March 27, 2021: Satellite Saturday

First, a high resolution satellite photograph from the European Space Agency’s Sentinel 2 satellite, taken on March 23.

You can click on this image to see it full-screen. You can zoom in enough to make out roads and big buildings.

Lake of the Woods north of the Alneau Peninsula and west of Whitefish Bay.

The south end of the lake was captured, but the full size image is too big to upload whole. You can view it online, here. Sorry, Sioux Narrows, the image cuts off just west of Whitefish Narrows. You got left out.

Okay, so this is a nice picture, but it’s four days old. What’s happening now?

Strange things. But first I have to explain a bit about the different ways satellites can image stuff. The ESA’s Sentinel satellites (there are six in active service) use powerful instruments to take pictures of a small part of the planet with high resolution, sort of like a telephoto lens. But because they’re working with such a tight field of view, they can’t cover everything all the time. So opportunities to photograph our lake only come up once in every several days. Throw in periods of cloud cover and darkness, and it’s actually more like once or twice a month.

This is why I usually rely on the NASA Terra and Aqua satellites. They’re designed to constantly take pictures, using a wide-angle view that lets them sweep over pretty much the whole planet every day or so, but in less detail. The pair of satellites have matching but complementary orbits, and the general situation is that Terra images LotW in the morning, and Aqua in the afternoon. The instruments they use to do this are called MODIS. That’s short for MODerate resolution Imaging Spectroradiometer. If you want to read more about it, you can see a brochure here. The key facts from an Ice Patrol perspective is that they scan the planet in multiple wavelengths, including separate bands of Blue, Green and Red light, two bands of Near Infrared, and two bands of Shortwave Infrared. For the true colour images, the blue, green and red components are blended to create a natural looking view, in a manner somewhat comparable to how your printer uses three ink cartridges to print full colour. For the false colour versions, a computer takes information from several wavelengths and processes it a bit differently. Today I learned that the images I use, which come from the University of Wisconsin at Madison, are processed using the same algorithm every day. That’s relevant, because some of the recent images have shown an unexpected development.

Here’s the image I posted a few days ago, from March 22.

Aqua satellite’s MODIS image from March 22, 2021, in false colour.

The ice is darkening all over.

And here’s the latest image, from March 25.

Aqua satellite’s MODIS image from March 25, 2021, in false colour.

Whoa! When I first saw this, (after I said “What the…?”) I thought maybe I’d accidentally pulled a picture from last year or something. Nope. So what on earth is going on? We can’t have suddenly created a foot of ice, because these pictures are from before our recent  drop to freezing temperatures. This picture was taken by the Aqua satellite on the afternoon of March 25, when the temperature was probably about 5°C. The overnight low the night before was a mild -2.7°C. We’d have been losing ice in the days before this picture was captured, not making it.

Okay, you know that TV show about the pawn shop? The one where they say, “I’m not an expert, but I have a friend…”? Well, I have a few contacts, and this time I reached out to Hilary Dugan, an associate professor at the University of Wisconsin at Madison, the place that publishes these satellite images to the web. She’s a limnologist (a freshwater scientist who studies lakes and rivers), and she knows a lot more about the MODIS images than I do. In fact, much of the information I dumped on you a few paragraphs ago comes from her. She was able to confirm that the image processing has not been altered. So she, and her colleagues who work directly with the public versions of the imagery that I use here, have two possible explanations. The lake looks different because of a change in lighting, or the lake looks different because the ice has transformed in some way.

I don’t really think it’s just the lighting, so there has to be some change in the quality of the ice. I thought that maybe the ice had candled, and Hilary suggested that it could have developed a lot of air bubbles, making it airier and whiter.

We’ll know more when we have a sunny day and get new pictures. In the meantime, I’m going for a little drive. Take a second look at that Sentinel image at the top of this post. You might have noticed that two little lakes just west of Kenora—Muriel Lake and Sandy Lake—are shining bright and conspicuously blue. I want to know why.

Okay, I’m back. Neither of those lakes is covered in blue snow. They’re shiny, and the ice is thin: there’s a prominent warning on the  Pellatt Community Centre sign. I went to the shoreline and photographed the ice. This is how it looked at Sandy Lake:

Shoreline ice at Sandy Lake.

And this is what it was like at Muriel Lake.

In both cases, the ice was granular, and very white from the air in it.

Maybe this is what’s happened to the ice all over Lake of the Woods.

I started work on this post this morning. Since then, today’s Terra image has come in.

Terra Satellite’s MODIS image from March 27, 2021 in false colour.

There are cute little cumulus clouds hugging the ground along the top and left of this picture, and high altitude ice clouds in the lower right. You can definitely see the shadow of the ice clouds altering the look of the lake ice. So I guess the quality of light does affect the colour.

I am reminded of the lyrics of Paul Simon’s song Graceland. “The Mississippi Delta was shining like a National Guitar…”

It’s sunny this afternoon, so I’m interested to see what Aqua‘s image looks like. It’s not ready yet, so I’ll publish this now, and update this post when the new image comes in. Revisit Ice Patrol and hit refresh on your browser sometime this evening if you’re curious too.


8 thoughts on “March 27, 2021: Satellite Saturday

  1. There may be a simple explanation of why the ice appeared so blue in the satellite image on the 22nd. We had rain on March 21st and the temps were above zero for several days in that period. So the satellite may be imaging the fact that there was a sheen of water on the ice.

    • Good point, but generally, water is supposed to look dark, not light. Water on top of ice might be different, though.

  2. I meant to say the temps were above zero – even overnight – in that period. Hence there was no refreezing overnight as has been happening lately. In our bay the open water (and remaining ice) have been freezing overnight and it usually into early afternoon before that skim of new ice thaws. And the existing ice is a frosty white before darkening in the warm afternoon temps. So, the morning photos should show a somewhat different colour than the afternoon, and nights above zero should result in a deeper blue because of melt water remaining on the ice.

  3. I wonder if it’s something like when trying to take a portrait of someone indoors while they stand in front of a big window. An auto-exposure light meter is often weighted to the center of the frame. If that happens to be the bright outdoors, the portrait can look more like a silhouette. If it happens to be centered on the person, the outdoor scene is overexposed and washed-out looking. Without a reflector or flash-fill, it’s pretty hard to get a decent exposure for both the foreground and background images with a camera. The wetware in our brains makes a better image, because it’s optimized to gather detail on both by putting together a series of “pictures” with differing “exposures” into what seems to us to be a single, better detailed, image.

    The modis pictures may have a similar limitation as a regular camera. It takes a single image of a region at differing wavelengths? On a clear day, the exposure shows good detail with the single exposure. On mixed cloudy days, it has to compromise. Detail on the brightly lit and/or partly cloudy areas gets lost through under and/or over-exposure. Software might be able to enhance the image somewhat, but if the information isn’t there in the source data, there must be limits to the enhancement. If it took lots of pictures to make the regional picture, it could set the “exposures” at a finer level of detail, but then the regional picture wouldn’t be a single point in time – instead it would be a sort of time-smear like what our eyes actually see.

    Anyway, just speculation.

    • I know, I know. But it’s cold, so we’re not melting, and it’s cloudy, so no sat pix, and none of my aviating friends have sent me anything, so there’s not much to say. I’ve seen gulls, though.

  4. The web cam at Red Wing Lodge (directly east of Bigsby Island) shows candled ice and lots of open water.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s