Søgeresultater for 'V339'

[nightskyDeltager Neutron star]

ATel #5378

Artikel 9 fra Steven Shore

The light curve has caused some amusing impatience; it’s been amusing to see some of the
comments. It’s perhaps always good to remember that in an environment as complex as we must
have in these ejecta, as in other time dependent structures, unexpected things happen.
This brings to the fore a comment, in the hope it will put some of the present results in
perspective.

The difference between a dynamical medium, like nova or supernova ejecta or even a
stellar wind, and a static structure — a stellar or planetary atmosphere — is “locality”.
An atmosphere is a very particular sort of medium, whether optically thick or thin (or
both).

It’s a gas that is constrained to satisfy a mechanical equilibrium constraint, that the
run of pressure and temperature balance the effect of gravity (whether self-gravity or a
thin layer above a mass). The gradient with radius of the pressure and density must every-
where counterbalance an acceleration. If radiation pressure is important, this effectively
reduces the surface gravity but as long as it is low enough, the radiative
acceleration will not destabilize the medium. At the same time, since the equation of
state links the temperature, pressure, and density, and both are connected with the transfer
of radiation through the layer, the luminosity enters the picture in a relatively simple
way, at least in the sense that the radiation must pass from the interior to the outside
through a “surface” and ultimately the only loss is radiative (OK, no magnetic fields, no
neutrinos, just to be basic). As long as the luminosity stays roughly constant, and there
is a very clear definition of what this means, the atmosphere will have a unique static
structure. “Constant” means nothing changes faster than the thermal timescale of the
medium (the rate of cooling) or the sound travel time (which sets the condition for
hydrostatic — mechanical — balance. This doesn’t mean that the only parameters entering
the picture are the thermodynamic variables. The whole business is governed by the
abundance of the elements since the different elements have different ionization
energies, transition energies (lines), and different ionization states. The mix
collectively produces the opacity and that’s the key coupling between the radiation and
the matter. But if the medium is stable and static, there is a solution to its structure
and you can predict a spectrum emergent from the medium as a function of only a few
variables (OK, knowing the metallicity).

A dynamic medium is a whole new problem. There’s no mechanical constraint to govern the
pressure and density. Those are imposed by the dynamics and depend on the driving. If the
medium is a wind, let’s be simple and take a steady outflow, then the condition that the
mass loss remains constant links the density and velocity at each distance from the star.
The mechanical requirement, that the velocity is determined by the driving, is the hard
part: you have to impose a mechanism (radiation, rotation, magnetic, whatever) that is
described by its own physical constraints, to determine the run of density with radius.
This, in turn, affects the ionization and — through that — the opacity at each radius.

Add to this that if a medium is static a photon leaves a line, or is absorbed, at a wave-
length that’s symmetric around the line center and not shifted by the differential (Doppler
effect) motion between strata.

Not so in a wind. A photon emitted at the center of a line sees ahead of it a reduced
opacity from the same transition because of the differential motion of the medium. On the
other hand, depending on the spectrum, another line may be shifted in the way. This is
the cross-coupling effect we’ve already discussed during the Fe-curtain phase of the nova
outburst. In a wind, except in the most optically thin case, this is inevitable. It is
the real reason for the absorption trough on the P Cyg profiles you’ve seen in windy
stars.

But there’s another thing about a wind that highlights the “non-locality”.

A wind is driven so it has to reach a terminal velocity. You don’t have infinite energy
to accelerate the flow forever. This means the rate of change of the velocity decreases
with distance from the star (or the driver, more generally). So in the outer wind, a
photon sees a longer effective absorption path length ahead of it because all of the
material above a certain radius is almost moving at constant velocity.

Not important that the density is decreasing, the velocity is constant. So the optical
depth can be large even though the density (locally) is small. This produces the deep
absorption that has a well defined edge in the P Cyg profiles of winds; the velocity
gradient goes to zero at distance and the effective path length approaches infinity.

OK, it doesn’t become infinite, but all of the matter is at the same velocity.

Instead, in ejecta, the stuff has been blown off at a moment and the driving is over.
Until there is a deceleration, and for matter far enough above the escape velocity this
can be very far from the star, the expansion preserves the velocity field. Eventually
there will be interaction with a background (if there is one, like the RS Oph-type novae)
or a small nonlinear gradient can appear, but this is late. In general; the matter is
going at its terminal velocity and each piece is independent. The formation of the
spectrum is, therefore, different than a wind — there’s no terminal velocity, just a
maximum — and the line profiles show this. Instead of well defined edges, the lines show
the change in the density along the line of sight, not mediated by a decreasing velocity
gradient. Unlike a wind, because there is no active driving, changes in the opacity just
change the spectrum and the thermal state of the matter but this is completely dependent
on the local densities and the highly non-local radiation field. So there is nothing to
constrain the density to be any particular value locally, and the structures remain and
don’t advect (move outward relative to the ejecta) with time as they do in a wind.

The ejecta are also very different than an H II region or even a planetary nebula. The
velocities are high and can be very different from one point to another, and very time
dependent. The recombination timescale in an H II region is long but the ionization
source remains on so the state of ionization of the gas can reach a balance. In a wind or
ejecta, this can be very different. The state of ionization of a gas can remain very high
even when there’s no longer an ionizing source because of the expansion. The decrease in
density slows recombination so even if the central source is extinguished the expansion
can decrease the density fast enough to effectively halt recombination. An H II region
doesn’t expand supersonically even though it DOES expand so the state of the gas is
governed by the collisions (rare though they are) and the ionizing photons (diffuse or
central). The same is true for a supernova but there the mass and energy of the ejecta
are so great that the shock from interacting with even a rarified interstellar medium is
enough to power the ionization and emission until the ejecta fall to the local sound
speed.

While this doesn’t have anything in particular to do with what’s happening now — at this
moment — in our favourite nova, I hope it puts the observations in perspective. There’s
another reason for this tirade.

Stellar classification is the reflection of the regularity of stellar atmospheres that
comes from this double equilibrium. Becau
se the radiation and matter are coupled in a
static medium, depending on the composition you would expect a particular spectrum to
arise from a run of thermodynamic conditions in the atmosphere. So there is a sort of
link between the features of the spectrum and the luminosity and radius — the latter are
combined in the “effective temperature”. This isn’t true for ejecta of any kind, and even
for winds. The spectrum is not unique for a structure, the worst non-equilibrium
environment you could fear dominates its formation: the matter and radiation are coupled
dynamically and locally and non-locally and may be time dependent. The desire to find a
classification system of any kind for phenomena that have such a wide range of variation
is potentially misleading and certainly much different than the basic taxonomy of stellar
atmospheres. One last example. The ionization of specific elements, line ratios among
ions of the same element, is used as a temperature indicator. In a static medium this
works. In a dynamical medium, as you know, this isn’t true. In novae, for instance what
you will see very soon, the same spectrum can show [Fe X] or He II and O I. They’re
coming, perhaps, from different pieces of the same ejecta with different densities at the
same velocities.

Range of interest

Oh, one more thing, not in the previous message: The universality of the optically thin
profiles, the best examples being the O I lines, makes for a useful tool — the peak at
+550 km/s is identical in all the profiles and just this one peak is enough to link to
the rest wavelength. I’ve checked this with about a dozen examples and it works almost
perfectly, meaning that the ionization and density structures are being identically
sampled by all available lines (mainly neutrals).

One example: next to Fe II 5018 there’s an emission that *could* be He II 5047. But
using the profile, it is almost certainly C I, like the 7115A line. The one for which it
isn’t working is the line around 6720A. That’s still a sort of mystery and since it’s
even visible at low resolution (< 3000) I recommend keeping track of its development in
the next few weeks. It could be something interesting.

Mere Q & A omkring det vi har observeret

Q: I’m rather confused about the CN 4216 line.

A: To confuse you more, The CN didn’t appear. It *may* mean it’s not a unique
feature of the precursor event — it was a hope to test that and the density of observations
did. There is only one clear report in any nova and that is only – I think – because DQ Her
was the brightest dust forming nova to have this early stage densely covered. It may not
be a unique signature, if dust does form here, the observation that CN was NOT detected
becomes the definitive disproof of uniqueness. Firm statements are definitely welcome
and that will be one. If dust doesn’t form, then we’re still stuck not knowing.

Q: Another question that puzzles me in my own low-res observations. If I want to
compare the continuum of the nova in visual light 3700 – 7000Å of all my observations, at
what wavelength area should I use normalize? Right now I’m using 6700-6920Å

A: This is a hard question. It would be best to normalize to the photometry. Even
though that’s dangerous you can use a region line that around H-alpha (say V or R) and
use that for scaling. Alternatively, if you can find lines that are not varying, that
helps.

A check is the DIBs (diffuse interstellar bands) that you can see in the higher resolution
spectra at 5780-5800A, for instance. These don’t vary. The same for the Na I D lines. The
interstellar value should be constant so a check on any contaminating continuum or
variation is to use that equivalent width. Otherwise it is best to go as far from the
line center as you can, at this stage it’s hard with the complexity of emission and
absorption features.

Opdatering til ovenstående spørgsmål:
ARAS gruppen (Buil og Leadbeater) er vist kommet frem med to metoder til at beregne
fotometri ud fra spektre og derved få absolutte tal. Derved kan amatør optagelser
sammenlignes med en uhørt præcision og give yderlig videnskabelig værdi til observationerne.
Et meget stort skridt fremad.
http://www.astrosurf.com/buil/calibration2/absolute_calibration_en.htm

Animation der fint viser ændringer i jern linjerne (Iron curtain)
Olivier Garde har lavet denne fremragende animation med alle hans data siden d. 14. aug.
(31 spektre). Vi ser Fe II linjerne ved 5169, 5198 og 5235 Å.

[nightskyDeltager Neutron star]

En kort opdatering fra tirsdag d. 10 sep. Egne observationer.

En hurtig, ikke dark/bias/flat, kalibreret optagelse hvor Ha emissionen går i mætning.

Ioniseringen af jern ses ganske tydeligt.

En bedre analyser kommer senere.

[nightskyDeltager Neutron star]

Lidt aha oplevelser fra Tvis omkring støv, Novaer etc.

I starten af denne for mig utrolige rejse med Nova Del 2013, undrede jeg mig over opfordringen
til at holde øje med Cn linjen 4216Å for støv dannelse. Er der en sammenhæng mellem dette
støv og noget jeg har i Tvis?

Så jeg rodede lidt rundt i det, bl.a. dette paper “Photo-Ionization Induced Rapid Grain
Growth in Novae” for at finde ud af hvorfor dette støv er så vigtigt. Selvfølgelig er det
vigtigt at vide hvad/hvorfor dette støv skabes, men er der vigtigere spørgsmål omkring
dette på en større skala? Interstellart støv, kometer, meteoritter, livets oprindelse?

Jeg tog mig sammen og spurgte til det.

Svar (Har forsøgt at sammenfatte og oversætte efter bedste evne)

Dette er et delikat spørgsmål. Jeg vil besvare det på en næsten epistemologisk måde. Vi ser
på en fast fase, der overlever de interstellare betingelser der er til stede, selv i
supernovarester og aktive galaktiske kerner. Dannelsen af faste stoffer er svært nok i
laboratoriet, men så tænk på problemet med skydannelse. I atmosfæren er du næsten på
grænsen ved termisk ligevægt, og du kan stadig bruge en slags ”forståelige” begreber til
at modellere overgangen fast – gas, men det er ikke altid tilfældet. Hvad jeg mener er: i
termodynamik glemmer man alt om struktur og bare balancere mængder med mængderne i samme
volumen. Hvis der er noget mikroskopisk, antages det at tage et gennemsnit, og man ser
f.eks. kun på makroskopiske tætheder, tryk og temperaturer. Det er derfor, du ikke kan
sige, hvordan en dråbe vil se ud, men du kan sige, hvornår vandet vil kondensere. På den
anden side, i den øvre atmosfære (ligesom natlige højtliggende skyer og polære skyer)
dannelsen af is er typiske for kinetiske processer, diffusion og sammenlægning af klynger
dannet af kollisioner og tilvækst fra molekyler i miljøet.

Det er ikke fordi du når en enkelt kritisk temperatur og “bum”
— Alting pludselig kondenserer (tænk på kondensations kamre). I stedet er den
gennemsnitlige energi af de kolliderende partikler lav nok til, at når de rammer
hinanden, kan de holde til det og ikke miste for mange andre atomer/molekyler/klynger.

Vi ved at dannelsen af faste stoffer – støv – sker i de tidlige stadier af supernova
ekspansion. Det sker i massive stjerners vind, røde giganter og superkæmpes vinde, og –
har du det – Novae. At det kan observeres i realtid (vores Nova Del 2013) i et medium,
som vi grundigt kan angive er næsten et laboratorium test af idéer. Det er grunden til,
at støvdannelse er så vigtig.

Men der er en anden, måske mere sublim grund, og jeg mener, at i en nærmest mystisk
forstand: i meteoritter vi finder pre-Sol korn/partikler, hvis sammensætninger vi mener
kommer fra røde super giganter, supernovaer og novaer. Der er særlige isotoper i disse
fragmenter, ofte kun 10 mikrometer eller mindre på tværs, isoleret inden for matricer.
Afhængigt af de omgivende temperaturer, sammensætninger, tætheder, forskellige
astrofysiske miljøer, vil der produceres forskellige isotoper Til dato er tre forskellige
sandsynlige kilder er blevet adskilt fra hinanden. Af disse en lille, men statistisk
signifikant antal er knyttet til novaer (som Nova Del 2013). De bedste artikler er af
Jordi Jose, Ernst Zinner, Andrew Davis, Christian Illiads, og der er en gruppe i Chicago
og en anden på Washington Univ. (US), der arbejder meget med dette problem.

Et sidste problem er, at tilstedeværelsen af støv i planetariske atmosfærer er afgørende
for nogle køling og blandings processer. I ISM, er støvet en kondens miljø, der opbruger
metaller (noget fra CNO og op, Zn ikke inkluderet, interessant nok).

Og det er den sidste grund
– Uden støv kan der ikke være dannelsen af molekylær brint.

Dette svar bringer straks en videre til meteoritter som indeholder dette pre-Sol støv:

Allende
Murchison
Tagish Lake
Og mange flere

Et foto fra samlingen – en stor oplevelse at kunne knytte en nova og de observationer jeg
laver med et ultra billigt instrument, sammen med meteoritter i min samling. I hånden kan
man holde og mærke et produkt af den voldsomme proces vi observer langt langt væk. En
proces som er med til at skabe betingelser for OS.

Fra en af Jordi Jose præsentationer:
Stellar Pyrotechnics: Classical Noave as Cataclysmic Events
http://pntpm3.ulb.ac.be/Trento/talks/pdf/jjose.pdf

[nightskyDeltager Neutron star]

Artikel 8 fra Steven Shore. Hvad vi kan forvente, De gamle Payne-Gaposchkin’s data, lidt
tips m.m.

An observation with the NOT has been requested for 7 Sept. (Saturday-Sunday) with the aim
of providing a second, independent calibration of all of your data and get another very
high resolution observation of the absorption features. The Na I velocities have remained
almost stationary but that may be a resolution effect. Two spectra, 10 sec and 1000 sec
have been requested. As I wrote, the agreement overall is excellent.

Once the nova is below V ~ 8 we will be getting Swift UVOT data and as soon as the XR
source turns on we’ll schedule Chandra (pre-approved).

We still have time for this period with NOT for three spectra, the next sequence starts
in Oct.-Apr. and a proposal is due for 30 Sept. for the period thereafter.

On the PI business, this is a relic of the proposal process at major facilities and
mainly because of proprietary practices. The same is true for Swift, they use the term
but we all know it’s fake because the data is immediately public. It should be
“Investigation leader” or something like that but the term was imported from other
observatories (e.g. HST, Chandra) where the data has a period of restricted access.

If the German group wants to keep their data that’s fine. It can’t be better than what
you’ve all produced.

Update: German group = VdS SPEKTROSKOPIE Fachgruppe
Pro/Am kampagne med observation af Wolf-Rayet stjernerne WR 134,135 og 137.
83 amatører og professionelle er involveret i projektet, bl.a. Knud Strandbæk AAF.
Der observeres hver nat over flere måneder fra IAC 80 på Teide
Teide holdene har haft mulighed for enkelte observationer af Nova Del 2013 uden at det
ødelægger WR kampagnen
Gruppens DATA er netop kommet med, heldigvis blev de små problemer løst

As a general comment about what can be done now — the rise and early decline would be
the key here, to cover the physics of the stage relative to other novae for which we have
the UV and emphasize the developments in the optical. For this using the low resolution
data is essential for coverage of the continuum and showing the relation with classical
observations (e.g. sequences of the spectra as images or, better, surface plots).

The continuum can be normalized “backwards” in time by using the available photometry so
the energy budget can be studied with the high resolution data showing the profile
variations and velocities. The sequences you’ve all produced showing the “retreat” of the
absorption on Halpha and the disappearance of the He I 5876,6678 lines are fantastic and
the lower resolution measurements can be directly compared with novae in the literature
from the days of photographic spectra. The density of coverage rivals a *very) few bright
20th century novae (I can recall Kepler myself!) done with photographic plates; my
thought for a project for anyone who might like to try it is to simulate what those
observers were measuring to see if it’s possible to calibrate their velocity curves. It’s
never been possible. Keep in mind that those characters were making microscope
measurements, or at best taking tracings (until the invention of the Grant machine).
comparing spectra with standards to get line displacements. It was very skilled but
subjective and individual plate quality varied far more than your data.

I was trying to reason one thing out today. The dispersion is not always discussed in
Payne-Gaposchkin’s compilation and for many I haven’t gone back through the literature to
see when the switch was made in different spectrographs and plate material. Those before
1940 were frequently prism spectra (at least at Lick and McDonald) so have the problem of
variable dispersion. The emulsions were usually rather coarse-grained, like Kodak 103aO
or AGFA, with grains of order 20-50 microns. Since the typical dispersion was 50 -100
A/mm, this amounts to 1-10 A/pixel equivalent; at H-beta this is 150-300 km/s per
resolution element (2 px). You see that even the low resolution data is of the same order
in the current database, and for photographic plates the maximum S/N ratio was never much
better than 20. You have, in some cases, 100 or better. So for these data even the lowest
resolution spectra provide information on the short timescale variations that none of the
classical studies could show. The coverage of the various “flash” stages can be used to
understand what the descriptions in the literature mean, this includes data from the ’60s.

And the time coverage is denser than I think has ever been obtained for any nova, nearly
continuous. So time series will be possible, for instance to look for short time
variations across the profiles (not just integrated values!).

One last thing. Thinking about exposures and calibrations, since we’re now about one
month (2 hrs LST) later, I should mention a caution.

Nogle vigtige tips

When your exposures are getting long, say one hour, for any object and it’s at high
zenith angle (high air mass), site change sin atmospheric absorption lines change the
line profiles. This is also a problem for the absolute calibrations. If you’re observing
a source at high air mass, and you have the time, it’s useful to get a standard at the
start and end of the exposure, no matter what you’re doing. Near the meridian this isn’t
a problem but since the atmosphere changes exponentially as you go away from that area
the structure (not just slope) of the continuum changes. For example, at [O I] 6300 and
H-alpha, water vapor bands sit on features that are possibly real in the profiles and you
need to be careful that the same features don’t appear in the standard. But this doesn’t
mean removing them, it just calls for care. These also change over time, from site to
site, so the extensive network you’ve built up is vital for inter-comparison. It’s not
only a problem for high resolution, the bands have structure that’s marginally resolved
even in the lowest of your spectra.

And this also changes the total flux, although for most lines < 6700A this isn’t a
serious problem. Remember that there’s water absorption even on Na I D, it is not only
confined to the red part of the spectrum, and when some emission lines appear (if they
do), like the [Ne III] doublet, these may have second order contributors in some spectra
depending on the spectrograph (e.g. [Ne V]3426). Alas, I have experience with this
problem.

steve

Link:
Gratis download af denne klassiker:
Cecilia Payne-Gaposchkin:
The Galactic Novae Dover Publications Inc. 1964
http://archive.org/details/TheGalacticNovae

Alle data fra det fantastiske samarbejde kan hentes her:
http://www.astrosurf.com/aras/Aras_DataBase/Novae/Nova-Del-2013.htm
Uhyre nemt at arbejde med disse data. Download Visual Spec eller ISIS og du er i gang.

[nightskyDeltager Neutron star]

Artikel 7 fra Steven Shore om Nova Del 2013

First, we’re nearly at the stage, t_3, where the optical spectrum usually goes through
another transition. The emission lines should strengthen, the continuum should quickly
fade, and emission lines of moderately ionized species should appear. That’s the standard
statement, that this timescale defines the nova event.

But as we discussed earlier, the timing of these events is tied to the structure of the
ejecta and the evolution of the underlying WD. In these spectra, for instance (And
Christian’s are also showing much of this) there’s a new feature. Look at the Ca II lines
(those around 8500A). There’s virtually identical structure on these lines, it’s not
atmospheric water absorption as demonstrated by the [O I] and Ca II 3933. These tiny
features, throughout the line profile, symmetric about zero, are signs of the ejecta
structure and the signal that these transitions are optically thin. The lines from
similar ions, or similar ionization/excitation conditions, should be the same and you see
the same structure on a forbidden line ([O I]) as the permitted (Ca II), from a neutral
and from an ion. The ejecta geometry, if we use a bipolar model, seems to fit a rather
high inclination but it’s also showing another effect. Notice in the second set of
profiles that the O I 8446 extends to higher velocity in the wings (like H delta) than
[O I]6300. The O I is connected to the ground transition O I 1302 in its lower state, the
upper state is fluorescent with Lyman beta, hence it looks like H-delta and the higher
Balmer lines that are weighted toward the inner part of the ejecta. The forbidden line
bleeds off the photons from O I 1302 so it’s a different profile, more like the Ca II
which are excited state transitions only. There are three of there, one of which is
nearly coincident with O I.

As the shorter wavelengths become more transparent, the profiles will become more nearly
the same. The next moment is when the UV starts to ionize the Fe and the curtain lifts,
when the [N II] 5755, 6548,6583 lines appear, and then when the [O III] 4363, 4949,5007
are excited.

The former are simple forbidden transitions, although with the same atomic configuration
as the O III. This is called “isoelectronic” in having the same state structures (recall
that N+ is the same number of electrons as O+2 but with a different nuclear charge, that
makes relatively little difference for the binding, hence the lines are near each other).
In the ejecta, since the O I 8446 line is formed by pumping, it’s intensity varies
linearly with density while the recombination lines, like Ca II (permitted and excited
states) form by recombination so the intensity varies as density-squared. To be more
precise, and I hope less technical, the formation of a line by recombination means that
electron capture takes place so the emission depends on the number of captured electrons
(one power of density) and the number of ions (the other power). Pumping depends only on
the number of ions to be pumped and the availability of photons, so it’s a different
density dependence. Now recalling that the density is lower in the periphery of the
ejecta where the velocity is highest (in this ejecta picture, but also for a wind), the
wings are weaker but extend to the point of invisibility. The [O I] is formed, instead,
by the 1302 photons being trapped and “leaking out” and that requires the inner region.
But there’s another important piece of information here, that the forbidden transitions
aren’t seen if the density is any region (for a temperature of about 10,000 K or so) is
too high so there’s an upper limit (about 1E9/cm^3) for the inner part. If we take that
to be about 1000 km/s, assuming what we know from other novae, then as a first pass guess
the mass of the ejecta is about 8E-5 solar masses (yes, you heard it first here). This
depends on the filling factor which, from the NOT observations and what you’ve seen in
the fine structure, suggests about 10% or 30% of the ejecta s filled with an aerosol of
filaments so this could be as low as 2E-5 M_sun.

This is a normal value for the ejecta and I’m assuming that the inner density is low
enough to produce the [O I].

The calculation assumes that we’re seeing this at 20 days with a velocity of 1000 km/s
for the inner part and about 3000 km/s for the outer, fiducial numbers. It doesn’t give
an abundance but it’s a start. The other is that the emission at H-alpha accounts for
almost 8000 L_sun if the nova is at 5 kpc and scales as (D/5 kpc)^2, so a lot of energy
is coming out in a single line.

It’s this last point I wanted to also mention because the ejecta are acting as a sort of
bolometer, or calorimeter. The energy now derives from the original hot gas and the
heating from the WD radiation. That will keep up until the nova turns off, when the
nuclear source collapses and the WD starts to cool. The rapidity of this stage is probed
by the direct measurement of the XRs, which will appear shortly if all is right here, and
by the appearance of very highly ionized species like Fe VII and Ca V, or even higher.
That’s still in the future but shouldn’t be very long. I haven’t heard whether the gamma
ray source is still on but it shouldn’t be, if the internal shocks are the powering
agent, but the radio should also turn on soon as the ejecta turn optically thin in the
centimeter wavelength range.

So that’s what’s to come, but the beauty of this stage is that we’re beginning the
transition when you get to see, like a tomogram of a body, the individual parts of the
inner ejecta becoming visible. I don’t know another stage, whether in stellar outflows
(like luminous blue variables) or even planetary nebulae (this is the last stage after
the superwind from the central star turns on) when you see the third dimension of the
universe so clearly.

As always, I hope these explanations are clear. If — and please always keep this in
mind — if you have any questions or comments I’m delighted to hear them.

steve

[nightskyDeltager Neutron star]

Lidt Q & A med Steven Shore – Nogle spørgsmål fra jeg selv og et par andre DK amatører

Now to your questions.

Q: We have looked at your model for the ejecta shape and the line profiles from
NOT you compare with the model. At first it’s looks like a reasonable match, but when looked
at in more detail, maybe there’s more to it. On both wings on the emission line there’s
absorption features that’s not in the model. These absorption lines, should they be considered
when evaluating the model?

A: The models are Monte Carlo simulations of the geometry and emissivity. The
description is in Paper II on T Pyx and Paper I on Mon 2012

http://arxiv.org/abs/1211.3453
http://arxiv.org/pdf/1211.3453v1 (side 6)

http://arxiv.org/abs/1303.0404
http://arxiv.org/pdf/1303.0404v1 (side 10-11)

They’re not purely qualitative but because they’re generic profiles expected for specific
emissivities they’re not to be interpreted without reference to the ionization state. In
this case it’s simple because the individual line profiles are optically thin. I’m
enclosing the comparison with O I in two lines, just as an indication.

The parameters are:

vmax (km/s); (density exponent); )(line exponent);
theta_out,theta_in,dr,prolateness,inclination

In this case, for a change, I used an inclination of 45 degrees to the line of sight. The
angles and some other parameters are different from the first pass model because this is
an illustration of some of the inherent uncertainties in this kind of model. You can’t
derive TOO much quantitatively. The “image” corresponding to these line profiles is also
produced by the code and I’m enclosing that for your amusement.

It’s for precisely the parameters listed.

Den omtalte IDL kode:
; IDL Version 3.6 (vms alpha)
; Journal File for PALADN::SSHORE
; Working directory: DKB500:[SSHORE.GHRSIDL]
; Date: Sun Jan 20 23:41:04 2002

plot,x,b*wt,psym=10,charsiz=1.5
!xtitle=’!4D!8v!3 (km/s)!3 (MBM40 !U12!D!NCO)
plot,x,b*wt,psym=10,charsiz=1.5
set_plot,’ps
plot,x,b*wt,psym=10,charsiz=1.5
device,/close
$laser idl.ps
$ftp avatar.kennesaw.edu
$laser idl.ps
$rename idl.ps mbm40-pdf-apj.ps
diff= small(10:109,10:135)-ss(10:109,10:135)
x=findgen(101)*0.01-0.5
d=diff
i=sort(d)
bins,d(i),d(i)*0+1,x,x*0+0.01,b,ws,wt

Q: The bipolar shape of the ejecta in your model looks very much what I/we would
expect from looking/observing planetary nebulas. When we started discussing details, we
didn’t have any ideas about how/why this shape occurs.

Eksempel på formen:

V445 Puppis

A: As William of Baskerville says in “The Name of the Rose”, “If I knew the answer
to that, Adso, I’d be teaching theology in Paris”. I don’t know. This is something that
has been a deep concern of mine now for about two years; I have some ideas but they’re
still to be worked through and remain conjectures (related to material transport to the
WD poles). It’s not a magnetic effect nor is it likely (for this case) to be the disk
alone. But it’s characteristic of virtually all nova ejecta, although some are more
spherical than others.

The planetary nebula case (ikke det samme som Nova ejecta) is more varied and nuanced.
Here we’re seeing an explosion, there it’s a wind that has timescales far longer and can
be continually structured in outflow (e.g. planetaries, LBVs, AGB stars).

Q: When reading your information about the WD and how the material transfers from
the accretion star to a spot on the surface on the WD and then ignite, we would expect an
explosion/burn on one side of the WD and not bipolar or spherical.

A: This is a beautiful point. Yes, you would expect that IF the accretion weren’t
from a disk. That’s why this is a different case. We don’t know the ignition conditions
well enough, not how the flame proceeds in a WD that’s differentially rotating.

Q: So why not just an explosion on the surface that ejects material in one
direction away from the WD as the WD rotates? Why sould we also look for a spherical
shape?

A: The layer is clearly mixed before the explosion, but premixing is a diverse
process. Some models use diffusion, slow and steady, others invoke shear mixing and
turbulence, it also depends how deep the mixing and spin-up go. The models we have done
to date in 2D and 3D are not there yet.

Q: Can you further comment on this from your paper?:
The catch here is an implicit assumption regarding the geometry of the ejecta, that the
central source is completely covered not relative to the observer but in the expanding
frame.

A: It’s that the ejecta are one dimensional in any sight line. The observer may
see the ejecta to the side but the material in the outer ejecta “looking toward” the
central object sees all of the garbage intervening along that line of sight. Think of
being in a fog. Looking up it’s clear, looking ahead it’s opaque. The ejecta look ahead.

Q: I’m I correct in assuming that this work will give us better “standard candles”
that will change our assumptions on distances to these objects?

A: Yes, at least I hope so, but that’s not likely to be true for the CO novae,
they’re likely from a broader range of WD masses than the ONe. This is only a sort of
secondary standard candle but it’s important for understanding the gamma-ray production
mechanisms and the evolution of the WD in the post-ejection stage.

I hope this helps!!

steve

Nightsky2013-09-08 22:04:08

[nightskyDeltager Neutron star]

Selv tak Frank.

Desværre kommer jeg ikke – en eller olding runder et halvt århundrede om fredagen.

[Frank LarsenModerator Super Nova]

Lars,

Spændende.

Tak for opsangen – min chef og min datter er desværre ikke enig i at jeg skal droppe søvnen ;(

MEN jeg forventer at du sidder klistret fast på min højre hånd fredag og lørdag nat i weekenden.
Jeg har en og KUN en dagsorden her til MAF starparty. Spektroskopi og fotometri på NovaDEL 2013.
Du er hermed udvalgt til at hjælpe mig.

Jeg har brugt de sidste mange aftener til at sikre mig at udstyret spiller. Optagelsen i sidste uge var hastet igennem efter at spektrografen har ligget adskilt.
Håber jeg kan få hentet LXD75 hos Lars M i morgen så jeg kan få sat to kikkerter op på den inden afrejse – ville helst have en EQ6 som tracker noget bedre.
Skal også liige have renset akromaten som bliver hoved instrument.

[nightskyDeltager Neutron star]

Steve har kigget på data fra NOT teleskopet og her er nogle spektra. Jeg har spurgt lidt
ind til hvorledes man kunne sige noget om hvordan materialet som er kastet ud (ejecta) er formet.
Kugleformet, ej kugleformet osv.

Han har kørt en model igennem computeren og fået resultatet nedenfor frem og det sammenholder
han så med ee faktiske data fra NOT teleskopet. I mine “amatør” øjne ser det jo ud som om ?????????

Modellen

Sammenligning med data fra NOT d. 29. aug.

Jeg har også rodet noget med kalibrering af data, og som eksempel har jeg dette fra NOT

…the first shows the reduction of the standard star, the second is the application
of the calibration to Nova Del 2013.

Lige nu hænger jeg stadig på med det alleryderste af neglene og tror jeg forstår hvad der skrives.
Men for den da jeg savner en astronomiforening hvor man bl.a. kunne tage lidt tungere emner op.

************************************
Se lige denne bemærkning, simpelthen imponerende:

Steve har han kigget på de data amatør astronomerne har samlet, specielt ARAS gruppen.

“I just did my first pass calibration and I’m checking those in the
database. This is really a spectacular pass ahead for the
group, something that will revolutionize the whole damned field!”

Amatør spektroskopi har virkelig en fremtid og kan bidrage signifikant med ny viden.
De data som er samlet ind matcher fuldt ud professionelle data eller mere.

En skam de super gode spektrografer som jeg ved enkelte danske amatører har IKKE bliver brugt.
Her kunne de virkelig bidrage med data.

Nightsky2013-09-03 01:24:06

[nightskyDeltager Neutron star]

Steve med det seneste omkring Nova Del 2013 udviklingen. En del om de linjer vi nu ser på
vore optagelser.

Endnu engang tak til Steve.

There’s been interest in some explanation of what developments are yet to come so here
are a few notes for the next week or so.

First, a word of advice. In thinking about what your spectra are telling you, it’s best
to “think like a photon”. By that I mean think about what a photon traversing a medium,
in this case the ejecta, will encounter and what will happen. In fact, this is the origin
of the Monte Carlo method, a technique for simulating the passage of a particle through a
very complex environment, subject to a wide range of processes and a wide range of
densities and states. You couldn’t find a better description for the ejecta. Recall
that the inner and outer parts, even were this a wind, have different outward velocities.
So a photon emitted in one place sees the rest of the surrounding gas moving — on
macroscopic scales — at different velocities and therefore differently Doppler shifted.
So if a photon is emitted in the outer parts, where the density is low, it most probably
escapes. If, instead, it’s emitted in the inner part, where the density is higher, it will
quite literally bounce around in both space and frequency (absorbed in a line center,
emitted in a line wing, encountering another atom in the line core, perhaps, and being
re-emitted there, etc). So in the initial stages, where the photons are actually from
the hot gas itself, the thinning of the outer regions is like the expansion of a wind and
the photosphere (an intrinsic one) moves inward. You see this in some of the film version
of the spectral sequences some of you have produced (especially for H-alpha). At first
the P Cyg absorption seems to move inward as the outer layers become optically thin, and
then the absorption disappears on that line (leaving a sort of dent) as even the
approaching material becomes transparent. The higher Balmer lines, on the other hand,
have a smaller emission/absorption ratio (the emission is formed further in) and the
absorption is progressively stronger. At the same time, you see with increasing clarity
and strength the structure of the whole ejecta, the various emission peaks, which signal
the thinning of the material at the highest distances and velocities.

But don’t forget the poor remaining white dwarf. It’s now in the supersoft phase,
although we don’t yet see that, burning the residual material from the explosion in a
source that reaches several 100,000’s K (of order 0.05-0.1 keV). The nuclear source is
deep, not at the surface, and has a photosphere of its own that depends on the newly
established structure of the envelope of the WD. This is inside the ejecta, at this
stage (as of 1 Sept) we don’t yet see that directly.

But we see another, important effect: the ionization produced by this source is gradually
advancing outward in the ejecta from its base as the ejecta thin and the photosphere
moves inward. This is the so-called “lifting of the Iron curtain” that’s happening in
the UV and the cause of the decline in the optical. Progressively more of the photons
can escape in the UV without being degraded through optical or IR transitions and the
continuum temperature increases as the two oppositely directed “fronts” approach. The
individual transitions from the ground state of neutral and low ions are in the UV and
some of them remain opaque although the continuum is increasing sufficiently to power
emission lines in the optical. Oxygen, in the form of O I, is the best example. The
[OI]6364 and 6300 lines are connected to the O I 1302, 1304 resonance lines. The latter
are still thick, so the photons knock around and finally emerge through “open channels”,
e.g. 8446 and the two forbidden lines. Their presence indicates the density is finally
low enough at the photospheric depth that the emission from forbidden lines is no longer
collisionally suppressed. The transition is abrupt in the optical, hence the term “flash”
used by the early observers, because when the right optical depth is hit, the transition
is almost instantaneous since the emission becomes local. The [O I] line widths, you
will have noticed, are lower than the wings of the Balmer lines so this is from the inner
parts. The O I 8446 was visible for a longer time. In the UV, we would see absorption
at O I 1302,1304 but that will gradually give way to P Cyg and then emission.

Something else to remember is that different elements ionize at different energies.
Oxygen, for instance, is slightly more bound than H, so the Balmer lines will be strong
when the O is still completely neutral. Once the O (and N) starts ionizing, they also
contribute recombination lines that can’t decay to the ground state directly because of
the blockage of the UV channels so they emerge where they can, at the exits marked “6300”
and “6364” and so on. The same for the C I and C II, and the N II lines. We are not yet
at the point where the N III 4640 lines appear but they will in due course.

The Fe II lines are now turning completely into emission as the peak moves toward Fe III
and higher and the UV lines turn transparent. The Fe-curtain will, once the ionization
reaches Fe^+3, disappear since that ion (Fe IV) has very few transitions in the part of
the spectrum where the UV is strongest. All of this is powering the decline of the light
curve and is what “the founders” didn’t suspect: the changes in the UV from the light
curve are timed to appearances of specific ions and transitions because the continuum
temperature continually changes, moving toward stronger UV and even XR, while the optical
is a passive responding medium. When the Lyman series turns transparent, and becomes
recombination dominated, the P Cyg profile disappears. The same for the He I lines, they
will reappear along with He II and other higher ions as the opacity in the UV drops.
Once the two fronts meet, that’s the nebular stage: the moment when the spectrum turns to
emission, we see completely through it, and the line profiles all look basically the
same. I say “basically” because density and structural differences leave their signature
on individual lines depending on their transition probabilities (forbidden or permitted,
as discussed a while back).

The nebular stage is a complicated period and very sensitive to the specifics of the
explosion. If the ejecta are spherical and smooth, all profiles will be basically the
same but differ in width because of their “weighted depth of line formation” (in other
words, recombination line strengths depend on density so the inner part always
contributes more, but it also depends on where in the ejecta a specific ion appears).
All of this changes quantitatively for nonspherical explosions, but not qualitatively.
The strength and velocities are those we see projected along a line of sight through the
expanding medium.

I apologize if this is staring to get heavy, it’s not intended. You have here a problem
of photons (motorcycles) weaving their way through traffic (cars, trucks) whose speeds
depend on where they are in the lane of traffic. If the ejecta are spherical the only
escape is along the direction of the flow. If aspherical, there’s a way out and free
escape by swerving to the side. This is something we’re just starting to deal with in
detail, and it’s your work that will illuminate it even more clearly for this
prototypical nova.

And as a last comment, one on the intensities/fluxes. In the next weeks, as the ejecta
change ionization and approach the sate of freeze-out (when the recombination’s are
independent of the WD illumination and depend only on
the rate of expansion), we will see
how structured the ejecta really are, the density and ionization stratification, and the
abundance inhomogenities. The absolute fluxes are the key, they tell you how much energy
is in each transition and therefore the number of radiating atoms. It seems, for
instance, that a few days ago H-alpha alone accounted for almost 8000 L_sun if the
distance is 5 kpc (less as1/D^2 depending on the distance). From this we’ll have a
first estimate of the ejecta mass, one of the key unknowns in any explosion and the
pointer to the conditions at the outburst. The other is that there is structure here in
the ejecta, you’ve already seen that in emission and absorption, and as different ions
appear that will link to the central engine.

steve

Nightsky2013-09-08 18:49:08

[nightskyDeltager Neutron star]

Mange tak Frank. Det vil være spændende hvis du får nogle spektroskopi data.

Vejret er jo fint derovre, klar himmel da jeg hoppede flyveren kl. 20.00

[Frank LarsenModerator Super Nova]

Flotte optagelser Lars!! Håber vejret holder så jeg kan få taget et spektre i aften. var forhindret i går.

[nightskyDeltager Neutron star]

Opdatering med egne observationer.

Styrken stiger stadig på Balmer linjerne. F.eks. er der en tydelig stigning fra d. 25 til 26. aug.

Sammenligning over den fulde flux fra natten mellem 14/15 til d. 26. aug. Første gang jeg nogen
sinde har fået så komplet et sæt data.

Nedenfor er skalaen ændret så vi bedre ser ændringerne. Det er ganske tydeligt at udover stigningen
i brint emission, kommer der også masser af absorptionslinjer. Denne fase med nye absorptionslinjer,
betegnes med “Iron Curtain”

På den sidste nedenfor ser vi også fint at der kommer nogle andre emissionslinjer.

I morgen kan jeg have kigget på nogle dybe optagelser af kontinuum.

Nightsky2013-08-28 21:36:21

[nightskyDeltager Neutron star]

Steven Shore har sendt endnu en artikel til os. Den forklarer lidt om hvad hvorfor eksplosion sker,
og hvad der vi ser nu.

*******************

Even at low resolution, many of you have caught the sight of narrow absorption features
at high velocity, that look like P Cygni profiles, on the metallic lines and also on He I.

As I’d mentioned earlier, the explosion is initiated on the white dwarf by the pile-up of
garbage from the companion, like a bad landfill that ignites. In fact, a silo explosion
(a grain storage facility) isn’t very different; the matter is compressed and heated to
the point where a chemical reaction starts that is fueled by the combustible material.
The thermonuclear reactions, mainly involving CNO processed by protons (hydrogen) from
the accreted material, triggers a mixing process at the interface between the accumulated
layer and the envelope of the white dwarf.

This is the part we can — so far — only model. The signature of the process should
remain in the explosion since the transformation from a flame to a shock is very fast and
unstable, leaving behind matter yet unburned and throwing off the outermost layers at
supersonic speeds. Because the expansion is above the speed of sound, pressure is
irrelevant for the structures that might be imposed and they remain preserved in the
flow. In fact, you’ve seen something analogous to this in everyday life. (the lovely
thing about hydrodynamics is that you can actually, physically, compare flows of very
different kinds when the processes are otherwise the same, a similarity notion). If
you’ve ever seen a waterfall or cascade, this will be familiar. Until the edge, the water
is flowing slower than the speed of a gravity waver (in other words, a water wave). But
at the edge it falls and decouples from any excitation, it’s in freefall and the bits
that start at a higher speed arrive ahead of those that were nearer rest at the start.
But the sheet of water preserves all of the structure imposed at the last point of
contact before the edge, the filaments and knots you follow downward that give a sense of
the speed of the fall. That’s what we see in the ejecta and that’s why these discrete
features, those now appearing, are tremendously important.

In the photographic era such lines were noted as “absorption systems” that appeared at
different stages of the light curve on the metallic lines. These were difficult to track,
often overlapping and highly subjective since the spectra were often poorly calibrated or
not at all, and the zero levels were poorly defined. All of the observers before the ’70s
clearly knew this but some were amazingly skilled at recognizing the different absorption
systems (and these were likely real, the most careful could distinguish multiple
components reliably like McLaughlin — who should be one of your heroes – and Payne-
Gaposchkin).

On the Fe-group lines, and the Balmer and neutral helium lines, these also arise from the
complex interconnections between transitions I’d mentioned earlier for the optically
thick stage. BUT the Na I lines — the D feature — is essentially different. It’s one of
very few ground state (resonance) transitions in the whole optical spectrum that isn’t a
forbidden transition (intrinsically very weak). In fact, this is one of the strongest
lines in the spectrum and also neutral. The Ca II H and K doublet is another but it’s an
ion. The K I line is in a terrible part of the spectrum, often (in many of your spectra,
for instance, inevitably!) hidden under a curtain of atmospheric water and molecular
lines and hence unusable. The Na I line is, instead, the unique tracer of the neutral
medium and the features that have now appeared on the D1, D2 components (together) are at
an intermediate velocity even with respect to the Fe II and Balmer lines. In other novae,
especially the work we’ve just finished on T Pyx (an old friend of some of you) the
velocities are intermediate but the same as we see in the later stages, more than a year
later the same feature is still showing up in other ions. This means the structures, the
density enhancements i the ejecta, are actually not moving with respect to the other gas
in velocity and expanding as a frozen-in feature, just like the waterfall. The striking
thing is that the velocities are intermediate, not the innermost of the ejecta and far
lower than the outermost (in other words, these are sort of imbedded in the ejecta and
“persistent”). Since the expansion is supersonic, they don’t “grow” spontaneously within
the ejecta – they have to be imposed on the expelled matter at the time of ejection.

Nova T Pyx

This point back to the explosion site itself, buried at the start under the mass of the
accreted layer. In T Pyx the broader narrow features (what a description, no?) dissolve
into an ensemble of filaments of widths no more than 10’s of km/s within a broader
envelope of a few hundred km/s but still far lower than the several thousand km/s of the expansion.

That these are seen in a certain stage is the result, it seems, of a recombination wave
I’d discussed earlier. But the most important feature is that being resonance lines from
a neutral species, these features trace the progress of the recombination better than any
metallic or Balmer lines. Now, in the last spectra sent by Christian Buil, you see the
two Na I feature but, if you displace to the first spectrum and use He I 876 (that then
disappeared after Aug. 15) you’ll see that He I also now has a detached feature. These
absorb at a specific position (radial position) in the ejecta and they have to be large
or we wouldn’t see the absorption.

In V705 Cas 1994 they formed as soon as the Na I emission peak strengthened. The same in
T Pyx. For V959 Mon we don’t know because it was hidden, and few other novae have been
caught at high time coverage (and also higher resolution, R > 1000) to make the evolution
clear. And taking the Ca II to Na I ratio at each component is a direct measure of the
ionization fraction (not just abundance since the Ca/Na elemental ratio doesn’t change
while Ca II/Na I will. As the wave progressed you will see different features appear on
different lines but always within the same intervals.

Now a quick word for the moment about CN and why this is so important.

One paper (!!) by Wilson and Merrill

http://adsabs.harvard.edu/abs/1935PASP…47…53W

reported this line and only in DQ Her. But they also discussed the NA I in another paper
and Payne-Gaposchkin discussed this also. The molecule, CN, is amazingly stable for a radical
(no, not a political comment). It has a high dissociation energy and can remain in stellar
atmospheres to hotter values than the Sun (> 5800 K). The same for CH and CO but we don’t
see those in the optical; they’ve been detected in the IR. The usual molecule is CO that
consumes almost all of the C w=o if that channel is saturated it means the C/O ratio is
high enough for other organics and hydrocarbons to form. The others, often quite complex,
are seen in winds from highly evolved stars. And the higher the C abundance the more is
available from which the solid phase — dust — can condense. Any isotopic anomalies
remaining from the nuclear burning will also remain locked in the dust so after a while
drifting through the Galaxy (shades of the Hitchhiker’s Guide, no?) they can be
incorporated through passage in a molecular cloud, into a star. The dust forms in a way
we don’t well understood but it is likely that molecular formation and grow
th is a signal
of the right environment for the appearance of grains. This may be purely chemical,
homogeneous condensation or nucleation”, or it may be induced (sorry, some of my own
work) but whatever the mechanism, it happens.

Therefore we can witness the dust formation process in a well constrained event and —
Holy Grail though it is — figure out what triggers the dust formation. Other molecules
have been detected in the IR, CO for example, but nothing from the cold matter in the
ground state.

In Nova del 2013, it seems that the CN has not appeared but it may yet and there’s every
reason to continue at all resolutions.

I’ll repeat, ad nauseum, that what you are producing is a legacy of depth and range
we’ve never had before. PLEASE don’t get discouraged or tired. I’m writing this as the
dawn breaks over Pisa thinking of all of you and I promise to explain more (as I’d
previously promised) about the origin of the binaries in this state.

Spektroskopi amatørerne bl.a. i på ARAS kan vist ikke få større anerkendelse.

http://adsabs.harvard.edu/abs/1935PASP…47…53W

Nightsky2013-08-26 18:17:02

[ulrik Planet]

Det er forstået Lars !

Ellers god observationslyst !

Ulrik