TeamSciMeetingMar09
Back to current versionRestore this version

EIS Team Science Meeting Agenda#

18th March

This page refers to one day of a closed meeting of the EIS consortium


1 Technical#

1.1 Report from Technical Splinter Meeting (previous day)#

Khalid reported on the instrument status, w.r.t. PZT non-linearity.

Also discussed the statistics of observations carried out to date.

Remote operations initial testing and model also reported on.

Another instrumental issue is the optimisation of the focus within the instrument. Harry Warren was called on to talk about the slit and grating focus. Google was apparently able to show that the grating focus happened in Aug 2008.

KFJ: Just to clarify there was an offset between the two channels, and now they appear simultaneously on the two detectors. HPW: True. We haven't quite seen dynamic events GAD: What about images on the same detector. Wasn't there a slight offset between Fe VIII and Fe XII. HPW: in my analysis, the offset is pretty well taken care of with PRY's formula. Extremely well correlated once you account for that.

Linear trend in the increase of the number of hot pixels on the detector with time. These are in the CCD detectors, approximately 1% of pixels. Sounds like a huge fraction of the CCD. What's the best way to treat warm pixels for scientific analysis. My thought from the start of the mission was to ignore them, and neglect them in the analysis. But HPW and PRY talked and found that interpolating actually gave good results. Wrote a document to be posted on this Wiki after the meeting presenting some of the results. Used the standard EIS_PREP processing. Then artificially inserted 30% bad pixels. So how badly were the fits degraded by ignoring these fake "bad pixels". HEM: Interpolatin after fitting? PRY: This is done by EIS_PREP. And is done in the solar_Y direction KPD: How do you know they're better fits? PRY: I have the original data, and identify the places where there never were bad pixels. Then I process these data by inserting fake bad pixels, interpolate, and then compare the line fit parameters. KPD: I think you're inventing data... PRY: That was my worry, too, but it seems to work! HEM: What about clusters of missing pixels? PRY: There are different methods of tackling these.

You could reproduce the original data 97%, but the bad interpolations tend to happen in the high-intensity areas. My recommendation is to use interpolated data rather than just ignoring the missing pixels. I was amazed how well you could do even with 30% warm pixels, but it seems we can continue longer than expected by a couple, even if we don't remove them.

It may be related to EIS's oversampling w.r.t. its resolution, so there is correlation of data which allows some justification of the interpolation: there's a contribution from neighbouring pixels anyway.

We use flat-field LEDs inside the instrument, and we also measure the solar radiation variation over time. We don't seem to be seeing substantial degradation in our sensitivity.

Contamination is also measured with Quartz Crystal Mircobalance sensors, which show no appreciable increase in contamination so far.

PRY: how about the event flags? Are they implemented? KFJ: Implemented in GSW, but not tested yet. They need testing on the Sun. BP trigger could be tested soon. PRY: Is there a problem with remote planning? We'll be more cautious about doing that? KFJ: If we sit together remotely, it may be easier.

JTM: are there problems with Data Volume DRW: It may be easier in the current system.

JLC: How about use of the XRT flare flag? Do we have plans to use it? HH: it is posible for us to use it HPW: He II may be better than XRT

HEM: What does the flag give us? Is it coordinates? HH: Yes, coordinates.

Discussion followed about how to test this flag.

1.2 Analysis Software#

1.2.1 Status of the reduction package#

AG went through the ordering of calibration steps.
  1. Dark current subtraction and options for this
  2. Marking hot pixels affected by CRs
  3. Hot pixel flagging
  4. Warm pixel flagging
  5. Dusty pixel flagging
  6. Absolute calibration (specific intensity units; photons)

The 1st step returns missing pixels, but they aren't filled in by other values. To do this, you have to allow the CR option (step 2?).

PRY: If I don't use /DEFAULT what happens? AG: If you don't select default, it should ask you to choose the DC file.

Computation of moments is available in XFILES -> XCONTROL -> Line Fit Can't construct moments in anything other than absolute units. Also doesn't work with negative values retained. Can then, in XFILES, use moments OR the Gaussian fitting routine supplied by PRY.

HEM: This is assuming no blends and a single line per window? AG: When you select a line, you have to then specify the line and continuum interactively. If you have two lines very close to each other, you can choose only a few pixels plus continuum to choose just that line. But if there is a blend then it won't work.

After the line and contin. are selected, the 0th, 1st, and 2nd order moments are computed. The last term 2*ln(2) is necessary to turn sigma into the FWHM. That's it.

XFILES also gives you the errors on these moments for intensity and velocity. The error on the line width hasn't been implemented yet.

HEM: is there documentation for this yet? What you said is really useful, but is there a written version of it.

Brief discussion followed about the desire for tutorials and guides.

1.2.2 Useful additional software#

(With descriptions or demonstrations)

EIS_AUTO_FIT and EIS_AUTO_FIT_GEN (newer) both available.

When you want to fit multiple gaussians the latter is better (?)

You might want to add many spatial pixels together to make a total or mean spectrum.

You can select sub-regions of the array.

eis_auto_fit, windata, fitdata, refwvl=195.12

  1. First choose the FITS file I want to look at.
  2. I then choose a pre-calibrated file
  3. extract the window data with EIS_GET_WINDATA
  4. choose one at 264 Å
  5. eis_mk_fit_template averages the spectrum over space
  6. go to the window, select the lines roughly and the continuum
  7. wanted to speed up the fitting for this talk, so...
  8. have a routine that takes the windata and rebins it by some factor which gives you that factor^2 less pixels to fit
  9. then read fit template eis_read_fit_template
  10. then start the fitting process, which takes a while, meantime...

The fit template stores the initial guesses as a text file which you can modify if you want to.

GAD: How accurate does it have to be? PRY: Not too accurate. In this case the lines are quite widely spaced, GAD: But how about if they're close together PB: in the case where there is a blend, then in some cases I found I had to restrict the parameters, e.g. fixing the widths to be the same PRY: You could try the lines having a narrow range of width? PB: what I did was to force the widths to be equal, which meant I had to change the code a bit PRY: sometimes you have to write a more specialised routine, because it's difficult to write something completely general? YKK: Is the continuum fixed as linear? PRY: yes, linear or constant

okay, the fit has finished:

  1. when yo do the orbital correction, have to specify a line -- in this case line 1.
  2. you can then view the quality of the fit with FIT_VIEWER
  3. see the I, V and dl maps. Click on a position and you see the spectrum AND fit.

It's available in SSW, and there's a tutorial on the web to show you how to use it.

That was the Gaussian fitting.

I also mentioned that sometimes you see a BP, for example, and you want to add all the pixels' spectra together. Arbitrary spatial pixels. It's complicated by the grating tilt and inter-channel offset, but that can be taken care of with this software.

I'll use the same dataset as before. Have a simple routine that makes an image in the specified line. eis_make_image

eis_pixel_mask to make the mask...

how is the pixel mask stored?

This performs the averaging and gives you a Channel A (long) and Channel B (short) spectrum.

In spec_gauss you can see the data quality with the yellow line (something to do with number of missing pixels?)

Just wanted to show the Gaussian fitting routines I wrote.

uses EIS_CCD_OFFSET for each wavelength in the spectrum to compensate for intra-channel vertical offsets.

Coffee#

10:40 to 11:00

1.3 Choosing data#

There are huge volumes of EIS data, but it would be good to have a discussion on how to go about accessing the kind of data you're looking for.

It seems like it's possible to cross-match the EIS and RHESSI catalogues already. RM has done this. And HPW will talk to SDO AIA people to discuss their requirements for integrating our information with the Heliospheric knowledge base.

Ultimately, though, it might be good to set up ways to make event catalogues; it could be possible for people to construct and contribute custom cataloguing/data-mining routines.

Seems like we have issues more with the Wiki than with Data Access per se, though.


2 Science#

2.1 EIS Science Achievements#

Each national PI will give their perspective on the scientific achievements of EIS in its first 2.5 years. These are personal perspectives on where we stand on two things:
  1. phenomena and scientific issues that were known before launch
  2. the discoveries Hinode has made with EIS
( they won't necessarily agree :-) )

11:30 Tetsuya Watanabe#

Magnetic reconnection, if it at all takes place for coronal heating, is far below the height of the reconnection point that causes major flares. Thanks to the high resolution of the EIS grating, we detected a huge number of Fe lines with various ionisation species, much less blended than in spectra taken ever before. E.g, Fe XIII in the short wavelength band. TW showed the EIS spectrum versus a lab spectrum taken by the National Institute for Fusion Sciences, but the latter's spectrograph isn't as good as an orbital spectrograph(!). Due to this high spectral resolution, EIS spectra are used to check data for the high-density limit.

TJ Wang commented that the Sakao flows may be waves already seen with TRACE.

11:50 Louise Harra#

Transient phenomena

So we

Last goal is energy transfer from the photosphere to the corona.

Things we need to do:

  1. Find out what lies above the kG fields in CHs
  2. Find reconnection inflow
  3. Relate different oscillations to the magnetic environment
  4. Link outflows in ARs to solar wind measured by STEREO/HI and ACE
  5. Find other examples of shock waves and understand their relation with temperature

Lunch#

12:30 to 13:30

12:10 George Doschek#

Took a slightly different approach than Tetsuya and Louise. I talked to NASA about giving press releases, and they want to know what major problems you've solved!

One of the probs at NRL we've been working on is AR loops. If we could say how loops are really heated, we could go far towards solving the coronal heating problem, and that's a press release that NASA would really be interested in. We're in the "middle" of modelling them. The real important thing we have is the density, so we can check models of heating and cooling of loops. Most of these loops are not in equilibrium. By measuring the densities in the cooling, we can really check models. Seems now that the most common model for things like Fe XII is that it's a loop around 1.5 MK. The kind of theory that best explains the observations is a loop made up of strands that heat and cool sequentially in some sense explains why they last longer than the cooling time. You can also measure path lengths when you have the densities, so you can tell if you have unresolved structures.

Warren et al. 2008 Apj 686 L131 The loops seem to cool together, so the DEM doesn't get very wide.

Loop filling factors are about 10%. The densities from Fe XII are about 0.5 dex too high versus Fe XIII. So it's critical to do the atomic physics for these ions better, and to check them against densities from other ions, like silicon. So we get these tiny fillin factors, but they seem to be doubtful. IN summary, though, we have tremendous progress, but we're left with some big questions and need to improve the atomic physics.

IUU made a slot flare movie: loop models; it would be good to make serious progress by the time of the senior review.

We observed Ca XVII and we can now separate it out from the Fe XI and O V blends. It can now be used by people much more reliably if you know what you're looking for. YKK has developed a procedure for unfolding these that soon non-spectroscopists can use to use this 5MK line, so it extends the DEM temperature range posible from EIS. We didn't appreciate this blend when we picked this line, but glad we picked it; we didn't appreciate the density sensitivity either. Didn't appreciate the fact that lower TR lines like O V and VI are reasonably strong in our data, because Skylab only saw O V lines in flares: change of principle quantum number, so it takes a lot of energy. Those things have been a surprise, but making progress.

Here's another example of unfolding the Ca XVII emission. Can extend our analysis of ARs to high T.

Louise and Tetsuya mentioned flows, and those come from regions that are normally dark. The Fe XII intensity shows wide, shifted lines in dark areas. XRT also seeing this sort of thing, but you don't know if you're looking at a flow or a wave. John and Harry thought they were waves, but they turn out to be flows. They're over extended regions. You also see them in these broad regions. When I sent a paper in on this, I think it was reviewed by people who worked in the heliosphere. e.g. Liewer, Neugebauer & Zurbuchen 2004 (SoPh 223, 209) who expect flows at the edges of active regions. Haven't been able to link the flows to colder material. They occur where the field is more or less monopolar, but don't know if these field lines are really going into the heliosphere or just form long loops.

They're also a f(T) and many active regions have a blue-shifted flow component. You therefore get a correlation between the flow velocity and the width.

John Mariska has found waves in coronal loops. These waves present a problem, becuase they don't fit a strand model: so how can you see a coherent wave feature across the whole loop?

Also in ARs, if you look at SUMER data in the corona and make a Jordan loci fit, you get a single temperature of 1.4 MK. But when you go on the disk, you don't see that at all. need to work on this to try to resolve it.

Quiet Sun has been really quiet. KPD finds <<1.0 filling factors in bright points. Really important ot understand what the densities really are.

We've been lucky to see flares. We've only seen one limb event (20061217). We'd like to see inflows. We really need to see a bunch of M flares. Observe the sites of evaporation and footpoints and see if they correspond to the general standard models that people have for flares.

When we see the footpoints, we can actually see evaporation.

We can probably do a lot more on polar jets; many more observations would probably be very useful to have as we're the only ones who can observe their properties.

Finally, EIS-SUMER campaigns: we wanted to calibrate EIS with SUMER if we could. With EIS and SUMER you get an isothermal corona. If you apply the calibration, they agree within instrumental errors. But the atomic dat for EIS are a little more messy. We have very little checks on ionisation balance. We measure temperatures well above the ionisation equilibrium temperature. The longer we can't figure it out, the more interesting it becomes. Perhaps non-thermal particles penetrate into the TR? That would be extremely interesting. Enrico Landi put recombination in, even though they're Li-like, and so simple. So we can make more measurements like this.

Finally (finally), something we can do is measurements with STEREO. Aschwanden is doing a lot with STEREO, but hasn't yet got involved in EIS.

A summary of what we ought to be doing.

There are several kinds of problems we can attack.

KPD: I think the flows were hinted at with Skylab GAD: True, but the new thing is that they're measurable, although it's difficult to say that we've seen antyyhing new so far.

PRY: is the density in the outflows different? PB: haven't done that yet

JTM: just a comment on all three talks: hadn't appreciated how much we've done. But we've never fulfilled yet the promise of integrating with the other instruments, and I suspect the real discoveries are going to be from doing that. I hope that when we come back next year, we'll have made a start.

HEM: any possibility of having any workshops? JTM: This could be a good time to start this.

GAD: coalingnments are very challenging for co-observing JTM: but it can be addressed in some meaningful way.

2.2 EIS Science Results#

Speakers are encouraged to put special emphasis on unsolved components in the context of what they talk about, and how they can be addressed using all three instruments on Hinode.

2.2.1 Active Regions#

Small active region close to the limb Can take the n_e ratio as we go from the disc to off-limb. We can also attempt to get the temperature in a simple way. Never subscribed to this before, but I'm showing a plot! We want to see how hot the temperature is. But if you take Fe XVI / Fe XV, you can see that the core is very hot around log(T) 6.7. We've done the same for Fe XIV / Fe XIII, we get a lot more, interesting structure, which we need to follow up and invesigate. We want to look for these higher temperature lines, but must be really careful about removing blends. The Fe XVII line is a good line to use.

When looking at Ca XVII, you can't use the CHIANTI Fe XI ratio, as it's wrong. But you can do it empirically by finding the ratio on the quiet Sun and then using that to remove the blend with Ca XVII.

Then we compared the XRT and EIS data and wanted to cross-calibrate the DEMs -- but they look nothing like each other! So we can't yet use EIS and XRT to have large-T coverage DEMs that we believe.

GAD: those limb-brightening curves about the limb: do you have O IV and O V? Cos that's a check on the scattered light? HEM: they do fall off quite sharply

PRY: For the AR, there was a paper last year that showed three separate isothermal components in the corona. HEM: that's difficult for me to believe.

Let's look at some CDS data taken above the limb, and it looks consistent with an isothermal plasma. Fairly well-established result. Now let's look with EIS. One of the things we found with SUMER is that the Si data seems to behave pretty well. Assume a delta-function EM and formalise the fit parameters. Get a pretty good fit. The density is being adjusted with intensity. The encouraging thing with EIS is that we get a similar result with the big strong lines that everyone includes in their studies. one thing that is unsettling is that the magnitude of the EM is different! This could mean Si / Fe ratio in the abundances of Feldman is wrong.

That's what can be done.

When you bring in Fe XIV - Fe XVI, they don't lie on the curves. So is there some dispersion in the temperature distribution? Then we end up with a fairly narrow distribution, but it doesn't account for the higher temperatures.

So another thing we've considered is MCMC used in PINTofALE. We also see a tail extending down to higher T. This accounts very well, but although tempting, it's not clear, because the ionisation fractions become very unclear as you get away from their peaks. Or maybe we really do have a tail in the DEM!

Ca XIV thru XVII help us to probe the high-T lines. Their contribution functions are narrower than the equivelant-T Fe species. Ca XIV and XV lines are fairly strong, and we should reconsider AR observations so that they include these lines.

GAD showed earlier an AR movie: I wanted to comment that the we show the Ko et al. deconvolution of Ca XVII.

As we move out from the core of the AR, we see the high-temperature emission coming down. Just because you dno't see any signal doesn't mean there's something wrong. Can still offer useful constraint for analysis. Something interesting is that the EM

Modelling evolving coronal loops with narrow-angle STEREO. Aschwanden has put a package on SSW for measuring loops' 3D geometry.

We're doing mutli-thread modelling of loops: we can calculate the expected lifetime of the loop, but it lasts much longer. We've put to gether such a multi-thread model. Good matches to TRACE loop lifetimes. Another independent check is the DEM. Can approximate observed lifetime and DEM.

The way we motivate the heating in a loop, the higher the density, the more energy you've to put in.

We try to reproduce the XRT loops, and can't find those loops. May be that we can't find loops in XRT due to the morphology.

Would love to see the EIS slit over a post-flare loop arcade.

An important goal for the future. So, to summarise:

HEM: I'm still confused by the isothermality... the heating models might predict a high-temperature tail. HPW: There's no physical interpretation, that's what's so difficult to understand. Essentially indpendent of height. Maybe there is this hi-T component. HEM: just not high-enough T lines? HPW: We have Fe XV. There are always probs with atomic data, like Li-like lines... KPD: One of the Fe XII lines could be being misidentified as Fe XI HPW: It's hard to see how, because the ratio is extremely constant. PRY: I remember fitting the lines, and the ratio was very flat.

GAD: Helen, you didn't get this isothermal plasma, but was it in an AR? HPW: I thought it was in an AR?

Discussion continued....!

Seems like almost everyone has spoken about outflows!

Seems a correlation between outflow speed and line widths. Can these widths be attributed to multiple components with different speeds? One such AR, we have observations as the AR rotates from east to west across the disc. The dominant outflow region seems to change position, but this could be a geometric effect.

Fit Fe XII 195, Fe XII 192, Fe XIII 202. Have used one of PRY's codes to fit a double Gaussian. Correlation between width and velocity remains, although it's not quite so clear for the smaller component at larger blue-shifts. These data are taken from each pixel in a box around the outflow region. Only done for the one active region, because the Gaussians were well separated.

LKH: does it change with time in the series of rasters you showed. PB: not by much

KPD: The interesting thing is in the change of one flow region from redshifts to blueshifts.

In the core of the AR, the blend to the red side of the line becomes more important at higher densities.

GAD: how did you determine your zero velocity? PB: Took a horizontal slice along the bottom

JTM: based on muliple simulations of loop flow problems, you have to be careful when fitting that if you choose two lumps of plasma: one moving and another static. I think Ken had the right idea when he used moments because you could use things like skew. that might be what this is telling you.

You can model an active region from the potential field. For SXT or XRT it works quite well hydrostatically. But in TRACE, all the emission is concentrated in the footpoints, but the model shows that there aren't loops. The intensities in the moss match, though, even if the warm, over-dense loops don't.

Antiochos & co. suggest that the moss emission in large active regions is consistent with steady heating. One thing I wanted to do is to look at dynamic active regions to see if there are any signatures of dynamics in the moss?

Use EIS at high-cadence to look at variation in the moss AR 10960. There were 15 M-class flares, 27 C-class flares. So the region is very flare-productive. I've made a movie of the moss region. It's not like the flares are happening away from the moss, they're happening close to or in the core of the region. But the pattern of the moss is very stable. De Pontieu et al. would argue that any variation in the moss is spicules in front of them. We used a study designed with 1-second exposures for a 75-second cadence. Have looked at the properties in little boxes in moss regions.

See that Vnt is very stable, as is the Doppler velocity and intensity. The flare doesn't really disturb the moss.

PRY: what's the cadence again? DHB: 75 seconds HEM: do you have enough counts to pick out changes in the moss? DHB: you do, yes.

Intensities are consistent with a steady or pseudo-steady heating, so the question is, is it showing contribution from steady heating?

HH: you mean the fluctuation of non-thermal velocity is very large? I would like to show a different result.

Hara: (Small-scale motions at footpoints: signature of coronal heating site) See large fluctuations in the line width map, where it is dark in TRACE. This region is quiescent, so there is no flaring happening at all.

2.2.2 Flares#

There have actually been a couple of flares observed with both RHESSI and EIS. Primarily interested in evaporation.

two scenarios for two diff't electron beam strengths:

  1. small blueshifts with gentle evaporations
  2. chromospheric recoil with explosive evaporation caused by much stronger electron beams.

EIS observations, fast rasters. Part of the loop has only filled at the Fe XXIII and XXIV temperatures. Only above 2MK do we see the blueshifted material. This shift increases with temperature. Fe XXIII may have an Ar blend, but the Fe XXIV seems okay.

Each line profile is symmetric and shifted, except for Fe XXIII and XXIV which still seem to have a dominant stationary component, plus a strong blue-shifted component.

Think there's a small loop with a much larger overlying loop. This paper was submitted about a month ago, and I've just focused on the impulsive phase and the Doppler velocities.

Was interested in looking at flow v as a function of temperature. This is a considerable progression from CDS. First, there's a linear dependence on temperature, and there is a very narrow-T divide between the blue and red-shifted components. The red-shifts are seen up to 2MK, much higher than before. Wei Liu took the Fisher model of a single burst of non-thermal electrons, where as Wei looked at more continuous dumping of energy, and also predicts high-temperature downflows.

In a previous micro-flare observation (B flare) published last year, we also see a high-T downflow. So this high-temperature downflow may be a regular feature in flares; more observations may help us answer this, or whether there's some spectral index dependence.

HEM: can you do densities? wouldn't that be helpful? RM: Have done the analysis but not the interpretation.

PRY: is the break to do with the detector you're looking at? RM: don't think so... but it's all in the paper.

2.3 Where to next for EIS science?#

14:45 to 15:15 Not only having listened to the contributions given this afternoon, but also having considered our progress in achieving EIS's main science goals — both pre- and post-launch — it would be good to have a discussion before and after coffee on what burning questions we think are obviously being left unanswered.

I'd like to budget 30 minutes for this.

Afternoon Tea#

15:15

2.4 Analysis Techniques#

The chances are in the last few years that if you've observed Fe XI 188.8, yo've probabluy also picked up 188.5 in the same window, so it's a little messy in that region of the spectrum.

If my ident of Fe IX 197 is correct, it's probably a good line for temperature diagnostics.

Can also "fix" the TRACE temperature response, so it's an important role for EIS to help better science from the imagers.

Based on the current ion-balance calculations, it puts Fe VII at logT 5.4. But it's clear that missed this
  1. 11 new line IDs made from EIS spectra (mainly in the LW band)
  2. 16 new or revised energy levels
Unfortunately no density sensitivity below log ne = 9.0

If you want to put Fe VII in your study, 196.22 is probably the best, and then 195.39 is a factor of two even stronger. Unfortunately, there are a lot of data which have problems and can be out by a factor of two between theory and observations.

Pretty fantastic that you can see 11 consecutive Fe ionisation stages from VII through XVII.

GAD: there's an Fe XXI and XXII line as well as Fe XXIII and XXIV

Also found there are Cr lines, too. Large EM between 5.5 and 5.8 enhances lines in this temperature range. have ID'd Cr VII and Cr VIII

Wanted to see what else we could get out of the slots. Compared consecutive slot and slit rasters: similar but different. Tried to reconstruct 40 gaussians from the slit profile. A LOT of free parameters. Each of these 40 gaussians has an amplitude, a width, a centroid, and a (constant) background. So I had to fix the parameters. I'm going to assume the same width because they all have the same temperature (it's just an assumption). Also assume that there are no Doppler shifts, that they're centred. So I just fit the amplitudes (justified on the basis of the consecutive raster).

So, we have the raster simulated from the slot. There are differences, but remember that the raster took time to make, and the differences could be of solar origin.

Using the slot plus TRACE and employing a MOSES-like technique. The velocities are very encouraging. HEM: so it's important that TRACE co-ordinates with EIS?

2.4 Observing Techniques & Strategies#

4 EIS Website & Wiki#

17:30 These are our public face for information. It would be good to talk about:

5 Action Items and Wrap-up#

17:45

6 終わり#

18:00 END


Confirmed Attendees#



Back to the EIS Team Meeting top page.