover across my video is as follows:

“The world a president has to grapple with: sometimes you can’t even imagine. That’s the job.”

Right, a world where people you are supposed to trust put innocents in harm’s way? Where those who aim to lead the United States lie, cheat, and steal? That world? Because, yes. Now I can, in fact, imagine that.

Brandon Clement

To license any of the content contained in this article legally, unlike some others …

Contact Brett Adair at

*Piece edited by Mr. Logan Poole

Weather Modeling: Can a model’s wrong forecast be a good thing?

If there was one idea in meteorology school that is the most mind boggling (and I assure you, there are many that ideas that boggle the mind!), for me it had to be the idea that a model being wrong can be a good thing.

Up and until you take a course on Numerical Weather Prediction (NWP), meteorology coursework is designed to teach you one thing and for you to understand that with every fiber of your being: the atmosphere is chaotic. Very tiny changes in the atmosphere go on to make very, VERY big changes down the road. This is one reason why it is important that meteorologists study as much math as we do, because you need to be as close to exactly right as possible. Because if your math, and thereby your forecast, is off by more than necessary for tomorrow, that error is going to magnified again and again through your forecast. By day 5, you will be horrendously off!
Some meteorologists decades ago even supposed that a wing-beat from a butterfly could one day be in some way responsible for a tornado on the other side of the world. This gave rise to the term “the butterfly effect”. In short, it’s a big, BIG deal to not mess this stuff up.

And then the poor unsuspecting meteorology student is thrust into a course on weather modeling and informed that sometimes a computer guess is better than real, observed data. This is the point at which hands become raised and stomachs feel sour. How could it be that a real, 100% factual observation could do more harm to a computer model were it to be used as a starting condition than the forecast from some other model? Well, it took me a while before I could bring myself to accept this to be true. And it’s really fairly simple and has HUGE ramifications for how a model works.

So: Data Assimilation.

Before a model can begin crunching numbers, it has to first have some numbers to work with. Simple enough. These numbers come strictly from real observations from that same (or a different model) model’s forecast for that time period from its previous run. Why is that you ask? Why not use real data instead? Well, there are almost always far more data points/grid points on a computer model than there is real data. If you have a surface observation once every 8 miles, but your model calculates a value for temperature every 4 miles, you can see why it would need a starting number at least once every 4 miles. The best way to get that is to use previous modeling.

After that, the model adjusts those starting values to move toward real observations. That is, it doesn’t outright replace those with real, factual data. Why not? Well, because if it did, not only could it have one point that looks vastly different from its previous (incorrect) forecast, but that point may not be representative of an atmospheric feature the size of which the model can understand. (Note this graphic is dated, but the ideas are still the same!)


Take for example a car with a shiny, candy apple red paint job. Except one spot. That spot, the size of a pinhead, has chipped and is now discolored. You of course love this car and want to get it touched up. You take it to a shop and return a week later to find they have painted the entire car the color of the chip instead of red! The chip color was real and factual, but didn’t represent the actual larger car. This is analogous to an observation that could be real, but perhaps was affected by a weather feature too small to be properly understood by the model. Perhaps a weather balloon was launched into a rain shower for example. The model, if it took the observation at face value, would “paint” an area much larger than that one rain shower as being equal to that observation. If the model can only “see” a square of 12 kilometers x 12 kilometers (144km^2), that tiny rain shower that maybe was only 4 kilometers x 4 kilometers (16 km^2) suddenly looks at least NINE times as big to the model! Some cases could be even worse with poorer resolution and data “smoothing”.

So, okay. We understand that it’s important to think about real observations in terms the model can understand. What does this mean for our forecast? We accept that the model has a fundamental error in its initialization. That’s going to result in increasing errors as time goes on, but it’s better than what would happen if we let that “real” observation get it in there and fooled the model altogether.

The most important result of this carrying-forward of bad data, though, is that it can sometimes result in good data being rejected because it disagreed too much with a model’s poor forecast! This means that a model that is doing badly will continue to perform badly. It may take several model runs before this bad data is corrected enough by real observations to start performing well again. This is often why a model will do poorly for an extended length of time before performing better. This is a trend that is important and is one of the things a forecaster can look for: how closely did the models “00 hour” forecast (it’s initialization) match up with the real observations? If they don’t match well, it can be taken as a sign that the model is not going to be performing well.

Further, this also means that a model that runs at a higher resolution could potentially expel errors more quickly. A model that runs more often will also have increasing opportunities to remove bad data. The inverse is also true, however. If a model has many runs in a short period of time, the duration of bad modeling may be shorter, but these periods may be more common as more opportunity for poor performance is given simply by running more often.

I hope that was clear, but I’ve included a few interesting links that hopefully will further illustrate these points. The first is the model diagnostics page from NOAA. It will give you an idea of what the NWS professionals think about which models are initializing the best. The second is the UCAR METED module on data assimilation. I recommend that particular module for meteorologists and enthusiasts alike as it is very informative regarding modeling!

Check back in next week when we welcome another guest blogger!

Meteorologist Logan Poole

Guest Blogger Mr. Ken Johnson: From El Nino to La Nina

One of the strongest El Nino weather patterns on record formed during 2015 and lasted into the spring of 2016. El Nino is where the water in the Pacific Ocean off the south American coast is warmer than average. The El Nino pattern we just observed is similar to the one experienced back in 1997.


El Nino’s impacts were felt across most of the United States. It brought much above average temperatures to a good part of the country. Above average rainfall was observed from California into Texas and parts of the Gulf coast states. The southeastern U.S. and Florida also received above average rainfall amounts. On the flip side of that, the northern U.S. saw warmer than average temperatures, along with below average rain and snowfall. El Nino creates a stronger subtropical jet stream, bringing weather systems out of Pacific ocean into Mexico, across Texas, the Gulf of Mexico, and into Florida.


One trend noticed was an increase in tropical activity and hurricanes in the Pacific ocean off the Mexican coast. Moisture from these systems was picked up by the upper level winds associated with the subtropical jet stream, producing heavy rain across the southwestern states and Texas.
Prior to 2015, most of Texas was in the worst drought on record. Ranchers were forced to sale off livestock, while lakes, rivers, and streams dried up. It was so bad, that the city of Wichita Falls main source of water, Lake Arrowhead and Kickapoo dropped below 20-percent capacity. This forced the city to recycle used sewage water for drinking purposes. A local church created signs that read, “pray for rain.” That’s one prayer that was answered.


The drought ended when more than 17-inches of rain fell across parts of Texas during the month of May. Lakes, rivers, and streams that were dry, were now raging with flood waters. Lakes filled to capacity for the first time in more than a decade.
The wet weather pattern lasted into the Spring of 2016 with more historical floods. The increased storm track across the south also lead to an increase in tornadoes in Texas and Florida, while places further north saw a decrease in tornado activity. The increase in strong winds further south also led to an increase in wind shear across the Atlantic tropical regions with below average hurricane and tropical storms. By the end of the 2016 spring, the water in the Pacific Ocean had significantly cooled off and El Nino was officially dead!

The cooler trend is forecast to continue into the summer and fall of 2016, leading to the opposite of El Nino, La Nina. La Nina is where the water in the Pacific off the south American coast is cooler than average. This leads to a stronger storm track center further north across the U.S. with wetter and cooler than average conditions in the northern and central U.S. The northward trend of the storm track, tends to favor warmer and drier than average weather in the southern U.S. Much like the current weather patterns, the strong El Nino in 1997 was followed by a very strong La Nina in 1998 and 1999. Drought conditions were experienced across a good part of the southern U.S. A La Nina weather is likely to blame for the historical drought in California and Texas prior to 2015.

Weather computer models are forecasting a strong La Nina as we move into the fall and winter of 2016. This will likely favor drier than normal weather from California into southern U.S. with an increase in temperatures.


Warmer weather may also help foster the chance for more late fall and winter tornadoes in the south central states from Texas and Oklahoma into Arkansas, Tennessee, and northern Mississippi. Weaker winds across the tropical regions of the Atlantic will tend to favor a more active hurricane season. For more on this up coming hurricane season and how it might be impacted by La Nina, make sure to read the excellent write up from Meteorologist Logan Poole.

Expect drier and above average temperatures across the southern U.S. wetter and cooler across the north. We may see a return minor drought conditions in Texas and continued exceptional drought across California.

Ken Johnson, KAUZ
(CBS) Chief Meteorologist

Ghost storms: What are they and what do they mean?

Our scheduled guest blogger, Mr. Ken Johnson, chief meteorologist at KAUZ 6 in Wichita Falls, TX will join us later this week, discussing what the transition from a strong El Nino to La Nina might be like. With model guidance, observations, and climatological reasoning all indicating a transition to La Nina, that blog will touch on topics that could prove very useful later this year.

In the meantime, isolated active severe weather along the northern tier of states continues as forecast by The Storm Prediction Center (SPC) has highlighted an area stretching from near Chicago toward the east as having a moderate chance for severe weather, including tornadoes, on Wednesday. The primary threat that day appears to be the risk for strong, damaging winds over a large area. As our last article discussed, this is the time of year for that area to take seriously any severe thunderstorm warnings that might be issued.

In the tropics, a weak tropical storm Danielle has come and gone in the Bay of Campeche, tracking into eastern Mexico. This storm was very short lived, and produced mostly rain and breezy conditions as its primary impact.

In the extended range (10-14 days) an interesting feature has been showing up on the Global Forecast System (GFS) model. After around 7 days, any forecast specifics become very vague and uncertain, but we can draw some useful knowledge at this range if we look at trends and consistency. This article’s primary focus will be to describe how best to utilize these very long range models when discussing tropical activity.

Often, models at very long ranges look somewhat similar to a climatological expectation. It’s not at all unusual for these models (particularly the GFS, which regularly forecast to 384 hours out) to churn out hurricane after hurricane that simply never verify. These storms are colloquially known as “Ghost Storms”. Ghost storms are the bread and butter of clickbait artists and are almost always available for an armchair meteorologist to point at and say “Hey, you guys, Katrina is coming back!”. I assure you that is not the purpose of this article.

Most often, these apparent ghost storms are fleeting. Here one run, gone the next. Other times, these storms seem to “hang out” at a certain timeframe beyond initialization, never actually getting closer to occurring. Often, a storm will continue to crop up at 12-14 days out.  It may appear on one run of the model, but as days pass, it somehow is still 12-14 days out. When either of these two things happen, you can be fairly confident that the solution is pure fiction. Even rarer, sometimes these storms appear at the extended range and come closer as you would expect in a “real” weather forecast. These are the most important to watch and the trickiest to forecast. These are the ones where the chance for a real storm may be possible.

So, let’s look at the GFS’s forecasts. Remember to look for consistency, and an approaching timeframe as discussed above:

16 0z 16 12z 17 0z 17 12z 18 0z 18 12z 19 0z 19 12z 20 0z 20 12 z 21 0z

Consistency: The GFS has trended stronger for the most part through time, but shows at least troughing in the Gulf of Mexico on each run. That’s good consistency at this range that there will at least be some general storminess in the Gulf at that range.

Timeframe Concerns: From the 16th through the 18th, the GFS struggled with timing. The storm was seemingly drifting between 348 hours (14.5 days) and 384 hours (16 days). By 00z on the 19th, the timeframe became accelerated. From yesterday into today, the storm stepped forward in time further, from 324~ hours (13.5 days) to around 288~ hours (12 days). This isn’t the scenario we want in order to become really concerned about a system, but it’s enough to spark interest.

Once we’ve decided that a storm is at least plausible (though certainly not likely at this stage), it’s prudent to compare to other guidance. Most extended range models (GEM and ECMWF) only go out to around 240 hours, so that doesn’t give us much help at around 288 hours typically. However, with tropical systems, development occurs on the order of days typically so if we look at the end of the run and compare it to the GFS at 240 hours, we can see if the impetus for potential tropical development exists across guidance.

00z gfs 21 mslp 00z 21 euro  mslp 240 hours 00z 21 cmc  mslp 240 hours

While the GFS is the most bullish at this stage, showing strong cyclonic curvature near the Yucatan, the GEM (CMC) and ECMWF (Euro) both show cyclonic curvature in the same general area as well as increasing wind speeds. This is indicative of a tropical wave. While that is no guarantee of development, a majority of tropical storms and subsequently hurricanes develop from the influence of these waves. These waves are also long lived, propagating from northern Africa. This increases the confidence in the forecast of wave positions compared to other storm modes because the feature already exists that is being tracked, even at 10 days out. So, we can say with some degree of confidence that in 10 days’ time, a tropical wave will be centered somewhere from the far western Caribbean to as far west as the Bay of Campeche. For now, we have little else to go on as far as tropical development, but we can look at the synoptic pattern to see where a storm located in that region might go were it to develop. Below, the ECMWF, GEM, and GFS 500 mb anomalies are plotted. This helps illustrate potential steering currents.

00z 21 gfs  500 mb anomaly 00z 21 euro  500 mb anomaly 00z 21 cmc  500 mb anomaly

All three models show significant ridging extending from the Southwest into the northern Plains, with additional ridging off the east coast. Troughing is shown across the Mississippi Valley and eastern states, which would act as a pathway for a storm to drift northward. This trough would probably lift north and east beyond 10 days, potentially closing that path, but should a storm drift northward and become captured by the trough or should the trough linger longer than is typical, impacts in the United States would be possible.

Beyond this, we should recall the underlying conditions for development. Are the seas warm enough? What is the expected shear across the basin forecast to be like this year? Is there significant dry air already ongoing? How have storms in that region fared thus far?

In the post discussing the hurricane season, we saw that seas were warm, wind shear should be below average, and that significant dry air intrusions were so far absent. We also have seen 2 storms (Colin, Danielle) develop in that same general vicinity. And as we discussed in the seasonal forecast blog, this can be an indication of genesis and track of later storms. This informs us that the underlying conditions for development should be more favorable than is typical.

After analyzing the data, my conclusion would be that the first 5 days of July are at a heightened risk for tropical impacts across the Gulf Coast. We can be confident that lowering pressures in the Gulf are likely during this time frame. Do note, however, that a tropical storm or hurricane landfall is still NOT likely.  With only 1.7 average hurricane landfalls per year ( and the assumption that the vast majority of these occur within the 6-month hurricane season, that means that any 1-week stretch (assuming equal chances through the season, a poor but workable assumption) has around a 7% chance of seeing a hurricane landfall. That being the case, with the data we have available from observations, climatology, our seasonal forecast, and model guidance there is around a 10% chance that a tropical storm or hurricane impacts the US during this time frame. And due to that a 90% chance that the week goes landfall-free.

In short, don’t believe the hype, but do be paying attention! Should circumstance change, I will either edit a note below or issue a new blog update. The post above will remain in its original form as an educational tool on the analysis of “ghost storms”.

Be sure to check back later this week for our guest blogger! I’ll have a new post of my own up sometime next week, or sooner if the weather dictates. Have a great week!

Meteorologist Logan Poole

Storm chasing, live streaming and weather extremes