Hydro-gravitational-dynamics (HGD) cosmology suggests that hierarchical clustering began at 10^12 s after the Big Bang, at matter radiation equality, and proceeded from the top down at the Schwarz viscous scale, progressively fragmenting the plasma realm into smaller clumps, beginning at the supercluster-scale and progressing to the cluster-scale and finally the galaxy-scale prior to the epoch of recombination. At recombination, Jeans instability fragmented proto-galaxies into million solar mass proto-globular-clusters. (Gibson 2006)
Baryonic dark matter (DM) cosmology suggests baryonic DM reservoirs in the form of self-gravitating planetary-mass globules of gas in hydrostatic equilibrium, which are a few astronomical units across. These baryonic DM globules are designated ‘paleons’ by Manly Astrophysics for their presumed old age. The evidence for paleons comes from scintillation of pulsars and quasars by foreground plasma, which can be modeled as spherical paleons with ionized outer shells that are ionized by plowing through interstellar gas at 230 km/s in their rotation around the Milky Way.
Paleons are suggested here to have to have been ejected from Population III protostars during coronal-mass-ejection chain reactions, which progressed around the equator at the rate of a magnetic reconnection shockwave, ejecting equatorial material which magnetically condensed into self-gravitating paleons. A similar process is suggested to occur today in the form of self-gravitating, planetary-mass cometary-knot (CK) ejection from late-stage asymptotic giant branch (AGB) stars.
In alternative baryonic DM cosmology, the epoch of recombination occurred later than the recognized date of 378,000 years after the Big Bang, when the universe had expanded by a volume factor of about 6 to the canonical density of baryons calculated by ΛCDM cosmology. Baryonic DM cosmology suggests that recombination occurred around 378,000 yr * 6^(1/3) ~ 687,000 years after the Big Bang, at otherwise canonical conditions.
ΛCDM cosmology is particularly robust in its evidence from the epochs of nucleosynthesis and recombination, but this standard model of cosmology is comparatively weak in its reliance on hierarchical clustering for the formation of structure in the universe, notably with the missing satellite problem of large galaxies, and the discovery of supermassive-black-hole quasars earlier than z = 6.
Additionally, dark matter (DM) concentrations in galaxy cores do not conform to models predicting a cuspy concentration, known as the ‘cuspy halo problem’. And the complete absence of DM in globular clusters requires secondary mechanisms to explain away its absence. Alternatively, baryonic DM that converts to stars and luminous gas in regions of high stellar density is predictive by comparison.
Structure formation by hydro-gravitational-dynamics (HGD) in the plasma epoch suggests that proto-spiral-galaxies formed by turbulent fragmentation, with the angular momentum of spiral galaxies naturally arising from eddy current vortices in the turbulence. While hierarchical clustering of ΛCDM cosmology may neatly explain the origin of dwarf spheroidal galaxies and the merger of giant spiral galaxies to form giant elliptical galaxies, it has no intrinsic mechanism to explain the typical angular momentum of spiral galaxies.
Pulsar and radio galaxy scintillation provide observational evidence for self-gravitating gaseous globules, designated ‘paleons’ by Manly Astrophysics, which are suggested to be the reservoirs of baryonic DM.
Finally, the planetary-mass ‘cometary knots’ in planetary nebulae today suggest a formation mechanism which can be extended to the suggested formation of their primordial paleon cousins in the early universe.
Alternative hydro-gravitational-dynamics (HGD) cosmology:
The ΛCDM cosmology standard model of cold dark matter hierarchical clustering (CDMHC) for self-gravitational structure formation is predicated on the 1902 Jeans criterion for gravitational instability, which neglects viscosity, diffusivity, and turbulence and which sets density to zero (the Jeans swindle) to derive the Jeans length scale. CDMHC suggests that hierarchical clustering only began after the epoch of recombination at 10^13 s, with gravitational structure formation proceeding from the bottom up, with small structures forming first and large structures forming last.
When viscosity, diffusion and turbulence are included in the analysis, HGD cosmology suggests that gravitational fragmentation proceeded from the top down at the Schwarz viscous scale, with the supercluster-scale fragmentation initiated 10^12 s after the Big Bang at matter radiation equality, followed by cluster-scale and galaxy-scale fragmentation in the plasma realm prior to the epoch of recombination.
HGD cosmology suggests HGD structure formation in the plasma epoch, between 10^12 to 10^13 seconds after the Big Bang, followed by Jeans instability at the epoch of recombination on the scale of circa million solar mass ‘proto-globular-clusters’.
Cometary knot (CK) formation by ‘coronal-mass-ejection chain reaction’ in AGB stars:
Thousands of cometary knots stream out from the stellar remnant in the Helix planetary nebula (NGC 7293) in a system where “the central star is about 6560 yr into its life as a star nearly liberated of its envelope.” (Capriotti and Kendall 2006) O’Dell and Handron (1996) give the density, mass and size of the neutral gas in the estimated 3500 cometary knots of the Helix nebula as, hydrogen density ~ 4 x 10^6 cm-3, with a CK mass range of ~ 4 x 10^25 to 4 x 10^26 g and radii of 60–200 AU, based on the distance to the nebula of 213 pc. CKs have bright rims facing the central star and cometary tails trailing away, caused by photoevaporation by the brilliant white-dwarf remnant.
The main body of the Helix nebula is an inner ring, roughly 500″ (0.52 pc) in diameter surrounded by a highly-inclined torus of 740″ (0.77 pc) diameter, with an outermost ring 1500″ (1.76 pc) in diameter. The CKs near the inner edge of the inner ring are traveling away from the central star, along with the ring material in which they are embedded. O’Dell et al. (2004) estimate an expansion age for the inner ring of 6560 yr, using an expansion velocity of 40 km/s and a present radius of 0.26 pc at a distance of 213 pc. In the interior of the inner ring, but not closer than 120″, CKs dominate the landscape, while beyond 190″, large clouds do, although, while the CKs in the inner ring are the most prominent, infrared observations have detected CKs in regions outside the inner ring in numbers a factor of 6 or so greater than the inner ring. The inner ring is the last of three major ejections, 6560 years into its life as a small hot very luminous star nearly liberated of its envelope. (Capriotti and Kendall 2006)
This alternative baryonic DM cosmology approach attempts to equate modern CKs with primordial paleons, makes two assumptions; that CKs are self gravitating objects, like paleons, and that no self-gravitating objects can form by direct collapse which are smaller than a Jeans mass, which suggests that CKs are ejected from the compressed outer layers of the star itself, rather than condensing from a diffuse stellar wind.
After helium is exhausted in the core of an AGB star, it continues to burn in a thin shell surrounding the core during the ‘early’ (E-AGB) phase. After the helium in the shell is depleted, a ‘thermally pulsing’ (TP-AGB) cycle begins. The star now derives its energy from burning a thin shell of hydrogen which converts to a thin shell of helium. The helium shell explosively ignites in a process known as a ‘helium shell flash’. The helium shell flash causes the star to temporarily expand and brighten, puffing up the star which lowers its temperature, extinguishing hydrogen fusion. The helium shell flash also induces convection (third dredge-up) which brings carbon from the core to the surface and also mixes hydrogen from the surface into deeper layers where it reinitiates hydrogen fusion to begin another thermally pulsing cycle.
The rapid helium shell flash lasts only a few hundred years in the life of a thermally pulsing cycle, where one thermally-pulsing cycle runs from 10,000 to 100,000 years. Our Sun may only undergo four 100,000 year thermally pulsing cycles before the contracting core is successful in ejecting its outer layers to expose a naked white dwarf. More massive stars, by comparison, may undergo many more closer-spaced thermally pulsing cycles than our Sun before fully ejecting their outer layers to reveal a degenerate white-dwarf core surrounded by a planetary nebula.
As the outer layers of a star expand following a helium shell flash, the magnetic field locked into the plasma attempts to enforce solid rotation during thermally-pulsing expansion, where expansion increases the moment of inertia of the expanding outer layers. If the magnetic corotation radius is forced below the surface of the star during an expansion phase, the magnetic field becomes twisted at this radius. When the magnetic field becomes twisted to the breaking point at the magnetic corotation radius, a spontaneous magnetic reconnection may occur, causing a coronal mass ejection. Magnetic reconnection and its accompanying coronal mass ejection results in a rebound shockwave which is suggested to set off a chain reaction of closely-spaced magnetic reconnections which collectively eject a filament of plasma from the equatorial region, designated a ‘coronal-mass-ejection chain reaction’.
If the average mass of a coronal mass ejection from the Sun is on the order of 1.6e12 kg (Carroll and Ostlie 2007), and if this mass is typical in AGB stars, then a chain reaction of something like a trillion closely-spaced coronal mass ejections would be necessary to create a single CK, suggesting an exceedingly-efficient process.
A suggested coronal-mass-ejection chain reaction of a planetary-mass filament would presumably clump magnetically into a self-gravitating CK, as it streamed away from its progenitor star.
While CK ejection likely occurs in each of a succession of thermally-pulsing AGB cycles, perhaps only those in the final cycle are illuminated in the subsequent planetary nebula phase. And since a large percentage of stars are intermediate mass (0.6–10 solar masses), which pass through an asymptotic giant branch phase, intermediate mass stars may make a significant contribution back to the DM realm.
Fragmentation at recombination:
In the plasma epoch prior to recombination, the Jeans scale exceeded the horizon scale, precluding gravitational fragmentation by the Jeans mechanism, due to the high speed of sound in plasma (on the order of the speed of light). At the epoch of recombination, the Jeans scale of neutral gas was on the order of 1 million solar masses, promoting gravitational collapse of the neutral continuum into proto-globular-cluster-scale masses. (Gibson 2006)
Additionally, Gibson suggests that HGD caused fragmentation into self-gravitating earth-mass ‘primordial fog particles’ (PFP) following the epoch of recombination, and that the PFPs have subsequently condensed to form earth-mass ‘Jovian planets’ (presumably designated ‘Jovian’ for their hydrogen-helium composition). And since the Jeans scale at recombination was on the order of one million solar masses, these PFPs were clumped into proto-globular clusters. These persistent Jovian planets constitute baryonic dark matter, explaining the missing baryon problem as 30,000,000 earth-mass rogue planets per star in the Galaxy. Additionally, Gibson replaces dark energy with hot dark matter, such as neutrinos, which only become significant in gravitational clumping at the galactic cluster scale.
I agree with fragmentation of the continuum at recombination into circa million solar mass proto-globular-clusters, but dispute their sub-fragmentation into planetary-mass PFPs. Instead, I suggest gravitational sub-fragmentation of proto-globular-clusters into circa thousand solar mass Population III protostars, where the Population III protostars efficiently eject their outer layers in the form of self-gravitating planetary-mass paleons.
Paleon formation in Population III protostars by coronal-mass-ejection chain reaction:
The suggested physical symmetry between CKs and paleons suggests formational symmetry, albeit with even-greater efficiency in the formation of primordial paleons.
Expansive cooling of the universe promoted sub-fragmentation of proto-globular-clusters, where the sub-fragmentation scale is suggested to have been in the range of multi-thousand solar mass cores. Population III protostars are suggested to have formed before continued expansive cooling could sub-sub-fragment still-smaller stellar-mass cores.
Non-turbulent freefall collapse is the exception in a turbulent world, with excess angular momentum forming a diminutive core surrounded by a much more massive envelope, partially supported by rotation. When a rotationally-supported overlying envelope is much more massive than its diminutive core, the system is suggested here to be unstable and susceptible to disk instability, with disk instability occurring by the suggested mechanism of ‘flip-flop fragmentation’ (FFF), as a catastrophic mechanism for projecting mass inward.
When a much more massive envelope, partially supported by rotation, surrounds a diminutive core and the diminutive core-to-envelope mass is insufficient to dampen inhomogeneities in the envelope, the envelope is suggested to be unstable, promoting runaway disk instability, causing it to catastrophically clump to form a new larger core, inertially displacing the (older) former core into a satellite status. This is the mechanism which is suggested to ‘spin off’ diminutive cores in prestellar objects in the form of gas/ice giant planets.
A contracting multi-thousand-solar-mass globule may have undergone repeated episodes of FFF to spin off sufficient angular momentum to form a Pop III protostar, ripe for further weight reduction by way of coronal-mass-ejection chain reactions.
Freefall contraction of an envelope to form a new core causes spin up, which likewise increases the rotation rate of the protostar magnetic field. Contraction also causes heating, with the ionization front moving outward from the contracting protostar core. When the magnetic corotation radius drops below the outward-moving ionization front at the ‘magnetic corotation radius’, the magnetic field becomes twisted, storing magnetic energy.
When the magnetic field becomes twisted to the breaking point at the magnetic corotation radius, spontaneous magnetic reconnection will occur, and if this results in a coronal-mass-ejection chain reaction, then planetary-mass filaments may be ejected with magnetically clump into paleons.
If coronal-mass-ejection chain reaction unwind multi-thousand-solar-mass Pop III protostars down to the 160 to 250 solar mass range, then the resulting Pop III stars may end their lives pair-instability supernovae which leave no stellar remnants, since there’s no observational evidence for zillions of Pop III remnants, in the form of white dwarfs, neutron stars or black holes.
To have converted some 5/6 of all baryons to DM paleons warrants an epoch designation, which is suggested as ‘Population III epoch’. To create such a high percentage of DM, the vast majority of the matter in the universe must have been processed through Pop III protostars, with a relatively-small percentage of baryonic matter becoming Pop III main sequence stars.
If ejected paleons escaped from the gravitational well of their Pop III stars, they may have remained gravitationally bound within their proto-globular-clusters, suggesting that paleons may still be grouped into circa million solar mass paleon clusters.
Extreme Scattering Events (ESEs) are suggested to be caused by the refraction of quasar radio waves by the ionized surface of occulting paleons, where the paleon surface is ionized by the shock of plowing through interstellar gas at around 230 km/s in its orbit around the Milky Way. Self-gravitating paleons are calculated to be on the order of a few AU across and in a number density of a few thousand per cubic parsec in the neighborhood of the Sun. (Tuntsov, Walker et al. 2015) Alternatively, the same scintillation effect can be modeled by anisotropic plasma distributions, such as a plasma sheet seen edge on without any accompanying self-gravitating dark matter component (Tuntsov and Walker 2015).
Manly Astrophysics calculates paleons to have a mass range of ∼ 10-7 to ∼ 10-1 solar masses, based on their stabilization by the condensation and sublimation of solid hydrogen (snowflakes). But since the ambient temperature of the universe has only dropped below the condensation point of hydrogen some 2 billion years ago, or so, hypothesized stabilization by hydrogen snow would be relatively recent.
But if paleons date from Pop III stars, then hydrogen snowflakes would have to be superfluous to their formation and survival. If hydrogen condensation has indeed increased the stability of paleons in the last 2 billion years or so, then perhaps this increased stability may be responsible for the discovery that galaxies today emit only about half as much light as galaxies emitted 2 billion years ago. Thus if the advent of the ‘epoch of hydrogen condensation’ increased paleon stability, it may have ushered in a new era of reduced star formation, giving rise to popular articles declaring that the universe is dying.
The suggested sedimentation of hydrogen snowflakes in paleons suggests still older sedimentation of less-volatile stellar metallicity in the form of dust and ice. And the sedimentation would tend to accrete to form a central solid mass within each gaseous paleon.
While paleons may have formed with Big Bang chemistry, contaminated by Pop III star metallicity,
they will have acquired (swept up) varying degrees of Pop II star and Pop I star metallicity in their 13 billion years of orbiting the Galaxy core, with more distant galactic-halo paleons having acquired less than those with orbits crossing the spiral-arm disk plane. By comparison, CKs are formed with highly-elevated levels of stellar metallicity, so paleons and (dark) CKs may vary more widely in metallicity than stars themselves.
An Earth-mass paleon with the average metallicity of the Sun (Zsun = 0.0134) may have a central solid object the mass of Earth’s Moon, while distant halo paleons may only have central solid objects the size of a typical Oort cloud comet.
Manly Astrophysics calculates a paleon density in the stellar neighborhood of ∼ 104 pc−3, which suggests that many hundreds may be passing through the outer Oort cloud at any given time. And with their relatively-large (circa 1 AU) diameters, paleons could sweep up dust, ice and microbes from comet clouds and debris disks surrounding stars, perhaps making paleon cores into rich panspermia reservoirs.
The extent to which paleons remain bound in their suggested primordial proto-globular-clusters is unaddressed, although their large diameters with readily distortable shapes may be considerably stickier than comparatively point-mass objects like stars, perhaps making ‘paleon clusters’ more stable over time than star clusters, of comparable size and density.
Flip-flop fragmetation galactic evolution:
HGD turbulence presumably instilled proto-spiral-galaxies with their specific angular momentum, or more likely with excess angular momentum that underwent galactic evolution to catastrophically project mass inward to form mature spiral galaxies, with their typical range of specific angular momentum.
Following recombination, Jeans instability is suggested to have fragmented proto-galaxies into circa million-solar-mass proto-globular-clusters, and with the loss of hydrostatic radiation pressure at recombination, proto-galaxies gravitationally collapsed to the point of Keplerian rotation, flattening proto-galaxies around their angular momentum vectors.
Proto-spiral-galaxies with excess angular momentum would have had diminutive cores, compared to the considerable galactic bulge of mature spiral galaxies. A massive disk overlying a diminutive core is suggested to be dynamically unstable, where the diminutive core is unable to dampen inhomogeneities in the disk from amplifying into runaway disk instability.
Runaway disk instability breaks the radial symmetry of the disk, causing the disk to clump to form a younger larger core that inertially displaces the former core to a planetary satellite status, in a galactic process designated, ‘flip-flop fragmentation’ (FFF), catastrophically projecting mass inward.
FFF was initially proposed as a catastrophic mechanism for projecting mass inward in prestellar dark cores undergoing freefall collapse, spinning off former cores in the form of gas/ice giant planets. (See section, STARS, PLANETS, MOONS, MINOR PLANETS AND COMETS)
In the Milky Way, the Large and Small Magellanic Clouds are suggested to be former diminutive, proto-Milky-Way cores, spun off in two successive generations of FFF.
In the final instance of Milky Way FFF, the clumping of the disk ended in the formation of a direct-collapse supermassive black hole, Sagittarius A*, with a central bulge sufficiently massive to dampen out disk inhomogeneities, preventing further disk instability.
Baryonic dark matter:
The absence of DM in globular clusters and the absence of a cuspy DM distribution in galactic cores has been called the ‘cuspy halo problem’, which requires secondary mechanisms to explain away in exotic DM theories. By comparison, the observed distribution is predictive in baryonic DM cosmology if gaseous paleons convert to luminous gas and stars in regions of high stellar luminosity/concentrations.
This alternative baryonic DM cosmology supports canonical state conditions (pressure, temperature and density of baryons) as calculated by ΛCDM cosmology at defining epochs, such as Big Bang nucleosynthesis and recombination; however, the timing would be shifted forward to allow Big Bang expansion to inflate the density of baryons (with baryonic DM) to the canonical density. So while the epoch conditions of baryonic DM cosmology are suggested to occur at the canonical density of baryons, the epochs would occur at circa 6 times lower overall matter density, in the absence of noninteracting exotic DM. The date of the Big Bang need not change in baryonic DM cosmology,
only the timing of those epochs which are dependent on the type of dark matter.
One note, ‘baryon density’ (Ωbh2) of the universe is defined to be a constant over time, whereas ‘density of baryons’, as used here, is simply the instantaneous baryonic-matter density, which decreases exponentially over time due to Big Bang expansion, so ‘canonical density’ at defining epochs refers to the instantaneous density of baryons, not the constant baryon density of the universe.
The Hubble expansion rate of the universe may also need to be altered in baryonic DM cosmology to reflect a later date for recombination. Direct measurement of expansion rates based on cepheid variables and/or Type Ia supernovae, however, should be free from this problem. Therefore the higher Hubble expansion rate figures (circa 72–73 km s−1 Mpc−1) directly measured from cepheid variables and/or Type Ia supernovae, which are agnostic as to the actual date of recombination, are likely to be more accurate than lower figures (circa 68 km s−1 Mpc−1) calculated from CMB Planck data and BAO scale in today’s universe, which are dependent on recombination timing. A Hubble constant based on an anomalously-young date for recombination would tend to reduce the apparent expansion rate, so low expansion rates calculated from CMB data are at least skewed in the expected direction.
Baryonic DM cosmology is agnostic with regard to the metric expansion of space itself, by way of dark energy or a cosmic constant.
If dark matter is baryonic, and if DM can convert luminous matter by way of paleon evaporation, and if luminous matter can conversely go dark by way of cometary knots streaming from AGB stars, then the relative ratio of dark matter to luminous matter may not be particularly significant, with the ratio varying from one galaxy to another and presumably decreasing slowly over time. The ratio does matter, however, in pinning down the actual date of recombination. For this conceptual approach a 6:1 DM:luminous matter ratio will be used for convenience, even though the missing baryon problem of ΛCDM cosmology could push the actual ratio higher than 6 to 1 and correspondingly push out the date of recombination as well. For a 6:1 ratio, a first-order approximation (of this conceptual approach) for the actual redshift of recombination is z ~ 1100/(6^(1/3)) = 605, around t ~ 378,000 * 6^(1/3) = 680,000 years after the Big Bang.
A recent study finds that early spiral galaxies (redshift z = 0.7–2.6) are heavily dominated by baryonic matter in the inner star-forming regions, with falling rotation curves (rotation velocities decreasing with radius). (Genzel et al. 2017) Lead author Reinhard Genzel in an interview for Scientific American with Charles Q. Choi quantified the baryonic dominance in terms of the “effective radius” (half-light radius) of spiral galaxies—the 50% light radius—where the effective radius is 50 to 80 percent dark matter in the Milky Way and other typical local spiral galaxies, compared to 10 percent for early (z = 0.7–2.6) galaxies.
The domination of early spiral galaxies by baryonic matter telegraphs and constrains spiral galaxy formation theory, along with the nature of dark matter. Paleon formation in the Population III epoch is presumed to precede catastrophic spiral galaxy evolution by way of FFF (disk instability), which is presumed to have evaporated paleons in the heat released during the gravitational collapse of disk instability. Intergalactic dark matter is gradually falling toward densified regions, i.e. galaxies and galaxy clusters, creating progressively-denser DM haloes around (spiral) galaxies, creating spherical dark matter halo distributions with low specific angular momentum. However, the inclined disk of satellites surrounding the Milky Way, including the Small and Large Magellanic Cloud as former spun off proto-Milky-Way cores, suggests that the Milky Way system may have been significantly twisted by external torque, perhaps caused by infalling intergalactic dark matter with non-zero specific angular momentum.
And presumably DM gravitationally clumps to form a cosmic web of dark matter, as predicted by computer simulations, explaining the numerous DM ‘sub haloes’ detected within the Milky Way DM halo.
Perhaps additional evidence for the gradual accretion of dark matter haloes comes from local (low-redshift) ‘passive spiral galaxies’, with falling rotation curves similar to those of high-redshift early spiral galaxies (Genzel et al. 2017). But passive spiral galaxies may be deficient in DM haloes due to crowding within rich galaxy clusters, rather than early-versus-late timing, where infalling DM may tend to form a global galaxy-cluster halo, rather than enveloping each member galaxy individually.
Capriotti, Eugene R. and Kendall, Antoony D., 2006, THE ORIGIN AND PHYSICAL PROPERTIES OF THE COMETARY KNOTS IN NGC 7293, The Astrophysical Journal, 642:923–932, 2006 May 10
Carroll, Bradley W.; Ostlie, Dale A., 2007, An Introduction to Modern Astrophysics, Second Edition
Genzel, R.; Förster Schreiber, N. M.; Übler, H.; Lang, P.; Naab, T.; Bender, R.; Tacconi, L. J.; Wisnioski, E.; Wuyts, S.; Alexander, T.; Beifiori, A.; Belli, S.; Brammer, G.; Burkert, A.; Carollo, C.M.; Chan, J.; Davies, R.; Fossati, M.; Galametz, A.; Genel, S.; Gerhard, O.; Lutz, D.; Mendel, J. T.; Momcheva, I.; Nelson, E. J., 2017, Strongly baryon-dominated disk galaxies at the peak of galaxy formation ten billion years ago, Nature 543, 397–401 (16 March 2017)
Gibson, Carl H., 2006, Cold Dark Matter Cosmology Conflicts with Fluid Mechanics and
O’dell, C. R. and Handron, K. D., 1996, Cometary Knots in the Helix Nebula, Astronomical Journal v.111, p.1630
O’Dell, C.R.; McCullough, Peter R.; Meixner, Margaret, 2004, Astronomical Journal, Volume 128, Number 5
Smith, Nathan; Stassun, Keivan G., 2016, The canonical Luminous Blue Variable AG Car and its neighbor Hen 3-519 are much closer than previously assumed, arXiv:1610.06522 [astro-ph.SR]
Tuntsov, Artem V.; Walker, Mark A.; Koopmans, Leon V.E.; Bannister, Keith W.; Stevens, Jamie; Johnston, Simon; Reynolds, Cormac; Bignall, Hayley E., 2015, Dynamic spectral mapping of interstellar plasma lenses, 2016, ApJ, 817, 176
Walker, Mark A., 2013, A snowflake’s chance in heaven, arXiv: 1306.5587v1