Lowitz in Antarctica ~ Images by Richard Corbett at the UK Halley Research Station on the Brunt Ice Shelf at approximately 75°S ©Richard Corbett, shown with permission.

About - Submit Optics Picture of the Day Galleries Previous Next Today Subscribe to Features on RSS Feed

Tobias Lowitz, professor of chemistry at St Petersburg started something in 1790 when he sketched faint arcs extending from the sundogs of a bright halo display. The arcs were not reliably sighted thereafter and their absence over Antarctica - a place for excellent halo displays and rare halos - strengthened doubts as to their existence. Lowitz arcs remain a rarity and these Antarctic images especially so. One thing we can definitely rule out. The ice crystals here were not nucleated by ski-resort snow machines, the modern source of many rare halos!*

The Lowitz arcs are visible in the upper unenhanced image as faint arcs reaching upwards and outwards from each parhelion (sundog). The lower image (levels shifted and unsharp masked) reveals two arcs from each parhelion and outside the 22° halo. Note the differences in apparent sky brightness each side of the outermost arc.

22° Lowitz arcs have three possible ray paths. Here we see 'upper' U and 'lower' L rays but modified because there are an odd number of internal reflections within the Lowitz oriented plate crystals. The odd number of internal reflections produce the so called 'reflected' arcs UR and LR.

The HaloSim ray tracing computation at left is an approximate fit to the display. 20% of crystals that were randomly oriented columns gave the 22° halo. 10% of plates produced the parhelic circle and parhelia. 70% of Lowitz oriented plates were needed to give the weak reflected Lowitz arcs. 70%? - Lowitz arc production is inefficient - one reason for their rarity.

The simulation used regular hexagonal plates with a c/a ratio (thickness ratio) of 0.1. They had limited rotation about the Lowitz axis AA with a Gaussian tilt distribution around zero of standard deviation 10°. The traditional Lowitz model has a uniform distribution of tilts.

The match might be improved by further tweaking of parameters and perhaps using non regular crystals. However, this was not done because experience shows that the fit rarely gets that much better. There are aspects of Lowitz displays that are not understood.

  *   But even in the Antarctic there might be other anthropogenic sources of halo crystal forming nuclei.
  **  The differences in slope of the right and left hand Lowitz arcs results from the sun being off-centre in the original uncropped image.