Best approach to model crowded cluster galaxies #161
Replies: 1 comment
-
Hi @joaquinhernandezg, I am very sorry about the late reply, I just noticed your post. I agree based on the the residual image that the fit is not correct or close. There are a couple of places where there might have been issues. Let me start by asking some questions:
Some of my own thoughts: My suggestion would be to break the image down into cutouts and fit individuals galaxies. Then transform the These are some high level suggestions but I could provide specific advice once I hear back from you. It would be quick to do this if I can see your image and PSF, that way I can send you a Jupyter Notebook with an example fit of one of the high index galaxies. If this sounds good, you can email me at |
Beta Was this translation helpful? Give feedback.
-
Hi, I write to ask what you think is the best approach to model cluster galaxies. I am interested in measuring sersic indexes and effective radius, and I am getting poor models when fitting galaxies with high sersic indexes (a ring-like structure at the center in the residuals). I am making stamps around each galaxy, calculating the segmentation image, then the parameter guesses using the Petrosian correction, and then fitting the stamp using a mask, and I suspect it could be an issue with the background subtraction and neighbors contamination, so I also tried adding a Const2D model to the fitting. I perform global and local background subtraction, and I tried several algorithms and combinations, and also tried single and multiple sersic profiles (for modeling bright sources in the stamps), changing the number of iterations in the fitting, applying weights (the inverse of the sqrt of the flux seems to produce better models in my case), but I can not make all fits converge to a nice model, although faint sources and sources with n<~3 are well modeled. Do you have a clue on what could be happening and how to tackle the problem? I am aware that this is not a simple problem, but any guidance would be helpful.
Here I attach some residuals from fits I did in individual stamps and then combined into a single image.
Beta Was this translation helpful? Give feedback.
All reactions