July 2018 GOES-17 update

GOES-S Logo

GOES-17 is looking up, as engineers continue to work on the malfunctiioning cooling system from 23,000 miles away. Logo courtesy: NOAA

Hello everyone,

A few months ago, I wrote about the issues with GOES-17, which is due to be our western satellite. To recap, the cooling system which keeps the camera and sensors aboard cool is working well below performance standards. This causes the sensors to overheat and deliver very noisy or useless pictures. The problem occurs on GOES-16 as well, but is nowhere near as bad, and is not causing any issues that NOAA can see at this point.

What exactly the problem is remains unclear. It is possible that GOES-T, destined to be launched in 2020, could be delayed, but nobody is willing to bet that until they can figure out what is happening. And, without cameras, it’s hard to figure out what is going on.

Even so, there are some measures NOAA can take that has improved the cooling system on GOES-17 some. And, it has had a significant effect. Per NOAA, the estimated performance issues, and this is all preliminary, are:

13 out of 16 ABI channels available 24 hours a day, with 3 channels available 20 hours a day near the winter and summer solstices;

10 out of 16 channels available for 24 hours, another 3 channels available for 20 hours, and the last 3 channels 12 hours before the spring and fall equinox.

So, that is a lot better than before, but obviously not optimal. In the meantime, design changes will probably be forthcoming for GOES-T and beyond for the cooling systems, and again, it’s too early to tell if these will delay the launch of those satellites. But it’s still good news for GOES-17. In short, this means that basic visible, IR and water vapor channels should be operational nearly all or all of the time.

Additionally, some mechanical slight-of-hand and additional tweaks could cause even more availability in the future, but we will have to see how this pans out.

Expect the first regular images to be sent this fall in test mode, including AllisonHouse Maps.

A “sucky” historical tornado season. Now what?

It has been a very slow year for tornadoes. We break down the numbers, but also remind you that tornadoes are just one aspect of severe weather, and we are entering the heart of severe weather season, and the start of hurricane season, now. Image courtesy: SPC

As of July 11, 2018, we are having an incredible tornado season. An incredibly low number of tornadoes, that is! Let’s review where we are, why we are, and where do we go from here.

First off, the spring pattern exhibited a “La Nina”-ish type pattern whereby the jet stream would crash into the lower 48 states from the northern Plains, to the Carolinas, and then offshore. This caused a couple of interesting things.

First, it caused a trough in the jet stream in the far western U.S., giving very welcome rains….until the flooding started. Cities literally got partially covered with mud, causing fatalities. The eastern Rockies remained dry, helping to set the stage for wildfires now. But in the Plains and Great Lakes, it was unseasonably cool to cold.

Northwest flow from Canada shut down the Gulf of Mexico, and kept many systems from developing and coming into the Plains. And when they managed to do so, there was so much cold and dry air in place, and with the Gulf scoured of deep moisture, even intense cyclones produced amazingly few tornadoes.

As Sam Lillo noted on Twitter, as of July 8, we only had 63 tornado watches issued by the Storm Prediction Center. Most of the time, that number is reached in March or April, and in some cases, early to mid-May. But not July 8! But this year, it took us that long to get to that point.

And then, in Late May and early June, it happened. The jet stream roared northward to the Canadian border, and flow across the central U.s. was weak. With weak shear and marginal instability profiles, only a few tornadoes were big and photogenic, like Bennington, KS on May 2 (really? Really?). You had to be at the right place at the right time, and many storm chasers missed that one.

So, how are we doing compared to, say, last year? Let’s check out a couple of benchmarks.

Number of days with at least one tornado across the U.S. (preliminary data for 2018) in:

…………………..2017 2018
January              10     4
February             5      6
March                14     13
April                   23    10
May                    27    27
June                   22    27

Now, by looking at this, you would think that April was pretty “dry”, but May and June were active, and that the primary tornado season was pretty typical, with a high number of tornado days in June. But the total *numbers* of tornadoes compared to, as an example, last year are staggeringly low:

Number of tornadoes each month (preliminary data for April 2018 onward):

…………………..2017 2018
January             142    15
February           115    48
March                176    55
April                   218  129
May                    299   89
June                   168   30

And here’s where 2018 went off the rails and into the ditch for severe weather enthusiasts. Thanks to persistent cold air intrusions from Canada, Alaska and, a few times, from the North Pole and Siberia through the spring, the necessary ingredients for tornadoes didn’t show up often, and when they did, they were marginal or very marginal situations at best, resulting in a handful (or less) of tornadoes, and very weak at that.

So does that mean that storm chasers and spotters have nothing to do for the rest of the year? Hardly. We’re entering derecho season. A derecho is a long-lived, large line of severe thunderstorms that produces widespread, significant wind damage, with 70-90 MPH gusts or even higher. For those, we need to see decent moisture in the ground for the crops to “sweat” it back into the air for these lines of thunderstorms to use. And, while there is modest drought in some areas, many areas have adequate moisture, or even too much. And, as of this typing, we have our third named tropical system, Chris, spinning inbetween North Carolina and Bermuda, has 105 MPH winds. While it won’t hit the U.S., hurricane season has started early this year, and it will be very interesting to watch as time goes on. You bet I’ll be watching, and you should be as well.

MRLE, part 2: the good, the bad, and the….potentially ugly?

Put on your thinking caps and put your flamethrowers away: this new radar scanning strategy could pose issues for the NWS and emergency managers! Image courtesy; SPC

Before you read this article, read Part 1, here:

Do you MRLE?

So, having said that, what does this mean for you? As a National Weather Service (NWS) meteorologist? As an Emergency Manager? As a storm chaser or spotter?

First, for ice and snow, and for just plain old rain showers, this scanning option generally won’t be used. MRLE is going to be used for severe thunderstorm events, and probably most commonly used for higher-end events, or events that involve a large area (think Mesoscale Convective Systems, or MCS’s, or Quasi-Linear Convective Systems, or QLC’s, or lines of thunderstorms).

Now, here comes some speculation based on facts, and my experiences with the dual-pole radar, and the SAILS2 scanning strategy. I think I’m right on this, but I would love feedback, either yea or nay.

We are going to see 34 second updates for 4 low-level scans in this mode. You’ll see them back-to-back halfway through the volume scan. What I think we’re going to see happen, as a result, is this: we are going to “see”, or catch, a lot more of these EF-0, very short-lived (2 minutes or less, most under a minute) tornadoes in QLCS’s, MCS’s, and derechoes (and even in weak or imbedded supercells). For all of the hype TV stations have with their own doppler radars, including those getting scans every 30-60 seconds, some of them MAY have had a case for saying that we can see tornadoes when the NWS couldn’t…but only for the very small ones, I realize with the SAILS2 scanning strategy, we’re seeing things the NWS couldn’t possibly have seen before, because up until recently, they could only get imagery updated once every 5-6 minutes. With SAILS2 and 2-3 minute updates to the lowest tilts, we now know some of the things the NWS was missing before. I assert that the new MRLE scanning strategy will allow us to see more miso-circulations (and other things) we couldn’t see before, that last a minute or less. And this is going to be a problem in some situations. Why? I see at least two issues:

1. Damage that was classified as microbursts in the past now might be understood to be an EF-0 tornado. These things may have only been on the ground for 10-60 seconds, and they might have only taken down some large branches for less than a block, or they could cause more damage because of lower quality construction of newer buildings (more on that in a moment). We used to just shrug and call them microbursts. Many of them are, but some of them may be brief tornadoes. In the grand scheme of things, this is a very small event that affects a relatively few people. But those people will be mad that a tornado hit their tree and dropped a branch on their car and there were no sirens, and the dreaded “there was no warning” phrase will be uttered in the media.

2. If you are an emergency manager, what do you do if you are not under a tornado warning, but suddenly get reports of damage on 2-3 blocks of your city, you can’t segment out the sirens, so you have to blow them for the entire city, or none at all, knowing that only a small portion of the city will have a shot at one of these weak tornadoes…if that’s what it is?

Now you’re saying…wait. If almost all tornadoes are supercellular in nature, then these are going to be rare events, right? Well, thanks to population growth and technology growth in our country, we’re seeing a lot more tornadoes, even the real small ones, and they are hitting a lot more buildings and things that weren’t there 40, 30, 20 10 or even 5 years ago. Furthermore, in 2017, 77% of Americans had smart phones per the Pew Research Center, and that number, I think, will only increase, albeit at a slower pace than previous years. But that means that if there’s a tornado, and there’s someone who can see it, it’s likely that a smart phone or dash-camera video will wind up on social media very quickly. And the armchair/social media “meteorologists” and the general public will howl when you didn’t blow the sirens, or issue a tornado warning.

There are other new things I think we will see, but I think this one is most obvious, and when we see these leading-edge rotations in squall lines/QLCS/derecho situations, we still have a responsibility to warn of these effectively, when possible. Here’s how I think we should do it:

1. There should be little pressure on the NWS to issue tornado warnings on these. Trying to do this is like playing “Whack-A-Mole”. I don’t care if you are a NWS Warngen god and can fly a warning out in 20 seconds with your nimble fingers, by the time it hits Jon Q. public, it’s either gone or just about to lift. I say this: do not penalize NWS skill scores for short-lived tornadoes (see #2 below in a moment) or are in QLCS/lines of thunderstorms. I’m not talking about “comma head” tornadoes in a squall line, I mean within the line itself. But if you MUST verify all tornadoes, then…

1B. A severe thunderstorm warning should have an option in Warngen to have, in the “Hazards…” section, a line which reads something like: “these thunderstorms are capable of producing very brief tornadoes”. If that line is added, a WEA alert gets sent out with this message: “Severe thunderstorm warning with XX MPH winds and/or large hail likely, and very brief tornadoes possible.”

2. In tandem with #1, NO tornado warnings on QLCS tornadoes unless you are seeing a persistent, significant, longer-lived rotation of over 5 minutes.

Why am I making a relatively big deal about this? Here’s why. In June of 2018, two weak EF-0 tornadoes hit Champaign, IL. There was a tornado warning in effect. No sirens were sounded because the EMA didn’t get any reports of tornadoes. Now, I would have sounded the sirens, but what if they were just under a severe thunderstorm warning? Would you blow the sirens if you have a block where a lot of branches are down, even if large? Maybe…maybe not. But it got EMA into trouble, and the public, which now has access to this data in real time, will be yelling as well. Finally, there’s one more thing. In the case of one of the two Champaign tornadoes, a roof was completely blown off and caused damage downstream as it broke up. It was learned that the roof was not physically attached to the beams of the house, as the nails to attach it all missed the beam when the house was built. That means that the roof was just laying on the house, and that’s why the tornado only got an EF-0 rating (should the EF scale be changed for newer construction? Hoo boy, that could be another blogpost for another day). That happened in one of the houses I purchased, and I made the manufacturer go in and re-nail it. They still missed the beams with quite a few nails, but they did at least hit the beams a lot of times. The problem is, if I jerk my head up close to the ceiling in my attic, I may get nails in my head! But, that’s another story. The point is, construction standards since the 1960 and 1970s have gone down at the expense of quality, so that homes and businesses can be built quickly. Thus, in a new subdivision, you could conceivably get a lot more damage from an EF-0 tornado, than, say, an older subdivision built with brick and a roof where it was attached meticulously. Now, most builders just air gun the nails in, using a lot more nails than needed, knowing some (hopefully) will catch on the beam, and knowing nails are cheap. Furthermore, we have seen many cases where only 4 bolts, one on each side, anchor the house to the foundation. Thus, weaker tornadoes hitting these houses can cause substantially more damage than on older homes. These 2 (or 2.5) suggestions I made above can handle these scenarios, along with public education of these brief spin-ups.

So, welcome MRLE, for better, and possibly, for worse for the National Weather Service and EM’s, not to mention, the general public. What do you think? Feel free to leave a comment!