Actually, mounting the antennae for horizontal polarization probably wouldn't help much with isolation from other WiFi signals. Most WiFi access points use polarization diversity, so the resulting transmissions are essentially polarization-independent. In buildings and urban areas, all the bounces and multipath wreak enough havoc with polarization that a polarized system wouldn't work - plus you'd have to hold your laptop just so.

Besides that, vertical polarization is preferred for frequencies at VHF and up because it propogates better over long distances. Horizontally polarized waves tend to get short-circuited by the earth as they pass over.

As far as pointing goes, if you can get a nice day with clear weather, a rifle scope rigged directly to the antenna does a nice job to help you boresight two antennae. The 5.3 degree beamwidth means you'll lose half your power if you're off by 2.65 degrees. Aligning the antennae by received power is tricky unless you happen to have proper equipment to provide a constant transmit signal and measure the received power.

Ice would definitely reduce the antenna performance, so some sort of radome is a good idea. Plexiglass would probably be a good choice. A good rule of thumb to keep the radome electrically thin (and thus transparent) is about a tenth-wavelength or less (1.2 cm in free space). Keep in mind that a wavelength in plastic will be less than the free space wavelength, so with a little hand-waving we'll say our radome should be less than 0.5 cm thick. 1/8" plexiglass will do quite nicely, and you could use some more of your angle iron to make a nice sturdy box out of it, caulk the seams, and have a pretty decent little shed. You'll take a slight hit in performance, but not very much. Any sort of plastic or fiberglass material should work well, so long as it's kept thin. You can even make most of the box out of metal, with just an RF-transparent window in front of the antenna, but you have to be a little careful not to make it into a cavity. Absorber (carbon impregnated foam) works well for this, but it's not really the sort of thing you can pick up at Home Depot...

This is valuable input from all three of you and will help me in my current testing and especially for the eventual permanent mountain-top installations. I had thought about snow on the solar panels (not sure about a good solution though) but not much about ice on the antennas. Building a small shed or box around the antenna may work well. Would it need a plexiglass or plastic window or would just an open side provide enough protection?

I agree with liquidsquid, you can get up to 30 dB isolation for well aligned polarization sensitive antennas. The parabolic reflector is essentially a wire-grid polarizer. They tend to be tricky, since they behave differently than our mechanical intuition would have us believe.

For metal to be an effective reflector, electrons have to be able to move at least a wavelength (12 cm at 2.4GHz). Since the rods that make up the reflector limit the horizontal motion of electrons to the width or the rod (~1cm), any horizontally polarized wave will pass through the reflector (or scatter). Vertical polarized waves on the other hand will be reflected back to the feed. On the other hand, our mechanical intuition suggests that anything vertically polarized would be able to pass right through.

As you mentioned before, I'm surprised there's not a better way to adjust the angle of the antenna. Looking at the specs, a few degree misalignment would kill your signal by a factor of 10. ouch....

Polarization will give you roughly 30dB of difference for horiz vs. vert. You can transmit on the same frequency at the same time with two polarizations and have two discrete signals intact due to this.

This is clearly a polarized antenna as it will not reflect signals in the vertical direction, and pass them right through behind it.

To change the polarization, rotate the mast and the antenna at the same time. I'm sure you can some up with something ;-).

On another note, you can see what kind of havoc ice accumulation can do by imagining those bars getting a fat coat of ice. You loose your polarization, and also your directional pattern. I was thinking if you have to put them on a mountain, it is likely best to put a protective "shed" over them. Something like a dome with heavy-duty plastic sheet over the outside so it is RF transparent. The plastic flapping in the wind can somewhat reduce ice and snow accumulation, and completely protect the sensitive elements inside.

You could build the "shed" like a silo shape instead, preferably round as to not present a flat side to wind. It would also keep the appearance discrete if you choose colors to match surroundings.

E-mail the guys on Mt. Washington about their antenna setups. http://www.mountwashington.org/weather/cam/deck/index.php
They have a wireless temperature profile network that runs on solar. It is pretty slick, and you could certainly learn a lot since they have horrible conditions for this sort of thing up there.

-Mark