Atmospheric hydraulic jumps--rare beasts...

The photo in this post was taken on the morning of May 8, 2008,  by an instrument filming in visible wavelengths on the GOES-12 satellite. An animated sequence and the source of this discussion can be found here. The feature of interest is the long straight line that runs from left to right approximately through the center of the image (more specifically, from WSW to ENE).  This line extends from the thumb of Michigan on the left, across lower Michigan, through Toronto, and along the northern edge of Lake Ontario. This image (and the movie that was available to the meteorologists at the time) caused great marveling and wondering--what caused that long straight boundary? What caused the cloud-free region? What causes the more subtle straight lines in the clear region?

There is not enough topography in the region for this pattern to have been caused by topographic variations. It doesn't appear to be determined by land-water variations. It is clearly more moist on the north side than the south (clear, dry air).  During the night before this image was taken, a cold front had moved through causing widespread rain. The boundary appeared after the storm passed.  The consensus was that this was an atmospheric hydraulic jump formed when a thin layer of faster northern air discharged into a zone of lower velocity air to the south. The thiner fast layer (presumably laminar flow) transitions through the hydraulic jump into a deeper layer of slower air, presumably more turbulent).

Fortunately, data on the temperature and humidity structure in the atmosphere was available from rawinsonde for three stations near this phenomenon. This data showed that there was a temperature inversion (colder air over warm moist air) at low levels, and that the cold air (at the 900 hPa level--I'm not going to try to convert that to an actual altitude, but I think that it's a few kilometers). When a fluid goes through a hydraulic jump, it typically loses its laminar characteristics and becomes turbulent.  The turbulence, in this case, would mix in cold dry air that would "dry out" the moister air, producing the clear air downstream of the hydraulic jump. You can also see some fine linear clouds running from WSE to ENE in the clear area. The satellite geeks interpreted these as a manifestation of an undular bore--a gentle form of another hydraulic jump in which there is not a sharp single change of flow conditions, but a gentle oscillating one, often with multiple crests and troughs.

If anyone has photos of other atmospheric jumps, I'd love to collect them and post them to the blog--there are very few out there, and some are not available for use because of copyright limitations.

What should I call this "snow"? Rimed? Ice pellets?

This morning when I went out to early to wheel my trash can to the curb, the temperature was about 25 F (after an afternoon/evening of freezing rain/sleet). The low sunlight was reflecting off a monolayer of brilliantly-shining "snow" on top of the garbage can. The most striking thing about this photo to me is that the pellets, which were a couple of millimeters in diameter, were rimmed with tiny crystals; I think that's what gave the layer such a sparkling brilliance. The only other hint that I can offer is that the trees were all coated with a fairly thin, but definitely noticeable, layer that we would normally associate with freezing rain rather than sleet. I've tried to figure out what to call this snow (ice pellets? rimed pellets?). I should have poked around in the pellets to see if the small crystals were on all sides, or just growing out of the top--oops, big mistake...  A quick Google search on "sleet", "ice pellets" didn't turn up any photos like this one! Viewers comments on what to call it and what it says about its history of formation are most welcome!

ADDED: Emboldened by Mr. Lockwood's comment that this looks like "graupel," I add some information about the formation of graupel. From that impecable source, Wiki:

Graupel (also called soft hail or snow pellets) refers to precipitation that forms when supercooled droplets of water are collected and freeze on a falling snowflake, forming a 2–5 mm (0.079–0.197 in) ball of rime. Strictly speaking, graupel is not the same as hail or ice pellets, although it is sometimes referred to as small hail. However, the World Meteorological Organization defines small hail as snow pellets encapsulated by ice, a precipitation halfway between graupel and hail.

The term graupel is the German word for the described meteorological phenomenon. Its METAR code is GS.

And, indeed, if you Google "graupel" and then "images", you'll find some with the overgrowths like I was seeing above. Mystery solved, thanks to Mr. Lockwood!

Razor Clams, Anchors, and Fluid Dynamics

After a 2-month break while I focused on relaxing, refreshing, and working on my forthcoming book "The Dynamics of Disaster," it took a Science NOW article to lure me back to posting here (i.e., spend some time trying to understand that article, which is really what the blog exercise is all about!).

Apparently biologists can measure the strength of a muscle, and have concluded that clams to not have enough muscles to plow more than 1-2 centimeters into the ocean floor. Winter and colleagues took an empty clam shell, filled it with epoxy, and tried to drive it into an exposed seashore, concluding that a clam could burrow no deeper than 2 cm.

Why is it then, that clam-collecting, is a rather challenging exercise--the clams shouldn't be able to get away from the collectors.  In fact, they can dig themselves into the sand at a rate of roughly a centimeter per second or more. How do they do this? Amos Winter et al. have proposed a localized fluidization mechanism (Journal of Experimental Biology, 215, 2072, 2012). The authors say that the clam technique could be a model for engineering self-burying machines.

The physics of the clam defense is illustrated in the frame attached. First, the clam extends a foot, and then it lifts is shell up. The clam, then contracts the shell itself, sending blood into the foot to inflate it and provide an anchor. Then, with the foot acting as an anchor, the clam drags the shell down, a process repeated over and over about every second. Winter discovered the key to the clams success: when it contracts the shell, it relieves pressure on the soil and this pressure change results in water flowing toward the body of the clam. This "liquifies" the sand in the immediate vicinity of the clam. This liquefaction, in turn, reduces the drag on the clam, allowing it to pull itself down. The liquefaction process is local and ephemeral as the water then migrates back into the sand and the clam has to repeat the maneuver, over and over.

Clam digging isn't a recreational option in the midwest where I live, but this makes me really eager to try my hand next time I'm at the sea shore!

Hurricane Sandy and baroclinic instabilities

Hurricane Sandy (right side) and the line of clouds marking

the cold air front coming in from Canada

NOAA image taken from Joe Zagrodnik's 

blog mentioned in the text

11/06/2012: Here's a link to a marvelous animation of the whole evolution of the storm, produced by Matt Shatley at the University of Delaware, and here's one to a NASA animation.  Here is also an interesting comparison between the two hurricanes, Sandy and Katrina.

The press has thrown out the idea that a "baroclinic enhancement" to Hurricane Sandy allowed it to develop renewed strength as it approached a Canadian cold front brought in by the jet stream, but I haven't seen that term "baroclinic" explained anywhere in simple terms. When I try to Google it, I get more equations than words,** so I'm going to try to explain it here. Take heed: I'm not a meteorologist, just trying to learn!
    Two "baro-" words occur commonly in meteorology. A "barotropic" atmosphere is one in which the density depends only on the pressure; these are typically in the tropics. (Note the word "tropic" in this term.) A typical barotropic region is the southeast U.S. in the summer, or the tropics, where everyday is about the same: hot and humid. There are no weather fronts in a barotropic atmosphere.
      A "baroclinic atmosphere" is one in which the density of the atmosphere depends on both the temperature and the pressure; these are typically in the midlatitudes.  Baroclinic atmospheres have distinct air masses of different temperatures with boundaries (frontal boundaries) between the two. There are density gradients at any level of the atmosphere in baroclinic environments. One website suggested that you can remember the character of a baroclinic atmosphere by the word "clinic": the atmosphere is out of balance just like a person who feels out of balance would need to go to a "clinic." In fact, "baro" refers to surfaces of constant pressure, and "clinic" to surfaces of constant density.  A flow is baroclinic if surfaces of constant density are turned at an angle with respect to surfaces of constant pressure, a point that I'll get back to toward the end.
    Baroclinic enhancement occurs when a low pressure storm system gets close to a high pressure system, and when there are strong temperature gradients between the two systems. (In fact, in some of the blogs and articles, 'baroclinic' seems to really just refer loosely to "conditions with horizontal temperature gradients.)" I found this blog "WXNOTES" by meteorologist Joe Zagrodnik at Florida International University to be particularly helpful. Written before landfall of Sandy, the October 26 post systematically goes through the four factors that came into play in this storm:

• 1. Hurricane Sandy
• 2. A mid-latitude trough in the polar jet stream over the Great Lakes (Rosby wave phenomenon)
• 3. An upper-level low over the North Atlantic (forming, with an upper-level high to its north off the field of view of the photo, a "blocking pattern"); part of the North Atlantic Oscillation
• 4. The subtropical jet stream

In the photo above, the cold front is marked by a blue line, Sandy by the red symbol, the subtropical jet stream by a green arrow (bottom center), and the blocking pattern by the "L" in the upper right. The subtropical jet stream is already deforming Sandy from the symmetrical shape expected of a classic hurricane into the "comma" shape typical of Northeasters. The trough in the polar jet stream lies behind (west) of the cold front.  The blocking pattern keeps Sandy from escaping to the east. Although not obvious on surface level maps, or even on the 850 mb level map below, Sandy is sandwiched between the polar and subtropical jet streams.

Here's where Zagrodnik's background information is helpful; some material like this has been in my earlier Sandy posts:

"There are two main types of temperature instabilities in the atmosphere: vertical and horizontal. The Frankenstorm draws energy from both of them. Vertical instability comes from the difference between the warm ocean waters of the gulf stream and the cooler, atmosphere above. Hurricanes draw their energy from the ocean and release it in their vertical convective updrafts, creating a “warm core” of latent energy release around the center (remember, warm air rises). This warm core is clearly apparent around Sandy in [the image to the left in this blog post], as the 850 mb temperatures are warmer around Sandy’s center than in the surrounding areas.

A forecast map of temperatures
at 850 mb for Monday morning,
Octrober 29 from Zagrodnik's blog referenced in text

The second type of instability, horizontal (or baroclinic) instability, is employed by mid-latitude, extratropical cyclones. Sandy’s warm core is in sharp contrast to the colder 850 mb temperatures over the continent. The potential energy between the warm and cold air is released when the cyclone draws warm air northward (the red arrow to the east) and cold air southward (the red arrow to the west). This storm is a hybrid system because usually weather systems only derive energy from one source or the other. Both of these instabilities tend to be at their greatest in the autumn months, and both are being combined in this storm."

     The 850 mb map shown here (a few km above the surface) shows these strong horizontal temperature gradients, as well as (small white arrows) the wind speeds. Colors are temperatures with yellows and reds being warm and blues and purples cold. In addition to the two red arrows mentioned above, the third one, farthest to the north, indicates where winds  coming in from the northeast are very strong. These winds are "squeezing" between the two low pressure systems (in the yellow-green region) and the high pressure system (blue-green region) to the north.
     (For more detail, see the discussion of the 300 mb and 700 mb information on Zagrodnik's blog.)

Unstable coin, from
Eastin's PPT

     Now, to get back to the baroclinic instability between Sandy, which has a warm core, and the cold air being brought down from Canada by the jet stream. Think first about an idealized situation: a circular mass of cold air (corresponding roughly to the purple and blue above) surrounded by warm air.  If the earth did not rotate, the cold air would tend to sink and spread out uniformly under the warm air.  Because the earth rotates, however, Coriolis forces cause the spreading cold air to gain a component of motion parallel (azimuthally) to the circumference of the circular mass of cold air. That is, the Coriolis force is balancing the gravitational force and prevents the column of dense air from spreading infinitely far away.  The circumferential wind around the edges of dense polar air masses is called a "thermal wind." If this wind is sufficiently dense, an instability develops in the wind, and large-scale eddies spin off of it and move toward lower latitudes. In so doing, they horizontally mix up the cold and warm air, and convert potential energy stored in the cold air into kinetic energy of the eddies.
     Another way*** to think of this is to imagine two columns of air that have the same pressure at sea level, a cold one with density r1 and a warm one with density r2, where r2 is less than r1.  At any height in these two columns, the pressure in the warm column will be less than that in the cold column, destroying the hydrostatic balance between the two columns. Cold air would flow in under, and mix with, the adjacent hot air until the pressure balance is restored.

Unstable/stable atmosphere
from Eastin's PPT

     A third way, through a PPT presentation that I found by Prof. Mathew Eastin of UNC Charlotte can be illustrated by the "coin physics" on the adjacent figure. Sitting on its edge, the coin is unstable with a center of gravity above the surface of the table.  There is a potential energy in the coin that can be defined by the position of the center of gravity. A small disturbance will cause the coin to fall (potential energy being converted into kinetic energy) to a stable position with lower potential energy when it comes to rest. Note that the warm air has indeed gone northward, and the cold air southward as needed to reduce temperature gradients across the hemisphere.
     Now, look at a cross-section of a hypothetical, four-layer atmosphere from the equator to the pole. Uneven solar insolation causes the air at the poles to be denser than the air at the equator and so the layers of the atmosphere are tilted at an angle to the isobars, lines of constant pressure. (Note, in the top part of this figure that although pressure increases downward, isobars are not horizontal but would be tilted toward the upper left). Each layer has a center of mass, as does the whole system (red dot). Each layer, and the system, has some available potential energy.  If the system is pushed by a small perturbation, just like the coin, it will move to a lower energy, more stable state with a lower center of gravity for the dense layers, and a higher center of gravity for the light layers.  The center of gravity for the whole system, dominated by the dense layers, moves down, reducing the potential energy of the system.  Another day of uneven solar insolation and the unstable situation is set up again.
   
   

_______________________________
P.S. Returning to Will Komaromi's blog that I referred to in an earlier post, he points out a measure of storm strength called the Integrated Kinetic Energe (IKE), a measure that combines size of the wind field with the strength of the winds. Sandy had the highest IKE of any tropical storm on record, with almost 4 times the IKE of Katrina. This measure, unlike the category classification, explicitly accounts for the immense size.

**An exception is here: http://www.tos.org/oceanography/archive/21-4_nadiga.pdf
***The text at this link was written by Yochanan Kushir, 2000.

Why did Sandy leave so much sand in Seaside Heights and even NYC?

Sand in Seaside Heights, New Jersey on 10/31/2012
Photo by Mario Tama/Getty Images published here

Coastal storms move sand in several directions: on-shore, off-shore, and sometimes, along-shore. I thought it would be worth reviewing some dynamics of sand movement during big storms in light of the inevitable decisions about moving around sand post-Hurricane Sandy. This topic is worthy of an in-depth discussion that would occupy a significant fraction of a coastal geomorphology course, so this is just a shallow overview. (Plus, I live in central Illinois, not exactly full of beaches where I could get first-hand experience!)
    While this is a particular and immediate case study, sand movement is an issue at many coastal cities. For example, project SANDAG is a $28 million regional sand replenshment project in the San Diego area; the Army Corps of Engineers proposes to spend $251 million on sand projects for Encinitas, Solana Beach, and San Clemente.

Schematic of sand movement

    Beaches are both formed and destroyed by wave action in the normal (note that word "normal") course of geologic processes. Waves move sand particles either in suspension or by saltation--bouncing along the bottom. These particles come from the erosion of rocks offshore, or from the erosion of sediments along the coast.  Sometimes the sediment is transported long distances--the white sand along Florida's "Emerald Coast" comes from the erosion of quartz in the Appalachian Mountains. Some waves, referred to as "constructive" waves move sand up onto a beach; "destructive waves" move it from the beach, forming a  slopes that disappear some distance below the surf. Wind often moves the sand on beaches into dunes.  Offshore, a common coastal landform is a barrier island, a long linear sandy island or series of islands that can extend over 100 km along a shoreline. Long Island is an island formed on glacial moraines, but has elements of barrier island geology where barrier islands of sand divide the lagoons south of it from the Atlantic.  The true barrier islands are Coney Island, Long Beach Barrier Island, Jones Beach Island, Fire Island, and Westhampton Island. In New Jersey, the famous beach spots of Seaside Heights, Atlantic City and Wildwood are on barrier islands. During the storm, water covered many of the barrier islands. As bulldozers and bucket loaders tried to clear the streets from sand brought inland, they were piling up like snow after a winter snowstorm, in some places there were piles up to 20' high.
      Beaches change all the time in response to the seasons and to individual storms. Change is the norm. During storms, waves attack parts of the beach system, moving sand from high-energy sites to lower energy environments. This process can take sand toward the shore, by the process known as "overwash" that appeared in the reports about Hurricane Sandy, or can take sand away from the shore, transporting it offshore and out, sometimes into sand bars. This process tends to cause storm waves to break further offshore and decreases the wave action that actually reaches the beach. As sea level rises in the future, beachs will migrate landward, somewhat like a tractor tread rolling over itself.  The transport of sand into New York City or Seaside Heights (photo above) is an example of the transport of sand inward, and a portent of things to come if the number or intensity of hurricanes increase due to global warming.
    Prior to human intervention, cycles of storms would work on beaches, repeating this process over and over on many scales, resulting in the beaches that we had in the early part of the 20th century.  As we humans have constructed infrastructure, often far from the beaches, the supply of sand to the coasts has been greatly altered, often reduced. Thus, the normal beach building processes have, for many decades, been altered by humans. Beaches are now, at great expense, often only replenished as sand is carted from one place to another in an ongoing intervention with where "Mother Nature" wants to put it. Fittingly, the first beach replenishment project in the U.S. was at Coney Island in 1922-1923.

Here's a link to a before and after image of the Mantoloking Bridge and damage near it where the hurricane cut a new connection across the barrier island connecting the Atlantic Ocean and the Jones Tide Pond. This bridge, built in 2005, cost nearly $24 million, with an additional $5 million for design and purchase of rights-of-way. It has been labelled as "unstable" now.

Frankenstein Storm: Pressure in the eye is 943 millibars--so what?

Hurricane Sandy NOAA

11/03/2012: This now includes a correction about where the storm originated.

Well, to answer the question in the last post, I am indeed stuck in Washington D.C. My first hurricane experience! So far, it's the lull before the storm (but I only finished this post after the storm). A gentle drizzle when we woke up this morning (Monday), increasing rain throughout the morning, a lull between rainbands of the hurricane around noon, and increasing wind and rain again now, mid-afternoon. We journeyed out to see what is going on in the stores, and found that at Target, there were four items that racing (or, rather, had raced) off the shelves: water jugs (small bottles still seemed to be in ample supply), bread, batteries, and camping gear.
         I've been watching the pressure in the eye of the hurricane with interest because two days ago, a I reported that it was 972, and that one prediction was that it might go as low as 950 mb.  The last reading I saw was 943 mb, but the internet in the hotel where I'm staying has the speed of a phone line, so I haven't been able to update my stats. (I saw 941, but the official record appears to be 943 mb. It broke the record of 946 mb from the 1938 "Long Island Express" hurricane.)
        I'm going to try to explain what has gone on as my way of learning it, but first I refer you to a very authoritative source, that of meteorologist Cliff Mass. Excellent discussion! I am taking my discussion from a number of sources including Mass's blog, and the Seattle Times new service, which in turn, compiled its information from the Associated Press, The Washington Post, and Tribune Washington Bureau.
        Unlike most Atlantic hurricanes, Hurricane Sandy did not begin as a wave off the coast of west Africa, but rather started in the Caribbean. A low-pressure air trough was spun into a slow counterclockwise rotation by weak winds around October 19, and thunderstorms over the next few days sucked up energy from the warm Caribbean waters. On October 22, it became Tropical Depression 18, quickly developing into Hurricane Sandy which started moving north. The Gulf Stream is as much as 5 to 9 degrees warmer than normal for this time of year. It grew into a suze where tropical storm level winds (39-74 miles per hour) extended over a diameter of 940 miles. It is the largest Atlantic storm known to have made landfall, at least since the government began keeping records in 1988, and very likely well before that time.
        The puzzles to experts, and amateurs alike, were "why did it come so far inland so late in the year, and what's going to happen about it since most hurricanes like this head back out into the Atlantic where they peter out over the cold waters?"  They are usually chased out to sea by "onshore weather patterns," (a very unhelpful phrase in the Seattle Times article!). As I mentioned in my previous blog, there was a high-pressure system in the North Atlantic Oscillation west of Greenland, and a huge low pressure trough moving eastward across the U.S. The two combined to push/pull it westward. 
        Even though Cliff Mass's last post was on Sunday (Oct. 28), a day before the storm made landfall, his description/prediction of what would happen after landfall is intriguing (and you really need to see the color images on his blog to visualize this).  The storm transforms from a tropical storm with a warm core into an "extratropical storm" with a cold core.
        First point: tropical storms generally form in the tropics where there are not large gradients horizontally in temperature (tropical to subtropical at most). They derive their energy from the warmth and moisture of the oceans.
        Second point: midlatitude storms form in regions of large temperature gradients. If you travel through the mid-latitudes at this time of year, you go from the climate of New Orleans to that of, say, Juneau, Alaska! Midlatitude storms get their energy not from the warm oceans, but from the warmth on their south side and its contrast to the colder air on the north side. This temperature gradient fuels our mid-winter storms. These are cold-core storms.
        Cliff Mass predicted that Sandy would go an "amazing transition" from a warm-core to a cold-core storm (through what is called an "Extratropical Transition"). Many tropical storms weaken when they go through this transition, but Mass thinks not for Sandy. The two energy sources (ocean energy source and latitude-dependent energy source) work together for long  enough to result in an overall strengthening and expanding system.
        In his earlier October 26 blog, Mass compared/contrasted the predictions of two major models, considered to be the most important ones: The European Center model (ECM) and the US GFS model. Up until late last week, the US GFS model was predicting a low pressure of around 950 mbar that would swing up over Long Island on Tuesday morning, and then hang around inland, eventually moving northward on Wednesday. On the other hand, the ECM was forecasting a sub-940 mbar low pressure making landfall on central New Jersey, then circles around there before moving northward. To quote Mass "The European Center and GFS models tend to overdo the deepening in such situations, so I don't believe we will see anything below 950-960 mbar." Last I saw, the measured pressure in the low was 941 mbar. Good job ECM!

       Let's summarize: An eye pressure of 941 mbar, sustained winds at 90 mph, and a diameter of nearly 1000 miles. To the right, I post one version of the Saffir-Simpson Scale, though many others only have wind speed. Using wind-speed alone, this would be a category 1 scale. But, using pressure in the eye, it would have been a category 4. Am hoping some readers will pitch in and explain this one! Meanwhile, we mull over that 20 years ago, Halloween 1991, the "Perfect Storm" unfolded almost exactly like this one.

Frankenstorm: Hurricane Sandy

Friday 2:00 p.m. EDT NOAA projection of the path
of Hurricane Sandy

I have a meeting in Washington D.C. until Monday morning and am wondering if I'll be able to fly out of here back to Illinois due to the fact that a monster storm, Hurricane Sandy, is creeping up the east coast of the U.S. If the NOAA projection to the left is correct, it looks like I'll make it, but if not... The head of the National Weather Service has been quoted as saying that he's estimating $1 billion U.S. damage (let's remember that last year there were fourteen billion dollar disasters in the U.S., most of them storm related (hurricanes, tornadoes, floods, droughts, heat waves, and wildfires).

This storm is getting a huge amount of press coverage, given that it's only a Category 1 hurricane at the moment.  It is getting close to full moon, which means high tides, which means greater than average storm surge as Sandy approaches. Surge and flood damage in the south east is already severe. It caused at least 21 deaths as it passed over Haiti, Jamaica and Cuba.

The mysterious thing about this storm is the strong hook to the northwest that you can see in the graphic above starting on Monday.  This is very unusual for an Atlantic Hurricane, and even more unusual, apparently, for an October hurricane. This is the tail end of the hurricane season, supposedly. So, what's going on with this storm? It turns out that it's not all that easy to dig out of the WWW and news stories.  (At the bottom of this post are some links that I found helpful.) However, a post by Will Komaromi, Ph.D. student at the Rosenthal School of Marine and Atmospheric Science, University of Miami, is excellent, and much of the material below is from that post.

Normally, as an Atlantic hurricane moves north, it encounters cooler water, drier air away from the tropics, and greater wind shear. In other posts, I've discussed the individual effects of these on hurricanes.  However, this year, the Gulf Stream along the Atlantic Coast is warmer than usual.

To address this, we need to look at what is happening out west.  There is a large cold weather system, currently bringing snow to Colorado, moving across the country, including a deep trough of low pressure (called a 'midlatitude trough'.) This storm is a large low-pressure system, so, very simplistically, it'll suck Hurricane Sandy in to the west. When the two systems interact, there will be a lot of action--thunder, lightning, rain by the bucketful, and snow. Very similar events occurred 21 years ago in the now famous "Perfect Storm" of October, 1991, that became the topic of a best-selling book and movie. There is an excellent graphic on Komaromi's WWW site above, comparing the two systems at the scale of the continental U.S.

Instead of weakening, the hurricane system merges with the mid-latitude trough (including strong jet stream winds that can provide extra energy to the hurricane). Sharp gradients in air temperature and density between the two systems works (creating sharp pressure gradients, known as the baroclinic instability) works opposit of the weaking effects described above.  Air moving out (diverging) from the hurricane system at upper levels combines with the baroclinic instability to create a low-pressure eye to the storm. The lower the pressure in the eye, the stronger the hurricane.  In the Perfect Storm, the pressure dropped to 972 mb in the eye; Komaromi says that it's possible that the pressure in the eye of Sandy could be as low as 950 mb.

A final factor in this is that the pressure system lying over the eastern North Atlantic, known as the North Atlantic Oscillation, is in a negative phase, meaning that there is high pressure to the north east of the hurricane that will prevent it from curving out over the Atlantic like most hurricanes do.  It's helping it to move westward (called retrograde by the meteorologists) right into the east coast of the U.S.  All-in-all, it looks like the northeast is in for a prolonged (days) period of severe wind and drenching rains, and that my childhood stomping grounds in western Pennsylvania are going to get a lot of snow from the cold part of this combined system.

I'll join with other meteorologists and bloggers to say stay safe, be prepared to be self sufficient, and to help your neighbors. Especially, check on elderly relatives and friends who may be vulnerable to early winter cold if the power goes out!  And, get out and vote early if you live in these regions; in most, early balloting is already underway. Election Day isn't until November 6, but some meteorologists are concerned that this storm could take down powerlines that are going to take a long time to repair, days to even weeks.

Other references:
http://www.washingtonpost.com/blogs/capital-weather-gang/post/hurricane-sandy-on-collision-course-with-mid-atlantic-and-northeast/2012/10/26/1f82c84c-1f7d-11e2-9cd5-b55c38388962_blog.html

http://www.nytimes.com/2012/10/27/us/east-coast-keeps-a-watchful-eye-on-hurricane-sandy.html

http://www.rsmas.miami.edu/blog/2012/10/25/perfect-storm-set-to-occur-on-21st-anniversary-of-original-historic-event/

L'Aquila trial, rogue climate experiment, and realignment of science-society relations

L'Aquila, Italy government office destroyed by
the 2009 earthquake. Image from
http://en.wikipedia.org/wiki/File:L%27Aquila_eathquake_prefettura.jpg

Scientists have been watching events taking place in a small medieval town, L'Aquila, Italy, unfold for about three years, and the world took notice yesterday when six scientists and one civil servant were convicted of manslaughter and sentenced to six years each in jail, and a total of over three million dollars in fines, amongst other penalties. This verdict came only a week after it was discovered by the Canadian environment ministry discovered that a California businessman more-or-less secretly spread 100 tons of iron dust into the ocean off western Canada to increase the growth of plankton to increase the salmon fishery. Something is changing in the relation between science, scientists, and members of the general public.
       One thing that these two events have in common is that a person outside the "scientific establishment" and associated government agencies was involved both in gathering data and interfacing with the public.  Here's a short summary of background as best I can infer from press reports (which, in the case of L'Aquila, are often misleading). see footnote**
     In 2008-2009, swarms of earthquakes shook L'Aquila, a town that sits in one of the most seismically active regions of Italy. Residents were unnerved by the constant seismic activity, but had a long history of living with it.  The older residents were used to leaving their houses and remaining outside, sometimes in their cars or encampments, until the activity subsided. Compounding the tension in 2009, a local resident who worked as a laboratory technician, was gathering radon data, and made unofficial predictions that alarmed the residents. He was cited for "procurato allarme"--essentially instigating public alarm, and was forbidden to make any public pronouncements. (There is no evidence in the peer-reviewed literature that radon measurements can reliably predict earthquakes and there is substantial evidence that radon emissions vary considerably due to many factors.)
       Because of the escalating tension, six scientists and one government official, members of the Serious Risks Commission,  were hastily, were called to a one-hour meeting in L'Aquila with the goal of furnishing the residents "with all the information available to the scientific community about the seismic activity of recent weeks". From here on, events are complicated, and the reader should see the account referenced below for details.  Briefly, the government official reported to the public after the meeting that the scientists had said the seismic situation posed "no danger", a point of contention throughout the trial. The scientists appear, in fact, to have said that the probability of a large earthquake was small, but that it was possible.
      Whatever the details, the residents of L'Aquila said that they felt reassured by the message conveyed and therefore, according to their account, did not follow the age-old tradition when the tremors continued, but remained in their houses. On April 6, an earthquake struck, tragically killing 309 people. 
     A year later, the six scientists and government official were charge with involuntary manslaughter for being negligent in giving advise on the risk to public safety. Note: they were not accused of failing to predict the earthquake as many news reports continue to assert.
     Yesterday, the judge handling the case convicted the seven on the charges and handed down a jail sentence that was 50% longer than requested by the prosecution. It may be three months before the logic behind this sentence is revealed. According to Italian law, the sentence can be appealed, and the seven will not be jailed unless the appeals fail. 
     The scientific community is in shock, especially the geologic community and those who work with hazards, risk management, and public safety, but there are obvious parallels running through other communities, such as the medical community. Scientists are already becoming wary of existing communication protocols, even in countries like the U.S. where much thought and experience, both in earthquake and volcanic hazards, has gone into defining how the interface between science and the public is addressed. As I wrote this, an announcement appeared on the WWW that the head of the Major Risks Committee in Italy and several of its top scientists have resigned saying that there are no longer good working conditions. There is a real danger that public safety will be compromised as scientists become afraid that any comment they make can put them in jeopardy of prosecution. I join with my many colleagues around the world in condemning the verdict--it helps neither the victims, nor potential victims of future events, nor those who try to ameliorate the effects of natural hazards and keep them from turning into disasters.
    But, I would like to ruminate on something else (since this is a rare diversion of my posts from pure science). Although it may seem unrelated, a story just reported last week initially by the Guardian, and then by the New York Times, raises concerns that are not entirely unrelated. Just as the convening of the meeting of the Serious Risks Commission was precipitated by the activity of a member of the general public, in this case a California business man, Russ George, has precipitated environmental concern because of a geoengineering experiment.  Geoengineering is an activity in which human intervention (the engineering part of geoengineering) is used in a large-scale attempt to alter the natural balances of the planet (the geo- part). The pros and cons of geoengineering are an issue of active debate in the scientific community. There have been carefully designed, peer-reviewed, small-scale geoengineering experiments of the kind that George performed, but this one was apparently carried out in secrecy, without peer review of the procedures, nor, apparently, a statement to the community of environmental impact. The argument that the experiment took place outside the territorial waters of Canada is a red herring.
     George was reportedly paid $2.5 million by the Haida people of northern British Columbia, to scatter 100 tons of iron dust in the Pacific off the coast of northwestern B.C. The Haida have relied for centuries on the salmon runs, and have been suffering after the crash of salmon nurseries in the Pacific. Their waters are depleted in iron, and the addition of iron in this experiment spurred growth of the plankton. Since plankton absorbs carbon dioxide and then dies and falls to the bottom of the ocean, it is one proposed scheme for burying atmospheric carbon.
The Haida apparently hoped that in addition to increasing the salmon fishery, they could take advantage of carbon offset credits.
     The problem with this experiment, and with many geoengineering proposals in general, is that if the experiments go awry, the can endanger the planetary system at a very large and potentially catastrophic scale.  The so-called "precautionary principle," under which many European countries operate, states that if a given action has the potential for harm, even in the absence of scientific consensus that the action is harmful, the burden of proof that it is not harmful rests on those taking the action. That is, potentially harmful actions should be avoided.
     The absence of transparency, as well as the fact that the experiment appears to have violated two international agreements that Canada had signed, has made scientists and policy makers as well, angry. (The international experiments are the United Nations convention on biological diversity and the London convention, which found that large-scale experiments in ocean fertilization are unjustified.)
      These seemingly unrelated cases bring a few points to attention. Each one involves an area of science (earthquake prediction; ocean fertilization) getting a lot of attention from scientists and policy makers. Considerable thought has been put into protocols, transparency and accountability for work in the subject.  There are methods for testing hypotheses and for dissenting with the prevailing paradigm: data are gathered and tested according to principles long established by the scientific method. Such visible issues attract a variety of people, including some ill-trained in scientific methods. In these two cases, there was/is a person operating outside the bounds of these methods and outside the protocols that have been established for use of the information. Scientists holler that either data do not support the rogue claim (the situation with radon and earthquakes and with previous ocean fertilization experiments), or that the design of the experiment and its results are not subject to scrutiny (the case with the current rogue fertilization experiment). In either case, the public is often subjected to ambiguous and contradictory opinions and recommendations. Continuous communication of data to the public and education of the public about the issues are necessary. Only in this case will individuals have the skills and then the resources to make informed decisions, to elect informed officials, and to evaluate both data and advice provided. In L'Aquila, such procedures should have led to improvements in construction and reenforcement of buildings; that did not happen. In the case of ocean fertilization, this should lead to an informed public that can decide if massive geoengineering practices should be attempted. 
     Some established procedures are better than others. Clearly there were flaws in the way that events were handled at L'Aquila. But even in places like the U.S. earthquake and volcanology programs where considerable thought has been given to crisis procedures, there are likely to be flaws that will be revealed when the system is stressed. Those who acted in good faith to establish the procedures, and those who work within the system acting with integrity, should not be made scapegoats when these flaws are revealed. It is imperative that failures be used to improve the system so that it will be better when the next situation arises. The relations between scientists, policy makers, and the public are changing, some even saying that that the L'Aquila trial is a "watershed case." (Tom Jordan, director of the Southern California Earthquake Center). Geoengineering may be a L'Aquila of the future.



***An excellent detailed account of the events by Stephen S. Hall can be found here in Nature:
http://www.nature.com/news/2011/110914/full/477264a.html

Here is another thoughtful discussion from one who has been involved with volcanic risk:http://www.nature.com/news/2011/110914/full/477251a.html

Felix Baumgartner: The physics of supersonic falling

Felix Baumgartner exiting his balloon to begin his descent
October 14, 2012

CORRECTED ON 10/16/2012 WITH THANKS TO ERIC LANE WHO SPOTTED THE ERROR (I HAD READ THE NEWSPAPER WRONG AND THOUGHT THAT HIS VELOCITY WAS 833.9 FEET PER SECOND, NOT MILES PER HOUR. HIS COMMENT IS ATTACHED.) 


Twenty four miles (128,100 feet) above Earth, a man encapsulated in a suit much like the astronauts wore on the Moon stepped out of his capsule to begin a death-defying plunge. Whether you approve or not, it's difficult not to admire the courage, imagination, and training that leads up to such a feat!

According to the press, he reached a velocity of 833.9 miles per hour (roughly 371 meters per second) and a Mach number of 1.24. Using the definition of Mach number (M = v/c, where v is his velocity and c is the speed of sound), the speed of sound was 298 meters per second. 

The speed of sound in a perfect gas depends on the square root of the temperature (in degrees Kelvin). The sound speed in air at ambient sea level temperature is about 330 meters per second, and so he would have still been falling at supersonic velocities, though his Mach number would have been slightly less.

Now, here's a thought exercise: If he had jumped through a bottle of bubbly champagne, and attained the same velocity, his Mach number would have been as high as 15 or 30 because the speed of sound in bubbly mixtures is greatly reduced by the presence of bubbles. It could be as low as 10-20 meters per second. The physics behind this is that the sound speed depends on the compressibility and density. A bottle of champaign with tiny bubbles has the density of the liquid (~water) but the compressibility of the gas (CO2), and hence a significantly lower sound speed than pure water and also less than pure air at the same temperature.

Even more extreme (and this is not a suggested experiment!!), if he had attained this velocity in a pot of boiling water, his Mach number could have been of the order 300!! Boiling water has the compressibility of champagne, but an additional factor that reduces the sound speed to only a few meters per second--latent heat transfer as water changes phase (liquid-to-steam-and back to liquid) as the tiny pressure pulses of sound waves pass through it. Here's a reference to the calculations:

Kieffer, Susan Werner, Sound speed in liquid-gas mixtures: Water-air and water-steam, Journal of Geophysical Research, v. 82(20), page 2895-2904, 1977. 

(Yep, I did publish in 1977, and yep, I'm that old! But, it was one of my first papers! I was working on the dynamics of eruption of Old Faithful geyser and it's boiling water, as an analog for eruptions of volcanoes.) 

Here's the abstract from that paper:

The sound speed of a two-phase fluid, such as a magma-gas, water-air, or water-steam mixture, is dramatically different from the sound speed of either pure component. In numerous geologic situations the sound speed of such two-phase systems may be of interest: in the search for magma reservoirs, in seismic exploration of geothermal areas, in prediction of P wave velocity decreases prior to earthquakes, and in inversion of crustal and upper mantle seismic records. Probably most dramatically, fluid flow characteristics during eruptions of volcanoes and geysers are strongly dependent on the sound speed of erupting two-phase (or multiphase) fluids. In this paper the sound speeds of water, air, steam, water-air mixtures, and water-steam mixtures are calculated. It is demonstrated that sound speeds calculated from classical acoustic and fluid dynamics analyses agree with results obtained from finite amplitude ‘vaporization wave’ theory. To the extent that air and steam are represented as perfect gases with an adiabatic exponent γ, independent of temperature, their sound speeds vary in a simple manner directly with the square root of the absolute temperature. The sound speed of pure liquid water is a complex function of pressure and temperature and is given here to 8 kbar, 900°C. In pure water at all pressures the sound speed attains a maximum value near 100°C and decreases at higher temperatures; at high pressures the decrease is continuous, but at pressures below 1 kbar the sound speed reaches a minimum value in the vicinity of 500°–600°C, above which it again increases. The sound speed of a water-air mixture depends on the pressure, the void or mass fraction of air, the frequency of the sound wave, and, if surface tension effects are included, on bubble radius. The admixture of small volume fractions of air causes a dramatic lowering of the sound speed by nearly 3 orders of magnitude. The sound speeds of the pure liquid and gas end-members are nearly independent of pressure, but the sound speed of a mixture is highly dependent on pressure. Calculated values for water-air mixtures are in good agreement with measured values. The sound speed in a single-component two-phase system, such as a water-steam mixture, depends on whether or not equilibrium between the phases on the saturation curve is maintained. Heat and mass transfer which occur when equilibrium is maintained cause the sound speed to be much lower than under non-equilibrium conditions in which heat and mass transfer are absent. The sound speed in a water-steam mixture may be as low as 1 meter per second.

Typhoon Sanba--and what is a "super typhoon"?

Sanba--NOAA image

Typhoon Sanba is churning away in the Phillippine Sea, having hit Okinawa and moving north toward South Korea and, eventually, China.  Sanba can't be aware of it, but it's got two friends in the eastern Pacific--Kristy and Lane-- and one in the Atlantic--Nadine--that NOAA is keeping track of. Sanba is going to encounter warm water for the first half of its trip between Okinawa and South Korea, but cooler water on the second half. That should at least be some help in moderating its intensity before it strikes South Korea, but it is still expected to be severe. China is already calling on residents to prepare for the storm. The possibility of landslides looms large there.

Sanba has been called (briefly) a super-typhoon. In a previous post, I discussed super-typhoon Tracy, which was so large that it would have covered the whole western half of the U.S. Sanba, like Tracy, is a huge storm.

Projected trajectory for Sanba. NOAA image.

However, sheer size doesn't make a super-typhoon. To be officially a super-typhoon, a cyclonic (rotating) storm must have winds that reach "maximum sustained 1-minute surface winds" of at least 150 miles per hour (65 m/s, 130 knots). It is the equivalent of a strong category 4, or category 5, hurricane in the Atlantic on the Saffir-Simpson scale, or a category 5 "severe tropical cyclone" in the Australian basin. As an example, super-typhoon Pongsona that hit Guam in 2002 had gusts to 173 mph, and sustained winds of 144 mph. This storm also was huge, with an eye almost 40 miles in diameter.

Sanba formed from a tropical depression on September 11, and fueled by warm water, had maximum sustained winds of 175 mph, with sustained winds to about 200 mph (information from earthsky.org here.) It made landfall in northeastern Okinawa early Sunday morning at 6:30 a.m. Okinawa time (5:30 p.m. east coast time in the US). The eye was nearly half the islands length in diameter. At this time, the maximum sustained winds were down to 120 mph, and gusts were measured up to 149 mph. (The Joint Typhoon Warning Center is the U.S. miliary agency responsible for issuing tropical storm warnings in the Pacific.)

Here are other posts on typhoons:

Megi: Taiwan
Tracy: Australia, Queensland
Irene: U.S. east coast
Yasi: Australia

Mirages

Mirage in Puget Sound
Image from the referenced Cliff Mass blog

A blog that I really enjoy is by Cliff Mass, a meteorologist, on the weather, climate, and meteorology of the Pacific Northwest. There is a really great discussion today (September 12, 2012) of mirages and the atmospheric conditions that produce them. Check it out!

Earthquake and possible tsunami near Costa Rica--criteria for a tsunami

Location of today's earthquake
From CNN.com

At 8:42 a.m. today, a magnitude 7.6 earthquake struck offshore 88 miles from San Jose, the capital of Costa Rica. It was located at 40 km (25 miles) depth. A tsunami alert has been sent out for the Pacific coasts of Central and South America.  Here is a link to the USGS description of the tectonic setting of the quake.

I'm not sure what all of the official criteria are for a tsunami, but here are a few factors that are important:

Not all earthquakes generate tsunamis, a problem that causes nightmares to tsunami forecasters. When should a warning be issued?  When should one not be issued?  To forecast whether or not a tsunami will form and how big it might be, geophysicists and oceanographers work together with complicated data analysis and modeling software on huge computers that continuously monitor seismic signals coming through the earth. It takes about 25 minutes for seismic waves from an earthquake to reach all stations of the global seismic arrays and another ten minutes for computers to crunch the data through models. (I recently discovered that the USGS sometimes learns about the occurrence of an earthquake via text messages even before the seismic waves arrive at stations! How cool is that?!!)

Officials can now issue a preliminary estimate of tsunami danger in less than an hour after a quake. During this hour after an earthquake, scientists are searching for four pieces of information that determine the likelihood of a tsunami: the geometry of the fault motion, the energy and depth of the earthquake, and the depth of water over the fault rupture zone.

Some earthquakes produce tsunamis, others do not. Earthquakes in which the dominant motion is lateral, so-called “strike-slip” like that on much of the San Andreas fault in California, do not generate tsunamis. They form, rather, during quakes in which rocks have some vertical motion along a fault, causing a bulge or depression to form in the water. Such quakes, known as “tsunamigenic” earthquakes, occur in the subduction zones of plate tectonics—primarily those bordering the Pacific Rim, where two thirds of all tsunamis occur, but also in the Indian Ocean, the Caribbean, and the Mediterranean. During the Tohoku earthquake, an area of nearly six thousand square miles--the size of Connecticut--thrust up as much fifteen feet, disturbing an enormous volume of water. Oceanographers need to know the depth of the ocean over the rupture zone because the more water above the displaced crust, the more water that moves out into the tsunami.

The energy of an earthquake and the process of transmitting the energy to the water are both related to the area of the displaced ground and how far it moves during the quake, its displacement. Both of these quantities are captured by the seismic moment, a quantity that is related to the Richter magnitude. This relation of magnitude to area and displacement of the crust is the reason that estimates of the magnitude of the earthquake are so important in prediction of the size of a tsunami. Boxers understand this piston physics well—a big fist thrown in a long punch is more powerful than a small fist with a short reach. 

The potential for devastation by a tsunami depends on the fault geometry in one another way. Waves from a fault rupture are bigger in some directions than in others.  Waves spreading away from the long sides of a rupture--that is, spreading perpendicular to the fault plane--tend to be much bigger than those that spread away from the tips of the fault, parallel to the fault plane. If two coastlines are at equal distances from a fault, the one perpendicular to it is more likely to experience damage than the one parallel to it. Because of this geometric effect, Bangladesh—though lying nearly at sea level—was spared much devastation from the Sumatra earthquake.

The depth of an earthquake matters because a tsunami is only produced if the sea floor moves into the overlying ocean. If the earthquake is buried too deep in the crust, the fault that caused it may never break through to the sea floor or, if it does, may be moving too sluggishly to give the water a good punch. Shallow earthquakes with substantial vertical displacements and high moment magnitudes generate the largest tsunamis. The depth of water over the displaced crust matters because it, along with the area of the displaced crust, determines the initial volume of water displaced and able to move into the tsunami.

**This is an excerpt from my forthcoming book "The Dynamics of Disaster", due out summer 2013 from Norton Press. It can be preordered from Amazon.com here.