p.s. The green was to the left
Thanks, but it was not to the left. I DID see the picture you posted and I would not have asked you to tell me what I saw.
The ZD back's native orientation is upright (portrait). The picture you posted was rotated (automatically, I guess); the green part is either at the top or at the bottom of the native image. The relevance of my questions is:
1. the *raw* data of the ZD back is not sensitive to bit loss. If you "loose" a bit, that pixel value will be somewhat off, and that's it.
2. the JPEG compression is highly sensitive to single bit corruption. If that occurs on the "wrong place" (which is much of the data), then the rest of the image or of one color channel can be compromized; the "rest" is from that point to the end of the row and down to the bottom edge. If the green was at the bottom, then the bit corruption was possible. However, if the green segment was at the top, then this JPEG corruption could not have been the case.
pps. As for my suggestion of a tint, I wondered if there was an underlying sensor problem that showed up in these circumstances.
The sensor as source is not impossible, for example a contact error between the sensor and the board. An error on the sensor itself, occasionally, is IMO not probable, but I have no authority in hardware issues. Of course, the miraculous healing of the image excludes the sensor's responsibility.