A new image of Eta Carinae and its Homunculus Nebula using recent observations acquired in early and mid 2018. I'll do my best to explain what's going on here, though my description may be inadequate.
Take special note of the blue parts of the nebula. Now imagine you are Nathan Smith, an astronomer who has been studying the Carina nebula and Eta Carinae in particular for years, now. His March 2018 observations used a filter designed to only allow light from a certain magnesium emission line to pass through, and this particular filter had never been used to look at it before. Surprisingly, this part of the nebula was in a different shape from previously observed emission lines. This is really cool, because maybe it can tell us more about the enigmatic process this system is going through.
But wait, in order to properly characterize this new layer of nebula, it had to then be compared to light from some other filters. However, the whole thing is expanding rapidly, and old imagery was too far out of alignment from the new imagery. Wait too long, and new data won't match, either. Enter the Director's Discretionary budget, which allows for somewhat unplanned, emergent observations to take place. Luckily for Smith, he got his new observations, and luckily for us, this means we get a new image of this amazing nebula!
I have some things to mention regarding the processing of this image. First, there were some pretty huge charge bleeds I had to either fill in with data from shorter exposures, or clean up with cloned data. Second, coverage was lacking around the left side and lower right corner of the image. Third, there were massive, circular filter ghosts everywhere, especially in the F658N data. I made use of older observations to help mitigate these artifacts, but as previously mentioned the bits of nebula are moving quite a bit, so I had to stretch and nudge them to get them to match. This all should be a reasonable approximation of how the nebula looked at this epoch, but if you are trying to recreate this image yourself, know that it took a lot of work and yours will probably turn out different.
I also want to mention that the smaller blue stars were very distracting and green, so I made creative use of the F658N data to help dim them to a much softer appearance.
Data from the following proposals were used to create this image:
Orange: ACS/WFC F658N
Green: WFC3/UVIS F336W
Cyan: WFC3/UVIS F280N
North is up.