|
Fig. 1 -Two photographs of the South Massif mountain
near the Apollo 17 landing site. Jack
White composed this particular juxtaposition. It has
been scaled down to fit the Clavius format. Here is the actual image composed
by White. (left: NASA AS17-147-22527, right: NASA
AS17-134-20435)
|
|
|
Fig. 1 (left) shows a
small mountain and a big LM. Fig. 1 (right) shows a large mountain
and a small LM. If the LM is smaller on the right because the
photographer is farther away, how can a genuine mountain behind it
appear larger, since it too must be farther away? [Jack
White]
The answer is quite simple. The relative sizes of the mountains
suggested by White's composition is not their true relationship.
White has cut out a smaller area of AS17-134-20435 (Fig. 1, right) and
a larger area of AS17-147-22527 (Fig. 1, left) and then enlarged the
smaller clip to be the same height as the larger one. Obviously this
will produce the illusion of dissimilarity. The reader does not
necessarily know that the photos have been cropped and resized. He is
likely to assume the composition has used the entire frame of each
photo.
The reseau fiducials can be used to
determine the difference in scale between White's two excerpts. In
fact, their primary purpose is to provide a normalized scale during
duplication and analysis.
|
Fig. 2 -The same composition as Fig. 1 with the
approximate position of the reseau fiducials emphasized. (left:
NASA AS17-147-22527, right: NASA AS17-134-20435. Annotations by Clavius.)
|
|
|
As shown in Fig. 2, the fiducial grid demonstrates that the photo
on the right has been enlarged to more than twice the scale of the
photo on the left. We can directly observe that the western end of
the South Massif summit is approximately one fiducial tall in both
photos.
|
Fig. 3 - The uncropped images used as the source for Figs. 1
and 2. The reseau grid has been used to normalize the scale of the
photos. (left: NASA AS17-147-22527, right: NASA AS17-134-20435)
|
|
|
Fig. 3 shows the two photos White used to compose his comparison.
They are shown here uncropped, except for minor cropping inherent to
the scanning. The reseau fiducials were used to normalize the scale
of the images. That is, the distance between fiducials is the same in
each photo. This guarantees that we are seeing the relative sizes of
the images as they appeared on the original camera transparencies.
|
Fig 4 -Images from Fig. 3 rotated, repositioned and
superimposed to show alignment of common summit line.
(NASA AS17-147-22527, AS17-134-20435)
|
|
|
In Fig. 3 the eye can still detect a difference in appearance
between what are supposed to be identical summit lines. This is
because the camera is tilted slightly differently in each photo. In
Fig. 4 the scale of the images is preserved as in Fig. 3, but the
photos have been registered to show the alignment of the foothill
horizon and the ridgeline and summit line.
It is common to assume that lunar mountains are nearby because
their rounded shape resembles hills on earth, and because the lack of
atmosphere denies the depth cueing that is provided by atmospheric
attenuation on earth ("haze"). Nevertheless the South Massif is
kilometers away from the landing site. Although the photographs are
taken from positions more than a hundred meters apart, that makes a
negligible difference in the appearance of mountains that are
kilometers away and on the order of a thousand meters high.
Thus we correctly expect to see a vast difference in the apparent
size of the lunar module, but little or no difference in the apparent
size of distant mountains. These photos do not present an anomalous
scenario. The apparent discrepancy was a product of Jack White's
composition and editing, not of an inherent inconsistency in the
objects being photographed or their appearance in the full frame.
|