Sea-effect Snow

Posted on January 13, 2015 by Curtis_Seaman

Take a look at this image:

Photo credit: İskender Şengör via Severe Weather Europe on Facebook

Is this picture from A) the Keweenaw Peninsula of Michigan in 1978? B) Orchard Park, New York in November 2014 (aka “Snowvember”)? or C) İnebolu, Turkey from just last week?

If you pay attention to details, you will have noticed that I credited İskender Şengör with the picture and properly surmised that the answer is C. If you don’t pay attention to details, get off my blog! The details are where all the interesting stuff happens! You’d never be able to identify small fires or calculate the speed of an aurora or explain the unknown without paying attention to details.

If you follow the weather (or social media), you probably know about lake-effect snow. (Who can forget Snowvember?) But, have you heard of sea-effect snow?

Areas downwind of the Great Lakes get a lot more snow than areas upwind of the Lakes. I was going to explain why in great detail, but this guy saved me a lot of time and effort. (I have since been notified that much of the material in that last link was lifted from a VISIT Training Session put together by our very own Dan B. You can watch and listen to that training session here.) The physical processes that cause lake-effect snow are not limited to the Great Lakes, however. Anywhere you have a large body of relatively warm water (meaning it doesn’t freeze over) with episodes of very cold winds in the winter you get lake-effect or sea-effect snow.

When you think of the great snowbelts of the world, you probably don’t think of Turkey – but you should! Arctic air outbreaks associated with strong northerly winds blowing across the Black Sea can generate snow at the same rate as Snowvember or Snowpocalypse or Snowmageddon or any other silly name that the media can come up with that has “snow” in it (Snowbruary, Snowtergate aka Frozen-Watergate, Snowlloween, Martin Luther Snow Day, Snowco de Mayo, Snowth of July… Just remember, I coined all of these phrases if you hear them later). Plus, the Pontic Mountains provide a greater upslope enhancement than the Tug Hill Plateau in Upstate New York.

One such Arctic outbreak occurred from 7-9 January 2015, resulting in the picture above. Parts of Turkey received 2 meters (!) of snow (78 inches to Americans) in a 2-3 day period, as if you couldn’t tell from that picture or this one.

From satellites, sea-effect snow looks just like lake-effect snow. (Duh! It’s the same physical process!) Here’s a VIIRS “True Color” image of the lake-effect snow event that took place last week on the Great Lakes:

VIIRS “True Color” RGB composite, taken 19:24 UTC 7 January 2015.

Wait – that’s no good! We need to be able to distinguish the snow from the clouds. Let’s try that again with the “Natural Color” RGB composite:

VIIRS “Natural Color” RGB composite, taken 19:24 UTC 7 January 2015.

That’s better. Notice how the clouds are formed right over the lakes and how the clouds organize themselves into bands called “cloud streets“. The same features are visible in the sea-effect snow event over Turkey (from one day later):

VIIRS “Natural Color” RGB composite, taken 10:36 UTC 8 January 2015.

Look at how much of Turkey is covered by snow! (Most of that snow cover is from the low pressure system that passed over Turkey a couple days before the sea-effect snow machine kicked in.) And – *cough* attention to details *cough* – you can even see snow over Greece and more sea-effect snow on Crete. There’s also snow down in Syria, Lebanon and Israel (Israel is off the bottom of the image), which is bad news for Syrian refugees.The heavy snow has shut down thousands of roads, closed schools and businesses, and was even the source of a political scandal.

But, on the plus side, the Arctic outbreak in the Middle East brings a unique opportunity to see palm trees covered in snow. And, how often do you get to see the deserts of Saudi Arabia covered in snow? (EUMETSAT has provided more satellite images of this event at their Image Library.)

Take another look at that image over the Black Sea. See how the biggest snow band extends south (and curving to the southeast) from the southern tip of the Crimean Peninsula? That is an example of how topography impacts these snow events. Due to differences in friction, surface winds are slightly more backed over land than over water, therefore areas of enhanced surface convergence exist downwind of peninsulas. The snow bands are more intense in these regions of enhanced convergence. There are also bigger than normal snow bands downwind of the easternmost and westernmost tips of Crimea, and extending south from every major point along the west coast of the Black Sea. This is not a coincidence. Land-sea (or land-lake) interactions explain this. Go back and listen to the VISIT training session for more information.

Sea-effect snow affects other parts of the globe as well. It’s why the western half of Honshu (the big island of Japan) and Hokkaido are called “Snow Country“. Japan was also hit with a major sea-effect snowstorm last week and, of course, VIIRS caught it:

VIIRS “Natural Color” RGB composite, taken 03:48 UTC 8 January 2015.

See the clear skies over Korea and the cloud streets that formed over the Sea of Japan? Classic sea-effect clouds. You can even see snow all along the west coast of Honshu in between the breaks in the clouds. Topographic impacts are once again visible. Notice the intense snow band extending southeast from the southern tip of Hokkaido/northern tip of Honshu similar to the super-strength snow band off of Crimea. And there’s another one downwind of the straits between Kyushu and Shikoku. Another detail in this image you should have noticed is the impact that Jeju Island has on the winds and clouds. Those are classic von Kármán vortices which we have discussed before.

Fortunately, 8 January 2015 was near a full moon, so the Day/Night Band was able to capture a great image of these von Kármán vortices:

VIIRS Day/Night Band image, taken 18:09 UTC 7 January 2015.

So, to the people of the Great Lakes: Remember you’re not alone. There are people in Turkey and Japan who know what you go through every winter.

UPDATE #1: While I was aware (and now you are aware) that sea-effect snow can impact Cape Cod, it was brought to my attention that there is a sea-effect snow event going on there today (13 January 2015). Here’s what VIIRS saw:

VIIRS “Natural Color” RGB composite, taken 17:29 UTC 13 January 2015.

According to sources at the National Weather Service, some places have received 2-3 cm (~ 1 inch) of snow in a four-hour period. It’s not the same as shoveling off your roof in snow up to your neck, but it’s something!

Beginning of Autumn in the Great Lakes

Posted on October 7, 2014 by Curtis_Seaman

Have you noticed it? The seasons are changing (for the mid- and high latitudes, at least). Days are getting shorter (or longer if you live in the upside-down hemisphere). This time of year, if you live in Alaska or Scandinavia or similar high latitude locations, you lose about 5-10 minutes of available daylight each day. (That’s between a half and one hour per week!) You may have noticed by the fact that your neighbor no longer mows the lawn at 11:00 PM because it’s still bright outside and hey, why not? I wasn’t going to sleep anyway.

Closer to home – in the mid-latitudes – loss of daylight is more like 1-3 minutes per day, which isn’t as noticeable. But, one day, you watch the sun set and look at the clock and realize that it’s only 6:30 PM and you think, didn’t it used to be light out later than this?

That’s not the only way to tell the seasons are changing. For one, there’s the arrival of snow. (Although parts of Montana, Wyoming and South Dakota received snow earlier this year while it was still technically summer.) And, for two, there’s what VIIRS observed on 27 September 2014:

VIIRS True Color RGB composite of channels M-3, M-4 and M-5, taken 17:57 UTC 27 September 2014

In case it’s not obvious, here’s what VIIRS saw earlier in the month (8 September 2014):

VIIRS True Color RGB composite of channels M-3, M-4 and M-5, taken 17:13 UTC 8 September 2014

Notice anything different between the two images? (Remember to click on the images, then on the “1735 x 1611” links below the banner to see the images in full resolution.)

That’s right – the loss of daylight leads to one of the benefits of autumn: fall foliage. VIIRS True Color imagery shows, quite clearly, that the leaves of New England and eastern Canada have changed color. Forests that were green in early September have turned orange, red and brown by the end of the month.

Another thing you may have noticed comparing those two images: the change from green to beige in the area around Montreal, Quebec. This is another sign of autumn: the fall harvest. This is a productive agricultural region in eastern Canada, and what you are seeing is the green vegetation (crops) being harvested, leaving behind bare dirt.

True Color imagery is useful for observing the changing foliage and the harvest because it is designed to reproduce what we humans observe on the ground. The red, green and blue components of the RGB composite are channels in the red (M-5, 0.67 µm), green (M-4, 0.55 µm) and blue (M-3, 0.48 µm) portions of the electromagnetic spectrum. When leaves change from green to red, the True Color RGB detects that.

Now, you’ve probably known since elementary school (or at least middle school) that leaves change color because of chlorophyll. And, unless you became a botanist, that is probably the limit of your knowledge on the subject. But, there’s a lot of interesting chemistry that goes on inside a leaf (and the whole tree) that determines it’s color.

Of course, leaves are green because they contain chlorophyll. Chlorophyll is necessary for plants to convert sunlight into sugar. Chlorophyll, by necessity due to it’s job, is highly absorbing of visible-wavelength radiation, although it is slightly less absorbing of green wavelengths. Green light is therefore preferentially reflected out of the leaves and into your eye, and the leaves appear green.

When the sunlight goes away and the air becomes cold, deciduous trees go into hibernation. They break down the chlorophyll in their leaves, and send the remaining nutrients down into the trunk and roots. This exposes the carotinoids that were in the leaves and these carotinoids have a yellow or orange color – they preferentially reflect yellow and/or orange wavelengths. Red colors come from a pigment called anthocyanin, which was recently discovered to be a sort of “plant sunscreen”.

Now, utilizing sunscreen when you get all your energy from the sun may sound silly but, recent studies have shown that anthocyanin protects the leaves from sun damage once the chlorophyll is gone so that the tree has time to extract all the nutrients out of the leaves before they fall off. Trees in poor soil conditions are more likely to turn red in the fall as a natural defense mechanism – they need to store all the nutrients they can from their leaves, since they aren’t getting them from the soil.

Oak and other leaves turn brown in the fall because of a buildup of tannin (link to PDF file), which is a waste product. Brown leaves are full of plant poo! Think about that the next time you go on a fall color driving tour.

Now, back to the satellite science before the biologists come after me for grossly oversimplifying leaf chemistry. I’ve often talked about the Natural Color RGB composite as being similar to the True Color RGB in many instances (except for the detection of ice and snow). So, what does that look like here?

Here’s the VIIRS Natural Color RGB from 8 September 2014:

VIIRS Natural Color RGB composite of channels M-5, M-7 and M-10, taken 17:13 UTC 8 September 2014

And here’s the same RGB from 27 September 2014:

VIIRS Natural Color RGB composite of channels M-5, M-7 and M-10, taken 17:57 UTC 27 September 2014

Why does the vegetation still appear green when the leaves have changed color? Because we’ve made vegetation artificially appear green. The Natural Color RGB uses the red wavelength visible channel (M-5, 0.67 µm) as the blue component. The green component is a near-infrared channel (M-7, 0.87 µm), where plants are their most reflective – leaves and other plant tissues don’t absorb radiation at this wavelength. The red component is a longer wavelength channel (M-10, 1.61 µm) where the water inside the leaves starts to absorb radiation and the reflectance goes down. Cellulose and lignin also weakly absorb at 1.61 µm. The bottom line is, plants are highly reflective at 0.87 µm regardless of how healthy the plant is, or what color the leaves are – so they will always appear green in the Natural Color images.

You might also note the one difference (apart from clouds) that shows up between the two Natural Color images is the lack of green surrounding Montreal in the 27 September image. This is another sign of the fall harvest: the highly reflective plants have been removed and all that’s left is dirt, which is not as reflective. That’s why those areas appear more brown in the later image.

If we look a bit further west in the True Color imagery from 27 September 2014, the fall color really stands out:

VIIRS True Color RGB composite of channels M-3, M-4 and M-5, taken 17:57 UTC 27 September 2014

Fall colors are visible from the Adirondacks of Upstate New York and Quebec to the Upper Peninsula of Michigan. The most vivid fall color is in Ontario – both in the area of Sault Ste. Marie and in the area of Algonquin Provincial Park, the oldest provincial park in Canada. Every autumn, the Friends of Algonquin Park post pictures of the fall colors, including this shot from 27 September 2014 showing just what VIIRS was seeing. Amazing colors!

We have sunny days, cool nights and plant survival techniques to thank for that.

BONUS:

Here’s a desktop wallpaper that’s zoomed in on the above image and cropped to the most popular screen resolution (1366×768):

VIIRS True Color RGB Composite Desktop Wallpaper (17:57 UTC 27 September 2014). This image fits monitors with a 16:9 ratio and is optimized for 1366×768 screen resolutions.

Make sure you click on the image, then on the “1366 x 768” link below the banner to get the full resolution image. Then you can right-click on the image and choose “Set as desktop background” to save it as your new desktop wallpaper.

Land of Lincoln Underwater

Posted on April 23, 2013 by Curtis_Seaman

The week beginning on 14 April 2013 was a big week for weather across the United States. There were 30 reports of tornadoes. (Make sure you click on each link, and look at the filtered reports.) And, when our home base of Fort Collins, Colorado was in the middle of being buried under two feet of snow, large parts of the Midwest received 4-7 inches of rainfall. This is a lot of rain for an area with saturated ground caused by recent snowmelt. Unsurprisingly, it caused a lot of flooding – including a sinkhole in a Chicago neighborhood.

Now, we know VIIRS is good at detecting snow. But, flooding is a bit trickier, particularly river flooding. First, flooding usually occurs when it’s cloudy. (Not always, of course, since you can have flooding from snowmelt or heavy rains that occurred upstream or caused by ice jams when it isn’t cloudy. And, as we saw with Hurricane Isaac, flooding may linger long after the clouds are gone.) Second, flooding can have a huge impact over a small area that your satellite might not have the resolution to detect.

Well, I’m here to report that VIIRS has the resolution to detect the flooding that occurred over Illinois last week. And the flooding lasted until well after the clouds cleared. Take a look at the image below from 21 April 2013, where the flooding is visible:

VIIRS false color composite of channels I-01, I-02 and I-03, taken 18:13 UTC 21 April 2013

This is a “Natural Color” RGB composite of the high-resolution channels I-01 (0.64 µm, blue), I-02 (0.87 µm, green) and I-03 (1.61 µm, red). If you click on the image, then on the “3124×2152” link below the banner, you will see the full resolution image. If you’re wondering where the flooding is, notice the rivers I have labelled in the image. Now try to spot those rivers in this image from two weeks earlier (5 April 2013):

VIIRS false color composite of channels I-01, I-02 and I-03, taken 18:13 UTC 5 April 2013.

Those rivers are a lot more difficult to see. The Illinois, Sangamon, and Mississippi rivers are the only rivers easily visible in the before image. A lot more show up after the heavy rains because they grew beyond their banks and became big enough for VIIRS to see. You might also notice that the vegetation has become much greener over this two week period. To make it easier to compare, here are those images cropped and centered on the swollen rivers, side-by-side:

False-color RGB composites of VIIRS channels I-01, I-02 and I-03, taken on 5 April 2013 and 21 April 2013

False-color RGB composites of VIIRS channels I-01, I-02 and I-03, taken on 5 April 2013 (left) and 21 April 2013 (right)

There are a couple of important things to note about these images that are related to how VIIRS and its satellite (Suomi-NPP) work. One is that Suomi-NPP has an orbit with a 16-day repeat cycle. Every 16 days it should (if it’s in its proper orbit) pass over the same spot on the Earth at the same time of day. The images above were taken 16 days apart, and as you can see in the captions, were taken at the same time of day. The only difference in the area included in the images is the result of the start time of the data granules being 13 seconds off. This means that VIIRS is viewing all the same spots at the same viewing angles.

This leads to point #2: the VIIRS instrument has a constant angular resolution (recall that it uses a constantly rotating mirror to detect radiation across the swath) which, when projected onto the surface of the Earth, means that it does not have a constant spatial resolution. (See slide 12 of this presentation.) The spatial resolution of the high resolution channels shown here is ~375 m at nadir, and it degrades to ~750 m resolution at the edge of the swath. In the images above, the center of the VIIRS swath (nadir) is near the right edge of the data plotted. The left edge of the data plotted is about 80% of the distance from nadir to the edge of the swath. The loss in resolution over this distance may be enough to prevent VIIRS from detecting all the flooding that is occurring. But, the important thing is that we are viewing all these rivers at the same angles and the same resolution. This gives the best comparison between the before and after images.

A few more things to notice in the above images: there is snow in the northern part of Michigan’s Lower Peninsula, with ice on Green Bay and Lake Winnebago (all of which are easier to see in the image from 5 April 2013). Does anyone living there still remember last year’s record heat wave? Many places in this region had already had a number of +80 and +90 °F days, but it seems like a distant memory now. This year, winter doesn’t want to end.

One last thing for today: If you focus on Michigan again you might notice another area of flooding. This one is large enough it wouldn’t be impacted by any resolution degradation (even though it is near the center of the swath where you wouldn’t be worried about that anyway). I’ve zoomed in on the area here:

False-color composites of VIIRS channels I-01, I-02 and I-03 from 5 April 2013 and 21 April 2013

False-color composites of VIIRS channels I-01, I-02 and I-03 from 5 April 2013 (left) and 21 April 2013 (right)

This is along the Shiawassee River near the Shiawassee National Wildlife Refuge, a few miles southwest of Saginaw. This area of flooding is confirmed by these aerial photographs taken on 22 April 2013.

Wild Week of Wildfires, Part II

Posted on June 20, 2012 by Curtis_Seaman

Last time on “Wild Week of Wildfires“, we looked at the Little Bear Fire and High Park Fire, two lightning-ignited fires burning out west that were so hot they caused saturation in the two 3.7 µm channels on VIIRS (I-04 and M-12). There was mention of the Duck Lake Fire, a lightning-ignited fire in northern Michigan, which VIIRS also saw, and I couldn’t resist showing some more images.

On 9 June 2012, the same day the High Park Fire exploded (figuratively speaking), the Duck Lake Fire finally reached 100% containment after burning over 21,000 acres. The next day (10 June 2012), Suomi NPP passed over the Upper Peninsula of Michigan, and it was actually a clear day. (This joke comes courtesy of 20+ years experience of living in Michigan.) Even with 100% containment, the hot spot of the fire was still clearly visible in VIIRS channel I-04 (3.7 µm) that afternoon:

Channel I-04 image of the Duck Lake Fire from VIIRS, taken 18:18 UTC 10 June 2012

The highest brightness temperature in the burn area in this channel at this time was ~331 K. As we saw before with the Lower North Fork Fire, the high resolution false color composite of channels I-01, I-02 and I-03 is useful in highlighting the burn area:

False color RGB composite of VIIRS channels I-01 (blue), I-02 (green) and I-03 (red), taken 18:18 UTC 10 June 2012

Notice the large, brown area that coincides with the hot spot in the I-04 image. The combination of wavelengths used in this composite (0.64 µm [blue], 0.865 µm [green] and 1.61 µm [red]) is quite sensitive to the amount (and health) of the vegetation.

You might have also noticed several other interesting features in the image that show up better when you zoom in:

False color composite of VIIRS channels I-01, I-02, and I-03 from 18:18 UTC 10 June 2012

The Upper Peninsula of Michigan was based on mining for most of its history, and several large mines and quarries still exist, which VIIRS can easily see.

If you didn’t know any better, you might confuse the iron mine southwest of Marquette, Michigan with a frozen lake, or miraculously un-melted snow leftover from winter, since that is just what snow and ice look like in this kind of RGB composite. Compare that with the true color view of the same area:

True color RGB composite of VIIRS channels M-3, M-4 and M-5, taken 18:18 UTC 10 June 2012

In this case, the iron mine stands out as a bright red. Why?

The true color composite uses wavelengths at 0.48 µm (blue), 0.55 µm (green) and 0.67 µm (red). The red channel in the true color composite is actually in the red portion of the visible spectrum. The blue channel in the false color composite (0.64 µm) is also in the red portion of the visible spectrum.

This example shows that the iron oxide (rust) produced at the iron mine is highly reflective in the red portion of the visible spectrum. That’s what gives it the characteristic rust color. Iron oxide is not nearly as reflective at shorter or longer wavelengths, so it shows up blue when red wavelengths are used as the blue channel (as in the false color composite) and red when they are used as the red channel (as in the true color composite).

Let this be a lesson to anyone who uses the false color composite as part of a snow and ice detection algorithm. Snow and ice are not the only things to show up that color. You may be looking at a really large iron mine.

JPSS (SNPP and NOAA-20)

VIIRS Imagery and Visualization Team Blog

Tag Archives: michigan

Sea-effect Snow

Beginning of Autumn in the Great Lakes

Land of Lincoln Underwater

Wild Week of Wildfires, Part II