04 December 2002  Total Solar Eclipse

Sunset Expedition to Fort Grey, Australia

Africa vs. Australian Coast vs. Australian Outback

The path of the total solar eclipse of 04 December 2002 spans two continents, Africa and Australia. The maximum duration of the total phase of the eclipse in Africa exceeds the appx. 1/2 minute one would experience in Australia, a factor of three on the Mozambique coast as the lunar shadow heads out over the Indian Ocean.  And, the maximum elevation of the Sun at mid-eclipse in Australia (at the coast) is only 9-degrees, and declines as the path progresses inland toward the outback, compared to a maximum of 45 degrees on the east African coast.  If duration and solar elevation were the two principal factors in selecting an eclipse observation site the choice would be obvious.  However, these are not the overriding criteria.

To maximize the probability of seeing an eclipse consideration of the climatology of the target areas, and the need for maintaining mobility and flexibility, are paramount.  Noting, of course, the caveat that "climate is what you expect, weather is what you get".  The later is particularly true as localized weather phenomena in regions affected by the decline of solar insolation during an eclipse are not well represented by statistical examinations of climatalogical (or meteorological) data.  None-the-less, it is quite clear that potential for near-sunset clear skies the inland regions of Australia at that time of year  exceed those found along the track anywhere in Africa.

It has been suggested that in a few isolated spots on the African continent, such as near Beithbridge in Zimbabwe, that the chance of "clear skies" might be as high as 60% based upon the normal mid-morning "mixing" model for cloud convection and formation.  During this time of year, the onset of the rainy season with higher humidity, cold-air advection brought on during the eclipse muddies the waters (if I may mix metaphors, if not the atmospheres) and the likelihood of cloud formation would increase over the nominal and usual scenario.  It would be wise for anyone considering an African observing venue to remain aggressively mobile. But, as many of us learned first-hand in our treks to the 21 June 2001 eclipse, where the path was very similar (in Africa) to that of this years event, mobility is particularly "challenging".  And, given the recently accelerating instability in Zimbabwe (in particular) this, sadly,  could make a less than desirable choice for foreign travelers.

In contrast, the inland regions of Australia eastward of the Torrens salt lake bed offer very dry conditions during the austral summer season, with mean cloud cover reaching as low as 20% (and best at the eastern extremedy of the path). On the Australian coast cloud cover may be "expected" at about 50%, though here too convectionally driven cooling of the late afternoon moist sea air would likely increase the likelihood of coastal cloud.  As one moves eastward along the path of totality in Australia the altitude of the sun declines, so the "likelihood" of a cloud along the line-of-site increases, which would bias these statistics.  When the low solar elevation is taken into consideration, inland sites are still significantly preferential to the coastal region.  For these reason, evaluating the likelihood of a view of the sun unobscured by intervening cloud, the eastern end of the path of totality in Australia "shines" as the preferential location for an eclipse viewing site.  With the ability to flee westward (i.e., remaining aggressively mobile) to react in situ to local weather conditions, this is where I will head for the 04 December 2002 eclipse.

Quality, Not Quantity

For some eclipse chasers the mantra is "duration, duration, duration".  For me, like a real estate agent (which I am not) it is "location, location, location".  At this point it would be useful to look at a more detailed MAP OF THE PATH OF TOTALITY OVER AUSTRALIA (2.3Mbytes, patience please). Electing to observe from Australia path means  accepting a duration of totality of approximately half a minute.  From the coast, near Ceduna, the centerline duration of totality is about 33 seconds.  At the  (refraction corrected*)  sunset point, near Tickalara, it is about 24 seconds.  One can look at this as either a half-empty, or half-full glass, depending upon your philosophy of life.  I.e., this can be viewed as a "loss" of nine seconds (27% of the duration of totality at Ceduna), or as a weather-driven opportunity to witness a total solar eclipse at (or very near) sunset.  This is something I have always wanted to experience, purely from the esthetic (some may say ethereal) perspective, but was unwilling to offer up a sacrificial eclipse.  Indeed, for all other eclipses I have observed the regions at or near the sunset (or sunrise) points were always contraindicated due to weather (as well as, secondarily, duration).  To see the totally eclipsed sun just over the horizon, looming large by the "moon illusion", in a conical wedge of darkness from the relatively narrow (appx. 25km wide) umbral shadow stretching through a nearly horizontal plane in the atmosphere is a tantalizing thought.  Then, at third contact the "lower" edge of the sun approaches the horizon near the point where the diamond ring from the soon-to-set very thin photospheric crescent will appear, and the umbral shadow lifts straight "up" into space.  Yup, I'll be there.  Any other takers?

* The position of the sun and moon in the sky [and hence the projection of the moon's shadow on the Earth] are affected by refraction through the atmosphere.  In computing local circumstances at the sites considered here I have applied a simple refractive "correction" model, which I have previously described HERE.  Such corrections are generally very small, but cannot be neglected for very low solar elevations.  This model used, however, is likely inaccurate at high precision at the most extreme eastern end of the path (near Tickalara) since the temperature (and pressure) profile of the atmosphere will likely be different than considered.  Further refinement is in work.

TOPOGRAPHIC MAPS (250,000:1), Path Lines and Local Circumstances

To facilitate discussion, the path of totality over australia has been overplotted on 250:000:1 topographic charts.  To see/download a chart click on the longitude range of interest.  Northern limit, Southern limit, and Centerline have been indicated (annotated red lines) on all charts.

LONGITUDE RANGE       CHART NAME   AREA COVERED (download size)
133° 00'E - 134° 24'E (Streak Bay) Denial Bay, Ceduna (1.7Mb)
133° 30'E - 135° 02'E (Childara)   Kalanbi, middle-of-nowhere, lots-of-nothing (4.5Mb)
135° 00'E - 136° 35'E (Gairdner)   Lake Gairdner NP, Wirraminna (4.9Mb)
136° 30'E - 138° 03'E (Andamooka)  Purple Downs, Andamooka, Lake Torrens (4.7Mb)
138° 00'E - 139 °14'E (Gammon)     Lyndhurst, Mt. Lyndhurst (3.8Mb)
139° 30'E - 141° 02'E (Callabona)  Mt. Hopeless, Lake Callabona, Cameron Corner (4.7Mb)
141° 00'E - 142° 30'E (Milparinka) Fort Grey, Strut NP (5.2Mb) {entire chart}
141° 00'E - 142° 30'E (Tickalara)  Old Naryilco, Tickalara (2.2Mb)

Details: The overplotted (red) limit and center lines on these charts have been computed conservatvely in accord with the system adopted by Espanak and described in NASA TP 2001-209990.  In particular, for this purpose only:

(a) a value of k=0.272281 is used, though this is in disagreement with the value of k=0.2725067 adopted by the International Astronomical Union.
(b) no corrections are made for the displacement of the lunar center of figure with respect to its center of mass.
(c) no corrections have been made for atmospheric refraction.

Thus the red center and limit lines on these charts should be in very close agreement to Tables 7 and 8 in the TP 2001-209990.

Elsewhere, in computing local circumstances for the particular places as discussed and tabulated on this page (and in links to this page):
(a) the IAU adopted value of k=0.2725076 is used.
(b) a correction of (+0.50", - 0.25") has been applied to the latitude and longitude of the moon to compensate for the displacement of the mean center of figure from the dynamical center of mass.
(c) the displacement of the path due to atmospheric refraction has been compensated for by a simple approximate correction assuming a "standard" atmospheric model and STP at seal level.  This approximation is likely to loose high fidelity at very low solar elevations for in situ air-path profiles which deviate from the assumed model.

The net effect of applying these corrections (b) and (c), and using the IAU value of "k" is illustrated on the Gammon, Calabona and Milparnika charts by the dashed black lines "above" and "below" the limit lines computed as per TP 2001-209990.  To generalize, applying these corrections results in limit line mean displacements of displacement perpendicular to centerline of appx. 0.6' to the north and 1.6' to the south in these regions (though, obviously, dependent upon topographic location).  The difference is small, but not negligible at/near the northern/southern limits.  For example, the airstrip at Binerah Downs (Milparnika topographic chart) is about 1km outside of southern limit by the TP 2001-209990 convention, but about 1km inside of southern limit as computed with the corrections described.  In either case, further location dependent corrections due to irregularities in the lunar limb profile (of about these spatial amplitudes) must be applied when high precision predictions are required.  The primary effect of (c) is to displace the path along the anti-sun azimuth and "extend" the path eastward by about 60km.
 

The "Plan" - from the ground up

Last year our small but dedicated band of die-hard Umbraphiles had such success in Zambia, that I advocated something similar for Australia.  Again, no touristy "if it's Tuesday it must be Melbourne" and "oh, by the way there is an eclipse" whirl-wind sightseeing trip with a lot of tag-alongs for whom the eclipse is of secondary import.  No.  Rather, a nucleus of us, the enlightened, those who advocate "eclipse or bust" and will stand by that until the last coronal photon wanes.  Of course we would welcome a "first timer" or two, if they are truly motivated by the eclipse.  Clearly some of us would want to avail ourselves of the opportunities to explore "down under" but not at the expense of, or until after, the Champaign cork has popped, the egg creams have fizzed, and we are scratching our heads about how we get to Antarctica in 2003.  So from hippos to 'roos.  Are you game?  The game is indeed afoot.

The "plan" in detail is still evolving, but in concept is now relatively mature. Michael Gill and I had been iterating this for quite a while and we were driven (motivated) by a desire to observe totality from very near sunset at centerline.  Recently, we fell in with Joseph Cali, an Aussie some of you may know from the Solar Eclipse Mailing List, who has first-hand familiarity with many of the logistics and transport "issues" which leap to mind.  And, Joel Moskowitz has said "count me in", and thus, this is now our small band of die-hards heading for the hottest place I know of outside of Tucson. By the way by "very near the horizon", I mean we target for the Sun still being above the horizon at third contact, by about 1-1/4 degrees.  Less than that, well, we're not absolutely crazy (though some may disagree) - but read on.

Before getting too specific (which I will) it is worth looking at the large scale path map now if you have not done so. This is a "road map" of the region covering the path of totality with northern and southern limits overplotted.  The path (which has been corrected for atmospheric refraction) "ends" at sunset just NE of Tickalara.  You will note a number of blue "+"'s in the path.  These are airfields in the path of totality, more on that later too.  You probably have noted the paucity of roads and the sparseness of the road network at the eastern end of the path.  Thus, the "basic" plan is that we will converge in Canberra (where Joe Cali lives) on November 30th.  There we will pick up (at least) two 4WD vehicles.  Joe has his own, and we are going to rent a camper van, either this model or something similar to travel to our "prime" pre-selected viewing site, which is on centerline near Fort Grey.  Two vehicles, of course, is a minimum. One would be downright fool hardy as we cannot risk a vehicle breakdown stranding us outside of the path of totality.  If more people (within limits!) end up joining us in this venture we will up the number of vehicles.

Conservatively, we would make our way to and  base in Port Augusta (or Broken Hill) two days before the eclipse.  IF at that time of our  departure to Fort Grey there is a clear and unambiguous reversal in anticipated weather conditions we would head westward toward Ceduna. IF not, however, we would take the following route, with drive times noted by Joe Cali (who has provided very helpful comments  regarding 4WD travel in the area which should be seen by anyone considering overland travel along the path of totality):

a) Port Augusta SE to Peterborough
JC: 120km About 1.5 .- 2 hrs
b) Barrier Highway NE from Peterborough to Broken Hill
JC: Good condition, sealed & fast 270 km 3 - 5 hrs
c) Silver City Highway N from Broken Hill to Tibooburra

Tibooburra is our "gateway to totality" (please be sure to see this link for more information).  It is one of (if not the) most remote, and hottest, towns in Australia.  Tibooburra is outside of the path of totality, but is the logistical "local" staging point for our final assault.  .

JC: Distance 340km. Time 4- 6 hrs. Condition - pretty good and sealed part of the way. Conditions get a little worse the further north you go. In a top heavy camper 4wd you probably shouldn't travel over 60-70km per hour once you hit the unsealed roads.

d1) Tibooburra to centerline near Fort Grey (about two hours drive time).

{Joe also provides additional drive time info for a more direct routing from Canberra.}

FORT GREY.  Fort Grey is in Sturt National Park. The elevation of the Sun, there,  is 1.3 degrees above the horizon at mid-eclipse and might offer a bit more comfortable "breathing room" against any horizontal obstructions than sites further East. Joe Cali had summarized "I checked the maps. The airstrip and the point where the road crosses the centerline have low horizons. FG airstrip: 10m high sandridges 11 kilometres away @ az = 243deg, Ft Grey: 10m high sandridges 15 kilometres away @ az = 243deg. There will be similar 10m sand ridges nearby." and indeed the Milparinka topographic chart confirms this.   We could climb the ridges for a bit of extra elevation if necessary. This site is accessible by a road and may be seen on a larger area road map also showing its location w.r.t. Tickalara.  Note: From centerline, 5' to the north of the Fort Grey landing strip along the road landing strip along the road, the duration increases by 3 seconds (to 26.2 seconds prior to limb corrections).

Local circumstances (left) for the Airstrip at Fort Grey, which has two 900 x 30 runways, heading 8/26 and 3/21, (information graciously provided by: Bill Harvey, Public Affairs, Department of Transport & Regional Services, (02) 6274 7860, Bill.Harvey@dotrs.gov.au, and Gary Oldman, Information Technology Manager, Royal Flying Doctor Service (SE Section), Ph  08 8080 1718, Fax 08 8088 4536, gary.oldman@rfds.gov.au). The somewhat off-centerline airstrip is in the path of totality and located at: Longitude = 141° 10' 24"E, Latitude = 29° 06' 48"S, with the advisory "sand hills in vicinity of strip".

d2) Continue North (unimproved?) road to centerline just east of Old Nariyilko

OLD NARIYLKO. "Bottom Narrabappa Tank", on centerline near Old Nariylko and is  is a possible viewing site.  However, with the sun only 1/2 degree above the astronomical horizon , and < appx. 1/5 degree above the local topographic horizon, at mid eclipse. That was a bit too close for us, and we vetoed this site in favor of the slightly more westerly Ft. Grey.  Do see the information on this, the Tickalara/Old Nariyilko/Bottom Naranappa site, though.  And, also, a 250:000,1 topographic map of the sunset end of the path of totality near Tickalara.

The Backup - The Escape (Plane)

Once at a specific site in the path of totality, particularly further toward the eastern end of the path, an overland "escape" to avoid unexpected clouds on eclipse day would be difficult.  The rate which one could back-track to the west by going overland off-road (with a 4WD of course) is  likely < appx. 10km/hr.  Indeed, fleeing even a "modest" distance from Fort Grey, off-road could take much too long.  We could backtrack through Tibooburra (see this MAP too) to reach another centerline (or at least in-the-path site, but we really would want to be able to move further faster.  Hence: the escape plane -  a light aircraft standing by at, or very near,  our PRIME observing site capable of transporting us from there to another in-the-path site on very short notice.  The plane would be pre-positioned at the airstrip at (or very near) our prime site and we would use it only if necessary to fly us to an alternate, pre-selected, site to escape clouds.  Of course this would mean leaving our vehicles at the prime site (or airstrip close to it) and observing from an airstrip where would would land as we would have no local transportation there.  If necessary we would fly westward back to the coast at Ceduna (and hence loose the "sunset" aspect of the eclipse, but a price worth paying for circumventing a cloud-out.).  Additionally, if the entire path were significantly "threatened" by low-cloud (or of low-cloud near the horizon erupts at our site very near third contact, we would use the aircraft for an airborne observation.

AIR/GROUND COORDINATION: We likely will arrange air charter out of Broken Hill, and in at least one scenario will be there on December 2nd.  Joe has summarized some of the options we are considering and has provided additional useful information HERE.

Aircraft Accessible Alternate Sites

The "alternate" air accessible sites in the path are limited.  I have attempted to locate all airstrips the path of totality (and a few within a few km of the limit lines, but outside of the path) on the 250:000:1 topographic charts and have computed local circumstances for those locations.  I have received specific information on the Tickalara and Fort Grey airstrips but further input, particularly from local pilots, property managers, or responsible government agencies would be most appreciated, and I am attempting to make such inquires.  Below, I list these sites and also give the U.T of mid-eclipse, duration of totality, and altitude of the Sun from these airfields.  The durations of totality are corrected for the path shift from a canonical model of atmospheric refraction, but are not adjusted for lunar limb profile effects.  The airfields may be located on extracts from topographic charts (along with additional local circumstances) by clicking on the site names below:

NAME/ID             E. Long.   S Lat.    Mid U.T.   Dur(s) Alt(d)
TICKALARA           142d 12.1' 28d 40.3' 09:11:38.7 22.6   0.3
BINERAH DOWNS EAST* 141d 33.2' 29d 01.2' 01:11:34.0 08.1   0.9
BINERAH DOWNS WEST  141d 29.8' 29d 01.5' 09:11:34.9 15.8   1.0
FORT GREY           141d 10.6' 29d 06.8' 09:11:36.2 23.1   1.3
WHITE CATCH COTTEGE 140d 59.2' 29d 13.0' 09:11:42.8 21.6   1.4
LAKE CALLABONA NW   139d 56.3' 29d 34.2' 09:11:33.9 27.7   2.4
MT. HOPELESS        139d 44.5' 29d 35.3' 09:11:35.8 26.7   2.6
BONFIRE BORE        139d 08.2' 30d 01.4' 09:11:32.7 17.8   3.3
MOUNT LYNDHURST     138d 42.6' 30d 10.1' 09:11:22.3 23.7   3.7
LYNDHURST           138d 20.6' 30d 17.0' 09:11:21.2 27.6   4.0
WIRRAMINNA          136d 13.2' 31d 11.3' 09:10:57.9 27.7   6.2
COONDAMBO**         135d 51.9' 31d 03.2' 09:11:07.1 04.4   6.4
LAKE GAIRDNER NP    135d 01.2' 31d 44.0' 09:10:38.9 20.6   7.4
CLEARVIEW STUD(?)   133d 59.8' 32d 06.5' 09:10:24.6 28.6   8.4
CEDUNA              133d 41.0' 32d 07.0' 09:10:25.3 35.5   8.6

* When the path is corrected for atmospheric refraction (and using a value of k=0.2725076 in agreement with the IAU adopted value, where [k sin pi] is the sine of the apparent lunar semidiameter, pi is the lunar horizontal parallax) the Binerah Downs East airstrip is north of southern limit defined by a smooth lunar limb.  The refraction corrected southern limit is shown in the linked topographic chart. Once limb corrections are applied, complete photospheric extinction is not expected from this site.
** At Northern Limit.
 

Finally, very near sunset, there are two airstrips at Naryilco just north of northern limit. Naryilco (a), Lon = 141deg 53.5minE, Lat = 28deg 34.5minS; Narylco (b), Lon = 141deg 55.0minE, Lat = 28deg 32.5minS.  From the more southern one (a), a 1 mile walk will bring you into the path of totality.

In the even of low-level obscuring cloud over the entire path, or "last-minute" cloud on the horizon at the prime site, the "escape" aircraft could be used to observe the eclipse from the air.  We are not talking a high altitude jet as we used in in 1986 or 1992, but here a small prop could save the day.

Ready, Set...

So, now our Sunset Expedition is a band of four hardy Umbraphiles.  I suspect not too many eclipse chasers would plan an observation only with the sun only 1.3 degrees above the horizon. But given the weather prospects, the weather contingencies we have put in place (and, if the weather dictates we could end up in Ceduna anyway where the vast majority seem to be headed), and this is something I have always wanted to do, it really could be glorious.  So my 23rd total solar eclipse may indeed be the "low point" of my eclipse-chasing endeavors.  I hope so, and I'll savor every of the 25 seconds of it.  If you are similarly inclined do let me know.  We are very seriously into the planning and arranges stages regarding resources, transportation, infrastructure, local contacts, etc. It would be wonderful to watch the umbra disappear at 09:11:48 UT "straight up" into space, and kiss it goodbye until we rendezvous with it again in the Antarctic.

A Few Useful Links to Other Sites:
 

Air Charter :

Glenn Schneider
This page is subject to change without prior notification.
Last update: 24 January 2002

Take me to Glenn Schneider's Home Page at Steward Observatory
Take me to Glenn Schneider's Umbraphile Page.