User Guide Why LTVT

Why LTVT?

There are many other products that produce very similar looking maps of the Moon, including the excellent and highly popular free Virtual Moon Atlas (VMA) of Christian Legrand and Patrick Chevalley. So why bother to create another?

The main reason is that the Lunar Terminator Visualization Tool (LTVT) has a highly focused and specialized purpose: to show where the terminator is expected to lie with the greatest possible precision.

The genesis of LTVT was the effort of one its authors to compare the famous 1609-1610 lunar drawings of Galileo Galilei to modern photographs taken under similar lighting conditions. That story is told in part on his website.

In brief, it seems theoretically possible to guess the dates on which Galileo's drawings were made by comparing the features he shows to those which would have been visible from his home on various nights. But it soon became evident that although VMA gave an excellent overall impression of what the Moon must have looked like in those days, it first did not show the expected position of the terminator very clearly (there are large digital steps in the arc), and second, even to that accuracy did not agree with other predictions of where the terminator should lie.

To resolve this confusion, the author had to resort to looking up the position of what is called the sub-solar point on the Moon (for the days in question) using the online JPL Horizons ephemeris system, and manually plotting the points at 90 degrees from it. This is very tedious, but it turned out VMA was not showing the terminator in the right place, primarily because it was not correctly illustrating, for these distant dates, the left-right and up-down rocking motions of the Moon called "librations".

It occurred to the author that to accurately determine the location of the terminator relative to the lunar features one need only know the longitude and latitude of the sub-solar point. Nothing else affects this. Also, to accurately know the centering of the image as seen from Earth one need only know the so-called sub-earth (or, more accurately, sub-observer point). These two points completely determine what we see.

Having had some previous experience with image processing software, and inspired by the example of Legrand and Chevalley, it occurred to the same author that it should not be terribly difficult to color a projected globe of the Moon with the appropriate known gray-scale intensities at each longitude and latitude, and then superimpose the exact terminator lines on it. Indeed it took only a couple of days to produce a functioning rudimentary program for doing this: the geometrical equations for projecting dots at known positions on a sphere onto a flat plane are not particularly difficult. But the fundamental philosophy was rather different. Legrand and Chevalley wanted to approximately simulate the appearance of the Moon around the present time, using simplified algorithms for calculating the geometry as seen from Earth. I wanted to draw it exactly, based on how someone more knowledgeable than myself told me how it should be oriented: namely allowing JPL to go through the highly complex and technical drudgery involved in computing the correct geometry.

As a result, a program which, for its purpose, was both simpler and more flexible than VMA evolved. For example, the manual input of orientation data makes it easy to generate views of the Moon as seen in projection, not just from the Earth, but from any other arbitrarily selected distant vantage point (such as from over the Moon's north or south poles, or directly over a crater away from the Moon's equator -- tasks that, as far as I know, is not generally possible in VMA).

However, the necessity of having to look up the correct geometry using the JPL on-line ephemeris system for each new time being considered soon became rather burdensome; so the author again asked if there might be a better way. After much further investigation he discovered that there were computer routines in the public domain for directly interrogating the machine-readable ephemeris files on which the JPL Horizons output is based, and also for doing the complex calculations necessary to compute, from the data in those files, the exact geometry of the Moon as viewed from the Sun. Moreover the files can be downloaded from JPL, and although they require a moderate amount of disk space (about 4 MB for each 50 years of interest) computing solar system positions by reading them is not only far more accurate, but actually faster than using the approximation schemes found in most ephemeris programs. Unfortunately the author also discovered that there are some very slight yet inexplicable (to him) discrepancies between the lunar geometries returned by JPL Horizons and the equally definitive printed Astronomical Almanac. The differences are very slight (at most a few hundredths of a degree in lunar longitude and latitude), and LTVT actually seems to estimate geometries somewhere between the two. Nonetheless the reason for the differences is puzzling.

As the result of the addition of this internal capability to estimate the correct lunar-solar-earth geometry at any given moment, LTVT has become more similar to VMA, yet it retains the original concept of being able to verify the results by manually entering the coordinates of the sub-solar and sub-earth points (which completely determine the view) from a reliable outside source such as JPL or the Astronomical Almanac. Hence the accuracy of the program in plotting the terminator and determining sun angles is limited only by the reliability of the source from which the geometry is obtained.

But the program remained something of a kludge, fully understood and operable only by its original author.

But now, with much encouragement and creative input from Danish pathologist and Moon enthusiast Hendrik Bondo, LTVT has evolved into the present product, which we hope retains the original quest for great accuracy, yet combines that with an easy to use intuitive bare-bones interface. So with all that said, we have chosen to jointly offer LTVT to you, not as a replacement for VMA, but as a supplement to that fine product.

There may well be other products that are capable of portraying the position of the lunar terminator at a particular time with equal accuracy, and perhaps with greater realism. We have not had the patience to try them all. Many are commercial products that cannot be tested thoroughly without purchasing a license. Most also appear to have highly complicated user interfaces that would require a considerable effort to learn. The ability in LTVT to manually enter the coordinates of the sub-solar and sub-observer points appears to be unique. We are not aware of any other product that allows the user to draw an accurate projection of the Moon based on a user-supplied viewing/lighting geometry.

We hope you will agree that LTVT is simple, accurate and easy to use. Best of all it is completely free!