Galaxies visible only in infrared?

[zed]

Are there galaxies which are redshifted enough to be invisible in the visible light but visible in infrared or microwave or radio spectrum?

I know some galaxies are invisible in the visible light because of the dust (Dust-Obscured Galaxies), but those would be visible without the dust obscuring them.

My question is about galaxies which are redshifted enough so we can't see them with the naked eyes, even when there is no dust obscuring them

[pbenni]

I wondered something similar. If available AP ccds are sensitive to the near IR range, and in combination to available IR pass filters, are there galaxies or other targets that could be imaged to show detail that otherwise we cannot see? One app I saw IR data on was webcam imaging Venus with an IR pass filter to see some atmosphere details.

thanks

Paul

[ronin]

The visible will move down to the IR but then the UV moves down into the visible, if further away then I suppose that the x-rays could move down to the visible and then gamma's above them if further still.

Because the EM spectrum is continuous and I am not aware of a maximum frequency then there shoiuld always be a wavelength/frequency that would be red shifted enough to occupy what we see/detect as visible.

[ollypenrice]

The new generation of telescopes will be very sensitive to the infra red, I gather. This infra red light started out from distant galaxies as UV and the professionals realized that they had not done much work on more local galaxies in the UV so they would not be well placed to compare distant redhifted UV with local native UV. As a result a new UV instrument was commisioned to try to learn more about local galaxies in UV.

At least this was my understanding of what a kindly French researcher spend some considerable amount of his time explaining to this dim English amateur when a friend wangled a tour of the Astrophysics institute in Marseille for us last year.

Olly

[zed]

I think that Galaxies do not emit in Gamma ray:

1. Galaxies emit all kinds of electromagnetic radiation, from x-rays to radio waves. - http://hubblesite.org/reference_desk/faq/answer.php.id=42&cat=galaxies

2. Natural sources of gamma rays on Earth include gamma decay from naturally occurring radioisotopes such as potassium-40, and also as a secondary radiation from various atmospheric interactions with cosmic ray particles. Some rare terrestrial natural sources that produce gamma rays that are not of a nuclear origin, are lightning strikes and terrestrial gamma-ray flashes, which produce high energy emissions from natural high-energy voltages - http://en.wikipedia.org/wiki/Gamma_ray

[Cath]

Their can be an atmospheric attenuation problem with trying to view the universe at anything other than visual frequencies , which I guess explains why biological eyes have evolved to use the frequencies they do ..

https://en.wikipedia.org/wiki/Skywave

[Cath]

I am not aware of a maximum frequency then there should always be a wavelength/frequency that would be red shifted enough to occupy what we see/detect as visible.

I wonder if their are particles/energy above the gamma ray frequency spectrum that we have yet to discover? .. or do we just class anything at gamma and above frequencies as gamma?

[JulianO]

Anything above gamma is still gamma basically, it needs a lot of energy to make such waves though (E=hf).

Most galaxies emit in the optical, and star forming ones in the UV, this gets red shifted into infra red once you're past a RS of about 2ish, so everything in the distant universe is basically IR and downwards.

Active galaxies, and odd astronomical events emit X-rays and gamma rays. Centres of galaxies too - but the flux is usually quite a bit less than at other wavelengths (counting a few photons a minute sort of thing) apart from events like gamma ray bursts.

[Knight of Clear Skies]

The visible will move down to the IR but then the UV moves down into the visible, if further away then I suppose that the x-rays could move down to the visible and then gamma's above them if further still.

This is correct, but the peak emission can be redshifted out of the visible range. For the most distant galaxies, it may be theoretically possible to detect them in visible light, but it is more practical to see them in infra-red.

The Hubble Ultra Deep Field was created by combining data from the ACS and NICMOS instruments: the latter is sensitive to infra-red and capable of detecting galaxies at greater red-shifts.

[zed]

Ok but does anyone know an example of such a galaxy and how far they have to be in order to be visible only in infrared?

[michael.h.f.wilkinson]

I'll ask my colleague Scott Trager from the astronomy department (Kapteyn Institute) about this. I'll see him on Wednesday.

[JulianO]

I think we worked it out the other day that something in the UV shifting to IR was around redshift somewhere around z=2, depending exactly what its emiting. Using the handy conversion table this comes out past 10 billion years ago.

[starminder]

Ok but does anyone know an example of such a galaxy and how far they have to be in order to be visible only in infrared?

Let's do a calculation!

Assume a galaxy has only Sun like stars. The Sun's effective temperature is 5800 Kelvin, this corresponds to a wavelength of 500nm at which it emits most of its radiation. If we were to redshift a galaxy that contained only Sun like stars we would have to make that 500nm light appear to be 700nm. This is a shift of 200nm. According to the redshift formula: change in wavlength / original wavlength, this is a redshift of 0.4.

So z = 0.4, a galaxy would have be travelling at:

v = z*c = (0.4*300,000 km/s) = 120,000 km/s due to the expansion of the universe.

We can now calculate the distance to this galaxy. v = H_{o} * d; here H_{o} is the Hubble Constant and d is the distance in mega-parsecs. The Hubble Constant is 67.80 +/- 0.77 km/s per megaparsec.

Therefore the distance is: d = v/H = 120,000 km/s / 67.80 km/s/Mpc = 1770 Megaparsecs or about 5.7 billion light years away.

This is the minimum distance a galaxy must be if it contained only Sun like stars for most of its light to be in the near infrared.

The vast majority of galaxys beyond redshift of 1 or 2 are all in the infrared. I'll post about the Lymann alpha forest and the Lyman break galaxies some other time.

[michael.h.f.wilkinson]

The sun does emit in the UV, so that is shifted into the visible. I am not aware of galaxies that are only visible in IR and longer wavelengths. In many the peak will be in IR, but a part remains visible.

[George Jones]

Let's do a calculation!

Assume a galaxy has only Sun like stars. The Sun's effective temperature is 5800 Kelvin, this corresponds to a wavelength of 500nm at which it emits most of its radiation. If we were to redshift a galaxy that contained only Sun like stars we would have to make that 500nm light appear to be 700nm. This is a shift of 200nm. According to the redshift formula: change in wavlength / original wavlength, this is a redshift of 0.4.

So z = 0.4, a galaxy would have be travelling at:

v = z*c = (0.4*300,000 km/s) = 120,000 km/s due to the expansion of the universe.

We can now calculate the distance to this galaxy. v = H_{o} * d; here H_{o} is the Hubble Constant and d is the distance in mega-parsecs. The Hubble Constant is 67.80 +/- 0.77 km/s per megaparsec.

Therefore the distance is: d = v/H = 120,000 km/s / 67.80 km/s/Mpc = 1770 Megaparsecs or about 5.7 billion light years away.

This is the minimum distance a galaxy must be if it contained only Sun like stars for most of its light to be in the near infrared.

The vast majority of galaxys beyond redshift of 1 or 2 are all in the infrared. I'll post about the Lymann alpha forest and the Lyman break galaxies some other time.

This is a nice calculation, but it is an approximation that is only good for small redshifts. My calculations indicate that: for a redshift of 0.4, there is an error of about 10%; for a redshift of 2, there is an error of about 66%

The relationship z = v/c is a non-relativistic expression for Doppler shift that differs substantially from cosmological redshift for z values that aren't small.

On Monday, I will try and post some detailed calculations for the redshift-distance relationship.

[rfdesigner]

One more point on redshifting.

A stars outpout can be defined in watts or photons. For our CCDs we care about photons, and this is where we suffer. A star may put out quite a bit of energy into the UV/X-ray/gamma region, but of course that doesn't equate with the same number of photons. Half a dozen UV photons have the same energy as loads and loads of IR ones, red-shift those UV photons into the IR and they don't start multiplying up, you still get just as few, making distant objects doubly difficult to find.

Derek

[acey]

In December 2012 a team of astronomers workign on the "Hubble Ultra Deep Field" reported a galaxy with redshift 11.9 (a new record). According to the following article, their method was "based on making images through ten different filters which cover the optical-to-near-infrared spectral region." In other words the galaxy was visible at optical wavelengths.

http://www.spacetelescope.org/news/heic1219/

[rfdesigner]

In December 2012 a team of astronomers workign on the "Hubble Ultra Deep Field" reported a galaxy with redshift 11.9 (a new record). According to the following article, their method was "based on making images through ten different filters which cover the optical-to-near-infrared spectral region." In other words the galaxy was visible at optical wavelengths.

http://www.spacetele.../news/heic1219/

Not so.

from the article: "The observations taken through the 1400nm filter, in which UDFj-39546284 was not visible, showed that the galaxy may be more distant than previously thought (as it only appears in the redder 1600nm filter)."

I suspect that they measured all available data, which included visible wavelenghts, and then used the data in the visible wavelengths to confirm there was no visible object, only a NIR object, i.e. the object needs to show the peak you get with high temperature black body radiation, but shifted into NIR, coupled with zero emission above that, rather than the flatter response you might get from a nearby NIR emitter such as an interstellar "jupiter" which might also show some visible response from reflected starlight.

I'm not an astrophysicist, but this is the sort of thing I'd want to make sure of before publishing.

Derek

[acey]

from the article: "The observations taken through the 1400nm filter, in which UDFj-39546284 was not visible, showed that the galaxy may be more distant than previously thought (as it only appears in the redder 1600nm filter)."

Some fuller quotes:

One previously-claimed candidate extreme redshift galaxy in the Hubble Ultra Deep Field was confirmed by the team. This is UDFj-39546284, for a while claimed to be the most distant known galaxy, at redshift 10 (heic1103). However, the improved and extended dataset has allowed the scientists to shed unexpected new light on this object, showing that it either lies at an even greater distance than previously thought (at a redshift of 11.9, handing it back the distance record), or must otherwise be a previously unknown type of extreme emission-line galaxy at much lower redshift.

There remains a small possibility that the object could be an extreme emission-line galaxy at much more modest distances (between redshift 2 and 3), but if this is the case then the object would have to be a previously undiscovered type of extreme emission-line galaxy. This hypothesis cannot be fully excluded until the object’s spectrum has been studied. This will not be possible until the the James Webb Space Telescope comes online in 2018. However, the balance of the available evidence favours the interpretation that this is the most distant object seen to date.

However, spectroscopy of objects as faint as those uncovered in this study is not feasible with current facilities. Instead, the astronomers have used what is, in effect, a crude version of spectroscopy, based on making images through ten different filters which cover the optical-to-near-infrared spectral region. This enables the broad spectral shape — but not the detailed features — of the light from a galaxy to be established. In this case, the method enabled the scientists to search for a characteristic sharp “step” in the galaxy spectrum, produced by the absorption of all very blue ultraviolet light by the neutral hydrogen gas which pervaded the Universe during most of the first billion years of cosmic history. The carefully chosen near-infrared filters exploited in this new study enabled this technique to be extended up to redshifts as high as 12 for the first time.

“Our study has taken the subject forward in two ways,” says Ellis [Richard Ellis, co-leader]. “First, we have used Hubble to make longer exposures than previously. The added depth is essential to reliably probe the early period of cosmic history. Second, we have used Hubble’s available colour filters very effectively to measure galaxy distances more precisely.”

[George Jones]

On Monday, I will try and post some detailed calculations for the redshift-distance relationship.

I finished the calculations yesterday, but I didn't post them. They are at about the level of an upper-level undergraduate physics student. Is anyone interested in seeing a detailed derivation of the redshift-distance relationship, including some explanatory comments and graphs? If anyone is, I will post my word-processed pdf file.

[nytecam]

Are there galaxies which are redshifted enough to be invisible in the visible light but visible in infrared or microwave or radio spectrum?

I know some galaxies are invisible in the visible light because of the dust (Dust-Obscured Galaxies), but those would be visible without the dust obscuring them.

My question is about galaxies which are redshifted enough so we can't see them with the naked eyes, even when there is no dust obscuring them

Galaxies shine predominantly via the stars they contain with a "Black Body" temperature which is the energy distribution across the wavelengths. The sun is a typical star which peaks @ ~500nm [see solar figure below] - cooler stars peak to the right and hotter stars to the left so there is a black body blend for any galaxy. At extreme redshift [effectively close to the Big Bang] this peak in radiation is shifted into the near-IR and so they are invisible to our eyes. Hence the need for large IR scopes to probe very deep.

ps: all gxys including the nearby bright Messiers, emit in near-IR and why I don't use an IR-block on my cam as such radiation adds more welcome photons to the image

[JulianO]

Yes please George!

[George Jones]

I have attached the pdf file for my calculations. If anyone sees any mistakes or wants any clarifications, let me know.

Redshift.pdf

[nytecam]

I have attached the pdf file for my calculations. If anyone sees any mistakes or wants any clarifications, let me know.

Thanks George but does it answer the OP thread title query or have I missed something?

[acey]

I think the crucial word is "visibility". An object is "visible" if you have a detector able to detect it, otherwise it is "invisible". Distant galaxies emit most of their radiation at infra-red, but will also emit faintly at shorter wavelengths (e.g. normal "visible light"). Whether they are visible at those wavelengths would depend on whether the telescope was powerful enough to detect that faint radiation.

The Hubble Ultra Deep Field Team detected a galaxy that was visible using a 1600nm infra-red filter, but was not detectable through a 1400nm filter. In other words it was invisible to that telescope at 1400nm (and at "visible light" wavelengths). A bigger telescope might have been able to see it at 1400nm, and a sufficiently huge one (with long enough exposure) might see it at "visible light" wavelengths.

http://www.spacetelescope.org/news/heic1219/