Hi John,

I've been asked about this from time to time, and when I read
your e-mail, I thought: I really ought to add this to the FAQ.
Then I looked, and saw that I added it a while back...

http://www.projectpluto.com/faq_curr.htm#riseset

...the text of which I've appended below.

I'm pretty sure this is the source of the difference you report.
Elevation does make a difference, but it's a consistent one (that
is, if you're high enough so that the sun sets five minutes later,
it'll rise five minutes earlier, too.)

Quite a while back, I did have Guide set up to work on a calendar
day basis. I don't think that survived Guide 1.0, for the reasons
listed below: it's too easy to get situations where you can _see_
that the moon is about to set (a few minutes after midnight), but
because the rise/set routine is restricted to the current calendar
day, it'll tell you that the moon isn't going to set at _all_.

-- Bill

-------- Text from the FAQ --------
Q: The rise/set/transit times I get when I click on the Moon don't make
sense to me.

A: In quite a few programs, when you ask for rise/set/transit data for
an object, you get the times that object rose, transited, and set on
that particular calendar day. In the case of the moon, on about 10% of
calendar days, one of these events will not occur at all.

Guide does things a little differently. When you click on an object,
you get the time of the nearest transit, the preceding rise time,
and the following set time. This has several benefits. The
events are all shown in chronological order. With the "usual" scheme,
if the moon rises in the evening and sets in the morning (as happens
around full moon), the order of events can be a little weird. Also,
the "usual" scheme can keep you from seeing that an event is about
to happen (or just happened), but on the following (or preceding) calendar
day.

For example: suppose I set up Guide with my site, at 23:40 local
time on 24 June 2000. When I click on the Moon, Guide will tell me
that the Moon will rise at 00:04 local time on 25 June. Obviously,
I'd want to know that the moon would be rising in 24 minutes. Had
Guide followed the "usual" method, I would simply be told that the
moon never rose on the calendar day 24 June.

Finally, under the method used by Guide, the rise/set/transit
times do not vary according to the time zone. The moon's actual
motion is not affected by time zones, of course, so it makes sense
that the rise/set/transit times should not vary by time zone.

The scheme used by Guide will be absolutely essential when the
rise/set/transit times are given for viewpoints on other planets.
(Right now, Guide doesn't show that data yet.) From the Moon, a
star rises and/or sets once every 27.3 days, so an event is unlikely
to happen on a given calendar day. (And I'm not sure what a "lunar
calendar day" would be anyway.)