Tulip Nebula 
& Cygnus X-1 Bowshock

Two things share this field, and only one of them is visible for what it is. The Tulip Nebula is a cloud of hydrogen about 6,000 light-years away in Cygnus, catalogued by Stewart Sharpless in 1959 and named for the shape it takes on a photograph. The other is a black hole, and what you can see of it is the wave it pushes ahead of itself through the surrounding gas.

Sh2-101 is an HII region — hydrogen ionised by the young, hot stars inside it, glowing as that hydrogen recombines. It is invisible to the eye and modest even on a photograph: the shape that gives it its name only resolves with enough exposure to separate the petals from the background. This frame runs 41.6 by 63 arcminutes, wide enough to hold the nebula and the field it shares.

Cygnus X-1 sits in the same field, and it was one of the first objects ever taken seriously as a black hole candidate, back in 1965. It holds about 21 solar masses, and it is not alone: it orbits with HDE 226868, a blue supergiant, close enough that the star's wind is being pulled onto it continuously. That is what makes the system visible in X-rays — infalling material heated to millions of degrees in an accretion disc. It is also what makes it visible here. The wind streaming off the supergiant ploughs into the interstellar gas ahead of the system, and the shock front it raises is the arc running through this frame. The black hole itself shows nothing at all; what the image records is the pressure wave it drags through the neighbourhood.

Ninety-four hours was the plan, not an accident. The shock front is the reason: it is faint, it is diffuse, and no amount of clever processing recovers a signal that was never collected. So 345 exposures went to OIII alone, and even that was not enough on its own. Pulling the arc out of the background took a hard stretch on the oxygen layer and a lot of masking, and the danger in that direction is specific. Stretch far enough and noise starts to look like structure — you end up showing oxygen that was never there, and the boundary between the shock and the sky begins to read as a drawn edge rather than a physical one. Most of the work was spent not on bringing the arc up, but on keeping the noise around it from following.

The colours here are assigned, not real. This is the Hubble palette: SII to red, Ha to green, OIII to blue — three narrowband filters standing in for the three colour channels. A narrowband filter records a single emission line, so what it captures has no natural colour at all until one is given to it. The stars are the exception: eight hours of red, green and blue were shot alongside the narrowband set and used for them alone, so every point of light in this frame carries its true colour while the gas around it carries a mapped one.