It's the moon, of course! On October 12 the moon is full, we are in eclipse period so the sun is near the orbit plane, and so is the moon. The RA and DEC of the moon at the two imaging times is given below:

UT	Right Ascension (degrees)	Declination (degrees)
0754	7.86	-1.85
2208	15.0	1.07

I calculated these using the script at : www.geocities.com/CapeCanaveral/Launchpad/1837/mooncalculator.htm

Using fuv_ptg.pro and calc_fuv_pointing.pro in fuview, the location of our object is right on in RA, and variably strange in DEC. That is no surprise. Now look for it in images from SI as well:

The moon is plainly evident in the SI-12 images. Hot hydrogen on the moon? Probably not, as Harald noted that scattered solar Lyman Alpha probably is is very broad. Since it is not resonant scattering (e.g. geocorona), there is little frequency redistribution.

One can _barely_ see the moon in the SI-13 image at 0754 UT. It is at the very top, basically colocated with the peak in the SI-12. This SHOULD be an indicator of what you might expect from a broadband view at 1400-1900 (WIC) except that since there is no N2 on the sun or the moon, maybe one would expect even less response relative to the Earth than that seen in SI-13. However, if rayleigh scattering of solar continum at shorter wavelengths has an effect, this starts to explain the unexpectedly bright moon (and Earth) in WIC.

The moon may be contributing to the limb in the 2210 UT SI-13 image, but it is difficult to say.

So, back to the Earth, and the altitude-brightness profile.