What is it about ice that makes it so reflective?

[kippert]

Hi,

So I've just been learning about albedo as part of my course and the question that strikes me is:
What is it about the structure of ice, particularly at the surface, that causes so much more light to reflect than with water?

The differences I can see are:
* Ice is less dense than water and therefore has more gaps
* Ice has less energy than water
* Ice is in a lattice type structure due to the hydrogen bonds

Here are the thought threads I've got on this atm:
* Photons are EM waves
* There is more hydrogen bonding in ice than in water
* Is there something about the hydrogen bonds which causes interference with the incident photons?
* What kind of EM field would cause reflection and deflection of photons?
* Do we need to distinguish between perfectly flat water and choppy water?
* Do we need to distinguish between pure ice and ice with lots of air bubbles in it?

Yeah, this one is puzzling me a lot, I feel like it's got something to do with the hydrogen bonds, but I can't figure out why at this stage.

Kipper

 

[Baluncore]

Welcome to PF.

Ice is solid water.
Ice is flat, water has waves.

Snow flakes scatter light, but ice can be clear, for example ice on a pond or black ice. Some icebergs show up on radar, some are invisible.

Metal surfaces make good mirrors. Light is reflected from changes in conductivity or dielectric constant. Liquid water has a very high dielectric constant at about 80, compared with most materials being between 2 and 5. Water should reflect better than ice. Solid water ice does not have a high dielectric constant.

 

[kippert]

Welcome to PF.

Ice is solid water.
Ice is flat, water has waves.

Snow flakes scatter light, but ice can be clear, for example ice on a pond or black ice. Some icebergs show up on radar, some are invisible.

Metal surfaces make good mirrors. Light is reflected from changes in conductivity or dielectric constant. Liquid water has a very high dielectric constant at about 80, compared with most materials being between 2 and 5. Water should reflect better than ice. Solid water ice does not have a high dielectric constant.

Thanks :)

so, the part I'm reading at the moment is specific to sea ice, so that might have a factor as it means the ice forms under dynamic and choppy conditions.

I did consider clear ice, I guess my question there would be does it still have a high albedo or does it actually just let photons through.

When you say "Light is reflected from changes in conductivity or dielectric constant." What do you mean there? Does incident light cause a conductivity change which then causes the light to reflect?

It kind of sounds like you'd expect the opposite of what I currently understand to be the case. I am currently under the impression that ice reflects more than water, but from what you are saying this, in all likelihood, shouldn't be the case.

 

[Baluncore]

I am currently under the impression that ice reflects more than water, but from what you are saying this, in all likelihood, shouldn't be the case.

I suspect you see ice that is often covered with snow. Maybe we need to reduce the complexity by narrowing the examples being considered. Is it scattering of light at the surface you are considering. Frost and snow have air mixed in with the ice crystals that scatters light efficiently.

The transmission and reflection at the boundary between two different materials is determined by the change in electromagnetic characteristics. When the materials have the same characteristics, the boundary does not reflect and cannot be seen.

You can measure the thickness of an icecap or an icesheet with radar. It is more difficult to measure the depth of water with radar.

 

[kippert]

I suspect you see ice that is often covered with snow. Maybe we need to reduce the complexity by narrowing the examples being considered. Is it scattering of light at the surface you are considering. Frost and snow have air mixed in with the ice crystals that scatters light efficiently.

Yes, I think this is key, and it's difficult for me to know simply because the course that I'm taking isn't being overly specific here unfortunately.

I think the impurities and imperfections are a likely candidate for what is being described to me, because I can't make sense of it in any other way. Ultimately the best test would be for me to try and create some perfect blocks of ice (possibly by freezing and melting water several times, along with using a syringe to try and remove as much air from the liquid water as possible by creating a super simple and small vacuum chamber. Then if I can shine a light at it and come up with some way of measuring how much is passed through, how much is absorbed and how much is reflected.

I do know one thing though, I'm glad that I'm questioning the material that I'm looking at at the moment because it seems clear that it's not presenting a full picture of what is going on.

Thanks for taking the time to respond, btw, it's much appreciated.

 

[sophiecentaur]

My thought is that the reflection that we see is Total Internal Reflection at Ice / Air interfaces. Water foam can look as white but it's seldom there for long and foam bubbles have two liquid / air interfaces at each surface and thin film interference can colour the appearance of foams. When viewed from under water, (snorkelling) air bubbles are extremely shiny too and that will be TIR. Air is trapped in snow and snow looks the whitest and stays for a long time. Also, the water in snow tends to be pretty clean in regions with no pollution so it will behave 'ideally'.

 

[256bits]

remove as much air from the liquid water as possible

you should try slow boil water, right from the tap water, sitting in the sun water, etc, just to see what makes a difference if any for comparison how clear the ice cubes are.

 

[sophiecentaur]

for comparison how clear the ice cubes are.

I wonder just how much air inclusions would make a difference to absorption of light through ice. Of course there will be scattering but, as the thing that makes ice (snow, actually) highly reflective and 'white' would not depend so much on scattering. The incident light on snow is not necessarily directional.

I'd appreciate some reaction to my idea of TIR as an explanation.

 

[sophiecentaur]

Water ice looks white but, there again, so does CO2 ice, for the same reasons. Basically most of the light that you see (reflected) has not traveled through a great thickness of the crystals. If you are buried in snow, it's actually quite dark because most of the light has been reflected. Our perception of these things is very subjective but even foam on top of a dark liquid looks pretty near white (Froth on dark beer for instance, looks near white because of the contrast.)

 

[Baluncore]

In this article... https://en.wikipedia.org/wiki/Albedo
The first diagram, left hand column, gives the albedo of snow, ice and water.
The first table gives the albedo of; Fresh snow = 0.8; Ocean ice = 0.5 to 0.7; open ocean = 0.06
https://en.wikipedia.org/wiki/Albedo#Water

I want to know what it is about the air filled 60 and 120 deg corners in snow flakes that makes them scatter light so efficiently.

Snow sparkle is first surface reflection.
https://www.wpr.org/science-snow-sparkle-explained-uw-weather-experts

 

[sophiecentaur]

In this article... https://en.wikipedia.org/wiki/Albedo
The first diagram, left hand column, gives the albedo of snow, ice and water.
The first table gives the albedo of; Fresh snow = 0.8; Ocean ice = 0.5 to 0.7; open ocean = 0.06
https://en.wikipedia.org/wiki/Albedo#Water

I want to know what it is about the air filled 60 and 120 deg corners in snow flakes that makes them scatter light so efficiently.

Snow sparkle is first surface reflection.
https://www.wpr.org/science-snow-sparkle-explained-uw-weather-experts

The first Internal Reflection between the ice and the air inclusions. The crystal angle can’t be relevant in this case because light comes from all around.
Scattering from ice crystals occurs at very high altitude where the crystals tend to lay horizontal as they fall down through stratified air. That keeps the angles all nearly the same and hence halos, sun dogs and various other pretty effects occur. There are often colored effects. Can’t give a link as I’m out in the sun. Google has it all, though.
PS I could well believe the ‘sparkle’ mechanism. A bit like quartz pieces in some minerals.

 

[256bits]

I wonder just how much air inclusions would make a difference to absorption of light through ice. Of course there will be scattering but, as the thing that makes ice (snow, actually) highly reflective and 'white' would not depend so much on scattering. The incident light on snow is not necessarily directional.

I'd appreciate some reaction to my idea of TIR as an explanation.

As if I would know.
But most crystalline materials, while transparent as a bulk lattice, when pulverized, such as table salt, do appear white. Even those materials appearing dark and opaque, such as shale, appear white when pulverized ( ie a streak across a shale surface ). It may not all be with TIR, but also reflection off at some angles, and including, from normal to the multiple surfaces producing the whiteness and diffuse radiation.

 

[sophiecentaur]

Even those materials appearing dark and opaque, such as shale, appear white when pulverized ( ie a streak across a shale surface ).

I would say that, if the bulk material is almost opaque then the internal path of the light must be short before reflection takes place. (That make sense to me.) The reflection must be internal, or a smooth face of the material would reflect like a mirror. The effect has to depend on internal gaps between small crystals.
You seem to be saying that the variety of angles on small areas of the surface produce good reflections in some directions. That would imply that a polished surface would act like a good mirror for just one incident angle. My choice of TIR as the explanation is based on the fact that prisms are always used (when possible) as reflectors because the Total is such an advantage in reducing light loss. Any random surface (except a polished metal or a dielectric coating - trendy sunglasses), in air, makes a pretty poor reflector. Snow, with an albedo of 0.8, is so much better than plain ice but, of course, it doesn't produce specular reflections.

Angle related reflections do occur in ice crystals but only under special circumstances like aligned ice crystals at high altitudes. Even then, I'm not sure that the halo images formed are actually very bright (just subjectively striking).

 

[256bits]

TIR as the explanation

Like I said I am making a guess that it would reflect off of the interface.
Would not refraction produce some colors?
Different frequencies would have different angles of refraction.
Maybe we don't see it since the crystals are orientated in all directions.
You certainly know more about light and optics than I could hope to.

 

[sophiecentaur]

Would not refraction produce some colors?

I don't remember seeing dispersion bands when looking through ice- even solid ice blocks. I should try to get a pyramidal ice mould, perhaps. But, as you suggest, a range of incident angles would probably cancel it out.
Only around the critical angle would there be wavelength selective TIR.

 

[Baluncore]

But most crystalline materials, while transparent as a bulk lattice, when pulverized, such as table salt, do appear white. Even those materials appearing dark and opaque, such as shale, appear white when pulverized ( ie a streak across a shale surface ).

Often simply wetting the surface hides the powder and makes the base material visible.
That suggests it is the difference between interstitial air or water that makes the difference. That all comes down to the comparison of refractive indices and the changes in the labyrinthine path followed by the light as a result.

 

[sophiecentaur]

Often simply wetting the surface hides the powder and makes the base material visible.
That suggests it is the difference between interstitial air or water that makes the difference. That all comes down to the comparison of refractive indices and the changes in the labyrinthine path followed by the light as a result.

Absolutely and it always goes darker when wetted. There will be much less TIR when the air gaps are filled with water. Incident light can go deeper and deeper. At the crystal / water interfaces it is very unlikely to be TIRed so down it goes.

 

[snorkack]

Clear ice is actually less reflective than water. Since ice is less dense than water, it has lower refractive index and reflects less light.
There are several possible scatterers in ice:
Air bubbles in ice
In sea ice, also droplets of brine, or at low temperature (under -20) salt crystals.
Clear black ice sometimes forms on ponds, but even on ponds ice is often white. What are the conditions that permit formation of clear black ice in nature, vs. conditions where ice is white?

 

[sophiecentaur]

clear black ice in nature

The surface must be very flat. Then, there will be just one good specular reflection of the Sun, in other places there will be much less light reflected. And you have to remember that all these descriptions are based on subjective observations. Our impression of "Blue sea", "Blue sky", "Black clouds" and many more, are very subjective and we can be easily impressed by these phenomena. People's description of a rainbow with strong colours is really not accurate. The rainbow colours are actually very de-saturated; it's mostly down to context.

I'd suggest that 'black ice' would be caused by ice that's fairly free of bubbles and major bumps etc and where the surface has just started to melt, giving the same effect as a showroom-ready waxed car body.

 

[Baluncore]

Black Ice. https://en.wikipedia.org/wiki/Black_ice

 

[snorkack]

Black Ice. https://en.wikipedia.org/wiki/Black_ice

From
https://unterm.un.org/unterm/display/record/wmo/na?OriginalId=aaa27375-9e55-4b4f-91f5-37a46be60f83
Note the part:

(1) Thin, new ice on freshwater or saltwater, appearing dark in colour because of its transparency, which is a result of its columnar grain structure. On lakes, black ice is commonly overlain by white ice formed from refrozen snow or slush

Transparent ice on still water, like one formed in cold but quiet autumn weather - no snow falling on ice. It would show the water underneath - light if the pond is shallow and has light coloured bottom, but deep water would be dark blue.
By Fresnel equations, the simple case of normal incidence, the reflection is (n₁-n₂)²/(n₁+n₂)²
Water has n about 1,334 at 0 degrees and 589 nm. This gives about 2,05% reflection
Ice is quoted to have n about 1,310 - not sure if this is the ordinary or extraordinary one. The difference is modest but relevant.
Sticking to 1,310, the surface between air and ice reflects 1,80 % of light - and the surface between ice and water just 0,008 %.