October 1%6 J. R. Blake 599 THE FAIRBANKS HALO OF APRIL 27, 1966 J. R. BLAKE Geophysical Instifufe, University of Alaska, College, Alaska ABSTRACT A description of a magnificent display of solar haloes visible over Fairbanks, Alaska, is presented and a photo- graphic rccord of thc nnthclic arcs is uscd to compare these arcs with the two principal theories of Wegener and Hnstings. The temperature range and interval of possible crystal formation is inferred from the halo forms and compared with aerological data at the time of the display. 1. INTRODUCTION of the ice crystals present, and on the solar altitude, h,. We refer the reader to the literature 12, 3, 5, 6, 8, 14, 16, The appearance of extensive displays of solar haloes is 17, 19,201 for further details, particularly to the exhaustive an uncommon event in most parts of the world away review by Visser [19], who describes over 50 different from the polar ice-caps. The d'splay reported here forms. The ice crystals producing the haloes usually constitutes one of the more impressive, appearances, occur in high, cirriforrn cloud, but sometimes, especially comparing favorably with the classics of history, such as over the polar ice-caps, the crystals occur very close to those described by Pernter and Exner [IS] and those the ground [4, 11, 211 giving rise to exceptionally brilliant analyzed by Visser [19]. It was visible from the general displays. vicinity of Fairbanks, Alaska and the observations here reported were made from nearby College (64'51' N., 2. DESCRIPTION 147'50' W.), between 1300 and 1530 Alaskan Standard Time (UT-10 hr.), on April 27, 1966. The accompanying photographs show the display The time of the first appearance of the haloes is not recorded by a Panon Widelux camera with a 140' horizon- known, but they were visible by 1230 and certainly well tal view (fig. l),and by an all-sky camera (fig. 2). The developed by 1300 AST, appearing in very filmy cirriform cloud covering most of the sky. The particular arcs that 1 The all-sky camera is designed for auroral photography. It consists of a convex mirror reflei tlng the entire sky via a plane mirror into the camera. The radial scale is not linear appear in such events depend on the type and orientation or cosine. FIGURE1.-Display seen by Panon Widelux camera with 140" horizontal view. High Speed Ektachrome (160 ASA) f:ll, 11250 sec. Unauthenticated | Downloaded 09/27/21 09:19 AM UTC 600 MONTHLY WEATHER REVIEW Vol. 94, No. 10 I FIGURE2.--Display seen by all-sky camera, with Pentax camera, 55-mm. Tachomar lcns, High Speed Ektnchrome at 1/500 sec. various arcs are identified in the following description by tween the 22O-parhelia and the sun, in comparison with its reference to figures 3 and 4. appearance elsewhere, even toward the anthelic point. At 1300 AST (h,=37.6'), the display was essentially that Superposed over the parhelic circle were both 120'- shown in figure 3. The most brilliant haloes were the paranthelia (lo), appearing as bright white knots of light, white, complete parhelic circle (4) , the colored 22O-par- with no arc estensions. At 1325 AST (h,=36.5') the helia (3) and their white tails superposed over the parhelic azimuth of these was measured at approximately 119' by circle, and the npper portion of the circumscribed halo (5) Prof. W. Mendenhall of the Engineering Department of and its intense white iTei1. As is frequently the case, the this University, using a surveyor's level, He also meas- parhelic circle was appreciably weaker in intensity be- ured the azimuth of &he parhelia as approximately 33' Unauthenticated | Downloaded 09/27/21 09:19 AM UTC October 1966 J. R. Blake 691 FIGURE4.-Sketch of display at 1510 AST (h,=28.8"). At this first observation, the 46"-halo (2) was not present at all, but the infralateral tangential arcs (8) were very bright and coIored, especially at their point of closest approach to the sun (the point of "tangency"). By 1320 AST (h,=36.2"), the 46'-halo was very weakly visible, and the points of minimum approach were indeed points of tangency. FIGURE%.-Sketch of display for period 1300-1430 BST (h,=37.6"- Extending below the parhelia (the focal lines of which 32.2'). Based on all-sky camera photograph by T. Ohtake. appeared to be inclined at some 60" or so to the parhelic circle, rather than normal to it) could faintly be detected the arcs of Lowitz (7), appearing as nearly straight lines, from the sun, at the same solar elevation, a figure agreeing perhaps slightly concave toward the sun, and extending excellently with that deduced from Wegener's [20] equa- toward the 22'-ring as sniooth extensions of the parhelic tions (33.1'). inclination. The arcs merged into the general glare and The circumscribed halo (5) was visible in two parts-the the point of contact could not be determined; likewise upper and lower contact arcs to the 22'-halo. Both parts the relative weakness of the arcs combined with the were very bright and colored, the lower one being con- overall glare prevented any determination of color. With siderably more diffuse than the upper one. The non- the exception of the parts very close to the parhelia, minimal refractions (analogous to the tails of the parhelia) the arcs did not persist clearly for very long, slowly produce a white haze on the outside of the focal line of this merging with the background illumination. halo, being exceptionally intense in the upper and lower Very faintly visible at this stage were portions of white solar vertical where the arcs are concave away from anthelic arcs (ll),converging approximately toward the the sun and the "tails" thus convergent. This accounts anthelic point, but not reaching it. The projection for the diffuseness of the sharply curved lower arc. The toward the sun was much more difficult to determine and upper arc did not at first reach the parhelic circle (4) but no decision could be reached as to a point of convergence later did so in a faint extension. The lower arc showed except t,hat it could not be much higher than the 22"-halo sufficient extension for the change of curvature to be in the solar vertical; it could however, have been as low detected, but did not reach the turning point back toward as the sun itself. The eastern arc was especially stronger the parhelic circle. than the other. The 22O-halo (1) was very weak, though always detect- By 1320 AST (h,=36.2") the 46'-halo had appeared able throughout the entire display. At first sight it ~eakl~7,in sections, strongest in the solar vertical. The appeared that the portion in the solar vertical was brighter, whole halo was faintly visible for a while. The anthelic whiter, and was intersected by the brilliant upper contact arcs (11) developed more strongly and extended farther arc. As the solar altitude decreased however, this portion toward the anthelic point, quite definitely converging separated from the 22O-hal0, and it was evident that Par- toward it although again not reaching it. They still ry's arc (6) was present, at first being very close to the showed no evidence of color and the extension solarward 22O-ring. It was later lost in the glare from the contact was still insufficient to define the point of convergence or arc "veil". Wegener's [20] theory shows a separation of mergence. <lo at a solar elevation of 37.6' and the observations are The whole display faded appreciably by 1345 AST thus in good agreement. (h,=35.4") , particularly the anthelic arcs and eastern Unauthenticated | Downloaded 09/27/21 09:19 AM UTC 602 MONTHLY WEATHER REVIEW Vol. 94, No. 10 120O-paranthelia. By 1430 (h,=32.2') however, the quite bright. Since the major part of Parry's arc is pro- anthelic arcs and 12O0-paranthelia reached their maximum duced by ice-columns with principal axes oriented from intensity and extent. The lower circumscribed halo, about 45" to 90" to the incident light, crystals producing arcs of Lomitz, and eastern infralateral tangential arc had a predominance of end reflections (i.e., anthelic arcs) over by this time gone completely, the western infralateral pure refraction (i.e., Parry's arc) would need to be short; arc and upper circumscribed halo were much weaker and however, short crystals would certainly destroy the very very diffuse, Parry's arc being lost in the diffuseness. The special orientation required for the production of Parry's 46"-halo existed faintly in the solar vertical. arc (a pair of prism faces must remain horizontal, thus By 1510 AST (h,=2S.S0) the display mas considerably presenting maximum air resistance and requiring long more diffuse and weaker ; however in the solar vertical, crystals and no turbulence). It thus appears that the a short, colored, but diffuse arc (9) was visible (see fig. 4), conditions for the production of Hastings' arc are mutually concave away from the sun, but in contact with the 46"- contradictory. halo which was still visible in this region. The point re- At least two other forms of anthelic arcs have been garding its tangency was specifically examined and it was reported. One (see Liljequist [11]) is not involved here. noted down as definitely tangential. Later calculation The other, however, frequently appears in observer's showed that the circumzenithal arc of Bravais (cf. [19]) reports as intersecting the sun [I, 161, although it is possi- should have been separated from the 46O-halo at this solar ble that some, at least, of these result from inaccurate elevation by about 2%".