Tag: satellite loop

Feeling The Weather

This is a step beyond being able to say rain/no rain.

As you might imagine I am a radar watcher. I looked at it today while on the sofa and said to myself, “That’s ugly.”

Obviously, it’s going to rain tonight–though that’s not a particularly blogworthy note.

What’s become more important to me is a recent increase in my own ability to look at remote sensing (like a radar or satellite loop) or computer models and understand at a very basic level just how that day will look and feel. This is a step beyond being able to say rain/no rain.

There’s a limit to how much of this detail I can actually impart in a forecast. Too much detail and the viewers will take away less! In some cases I’m not sure there are ways to quantify these physical feelings.

It’s frustrating to know more than you can say.

Another Evening With Frances

There is one thing that has been established beyond the shadow of a doubt this week. Everyone has a connection to Florida. Whether it’s a friend or relative, someone living there or just visiting, we all have an equity stake in Florida.

Wherever I go people ask me about Hurricane Frances. We’ve all seen what happened on the West Coast of Florida, and this storm promises to be stronger. It’s no surprise that it scares the daylights out of normally unflappable people

Today, for the first time, the computer guidance is beginning to agree. I’ve been pointing to Jupiter/Hobe Sound and the official pronouncements aren’t far off that mark. Of course the hurricane actually has to perform as forecast… to ‘verify’ in the vernacular of meteorologists, which is never guaranteed.

A few things struck me this evening.

On-the-air, we played an ABC report which included an interview with, what I suspect, a government official in the Bahamas. He complained that maybe they had underestimated the storm.

What planet is he on? The predictions for the Bahamas couldn’t have been more dire if we had said a fiery meteor was plunging their way! The Hurricane Center, which cooperates with the government of the Bahamas in hurricane prediction, went out of its way to scare the crap out of Bahamians – and for good reason.

Unfortunately, areas with a lot of tourism often underplay warnings and later downplay damage. It’s not good for business. Not many people are going to want to go to San Salvador Island after today’s report of 120 mph sustained winds. Nassau might get a close scare. Freeport could get a direct hit.

I really miss having radar that sees Frances at this stage. Tonight the satellite imagery started showing some ‘weakness’ on the hurricane’s western flank. I commented to my friend Bob that I thought the storm would be downgraded… and it was at 11:00 PM&#185. Now Frances is Category 3.

It’s funny, but when satellite imagery begins to show a change, it doesn’t strike me as soon as the image actually comes in. It usually takes a while, staring at the satellite loop, before the trend takes hold. This is most frustrating, especially during winter storms, when I go on the air then look at the same data after my weathercast and begin to question impending changes.

The fact that Frances is weaker tonight doesn’t mean too much of anything. Storms naturally get weaker and stronger in response to their immediate environment. There are guesses why it happened, but no one knows. Hurricane experts are baffled by unknown forces all the time. And, for some unknown reason, hurricanes only have a finite amount of time they can spend as major storms. Again, no one knows why nature works this way.

Since all of weather is guided by the laws of physics, we should understand all the forces at work. We do not.

The official forecast is for Hurricane Frances to regain strength in its final march over open water to Florida. The Hurricane Center’s number for Saturday at 8:00 AM EDT is 140 mph, equaling Hurricane Frances strongest point.

It really doesn’t matter. The difference between 125 mph and 140 mph isn’t all that much in the general scheme of things. Even a minimal hurricane will cause significant damage.

More than the wind, I am worried about Frances losing her steering currents and wandering aimlessly, or at a very slow speed, in the warm Atlantic waters between the Bahamas and Florida. An extended period adjacent to land might be worse than a quick, but direct, hit. There will be that much more time for flooding and tornadoes and wind. The forecast will become exponentially more difficult (and less accurate). There will be that much more terror.

&#185 – I have no idea how this happened, but the Hurricane Center issued its 11:00 PM bulletin with the wrong wind speed! Frances was called Category 4, though it had been downgraded to Category 3. You would think something like this would be vetted.

What I’ve Been Up To At Work

The past few weeks have been spent getting ready to use some new equipment at work. Our very dependable, SGI based, Liveline Genesis system has been replaced by Weather Central’s :Live.

Actually, replaced is not a good word, because :Live is really an add-on which extends the system. We’re still producing some graphics in Genesis but it now it doesn’t go on-the-air.

The SGI system we were using has to be at least 10 years old. These systems run slow by today’s standards. Our hard drive was only 4 GB! From time-to-time I had to go in an mercilessly blow out perfectly fine work created by the other guys in the weather department because we just didn’t have enough room.

Computers and homes are very similar in that you can never have too much closet space. And, of course, the hard disk is the closet of computing.

The problems with the SGI system were legion. It never handled the look of fonts correctly. Its interface, developed in he dark ages of computing, was anti-intuitive and often different in different parts of the system. It took long amounts of time to render animated segments, like a satellite loop or fly through, before they could be shown on TV.

On the other hand, it was nearly bulletproof. The system hardly ever crashed or locked up.

Because the SGI system was based on the Irix operating system, from time-to-time you’d have to delve into the **ix environment to attack a problem. It is a bit scary to do, because it is so foreign to most computer users. Over the years, as I have become more conversant in Linux, another **ix language, Irix has become more understandable.

Every time I have a problem, and work with one of Weather Central’s tech support people, I wonder how they do this with computerphobes? Often we can skip the first 5 or 6 steps. Imagine trying to describe this obtuse text oriented operating system over the phone!

The new :Live system allows us to show animations with no rendering time (though files still have to load from the hard drive to memory, which does take some time). It also integrates multiple layers of animation and still images, which makes it much more flexible. The most interesting part is the ability to stand in front of my green chroma key wall and use my finger as a mouse, drawing or placing objects on the TV screen in real time (or :Live, I suppose).

Right out of the box, it looked much sharper, cleaner and modern than what we had been doing. Simple things, like forecast pages, now run with animated backgrounds. Maps and icons look crisp. The satellite imagery is a little blockier and pixelated than what we were using, especially when viewed at a regional or tighter level.

Because the commands to create each graphic element are programmed in quasi plain text, I have started to write some new ‘scenes’ to suit our needs.

The downside is, this is a Windows based system – Windows 2000 to be exact. It has crashed more than once. So far, not while on-the-air, but awfully close. It also seems to have memory leak problems, not a surprise in a Windows environment. That means, if you run a sequence through, to check it out, you may be pushing the car closer to the edge of the cliff with each mouse click.

I already see some changes I’d like added to the system, which is probably a blessing and curse to those who designed it. I will help them make it better, but probably at the cost of being a pain in the ass.

At the same time we added :Live, we’ve also begun running our own, locally produced, high resolution, computer forecast model. I’ll get into that later.