Continuous marking of the sun's noon position revealed
an figure eight-shape.

1998: November 9 at 12:07
h

Above: The difference between the constructed N-S line and the compass is called magnetic deviation; here app. 2.5 degrees.
 

     At equinox the shadow of the day-line of the shadow followed an apparently straight east-west line (here called the equinox-path). Now having an east-west line the construction of a north-south line is simple geometry.
     A 'compass-rose' depicting the cardinal directions (on the photos south is up) was constructed on the north-south line. A compass placed on top of the constructed 'compass-rose' showed a deviation in the range of some 2 degrees, which first was seen as an acceptable error-margin, but later on was realized to be the difference between geographic north and magnetic north: Magnetic deviation.

     Around winter solstice I was back again on the sundial (the full-drawn line) for some weeks. The shape now became evident. Before leaving in February it seemed like the half-hour lines overlapped one another one to two minutes. On the Internet I later found out that the figure-eight-shape is called an analemma. 

The gnomon
 

     The analemma marking continued in vacant periods in 1999, 2000 and 2001 until a double lining of the analemmas  made it clear that the construction of the wooden terrace, on which the sundial is drawn, was unstable. In 2001 a termite attack on the base of the gnomon made the this unstable too. All marking were then suspended.
     In the beginning of March 2002 the construction was reinforced in preparation for the coming vernal equinox.
     The analemma is of no significance for the studies of orientation of Khmer temple structures as the ancient Khmers did not use mean time. Therefore the marking of analemma was suspended. Emphasis was put on marking day-lines (declination lines) - especially the paths of the shadow at the equinoxes and solstices, and the days when the sun passes zenith and nadir.