Why the Interest in Eclipses?

mega-what |  The prehistoric people of north-west europe watched the rising and setting positions of sun and moon against the horizon very closely. They developed techniques for fitting the shape of the landscape to celestial cycles. Their monuments were built in places where the earth was in harmony with heaven in a very practical way. 

Lunistice positions on the horizon as measured by the Prehistoric Lunar Calendar are an indirect pointer to the position of the nodes of the lunar orbit and thus to the time of year at which a lunar eclipse may be expected to occur, as measured by the Prehistoric Solar Calendar.

In this schematic diagram:

• The 18.6 year lunar nodal cycle is repre­sented by the outer circle and reads clock­wise. Each of its divisions is centred on the appropriate lunar event and represents a period of 1.16 tropical years, about 14 months.
  
  
• The annual solar cycle is repre­sented by the inner circle and prog­resses anti-clock­wise.
  
  
• During any lunar period, visible lunar eclipses may be expected to occur only in the months centred on the events at either end of the adjoining solar axis.
  
  
• The fact that the "eclipse months" of the Prehistoric Solar Calendar overlap by a quarter-month at each end is an important feature that allows the system to generally produce the correct answer even though the natural cycles do not reconcile neatly.
  
  

Simple really! Observe lunistice rise / set positions to understand the time of year when eclipses can happen. 

Ancient observers, having solved the problem of knowing when a lunar eclipse might occur, would know from experience that there must either be a total eclipse, a partial eclipse or no visible eclipse. Could they have known which was most likely? Not completely impossible for, timing aside, on study of the data it would appear that eclipses occur in semi-regular patterns of fairly short duration. The typical, underlying, pattern seems to be two seasons of non-visible penumbral eclipses followed by two partials, two totals and then two more partials to end the sequence. Occasionally the penumbrals extend to three seasons in a row or reduce to one. The sequence of six visible eclipses sometimes reduces to five and may contain one, two or three totals, though it always begins and ends with a partial. 

The main problem for a naked-eye observer with no theoretical knowledge is that not all visible eclipses can be seen, because they will sometimes occur when the moon is below the horizon. It must be said though that in any sustained period of observation and certainly several times in a human lifetime, eclipses would be observed that had started before the moon rose or were not completed before it set. 

Therefore it would be possible to deduce that eclipses can happen when the moon is below the horizon. It was an essential part of these people's methodology that one can change the time and place of a rise or set by changing one's position on the earth's surface and, if desired, the same event may be observed repeatedly at different times from different places. So, while they could not be certain that an eclipse would actually be visible, they probably had a good idea of what to expect and they certainly knew when to expect it. Their methodology could not predict every eclipse perfectly but the overall pattern was mapped very well and it may be that their observational experience would have given better performance than seems immediately obvious to us.