Stockholm 1535 halos

The V�dersolstavlan, �Sundog painting�, represents an ice halo display seen over Stockholm on April 20, 1535 (Old Time, Julian Calendar) and is a revered earliest image of the mediaeval city.

The painting is a 1636 copy (and from its halo depiction evidently an accurate one) of the now lost original thought to be by Urban M�lare (Urban the Painter) and commissioned by Bishop Olaus Petri. The interpretation of the sky events and �six suns� was the subject of much controversy between the church and the elected King of Sweden Gustav Vasa.

Controversial or not, the halo depiction is a scientifically faithful one. The noon solar altitude on the day was 45� and ray tracing simulations show this was too high for the scene. A solar elevation of 36� best fits the display and corresponds roughly to 9am or 3pm.

Interestingly, the painter has rendered the sky not as would be seen by a normal eye view. Instead he shows the display as it would be approximately reproduced by a modern fish-eye lens. He shows the entire sky in an azimuthal or possibly a stereographic projection. This is in sharp contrast to the portrayal of Stockholm itself and the landscape. The actual halo shapes are all represented by arcs of circles, a convention that persisted long after this painting was made.

A bright and complete parhelic circle dominates the sky with the sun placed at upper right. A matching HaloSim simulation (zenith centred azimuthal projection) is oriented the same way. Like beads, the real sun and five extra suns are strung around the parhelic circle. How do they correspond with a modern simulation? The two closest to the sun are common sundogs from plate oriented crystals. Within a few degrees of their correct sky position are shown two 120� parhelia � also from plate crystals. Opposite the sun is another brightening which is shown also in the simulation. This is the anthelion, not a true halo but a bright area where Wegener arcs from column crystals intersect the parhelic circle.

The painting�s representation of the 22� halo is curiously off-centre. This might be the result of following a stereographic projection while insisting on using circles to depict the arcs.

The two outer arcs stretching from the 22� parhelia and intersecting above the sun are stylised. Half close your eyes and they are seen to represent the �gull-wing� top of the circumscribed halo of the simulation. Evidently the skies that day contained plenty of column crystals.

The placement of the two sundogs between the 22� halo and where the circumscribed halo intersects the parhelic circle is impressive. It speaks of acute observation and allows the solar elevation to be fixed precisely.

The large arc at lower left could be an infralateral arc from short horizontal columns. Its position is very approximately correct but the arc never crosses the parhelic circle.

The object at the zenith, the parhelic circle�s centre, is problematic. It is often said to be a crescent moon but the moon was not a crescent on April 20, 1535. It could be a circumzenithal arc from the abundant plate crystals but that implies that, contrary to the other arcs, the CZA was depicted as it would have appeared at much lower sun and earlier in the day. Early halo drawings, even the famous later one of Tobias Lowitz, were often an amalgam of halos seen at various times of the day. This depiction could be the same except it jars with the otherwise rather precise single time portrayal of the sundogs and circumscribed halo.

A painting of originality, close observation and much beauty.

For other aspects of Vädersolstavlan see the comprehensive article by Mats Halldin.