Regional and Mesoscale Meteorology Branch

00:00:00:13 – 00:00:27:10
Speaker 1
Welcome to today’s visit Satellite Chat. For today, we’re going to focus in, Alaska. And to help us do that, we have a couple, guest with us here today. First we’ll be talking about, synthetic imagery from the Nam nest over Alaska. And we have, Dan Lindsay, who is with, Noah knows this. So at this point, I will turn it over to Dan.

00:00:27:12 – 00:00:58:11
Speaker 2
Sorry. Thanks, Dan. So a little background here on synthetic imagery. Several years ago, whenever we started preparing for the launch of Goes-r. We started simulating it was our API bands, from high resolution model output. We originally were doing it from the NASA, so we’re from all over the continental U.S. and, it turns out that doing this not only is a good way to provide proxy data for the API, but also it’s a really nice way to visualize clouds.

00:00:58:21 – 00:01:24:03
Speaker 2
And other things. From that you can see in satellite imagery from high res models. So we, we also began doing this with the Nam Conus nest a few years ago. And there was a request from many offices in Alaska, including Carvin Scott and others, to do this from some model over Alaska. So starting just a couple of months ago, we we began doing this from the Nam Alaska Nest model, which is a five kilometer model.

00:01:24:11 – 00:01:46:06
Speaker 2
I don’t know if you guys get this in the works or not, but it’s one of the, operational models that EMC puts out. So what you can see here is our program page. I’ll go to the top so you can see what it looks like. If you go to the Ram Be Home page and click on the Rams just online it and then click because our proving ground, it’ll bring you to this page.

00:01:46:13 – 00:02:08:04
Speaker 2
And you can see there’s a lot of different products and the synthetic imagery from the Nam Alaska nest, of course, is what I’m talking about today. And we’re simulating currently five bands. These are again, all bands are going to be on the Goes-r API. All three water vapor bands at 6.2 micrometers, 6.95 and 7.34. Remember that Goes 15 currently has a band around 6.5.

00:02:08:04 – 00:02:31:12
Speaker 2
So sort of between these two. So we don’t really have a band that matches up exactly with the current goes. But but now we have three to choose from. The differences are that, the lower the wavelength, the actual, the higher you see in the atmosphere, because the weighting function is picks higher up. So when you’re looking at really upper tropospheric, you want to go to 6.2 and more mid to lower tropospheric say around 6 or 700 millibars.

00:02:31:12 – 00:02:51:20
Speaker 2
You want to go with the 7.3. And then in between would be the 6.9. The 8.5 band is useful mostly for this another water vapor band, but it also is good for SO2. For which of course, you guys probably have to deal with sometimes with your volcanoes. Unfortunately, the model doesn’t predict SO2, so it’s not so useful from that perspective.

00:02:51:22 – 00:03:12:18
Speaker 2
And then finally we’re doing one of the infrared window bands at 10.35. So I just wanted to give you a real quick demonstration on the best way to use this information. We’re doing this only from the ozone run of the Nam Alaska nest each night, and that model goes out to 60 hours. So it’s a 0 to 60 hour for actually, we’re starting with a nine hour forecast and going out to 60 hours.

00:03:12:20 – 00:03:30:09
Speaker 2
So just as a quick the way it looks, if you click on HTML5 loop, it’ll bring up this loop. And I’m going to try to step through this so that you can actually see every time. Step down here at the bottom is the time. This is what hopefully the 10.35 micrometer band from, from goes-r will look like.

00:03:30:11 – 00:03:48:23
Speaker 2
And the color tables down here, I don’t know if this is the same color table you guys use or not, but hopefully you’re familiar with the idea where the colors represent the colder cloud tops. As you step forward in time, you can see this big cloud shield associated with the low down and the south. Sort of the western Gulf of Alaska moving across.

00:03:48:23 – 00:04:08:14
Speaker 2
And you can see the clouds shield moving up. The current time I guess is around 2030 or something like that. That would be at this time. And if we step forward, you can see the model brings the cloud shield associated with that low up over, I don’t know the terms of all of your geographical areas, but I’ll call these the islands that Juno is near.

00:04:08:20 – 00:04:34:10
Speaker 2
In fact, what do you call these islands? Do you know, it’s theater channels. It’s kind of fairly calm. Yeah. It’s okay. In our channel. Okay. Okay. The inner channel. So you can see the clouds moving up over that area. Another thing you can see from this, which is interesting, maybe not as relevant for you guys, but, you can sometimes see the cool, cool ground signature, up here in the northern part of Alaska.

00:04:34:16 – 00:04:55:24
Speaker 2
I guess it’s relatively warm currently with the really cold air over eastern Siberia and Alaska is more moderate at the moment. So the cold ground signature is really confined to the Barrow area currently. And then you can see we can go all the way out to the 60 hour forecast, which is valid on 1230 on on Friday, which is at this time.

00:04:55:26 – 00:05:11:23
Speaker 2
So what I think this is good for is actually cloud forecasting, because you guys have to populate your sky cover grid. So if you were in Juneau, you may look at this model and say the timing of when these high clouds come in. And it looks like it’ll be sometime tomorrow morning in order to populate your grids.

00:05:11:25 – 00:05:33:18
Speaker 2
I want to show one other example, the 6.95 band if you click on this link here. But I actually made a different version that zoomed out and I have a comparison to the current goes, In order to show you how to evaluate the model on the left here is the, the, synthetic imagery, the same thing we were just looking at.

00:05:33:18 – 00:05:52:14
Speaker 2
I did put the lat line lines on. It’s a little bit hard to see Alaska. The, the border is up here. Juno is down in this area. And you can see the low here, in the western Gulf of Alaska and that this is the nine hour forecast. And if we step forward, I want to focus on get to the 16 is your time, which is there.

00:05:52:14 – 00:06:14:16
Speaker 2
That would be 16 of these this morning just a few hours ago. Now on the right here I have the observed goes imagery from the water vapor band. And I’ll step forward is a different projection. So you have kind of have to use the lat long lines to orient yourself. But if you step forward to 16 v to where the times match up, one thing you can do is compare the location of certain features in the model versus the observed imagery.

00:06:14:19 – 00:06:35:03
Speaker 2
In order to get an idea whether the model is on track or not. So for example, you can see in the observed imagery you have this cloud feature, which I guess maybe associated with the cold front. I’m not completely sure because I haven’t looked too carefully at the case, but, a very distinct cloud line associated with the low as it moves to the east.

00:06:35:05 – 00:06:52:23
Speaker 2
And, one thing to note is it’s to look and see exactly where it is, and you can see that it looks at somewhere around maybe 140, 748 degrees west longitude, something like that. But if we go over here to the corresponding time in the model, you can see that that cloud line is a little bit further to the west.

00:06:52:23 – 00:07:14:22
Speaker 2
It’s located closer to about 150 degrees west. So what this tells you is that the model is actually a little bit too slow. With the advancement of this cloud front, or this cold front, rather in the clouds associated with it. And so if you step forward for the rest of the forecast, that may tell you that this particular model run is just going to be a little bit slow with the timing of the system as it moves to the east.

00:07:14:22 – 00:07:38:22
Speaker 2
So it’s best maybe not to use the, the cloud grids, the cloud field. Exactly. But you can adjust it by a couple of hours. So it’ll probably be, the clouds will probably arrive a couple hours earlier than that. What this forecast is saying, if we go back to 1680 again and compare this cloud shield, which is moving up over the Aleutian Islands versus what is observed, you can see that it actually matches up pretty well.

00:07:38:24 – 00:08:01:15
Speaker 2
So, the timing of that would be pretty good as far as the location of the low itself, you can kind of see in the simulated imagery, it’s somewhere in this area compared to here in the, in the observed imagery. So it looks pretty similar. Now, of course, if you have the model drift themselves, you can just plot up the MSL to see where the surface flow is located, or perhaps the upper low.

00:08:01:15 – 00:08:21:05
Speaker 2
If this is, this is maybe more of a an upper low than a surface low. So as far as a UPS and a UPS to goes, we just got this online, not long ago. And so, we are working right now to get it into a website to working with Melissa Quiller up in, in Fairbanks.

00:08:22:08 – 00:08:46:12
Speaker 2
Do you guys know when you’re going to be transitioning to a web? Two? Is it going to be soon or a while from now? We’re going to be installing in July, 15th July. Okay. So at this point, I don’t know if you’re interested in getting this in your area. It might be best just to kind of wait until it works to comes along, because by the time we got the Awacs one stuff going and set up, you guys would probably be transitioning away from it anyway.

00:08:46:15 – 00:09:04:10
Speaker 2
So what I would recommend is if you want to get this imagery in your Awacs, or it works to that, maybe we can wait until the summer and work on it then. Yeah, that sounds good. Yes. Okay. And you guys have any questions? Is this all pretty clear on what what’s going on as far as the synthetic imagery and how to use it?

00:09:04:13 – 00:09:28:01
Speaker 2
No, it’s pretty clear that the big problem we have with, imagery is, or trying to forecast capacity of the clouds, not just cloud cover. You know, how much light actually gets through and all the models, you know, they basically are 1 or 0. Either it’s South cover or it’s not. Right. Yeah, I can see how that would be a challenge.

00:09:28:15 – 00:09:45:10
Speaker 2
I mean, in some respects, this type of imagery could help with that if you believe the model, because, the colder the brightness temperature, in this case, you can see the greens and the blues here. This is going to be the thicker cirrus clouds in the model, compared to say, over here where you may have the thinner cirrus clouds.

00:09:45:12 – 00:09:56:18
Speaker 2
So I think it probably could provide some guidance in that respect. But again, as is true with any forecast, you’re always at the mercy of the model.

00:09:56:21 – 00:10:22:11
Speaker 2
Any other questions about the, synthetic imagery before we move on to cold air aloft? Can I just have a quick question? Is that the domain you’re showing there on the left? That’s the actual domain of the Nam nest that’s running or. Yes. Okay. That’s correct. It’s built I believe it’s initialized from, actually, I can’t remember if it if they initialize it, if the actual Nam 12 goes this far to the west, or if they, they may initialize it straight off the GFS.

00:10:22:14 – 00:10:33:29
Speaker 2
Okay, I’m not sure, but this is the this is at least you can see the boundaries down here of the domain of the Nam Alaska nest.

00:10:34:01 – 00:10:42:09
Speaker 1
So so Dan, in the Goes imagery we have this limb effect as you go further to the north. How do we interpret that in the, synthetic imagery?

00:10:42:11 – 00:11:04:04
Speaker 2
I think that the model, the radial transfer model we use to generate a synthetic gradients is supposedly takes that into account. So you can see in the observed imagery it’s it’s generally a lot colder up here on the limb. And that’s because, you know, the, the radiation is passing through more atmosphere since it’s on the limb. And you can see that’s generally true up here in the simulated imagery as well.

00:11:04:04 – 00:11:24:19
Speaker 2
But it may not be modeled exactly right. You may get a slight difference. So it’s, you know, any time in northern Alaska, you’re going to have challenges with ghost imagery simply because it’s so far on the limb, which is actually a nice lead into the polar imagery, which is, what I think Jack is going to talk about.

00:11:24:21 – 00:11:33:05
Speaker 1
Okay, let’s let me get control back to my screen here.

00:11:33:08 – 00:11:47:22
Speaker 1
And next we’re going to have Jack diastolic, who is, here, Sarah, talk about another, Alaska product called the Cold air aloft product. And at this time, I’ll turn it over to Jack.

00:11:47:24 – 00:12:14:05
Speaker 3
Yeah. Okay. So, hi. I have a couple things to show off. First of all, is, just make this a slide show. First of all, a little background. So we had a site visit by the satellite liaisons back in September, and I was talking with Eric Stevens of Gina in Fairbanks about some forecast problems that they deal with.

00:12:14:07 – 00:12:44:14
Speaker 3
And he mentioned that one issue that they have at the far northern latitudes is that in the winter, you can get these, cold air gathering up high where it can be a problem for the gelling of jet fuel and passenger and cargo aircraft. And he said that, you know, there’s only a few soundings in Alaska, and there’s not much around the Arctic to keep track of these regions of cold air.

00:12:44:21 – 00:13:11:13
Speaker 3
So they use the models and some aircraft data when they can. And they were looking at, also what polar orbiting satellites can do. In particular, they were looking at the new Caps data. So I mentioned, well, you know, we don’t have new caps here yet, but I can help you out with mirrors data, which is another retrieval algorithm that operates on the Am Su and also the ATMs instruments.

00:13:11:15 – 00:13:41:29
Speaker 3
So I put together a couple slides for him and got into contact with a group of various organizations that is working on this cold air aloft problem. And so here is kind of what we’ve developed here at sera in conjunction with Eric and the other group as well. So what I have here in the PowerPoint is an example from the website, although this is from earlier this year, 24th February of 2014.

00:13:42:01 – 00:14:04:17
Speaker 3
So we’ll go through this case where there was some cold air over Alaska. And then we’ll look briefly at what’s going on with the website today. So I have this annotated. And what you have here is a northern 19 pass over the northern hemisphere obviously. And you can see that, there’s some color code and some plus signs.

00:14:04:17 – 00:14:46:02
Speaker 3
So what this is, is I’ve taken all the mirrors, retrievals along the swath and looked at them to find the coldest temperature recorded by the mirrors, retrieval from the surface to 100 millibars. Then I color coded that according to the scale in the upper left corner of the plot. Now, turns out that -65 C is more or less the cutoff point, or kind of the critical temperature, if you will, of when airplanes need to start avoiding air mass of that cold because of the potential for gelling the jet fuel.

00:14:46:05 – 00:15:10:13
Speaker 3
So I have just a simple scale here. Anything that’s gray is greater than -60, implying that there’s really no danger. The light blue is -60 to -65, indicating that, you know, you’re still okay, but you may want to kind of see what else is going on in there. And then once you get into the blues, you’re below the -65.

00:15:10:13 – 00:15:43:29
Speaker 3
And then when you get to the pink it’s below -70°C. Again, that’s the coldest point as seen by the mirrors retrieval between the surface of 100 millibars. Now where these plus signs are located indicates that this cold layer of air extends below flight level 450, which I have indicated there in the upper right corner. So the idea is that, you know, the airplanes use these flight levels to fly on.

00:15:44:02 – 00:16:09:01
Speaker 3
And flight level 450 corresponds to roughly 45,000ft. So what that means then, you know, because you could have a cold air aloft signature such as, let me see if I can move my cursor over here. Right here in this little region here, and then sum over further to the other end of the swath where you have cold air below -65.

00:16:09:03 – 00:16:33:08
Speaker 3
But it’s at a elevation or an altitude that really planes don’t fly in that region anyway, so there’s no problem. But where you have all these plus signs here, pretty much most over northern Russia and so forth, and then into the Arctic Ocean. That’s where you have very cold air. That’s also at a level that would cause problems for jet aircraft.

00:16:33:10 – 00:16:55:23
Speaker 3
You can see then as the swath continues down in this direction, that some of that cold air made it to northern Alaska. And seeing that we were kind of talking to the Alaskans about this first, this is the area that we will focus on, even though it’s not the coldest part of the swath, there are some some cold air signatures here in northern Alaska.

00:16:55:25 – 00:17:17:29
Speaker 3
So I’m going to go forward here to the next plot, which is a regional view of Alaska. So I have a few annotations here to the Arctic. Here’s Barrow, Alaska. And then here’s, you know, just the states. So this is the coastline here. Basically. So this is sort of the blown up view of what I showed in the previous slide.

00:17:18:01 – 00:17:55:02
Speaker 3
Now in this particular case though I have some extra information wherever you see the plus sign is again where you have regions of air colder than -65 C that extend below flight level, 450. Now I have annotated on the right hand side of the plus the levels that bound the layer over which that cold air exists. So for example, in this boxed region just south of Barrow, the cold air extends from 35,800ft to 39,100ft.

00:17:55:05 – 00:18:25:02
Speaker 3
So that gives the forecaster or the pilot some idea the vertical extent of this cold air. So this is what I’ve chosen to display kind of a three dimensional view on two dimensions, which is a little tricky. But this is what we have for now. So what I have here, of course, is this boxed retrieval. And right at about this time there was a sounding over a barrel.

00:18:25:09 – 00:19:00:09
Speaker 3
So we’ll take a look at the next image to do the co-location between the radio sound and the satellite retrieval. So just for reference, in the upper left corner I have that same picture and just blown up here is now the sound that we have in blue is the Barrow radio sound from zero Z 25th February 2014. And then in the black is the co-located roughly temperature retrieval from the mirrors off of 919.

00:19:00:12 – 00:19:26:12
Speaker 3
Then I have in the red line here is the -65°C line. Kind of to focus in on that, to see if it captured, as much as the radio sound did. And you can see that in the broad respect, the satellite retrieval did quite well. Now, you’ll never get these kinds of wiggles from the satellite retrieval, but in general, it does do a pretty good job.

00:19:26:12 – 00:19:59:18
Speaker 3
And in particular, it does show that a spot of air that extends below 65 degrees low -65 Celsius here, around 200 millibars. So that is some nice confirmation of these retrievals. As, as we’re kind of making this product. So what I’m going to do now is I’m going to get out of this mode and look at our website.

00:19:59:21 – 00:20:30:19
Speaker 3
So again, it’s off the same Ram Ramsey’s online. It’s on this cold air aloft. And here’s what you see. So I’ll click on this. And this is the loop from today. Basically. Now one of the nice things about using the mirrors is that it’s running on several satellites. So we have five satellites that are on display here. There’s no 18, no 19, Mediaset one and Mediaset two and the Dmsp 18 satellite.

00:20:30:20 – 00:20:57:23
Speaker 3
Let me just stop so we can kind of slide through here. So today oh I shouldn’t mention this Mediaset one is going to look a little different because they use 90 footprints for each scan line. Or these other ones use 30. So the plus signs will be packed closer together. So that looks a little different. And the Dmsp satellite.

00:20:57:23 – 00:21:18:22
Speaker 3
Let’s see if I can get an example of that. I can’t get everything on the one thing here it looks a little different because it’s a conical scanner. Let’s see if I can find an example. Yeah. Here. So you get these kind of features because the instrument scans a little differently. But the bottom line is the same product.

00:21:18:25 – 00:21:52:05
Speaker 3
So as it looks right now today looks like most of the cold air is. Maybe directly over the pole and kind of closer to Russia. I haven’t seen a big cold air outbreak over Alaska this year. So let me get to the latest okay. So here’s the latest image. And you’ll note that there is a problem. Oh excuse me with latency for these data.

00:21:52:07 – 00:22:19:06
Speaker 3
Sorry but that’s just kind of something we got to work around. Also another caveat is that this mirrors retrieval. The temperature retrievals aren’t technically operational overland, although I have seen some results that they’re pretty good overland in some cases. So just because they’re not officially sanctioned, so to speak, doesn’t mean they’re not useful. And certainly over the Bering Sea, that would be water.

00:22:19:06 – 00:22:41:03
Speaker 3
And there’s quite a bit of air traffic coming into Anchorage from Asia and so forth. So this is kind of a new product, and we’ll see how it works over the upcoming week or upcoming winter. Let me go back real quick here and just show you that I have this Bering Sea regional view, but there’s not much going on there.

00:22:41:09 – 00:23:13:23
Speaker 3
As I mentioned, we haven’t had really big cold air aloft issues in this area yet this winter. I think there was some maybe earlier in December. Yeah, mid December there was some, but nothing too major. So we’ll kind of keep an eye out on that over the coming winter. So I don’t know if Juneau if you have responsibilities for these aircraft forecasts, I know that there’s the US at the US you or something in Anchorage that does that.

00:23:13:23 – 00:23:21:05
Speaker 3
So I don’t know if they do that solely or you also have some responsibility for that.

00:23:21:08 – 00:23:27:06
Speaker 2
And awoo. And Anchorage has the main responsibility for aviation.

00:23:27:06 – 00:23:43:22
Speaker 3
Across the region. Okay. So use okay. So that would be, their responsibility. Okay. Well, that’s that’s fine. And I don’t know if this would be of any use otherwise. But it’s out there in case you ever need that.

00:23:45:11 – 00:23:51:13
Speaker 1
So any questions? Any questions? On this product.

00:23:51:16 – 00:24:16:18
Speaker 2
I guess this is Bill Schober. Just, just to reinforce, what what John has said is that, you know, we have a team that is focused on this a lot of a lot of stuff going on and, more more to follow. There’s, a lot of different, ways of looking at this. And we’re working with the would to get some impact from it.

00:24:16:18 – 00:24:17:14
Speaker 2
So.

00:24:17:16 – 00:24:38:26
Speaker 3
Oh, yeah. I’m glad you mentioned that, Bill, because yeah, I’ll add a little bit more. So yeah. So this team ultimately they want to get a product like this into a website too. And again the how it exactly will look is still being decided. But our role sort of at Sarah was to say, well, look, we can get something up on the web quickly.

00:24:38:29 – 00:25:06:03
Speaker 3
And so and then it would serve those who may not have a website to and it could also serve as sort of a prototype. And we get forecaster feedback as to whether these kinds of displays are useful or they’d like to see something else or additional information or fewer or less information. So yeah. So this is kind of a more of an interim and prototype solution to something that ultimately we want to get de to.

00:25:06:04 – 00:25:10:12
Speaker 3
Yeah. So thank you for mentioning that though.

00:25:10:15 – 00:25:21:04
Speaker 1
Okay. Any other questions or comments on anything that we’ve looked at today before we leave here?

00:25:21:07 – 00:25:23:22
Speaker 2
No questions for Gina.

00:25:23:24 – 00:25:27:13
Speaker 1
Okay. Well thank you everybody for participating and have a great day.