Converting from Observed to Rest-frame Wavelengths in SpecViz

[kassin]

From @bethanjames:

"It would be very useful for the user to apply a redshift value near the wavelength selection tool, that way you could offset the wavelength array and move through the cube in the rest-frame wavelengths."

@nmearl I know this is implemented in the stand-along SpecViz. Do you have a suggestion as how it would work in CubeViz?

@brechmos-stsci

[brechmos-stsci]

@kassin There is some functionality that Jesse is working on for the unit conversion stuff that we could leverage for cubeviz. Is it just a matter of the user having the ability to define an offset the redshift?

[kassin]

Yes, the user would enter the redshift and the spectrum would be shifted to the rest-frame.

[brechmos-stsci]

So, from the non-specviz side of this, I think it would just end up being a change in the internal wavelengths variable for display.

Any change in the redshift, though, we would have to propagate down to specviz.

@nmearl I suspect we are going to want a Message sent from cubeviz or specviz about a redshift change so that both can operate on it. Does that make the most sense?

[nmearl]

@brechmos-stsci indeed! The redshifting behavior only requires that a single float value be provided, so something like

specviz_dispatch.change_redshift.emit(redshift=0.01)

Would be all that's needed from cubeviz.

[brechmos-stsci]

@kassin @hcferguson Is there a specific range of redshifts which are acceptable? (z>0? should it be less than a certain value? or is any numerical value acceptable?)

[brechmos-stsci]

@nmearl I don't see a reshift_changed (or redshift_changed) in specviz. Is that something I need to add or am I just not seeing where it is in specviz?

[nmearl]

Do you want to this change the non-redshift x-axis wavelength values, or just activate the top axis red shift display mode?

[brechmos-stsci]

@nmearl I would assume they are going to want to see the top axis red shift display mode otherwise I think users will find it confusing. What is the name of the dispatch message? I didn't see it.

[brechmos-stsci]

@kassin When you have a minute, could you please see the question above?

[nmearl]

@brechmos-stsci Try pulling specviz master; I think it wasn't exposed properly. But the call should look like

specviz_dispatch.change_redshift.emit(redshift=0.01)

[hcferguson]

@brechmos-stsci While the extragalactic case usually has positive redshift, other science can have sources with both positive and negative radial velocities. For those cases, usually the shifts are expressed in km/s rather than z, though, and those can be positive and negative.

[brechmos-stsci]

@nmearl I think I am not understanding the parameter to change_redshift. Is this the z value in that rest_wavelength = obs_wavelength * (1 + z)?

If I put 1 in for the "Axis Mode" Redshift... the bottom axis shows 0.6, I would have though the top "redshifted" axis would be 0.6 * (1 + z) = 1.2 (with redshift z = 1). (I understand that 1 is a huge redshift but it makes the math easier in my head.)

Am I mis-interpreting the redshift value?

[kassin]

@brechmos-stsci The spectrum is at a given redshift. This "deredshifts" it so that it is shown in rest (not observed) wavelengths.

Btw, z=1 is not a large z. I am on a JWST proposal to observe a z=11 lensed galaxy.