Published: 2009-04-07 08:05:00
Updated: 2009-04-07 08:07:10
Posted April 7, 2009
By Mike Moss
You may have, though it could have been named something like "pictures of earth" or another similar file name. There are a few versions of the PowerPoint file that seem to be sent around every few years, and I just received another one in the last week or so and thought I would address it here. An earlier version I received stated that it was a collection of photographs taken by astronauts on the space shuttle Columbia, while the one I most recently received said the images were taken by astronaut Sunita Williams, who flew a few years later on different shuttles, en route back and forth to a long stay on the International Space Station.
The PowerPoint file does contain some very nice images of the earth, several of which appear to have been taken with unmanned satellites and a few of which may be astronaut photos (you can see a version of the slide show on YouTube, which I've provided a link for). However, about half of the images were NOT taken by either of those methods, but instead appear to be simulations generated using a computer graphics package not unlike the one we use here in the WeatherCenter to create the map backgrounds we use behind some of our computer model data or satellite and radar imagery.
In particular, there are a couple of images that purport to show a photo of night falling across Europe, the U.S. and South America, in which you can see a portion of the earth in sunlight and a portion in darkness, with the lights of cities shining up from the darkened areas. These are certainly eye-catching images, but they were not taken by an astronaut or even by a satellite. This is evident by (1) the lack of clouds over such large areas, which is never the case on the real earth, and (2) the fact that no camera or satellite sensor would properly expose both the sunlit and darkened areas at the same time. A photo that allowed enough exposure to show the city lights would vastly overexpose the sunlit parts of the earth, while one that properly exposes the sunlit areas would simply show darkness for the remainder of the image. In addition, the "terminator" separating light and dark areas typically appears a little wider and more diffuse in real photos from space.
Likewise with the images of "our planet in the middle of the universe at night," which are simulations. Again, no clouds anywhere and, even more noticeable, one of these images is shown from directly above the north pole, and you can see "city lights" in the U.S., Europe and Asia all at once. Even in winter, those three areas are never all in darkness at the same time and some portion of that view would be brightly sunlit.
As an experiment, I took our weather mapping software program and played around with the location of the "light" shining toward the its virtual earth, placed that light in an unnatural position where most of the visible portion of the planet would be in darkness, and overlaid a file that outlines all municipal boundaries in the U.S., then filled in those outlines with yellow shading, which does a pretty good job at re-creating the appearance of lights in the two images (in my case, I did not have city shape files for other countries, so the only "lights" appear over the CONUS, Alaska and Hawaii).
Incidentally, there is one meteorological satellite family that is tuned in such a way that it can produce images of lights at night. It's called the Defense Meteorological Satellite Program (DMSP), and I've included links to a couple of those "real" nighttime satellite images. One is a neat composite showing the effects of a lunar eclipse on what the satellite can "see" from one orbital pass to the next at night, ranging from clouds that appear brightly lit while the full moon illuminates the earth to almost nothing being visible except for city lights while the moon is in total eclipse. The other is an interesting look at the differences in nighttime lighting between North and South Korea, as seen from the DMSP satellite.
None of this is to suggest that the "Blue Beauty" file shouldn't be enjoyed for what it is, only that it should be viewed with the understanding that the sources of some of the photos may be misattributed, and that several of the images are nice computer creations rather than photographs.