Monday, July 9, 2012

One Afternoon's Notes On Physics Of Snow

notes from: "Field Guide to Snow Crystals", by Edward R. Lachapelle

(if i get ambitious i'll find some pics to accompany the terms)

I'm always so fascinated that no matter where you look in this universe, complex heterogenous structures are always forming, always these curious complications.  all of this is just water, the response of trillions of trillions of water molecules to their their environment and each other.  or maybe it's mathematics...

snow crystallizes in atmosphere into plates, hollow prisms, needles, stars etc... and some combos.  as conditions alter and snow crystals fall to diff layers, new forms crystallize on them.  they also get hoared.

secondary processes occur in atmosphere and after falling:

1) close to 0degC metamorphism where details are lost and roundness develops

2) riming if it falls through layer of supercooled water droplets makes crystals rough and blobby

3) temperature gradient metamorphism which ultimately leads to totally restructured depth hoar, layered cups and needles

4) compaction

associated process is hoar:

1) freezing of microscopic supercooled water droplets: hoar needles and feathers into the wind

2) freezing of water vapor.

these processes can take a day to 6 years and more.  eventually all original form is lost and hoar develops or under compaction, random connected ice crystals -> glacier ice.

snow is from 0.15g/cm^3 to ~1.

looking at a handful of snow you can try to deduce it's history in the atmosphere and on the ground.  and predict skiiability and avalanche potential.

clouds are

1) water droplets (low)

2) supercooled water droplets

3) ice crystals
      ice crystals form around nuclei of favorable particles, even microscopic salt crystals.  if they fall through supersaturated air they will grow.  conditions change they even may be caught in updrafts.

     if they fall through microscopic water droplets they collect rime.

symmetrical snowflakes are a rarity contrary to the picture books.

Nakaya, U. "snow crystals: natural and artificial" cup '54  grew crystals in lab conditions 0deg to -25 and supersaturation wrt ice from 100 to 140.

needles seem to grow in distinct regions of 0 to -5, -5 to -7.,

then a region of scroll or cup from -7 to -10 but above the saturation  vapor pressure wrt supercooled water line

dendritic region is between  -12deg to -17deg and from below the supercooled line upwards

plates and sectored plates occur -10 to -20 along with thick plates though thick plates mostly below the line

plates perpendicular to the plane of crystal occur in a small region between -20 to -25 blow the line

columns occur -7 to -22 way below the line.

some regions contain only one kind of crystal.  others contain mixed.  it is VERY strange.

Magono, C. and C. W. Lee "meteorological classification of natural snow crystals" jnl of the faculty of sci, hokkaido u. ser VII (Geophysics), II, no. 4 (nov '66) p. 321-355

extended this with in field observations.  it kind of matches, they looked all the way down to -40degC.  they found more columns and combined columns and bullets down there.

so then he says that because snow crystals have high surface area to vol ratio, they are very thermodynamically unstable [[WRT TO WHAT?]]


i guess they are at equilibrium with supersaturated air.  now what's that?  air cooled below the dewpoint, so...   i don't understand it microscopically, but i guess the water molecules have energy to give up.

try supercooled water.  when you take heat from water it cools, then it will solidify.  if you take this heat from it to supercool it to the same temp, ... is it you drop to lower temp by taking LESS heat from it?  what measurements do you get from exp?  temp is avg kinetic energy so less molecular motion but molecules just havn't had chance to orient...  when it touches a seed... ah...  we get enthalpy AND entropy to think about.  i don't understand ANY of this.

at any rate, in dry air... i guess molecules at pointy spots evaporate and migrate to flatter spots.

this is called equitemperature metamorphosis and happens fastest close to 0deg and slower towards -40, it does not happen colder than -40.  which appears to be the temp limit of snow formation  in Magono and Lee's diagram.  so maybe colder than that it just evaporates.

i guess if there is a temperature gradient, the water will sublimate off in one region and condense in another region...  so one region will be...

equitemp metamorphosis will continue till random packed uniform spheres about .5 to .6 g/cm^3  (he says random uniform sphere packing is .580  i don't understand what that might mean)  anyway i guess the water is not traveling far, because no gradient so it simply travels from pointy places to nearest smooth places.

if there is a temp gradient, water is continuously evaporating from warmer regions to colder regions. in this case the vapor deposition is like snow formation and will form larger and larger crystals up to 1cm of cups and paralleled joined needles and hollow needles, the crystals loosely packed.  this is called depth hoar.

the original crystals are lost and completely new crystals form.

(Q3) how is this different than formation of firnspiegel?  oh it's migration of vapor from one crystal to adjacent crystal, not capillary action of water.  fascinating.

so in what environments does this happen and which way does the gradient go?  warmer closer to the surface or further or any number of situations?

this happens when gradients are generally more than 0.1degC /cm

he says usually warmer closer to the ground, cooler above.

(Q5) one question though: do ALL the layers recrystallize or do the crystals at the warmer layers slowly waste away to tiny spheres at the expense o the growth of larger depth hoar above?  [[and then the lower layers will slowly compact and that will change the process...]]

[[there must be similar process in geological settings... all kinds of dissptv can be happening...]]]

in equitemp metamorphosis, the rounding grains are being sintered together and bonding eventually to continuous glacier with micro bubbles.  in temp grad met, the larger crystals do NOT bond, and the bulk strength of the snow weakens. 

yes he says sometimes it results in entirely evaporated empty pockets below, and weakly packed snow above and avalanches.

the depth hoar photos look like the way  soil freezes into those vertical bundles of crystals.  similar process?  in soils the crystals push the soil up and and make it fluffy.

this mostly happens in cold continental climates in shallow snow.  not deep coastal snows.

the sign of depth hoar crystals is: "layered structure that appears externally as a stepped or ribbed surface on certain crystal facets".  "tendency for the crystals align in faint columnar patterns parallel to direction of vapor diffusion"  hmmm

of course once the temp gradient is removed, the depth hoar reverts to equitemp metamorphosis and decays..

is the deposition of supercooled  water droplets from storm winds onto cold surfaces.  the deposits are bulk, needles or feathers that grow  into the direction of the wind.  like those ice vanes i found on grassblades growing towards the falls in ithaca.  (on the grass about 50 feet from the bottom of some high falls one winter, cold mist blowing steadily against the grass from the falls and each blade had a grass blade thin ice vane sticking out from it TOWARDS the wind.  some were more than half an inch long.  was surreal)

the deposits are made of individual tiny drops, even the needles and feathers, are not single crystals.

(Q6) why on earth do NEEDLES form out of compacted drops?  how do the air currents and freezing regimes contrive to make the drops condense on the TIPS of the needles and not fill in between?  or just HOW do they form?  very strange!!!!

the photos of the feathery forms  seem to hint at hexagonal angles? 

rime also forms bumps on falling snow

is deposited from water vapor cooled below dew-point onto cold surfaces.  it forms flat stepped crystal forms or flat feathers (kind of reminiscent of those artificial bismuth crystals but not 3d screw dislocations).  when it forms on the surface of snow it glitters and a layer of it forms excellent lubrication for slab avalanches.

it is either surface hoar or depth hoar.

it usually forms on cold clear nights with high humidity causing the air to cool and humidity to condense out.

two processes can lead to firnification:  compaction or condensation from water coming in due to CAPILLARY ACTION

(god the complexities about!!!  that must be how firnspiegel forms.  light travels through upper layer, gets absorbed deeper in and eventually causes some melting, then THAT water is drawn up through capillary action [WHY? (Q2) because lower is more compact and upper is more capillary space to draw the water? hmmm] where it is deposited on cooler upper layers and forms eventually a thin shiny sheet of ice at the surface. 

it's a kind of convection!!! so cool!)