Establishing A Physics Identity

Front cover of my 400+ page thesis, and me (ca. 1981) recording magnetic classifications of sunspots, estimating their areas using a solar graticule.

In a recent issue of Physics Today (May, p. 47) we learn ('Developing Physics Identities') of the various tracks to establishing a "physics identity".  The one that most caught my attention was "When you have your own ideas and your own project. Basically you're working on something not having been spurred by someone else telling you to do it."

After passing two 3-hour physics qualifying exams for the M. Sc. degree,  I was asked by my adviser whether I needed "ideas for the thesis". I told him 'no, thanks', as I had already been working on a problem of my own for the past two years. This concerned how one might use morphological and magnetic features of sunspots to predict geo-effective events, specifically flares that incepted sudden ionospheric disturbances or SIDs.  Thus, my thesis was born, leading to no fewer than six published papers in peer-reviewed journals- twice as many as any other staff members in the time.

The title of the thesis became: 'Can Sunspot Morphology Be Used To Predict The Occurrence of SID Flares?'  and had been subsequently advanced for  Ph.D. submission at my advisor's request after the publication of the first 3 papers. Exemption from the oral exam followed but the conferral of the Ph..D.  was delayed by one outside examiner (from Sweden) to "enable further observational work via the Solar Optical Telescope" (Not the Hinode one!).

Alas, after 1 1/2 years the SOT was terminated in the NASA budget under GOP pressure - and with Reagan's military expenditures already reaching $1. 7 trillion, leaving the Ph.D. stillborn. See e.g.

http://brane-space.blogspot.com/2011/06/telescope-that-never-was.html

An M.Phil. (Master of Philosophy) higher research degree, next thing to a Doctor of Philosophy or Ph.D.,  was awarded instead.  This despite the fact my thesis had produced no fewer than five papers based on original work and insights. And at 401 pages (including 68 pages of Appendices) the thesis was more substantial than most doctoral dissertations - and with more published papers engendered.

The strategic basis and motivation for my work was originally published in a paper appearing in The Journal Of the Royal Astronomical Society of Canada, e.g.


http://adsabs.harvard.edu/full/1983JRASC..77..203S

The stand-alone authorship is especially important because it discloses the person's ability to write a full scientific paper on his own, and have it published professionally. This proof alone qualified me for Full Membership  (as opposed to 'Associate') in the American Astronomical Society, despite not having the Ph.D. (My 2011 monograph, 'Selected Analyses in Solar Flare Plasma Dynamics' is the equivalent of the doctoral dissertation that never was, i.e. had the thesis been so accepted without the need for the added SOT inputs).

Not long after writing the first four chapters of my thesis, they appeared in distilled form, in two lengthy papers published in Solar Physics, e.g.

http://adsabs.harvard.edu/full/1983SoPh...88..137A

And a somewhat shorter paper that followed, e.g

http://adsabs.harvard.edu/full/1984SoPh...92..259A

Given all the material in the papers was basically from my thesis, the only elements not prepared by me (in the 2 co-authored papers) were the assorted final (publication ready) graphs and line diagrams. These were prepared by my thesis advisor (Dr. A. Achong) since he had the special technical implements to do the quality figures required. (This was for Solar Physics. ) The JRAS - for my original paper given above - didn't have these technical drawbacks and could convert ordinary black ink line drawings-figures to page compatible ones, hence I drew all those myself).

At times it was amusing leading up to the papers' submission, because I had to bring Achong up to speed on the statistical basis for the papers (e.g. Poisson distributions in the 2nd) and the physical basis for the interpretation in the first, i.e. using vector magnetograms  to support a mutual polarity intrusion leading to enhanced magnetic field gradient and higher probability of solar flare. However, Achong insisted his name go first because he was my advisor and "How would it look otherwise?"

Never mind, the ability to conceive and forge a project of my own - an extended, original work with numerous data sets and parameters - showed me I irrevocably qualified as a solar physicist. This project also sealed my identity along with my contribution to perhaps the most monumental space and solar physics project ever undertaken, The Solar -Terrestrial Predictions Workshop e.g.


The massive effort saw the input from over two  hundred solar and space physicists covering every aspect of the problem of solar-terrestrial interactions, including: long, medium and short term solar  forecasting, associated  geomagnetic activity and auroral (substorm) forecasts, as well as ionospheric predictions.

 My paper on the statistical limits to solar flare prognostication appeared on pp. 276-284 of the Proceedings and received much attention from the other contributors - since of course it impacted in multiple ways on their work as well.  The paper followed from the last  two chapters of my thesis and was based on the concept of a Poisson-based “delay time” for build up of magnetic free energy,  first postulated by me in early 1984,  for application to “SID” (sudden ionospheric disturbance-generating) flares, with the release attendant on a change in initial free magnetic energy (E _m = B²/2m ) given by[i]:

¶ / ¶  t  {ò _v  B²/2m  dV} = 1/m  ò _v div[(v  X B) X B] dV 

 -   ò _v  {h_an | J_ms |² }dV       

where the first term on the right side embodies footpoint motion, and the second, joule dissipation, but with J_ms the current density at marginal stability – since the marginal stability hypothesis is required for a driven process, and h _an  is the anomalous resistivity. In the same paper, it was shown how the flare distribution corresponds to a Poisson process of the form P(t) =    =   exp (- l)   l^t  / t!, where theoretically the Poisson mean rate of occurrence is: l_m =   l Dt, , with Dt  = t,  assuming the time interval Dt  = 1d. In reality, measuring constraints (say achieving uniformly equal time intervals between successive Mt. Wilson magnetograms), will usually ensure  Dt ¹ 1d, so  Dt  ¹  t  thereby introducing a selection effect variability, complicating computation of P(t).

 It was also suggested, but not proven, that variability in l arises from  variability in vertical magnetic gradients (B_z) and critical changes in the associated current density at marginal stability (Jz_ms) such that:  d(Ñ(+ B_z) ) Þ d Jz_msÞ dJ_z  ( dt)/ dt where J_z  is the vertical current density associated with putative footpoints magnetic induction (+ B_z ) and rate of change in |B_z| modulated by significant evolutionary changes (dt) in the lifetime of the magnetic field, especially critical if   dt < Dt.

Since magnetic gradients and associated scale lengths (ℓ_B) also will change in time, there would be scope for accepting a Poisson process of form P(t) =  =   exp (- l)   l^t  / t!  which would embody an energy modulation with some inbuilt variance, with the latter having to be known to determine how much energy might be released and when. In other words, the differing scale factors inevitably introduced variabilities that were difficult to account for. The Poisson statistics therefore had to be able to take these differing modalities into account.

Below: Photo taken after being awarded the M.Phil. in Physics:

-------------------------------------

[i] Stahl, P.A. and  Achong, A.: 1984a, Proceedings of the Second Caribbean Physics Conference, Ed. L.L. Moseley, pp. 1-11.

Accounting for the Solar Flares and Events In Solar Active Region 2017

The X-1 class flare (extreme upper right) that erupted on March 29, likely in a quadripolar sunspot group.

By now, most solar researchers are aware of the events associated with Active Region 2017 last month. Most remarkable was a major (X-1) class flare that erupted on March 29th, starting at 17:35 UT, peaking at 17:48 UT and ending at 17:54.  The area of impact on Earth affected large portions of its sunlit side especially at the sub-solar point. The flare was geo-effective, with high frequency radio communications blacked out for over an hour.  Other events associated with the flare also occurred, including:

- A 10 cm radio burst  with a peak of 360 solar flux units (SFU) from 17:45-17:48

- A Type II radio burst was observed at 17:53  with an estimated velocity of 508 km/sec.

In addition the region has been replete with M class flares (the x-ray scale is ascending from C to M to X) and the characteristic of the region is that it contains a delta class sunspot group. This is a category of group I've posted on previously, in conjunction with my own research.  Much of this focused on the events associated with this delta class group - which I photographed on 11/4/ 80 at its maximal extent:
The group itself marked the return of McMath plage region 17181, and first appeared on the sun's east limb on Oct. 30, 1980 with an area of 70 millionths of a solar hemisphere (msh). In other words, the area increased more than a factor 20 within a span of five days. By 1st November, the group was in the complex delta class, which is determined by the presence of mixed magnetic polarities in the same region. (The active region itself was given the designation AR2776).

Most conspicuous in the course of the group's evolution was the leader spot's pronounced, elliptical penumbra (outermost, lighter sector). This appeared to fragment from Nov. 6 which may have led to a steady decrease in total group area from that date. The emergence of a defined vortical structure in the penumbra occurred thereafter and was accentuated by the formation of a "light bridge" within the penumbra.

At issue, however, was whether it was indeed the delta morphology that led to the high flare occurrence, or something else.  Note that the qualifying entry to the delta class is umbrae - dark central  regions - separated by less than two degrees within one penumbra having opposite polarity. This condition leads to what we describe as a steep magnetic gradient, viz.

grad B = [+B_n - (-B_n)] / x

where the numerator denotes the difference in the normal components of the photospheric magnetic field (between opposite polarities of the active region) as measured by vector magnetograph, and the denominator is the scale separation (x) between them. In several iterations of the Solar-Terrestrial Workshops and Predictions Proceedings, it has been noted that when grad B exceeds  0.1 Gauss/km then a flare is 96% probable within 24 hrs.  Alas, it is not possible to forecast the specific x-ray class.

 Subsequent investigations I conducted disclosed it was more likely the presence of a magnetic multi-pole in a delta class spot that gave rise to large flares such as the one observed on March, 29 this year.


In the diagram shown, such a multipole sunspot model is depicted such that the most intense magnetic field locus is centered in the largest sunspot and is "line-tied" , i.e. it represents the anchored ‘foot’ of one magnetic arch with the other foot connected to a 2nd polarity (minor spot) executing proper motion around it.   In this case the dipole undergoes an approximate 28 degree proper motion over time interval (t2  - t1) during which the (-) polarity footpoint is displaced from a1 to a2.  Treating the region as localized in the complex plane (note the defined Re axis) it is possible to do extensive analysis of motions, shears, and behavior of the respective poles and spots. Such analyses are too comprehensive to go into here, in a blog post, but are described in my book, Selective Analyses in Solar Flare Plasma Dynamics - also available as an E-book.

Of special interest for really large, energetic solar flares are quadripolar sunspots. In one quadri-polar  analysis by Somov et al (Solar Phys., 1998) a series of magnetic charges was specified as:

(+B) <-> e (N)

(-C) <-> e (S)

+A <-> e (n)

-A <-> e (s)

Where the capital letters on the left denoted the identified loop foot points and the  'e' identifiers on the right specified the respective magnetic "charges"  or polarity centers. By establishing a "characteristic dynamic length"  ℓ  for the region, the researchers were able to use Poisson statistics to predict what the average flare production would be over time.

Much more work remains to be done, and plausibly we won't really have a major breakthrough until we can get a high resolution solar optical telescope. This would be of the type that had originally been planned in the 1980s (SOT-1),  designed as a "Hubble version" of a solar telescope dedicated entirely to fine-resolution optical observing but also equipped with:  a tunable filtergraph, a system of state of the art spectrographs, and a photometric filtergraph. All would be contained in a single housing called the 'Focal Plane Instrument Package'. This SOT-1  would have been able to resolve coronal and other structures to within 0.1 arcsec or better, and deliver spectacular real time photometric and spectrographic imagery and information.

 The SOT-1 was scrapped by the budget cutters because they believed mounting a "Star Wars" missile defense system was more important (never mind physicists later showed this program was useless, your classic 'white elephant').

Maybe the budget obsessive nitwits - who always seem to be able to find money for more wars or giving $1b to the Ukraine- will wake up when a large CME strikes and knocks out our power grids. At least one hopes they will, but who knows?

The Telescope That Never Was

In the blog post before last I made passing mention of the SOT or Solar Optical Telescope. By 1984-85 it was being portrayed as a "Hubble version" of an optical telescope dedicated entirely to fine-resolution solar observing. And not too soon! Solar cycle 20, one of the most active in decades, was already nearing its end and solar physicists remained frustrated by a persistent lack of instrumental resolution that prevented identifying the initial flare instability site, presumed to be located near the tops of coronal arches or loops.

Oh sure, many had their suspicions, but without the actual evidence that was what they remained. At the time I was nearing completing of my (intended) Ph.D. Thesis which sought to perfect the forecasting of a particular type of flare I had discovered (the "SID flare") and reduce it to a time frame of 24 hours and no more. Also, the intent was to scrub the forecasting foundation of predictions of "no flare" days entirely, since these merely inflated the prediction success rate.

I had managed in several trials leading up to the final thesis presentation, to raise pure forecast scores to near the 61% threshold for solar active regions (ARs) for which most data were known. For example, I had the use of magnetograms and vector magnetographs that enabled very high estimates of the magnetic gradient:

grad B = [+B_n - (-B_n)] / x

where the numerator denotes the difference in the normal components of the magnetic field (between opposite polarities of the active region) as measured by vector magnetograph and the denominator the scale separation between them. If I calculated (based on the magnetograms) that grad B = 0.1 Gauss/km then I knew a flare was 96% probable within 24 hrs. In 11 of 14 cases for which I applied the gradient test to SID flares, they occurred within the time frame.

The optically-based forecasts were more problematic, and I mainly had white light sunspot photos (e.g. such as that shown, taken by me with a catadioptric 200mm telescope in 1980) and also H-alpha films related to the sunspot regions. The problem was that the level of resolution simply wasn't adequate. While on the best days (say for white light photos, or even H-alpha views of the Sun at the Harry Bayley Observatory) one might approach 1 arcsec resolution, you really needed at least an order of magnitude better, or 0.1 arcsec to resolve many of the active filamentary structures, for example in the solar penumbra and surrounding regions in H-alpha line of 656.3 nm.

Without being able to penetrate the optical barrier, we had mainly semi-empirical models - which allowed no consistent quantitative predictions based on optical manifestations, such as shear angle (estimated from the angle of penumbral filaments to assigned coordinate axes) or the emergence of the critical 'delta' sunspots which grew rapidly and with mixed magnetic polarities.

Obviously a dedicated space telescope would be supremely useful! For example, in the case of the grad B index forecast, if on a particularly cloudy or overcast day one is unable to obtain vector magnetograms, then one cannot measure the intensity of the normal vectors. If so, one cannot obtain the difference [Bn(+) - Bn(-)], and hence no grad B. And no prediction. Same thing if clouds covered the Sun and one couldn't see twisting fibrils in the vicinity of growing sunspots at all.

The need for a Solar Optical Telescope was self-evident. The first flight of the NASA-engineered SOT (designated SOT-1)was put back a ways to 1986-87 but by the time I submitted my Thesis in early summer of 1984, this was believed (by at least one Swedish examiner) to have worked out for the best. He envisaged me taking the thesis to an American unversity and completing it there in time to make use of the much more refined solar optical imagery, which in turn would refine my solar flare forecast quality. It made eminent sense, though I did worry (and mentioned it at the time) of the talk circulating about budget cuts, especially as Reagan's tax policies had bled down not only U.S. domestic programs but its space missions too (especially radically after the Challenger disaster, in January, 1986). There was also talk of budget trimming for the SOT as early as the summer of 1984.

The key component of the SOT was its CIP or 'coordinate instrument package' - which looked from afar like the top of a giant bus or trailer, to be mounted to the main telescope as shown in the graphic. It's module contained: tunable filtergraph, a system of state of the art spectrographs, and a photometric filtergraph. All would be contained in a single housing called the 'Focal Plane Instrument Package'.

The tunable filtergraph was particularly important, as it was to produce very high resolution digital intensity maps of the solar atmosphere in white light. The components included: blocking filters, a tunable birefringent filter, beam splitters, enlarging optics and two CCD cameras - one of high resolution (with field of 61 x 61 arcmins) and the other a background camera- with a field of 122 x 122 arcmins. The operative wavelength range would be 460-760nm or essentially covering the whole visible spectrum. Meanwhile, the filter bandpass ranges would be from 0.004 nm at 460 nm to 0.0125 nm at 760 nm. Enough to make the budding solar forecaster's mouth drool!

The spectrograph system, meanwhile, consisted of an ultraviolet (UV) and visible spectrograph integrated into one structure. The first used two UV gratings and 2 CCD cameras located at its focal plane, while the second used an Echelle grating and 4 cameras at its focal plane. All three gratings were to be mounted on a single rotating assembly.

Meanwhile, the spectrograph slit and collimator were designed to move in tandem enabling the slit to scan across the solar image without affecting the angle at which light strikes the grating. Hence, the spectrum remained fixed in the focal plane. The wavelength range for the visible spectrograph was 270- 1000 nm, and for the UV instrument 210-200 nm. The slit dimensions were 0.08 arcsec by 61 arcsec.

The photometric filtergraph (see Fig. 2) was designed to record high resolution, wavelength -disriminated images of the solar atmosphere on photographic film. It included two camera systems, each having 1000' film transports, focal plane shutter, film annotation device and filter wheel. Light then striking an indexing mirror (Fig. 2) is then directed to either camera. The operative wavelength range was 220- 668.7 nm, with a field of view of 120 x 150 arcsec at f/24. Exposure times would range from 1 millisecond to 10 seconds, and filters for the instruments would have included: line blanketing, H- opacity, Hg continuum, Balmer continuum, Ca K, H-alpha and UV continuum.

On this telescope my completion of the Ph.D. depended, but alas, by 1986 it was cut - the $2.1 billion deemed to be prohibitive given the U.S. budget realities, already bleeding $1.7 trillion into the red -mainly thanks to the combo of Reagan's tax cuts, along with out of control military spending. Including foolishness like the 'Strategic Defense Initiative' (or 'Star Wars') which monies would have gone a long way toward making the SOT-1 a reality, along with my Ph. D.! But it's too late to cry over telescopes that never materialized.