About the Photographs

It’s a great excuse to visit Rome. Compiling the pictures in the book “Beyond: Visions of the Interplanetary Probes” involved hundreds of hours spent combing through archives devoted to the landscape photography of the deep-space probes, both on-line and among the innumerable bound hard copies stored at NASA’s PIRF, or Planetary Image Resource Facility for southern Europe—which happens to be in the basement of a large research institute just outside the eternal city, in a vineyards-dominated suburb called Frescati. The planets are all named after Roman gods, of course, but the Solar System is a far older place than either Olympus or the Coliseum, though its vistas are quite new to us. Not very far from the National Institute of Astrophysics, a papal warrant was issued for Galileo’s arrest; just up the autostrada, Leonardo climbed Tuscan hills and made the first drawings of landscapes viewed from above. Both they and Galileo’s careful pen-and-ink renderings of Jupiter with its moons are the direct antecedents of the images reproduced here.

Most of these pictures came from extended on-line excursions along the image trajectories of various missions: for example, I went through every one of the many tens of thousands of shots taken by the two Voyager probes of the late 1970s and 1980s as they passed Jupiter and Saturn. Ensconced in that basement room in Frescati, surrounded by space-probe models and schoolroom-style globes of Venus and Mars, I also looked at every picture sent home by the two Viking orbiters that reached Mars in 1976, and every shot of the five mid-1960s Lunar Orbiters. It’s no doubt a rank cliché to call these little virtual jaunts through the Solar System “voyages of discovery”—but it’s also practically the only way to describe them. When you follow the image trajectory of a spacecraft that can send one picture to Earth every forty-eight seconds, you are in some senses on that journey, its long stretches of tedium periodically punctuated by moments of sheer amazement at the awesome unfolding planetary views. In the end I felt as though I had actually been a passenger on these interplanetary rides, certainly one of the most valuable experiences associated with the making of this book. If a part of that sensation was conveyed here (well, the amazement part, not the tedium part), I’m satisfied.

Some of these pictures are from web pages that had already been partly filtered, for example NASA’s spectacular on-line resource “A Planetary Photojournal”[1] or the image archives of the Mars Global Surveyor, which contain a vast number of pictures, but far less than the spacecraft’s full output (though I put a dent into that as well). A few others came from the smaller and more personal selections visible in places like Calvin Hamilton’s excellent website “Views of the Planets.”[2] Many of the images in this book are multiple-frame mosaics—the result of finding remarkable contiguous single frames and collaging them together, a process leading, sometimes literally, to the horizon. (The raging Martian dust storm reproduced on pages 135 through 138 of the book materialized in this way; the 60-part mosaic of Jupiter and its oceanic moon Europa on pages 240 and 241 likewise started out with the discovery of several spectacular single images.)

So the choice to use a shot was always my own; in most cases it came after logging time with primary sources; frequently it led to the creation of mosaics; and sometimes it was helped along by the keen eyes of other image curators who had already made a partial selection from the huge quantities available. I owe them my thanks.

Almost all of these photographs required substantial amounts of digital processing. Many had never been rendered into color before, or if they had they’d long since vanished into the databanks of various research institutes. I was very lucky to work with one of the best image processors in the planetary sciences community, Paul Geissler, who helped me transform many of the black-and-white single-frame source pictures I had chosen into color. And no doubt this requires some explanation.

Almost all of the space probes until the mid-1980s Galileo Jupiter mission carried camera systems called vidicons. Employing a video tube, they were very similar to black-and-white TV cameras, although they produced stills. In order to create color pictures the vidicon had to use a filter-wheel system, in which a series of filters accepting light at various wavelengths were passed in front of the lens. To make a color picture, the probe recorded separate images with three different visible light filters (for example orange, green, and blue), which could later be superimposed back on Earth to create an acceptable color composite. Later probes replaced the vidicon with a semiconductor CCD, or charge-coupled device—the very same ultra-sensitive piece of solid-state technology that has revolutionized amateur astronomy over the last decade or so.[3] But even with the CCD the principle of how to make color pictures is the same, only here the various layer of pixels are coated with colors allowing light through at different wavelengths.

Why doesn’t the probe simply combine those three visible light wavelengths light itself, before transmitting a color picture to Earth? Mostly because each wavelength, including ones not visible to the eye, such as infrared, reveals different things about a photograph’s subject—and scientific curiosity, needless to say, ranks higher than appreciation of fine color photography among the people designing and running these missions. (Which is not to say color isn’t important: it’s very useful in trying to determine the surface composition of these moons, planets, and asteroids.)

If the need to take at least three exposures of a given subject in order to get one color shot seems cumbersome enough, consider that these probes also move at extremely high rates of speed—in the case of the Voyager fly-bys, something like 35,000 miles per hour, which is a good deal faster than a rifle bullet. Viewing geometry thus inevitably changes between exposures, frequently quite radically. To get three frames to align, they usually must be transformed with digital tools so that their perspectives appear almost identical—and even then, in many cases, only a portion of a given shot can be rendered into color, due to a lack of overlap with its neighbors. And as if this weren’t already enough, as already mentioned, many of the color pictures in this book are mosaics, creating a truly complex situation, since each individual frame in the mosaic is comprised of at least three shots, and these in turn have to join at the seams so that the changed viewing geometry apparent to the spacecraft as it whizzed by isn’t distractingly apparent to the viewer!

In these cases I always started out by making a workable black-and-white mosaic, and only then—assuming I thought it needed to be in color in the first place—would I ask Paul Geissler for help. All of the shots in this book that weren’t picked up already rendered into color on-line—and more on that in a minute—were composited by Paul, who frequently had to come up with some serious hocus pocus to underpin my monochrome mosaics with color. Sometimes a third filter was picked up from an orbit that took place months later; or if the chosen image was from the narrow-field camera, a filter or two was borrowed from the wide-angle camera that operated at the same time (or vice versa); in some cases all the color information was imported from another orbit entirely, and in two cases a small amount of color information was taken from another mission. Paul would then frequently send me two pictures: a color shot in which my black-and-white image had been used as a top layer, with the color information added beneath (a technique that, when done right, can result in a seamless color image); and a composite that had gone back to the original frames and re-combined them, not incorporating my monochrome image (ditto, depending on various factors—see the Mars polar cap mosaic at the top of this article). I would then make a choice and usually set to work some more on that image.

With some notable exceptions (all of which are identified in their captions—and any errors in that regard are mine), every reasonable effort was made to print “true” color here. Although I recognize the necessity sometimes to exaggerate and “stretch” color for scientific purposes, I never saw the reason to do so when presenting pictures for aesthetic reasons, particularly—and this is key—when the subject matter is so inaccessible to direct experience. The Solar System is already spectacular enough without pumping artificial colors into it. I was particularly concerned that objects like Jupiter’s awesomely garish Io, which is essentially a fuming volcanic ball of yellow and red sulfur, wouldn’t be portrayed in this book—as it is in so many on-line pictures—in an exaggerated way. Io is already a true-color eyeful. Here’s Paul on the subject:

My best impression of Io’s true color is at http://pirlwww.lpl.arizona.edu /HIIPS/EPO—see in particular the section on contrast enhancement. Your own perception of color is a fascinating topic and it depends on the time of day. For Io, the important point is that the reds are RED! Voyager was blind to red so we had an incomplete understanding of Io’s colors until Galileo arrived. Red on Io indicates sulfur that has been disrupted by heating or by radiation. If it weren’t for ongoing eruptions, the reds would all convert to yellow.

Once I had an image on the screen, I tried not to do more than what a good darkroom printer would: essentially making sure the blacks are as advertised and the whites likewise, increasing the contrast of what generally start out as very flat images, trying to make the colors conform to our best understanding of what they would really look like to the human eye based on various criteria, including telescopic observation, and removing many thousands of gnat-like reseau marks—the black spots that speckle all the primary Voyager and Viking images.[4] This ordered swarm of dots was put there intentionally and was used to make sure the images conform to a fixed grid on Earth, a process necessary due to the slight variations that would creep into the vidicon pictures. After a while, I saw them (and cloned nearby pixels to erase them) in my sleep.

One exception to this self-imposed true-color rule is the Sun, which should never be looked at with the naked eye anyway and which was imaged in non-visible wavelengths (often ultraviolet) in the first place. The Sun is pure energy; the pictures in this book accurately reveal many of its structures, from magnetized filament loops to gigantic eruptions; the colors I used no doubt reveal my own earthly bias about what flames should look like. (All of which is to say that I frequently modified the colors offered by the SOHO and TRACE imaging teams, sometimes drastically and sometimes subtly; I apologize if I offended anybody as a result.[5]) Overall, though, I tried to be as true to the human eye as possible, and I’m the only one to blame where failures occurred.

At this stage I should point out that a fair number of the color pictures in the book were already available in the on-line archives mentioned above. But even these frequently required additional processing, either to modify their colors according to the best information currently available, or to remove seams between mosaic frames, or to clone over a speckle of uncorrected bad pixels or other transmission artifacts. And some, of course, simply slid into place without needing additional work—a testament to their talented processors, who often remain anonymous, with their pictures simply credited to JPL or NASA. I also owe these unsung individuals my thanks, and would put their names in this book if they were easier to locate.

When it comes to the black-and-white shots, a similar thinking applied. I always asked myself what I would do in a darkroom if I had taken the image on Kodak Tri-X film in natural light—unlikely, but possible to imagine—and now had it loaded into my enlarger. Again, there are some notable exceptions to my “visible light” rules, however. The radar pictures of Venus sent home by the Magellan probe in the early 1990s are obviously one, because with radar-image creation the texture of a given piece of topography makes the shot—not its ability to reflect or absorb visible light. The darks in those Venus pictures are a result of rough, radar-scattering terrain; the vivid areas are either the smoothest or are comprised of materials that tend to reflect radar pulses back towards their origin. Still, the results are close enough to visible-light photography to make readable images (and distant enough to account for their dreamlike feel). For other reasons, the same goes for the very few black-and-white pictures not taken in visible-light frequencies. In every case where a shot isn’t from visible light frequencies that fact is mentioned in the accompanying caption; in all cases where an image is actually a mosaic, that’s also mentioned.

Finally, the black-and-white images of the Moon sent home by the five Lunar Orbiters of 1966–67 are a special case. Each picture originated on 70-mm film that was automatically exposed and processed in lunar orbit, and then scanned and sent as a radio signal to Earth, where the individual scanned strips were reconstituted. (Early spy-satellite technology, no doubt, here used to scout potential Apollo landing sites.) The results were spectacular hard-copy prints that also suffered from the pronounced “Venetian-blind striping” effect. These prints were drum-scanned, more than four decades later, at an exceedingly high resolution for this book, then subjected to partial processing by Paul Geissler using software adapted from the Lunar Orbiter Digitization Project at the US Geological Survey (a redoubtable outpost of exquisitely rendered planetary maps and other materials based on the work of these probes[6]). All of the Lunar Orbiter shots then required many hours of individual work by hand, using digital tools, to remove some if not all of the remaining artifacts left by their unique production method.

Finally, a few words about digital cloning and fidelity to the truth. Apart from disguising the above-mentioned reseau marks, in many other cases cloning techniques were used to cover data gaps in a picture due to interruptions of the probe’s signal or to other glitches during the imaging process. This is easy with contemporary image processing programs, but I did so as minimally as possible. I also sometimes increased the amount of black space around a solar system object—but only when I knew with reasonable certainty (frequently by consulting wide angle images taken at the same time) that there was nothing else visible in that space. And in one case, specifically at the bottom right corner of that multi-frame Mars dust storm on pages 135 through 138, I actually extended a planetary limb a bit in order to match the one defining the opposite horizon. Apart from believing that this would improve the picture, I felt justified in doing so because from the angle of the Viking Orbiter when the shots constituting that mosaic were taken, that region of Mars was swathed in an opaque reflective atmospheric pall anyway. (In any event the digitally extended area represents less than five percent of the total image.) Lastly, full disclosure: in two places I used digital techniques to add a planetary terminator—the boundary between the day and night side of a planet or moon—to a pre-existing image mosaic. In each case the beginnings of that terminator was already visible in exactly the same location, but the hard cut-off of a part of the mosaic edge intruded somewhere within that area. These two pictures are visible on pages 114 and 179 (two version of the same Mars mosaic) and 288-289 (the best single mosaic of Neptune’s moon Triton).

Anyway, all this stress on virtuality and long-distance representation shouldn’t be misinterpreted. One of these days I’d really like to go myself—though I think I’ll exchange that Tri-X for color-slide film before I strap on my boots and lower the visor. As Robert Heinlein once put it: “Have space-suit, will travel.”

Michael Benson
Ljubljana Slovenia

1. See http://photojournal.jpl.nasa.gov/
2. See http://www.solarviews.com/eng/homepage.htm and also http://www.nineplanets.org/
3. For a superb account of this, check out Timothy Ferris’s recent book Seeing in the Dark: How Backyard Stargazers are Probing Deep Space and Guarding Earth from Interplanetary Peril, Simon & Schuster 2002
4. For very deep archives of raw images from many interplanetary missions, go to http://pdsimg.jpl.nasa.gov/Atlas/Atlas.html
5. See http://sohowww.nascom.nasa.gov/ and http://vestige.lmsal.com/TRACE/
6. http://wwwflag.wr.usgs.gov/USGSFlag/Space/GEOMAP/PGM_home.html