3.Now we have figuratively and literally reached our destination. I have previously emphasized that you should pay close attention to Kilroy's rock and notice how it's appearance evolves over the course of the six days Spirit spent at Bonneville. You have already seen this rock on sols 66, 67, and 68 and you should now have a decent idea of how on each subsequent day a greater amount of Kilroy became visible. In particular you have seen how Kilroy went from being fully hidden to barely visible, depending on the time of the image capture on that first day, to being clearly visible in the space of three days.
In this article our focus turns to the determination of why his appearance changed and whether it continued to change in the days beyond sol 68. [Spoiler Alert: It Did!]
Ch-Ch-Changes…
On sol 67 Spirit had moved considerably closer to the rock, Kilroy was now visible along the rock's left hand side. This could be in part due to the change in lateral viewing angle from one day to the next. However, he could no longer be seen extending above the rock from behind despite the picture being taken from virtually the same relative elevation. Remember that the image on sol 67 was taken after it had finished the climb of that day. However, because the camera used was mounted closer to the ground much of the elevation change from the climb was offset. For reference, the PanCam and NavCam are mast mounted 1.54 meters above the Martian surface while the HazCam is located 52 cm above ground.
Whether the disappearance from one day to the next of something extending above the rock is the result of a small relative elevation change seems unlikely. While it is possible that the decreased resolving power from the use of a different camera this also seems unlikely. Most of the difference in the resolution from one camera to the other would be largely offset by the decreased distance.[1] The substantial change in the visibility of Kilroy on the rock's left edge the following day, sol 68, makes it extremely unlikely that this could be attributed to elevation changes, differences in resolution nor is it due to a lateral change in the rover's position.
In contrast with the small relative elevation and lateral changes in position during the previous days, on sol 71[2] in preparation for capturing a series of PanCam images used in conjunction with the miniature thermal emission spectrometer (Mini TES), seen in Figures 2. and 3., Spirit continued towards the crater floor and stopped next to Kilroy's rock. Perspective has clearly changed with the rover now positioned directly next to and above where Kilroy has been observed.
This turns out to be sensational as is demonstrated in Figure 1 (D). In every image that includes this rock during sols 67-70 Kilroy is clearly seen at its left side. Now with Spirit standing next to where Kilroy was he cannot now be seen. This is despite the fact that we are directly seeing the rock's left side and we are viewing it from 1.54 meters above the Martian surface looking down.
Let me repeat this, looking at the place where Kilroy has been clearly seen in every photograph for days, he is now clearly and unambiguously absent.
In Figure 3 I have reduced the image from Figure 2 to grayscale in order to minimize the effects of the overlay and I have cropped it somewhat to center the rock seen in Figure 1 (D). The rock is located in the foreground. This photograph was taken with a mast-mounted camera that rotated in place to create this panorama. Objects located at the bottom of the frame and close to the rover will appear somewhat changed in orientation relative to the same image taken further away.
To clarify this, picture a group photo with a landscape background taken at a distance of ten yards or meters. Now consider how the image would look if you were standing amongst the group and rotated in place to create this group shot to create a panorama, as you can do with modern smart phones, e.g. Apple's iPhone. The people in the resulting panorama would look different depending on where they stood relative to you. Some of your friends would now be seen from the side or the back while the background would look largely the same. This is the effect that can be observed in Figures 1 (D), 2 & 3 but with rocks and not people.
References and Notes
1. The angular resolution of the NavCam is 0.82 micro-radians/pixel (mrad/pixel) with a field of view of 45º x 45º, the PanCams are 0.28 mrad/pixel and have a 16º x 16º field of view, while the HazCams are 2.1 mrad/pixel and a 124º x 124º field of view. [Maki, et al., 2003] The obvious bit of information is that the pancam has the highest resolving power, the NavCams less, and the HazCams the least. This is as you might expect given the different task each set of cameras was designed for. The resolving power of the MER cameras expressed in terms of micro-radians (0.0000573 degrees) per pixel is a measure of resolution relative to angle. Thus, for a given angle as distance decreases so too does the size of the area imaged per pixel. That is, for all other things held constant, the closer the object is to the camera the greater the level of detail that can be resolved.
2. In all other cases when images for use in a panorama are taken we have access to the underlying raw images. This time, with what is certainly one of the most important images taken, they are not publicly available. The date sol 71 is based on the knowledge that the rover was moving generally towards the crater each day of its visit to Bonneville. The image progression from sols 66-68 demonstrates the direction of movement up until this image was taken. Consulting with the illustration of the rovers path (Figure 2, Welcome to Bonneville) supports this.
In photographs from as late as sol 70 we can clearly see that there are no tire tracks which would be visible had the rover continued past the rock and had then returned. There are no public references to when or from what position the panorama was captured. This can potentially be explained due to calibration issues that were later corrected [Ruff et al., 2005] and they therefore would not be included in published reports.
Trying to identify when the images were taken is complicated by the fact that NASA has very likely misrepresented when some of the raw images were taken, not published others that clearly must exist and cropped others in an obvious way. Whether these were simply errors or a variety of technical problems is not known.
Compare Figures 4. & 5. which were, according to their file names and NASA's web site, taken two days apart: 2F132402222EDN1800P1161L0M1 with 2F132591409EDN1800P1161L0M1.
 |
 |
If the time stamps are correct then the rover did not move for two days as it is clear from the two photographs. Except for a tiny change in the position of the robotic arm the only thing that has changed is the angle of the sun.
The use or the rotating mast for the capture of panorama images with either the PanCam or the NavCam provides a somewhat distorted view of the area near the rover when compared with the same seen taken at a greater distance from the crater's edge. The result is that this rock, because it is right next to the rover, any imaging of it would be from the side. Hence the view we would see would be a side view not a front view.
Maki, J. N., et al. (2003), Mars Exploration Rover Engineering Cameras, J. Geophys. Res., 108, 8071, doi:10.1029/2003JE002077, E12.
Ruff, S. W., P. R. Christensen, D. L. Blaney, W. H. Farrand, J. R. Johnson, J. R. Michalski, J. E. Moersch, S. P. Wright, and S. W. Squyres (2006), The rocks of Gusev Crater as viewed by the Mini-TES instrument, J. Geophys. Res., 111, E12S18, doi:10.1029/2006JE002747.
An analyst born 53 days before NASA. A midwesterner now living in the southeast. 
