2013年11月3日混合食天气估计
Weather Prospects for the Hybrid Eclipse
2013 November 3
Eclipses are wonderfully predictable events whose predicted coordinates are the culmination of three millennia of astronomical and mathematical discovery. Utilizing the number-crunching clout of computers, eclipse tracks and times can be reliably calculated for centuries into the past and future, and with less accuracy, for millennia in both directions. For the upcoming hybrid eclipse in November, 2013, the sweep of the Moon’s shadow can be located to within a kilometre in position, and to within a part of a second in time. Only the weather remains uncertain. The track of the 2013 hybrid brings an imposing meteorological challenge. Its beginning seems promising, in the relatively sunny skies beneath the sub-topical highs of the Atlantic, south of Bermuda (Figure 1, below). It is an annular eclipse in this region, and so most enthusiastic eclipse travellers will elect for a position farther down the track where the Moon covers the entire solar disk and the eclipse becomes total. Alas, as annular turns to total, the track moves out from under the favourable anticyclonic skies of the mid Atlantic and into the influence of the Intertropical Convergence Zone (ITCZ). The ITCZ is a region where winds from the Northern and Southern Hemispheres converge, squeezing the humid tropical air upward, and germinating a band of frequent heavy showers and thundershowers that girdle the globe.
http://home.cc.umanitoba.ca/~jander/tot2013/cloud1.png
Figure 1: Map of average cloud amount in tenths with the eclipse track superimposed. Data CIMMS/NOAA.
The shadow path is completely over water for the first two-thirds of its length, and then comes ashore in equatorial Africa for the remainder of its transitory journey across the Earth. Since more than half of the track – about 60 percent – is within the influence of the ITCZ, clear-sky prospects are generally meagre. From the satellite-based cloud-cover charts in Figures 1 and 2, four spots show as being relatively more promising than the rest: a location on the water near the point of maximum eclipse; a site in northwest Uganda; one in north central Kenya; and a fourth in southern Ethiopia.
The satellite-derived cloud cover charts are not exactly comparable to measurements taken from the ground by humans and machines. Observers and machines record from one spot, with an oblique view toward the surrounding horizons. Satellites observe from altitude, but also have oblique views toward the horizon, which lies much farther away than for a ground observer. Machines see only the cloud directly overhead and assemble areal measurements by keeping track of how the overhead sky changes with time. Detection of cloud from satellites relies on complex algorithms that compare infrared- and visible-light radiation measurements to make a judgement about the extent of cloudiness in the image pixels. Both ground-based and space-based observations are concerned with cloud amount, while the important parameter for eclipse-watchers is the opacity of the cloud - does it block the Sun? In the extreme, a sky that is covered by transparent cirrus cloud could be observed as overcast (in actual fact, it would be recorded as "thin overcast") though it provides little impediment to viewing an eclipse. The upshot is that satellite and surface-based observations are probably biased to report a heavier cloud cover than actually exists. The cloud charts in Figure 1 and 2 should then be used to compare sites with each other rather than to derive an exact probability of seeing the eclipse. That statistic is best represented by the percent of possible sunshine in Table 1.
At the Point of Maximum EclipseMaximum eclipse (1m 40s) occurs at 3.904°N 11.694W, 325 km SSW of the Liberian capital, Monrovia. Figure 1 shows that the cloud cover is 10 to 20 percent lower than areas to the east and west along the track, making it an attractive alternative to land-based expeditions. Most of the cloudiness in this region comes from large clusters of thunderstorms and other types of convective cloud. Examination of individual satellite images around the location of the maximum eclipse showed a highly variable cloud climatology from day to day. On some days, the cloud is barely present, with only cumulus and small shower clouds dotting the ocean. At other times, the cloud shield is large and solidly overcast, requiring considerable movement to escape into sunshine. Encouragingly, in no case was the track ever completely clouded out.The main advantage of a site near the point of maximum eclipse comes from the easy mobility of a ship, which can move out from under the cloud shield and into more promising weather. Because this may require a considerable diversion, the vessel should have a relatively flexible schedule and access to routine weather forecasts.
http://home.cc.umanitoba.ca/~jander/tot2013/cloud2.png
Figure 2: High-resolution map of average monthly cloud cover in tenths across Africa, based on 24 years of satellite measurements. Data CIMMS/NOAA.
Gabon, Congo, and the Democratic Republic of the Congo
Gabon, where the eclipse track first comes ashore, is a tempting destination for eclipse watchers who want a stable site for photography and the longest possible eclipse (slightly over 1 minute). Alas, when the track leaves the water and crosses onto land (Figure 3), two additional cloud-making factors come into play. One is the greater daytime heating of land surfaces compared to water; the other is a general increase in humidity as a result of the transpiration of water vapour by the jungle vegetation in this region of the globe. Both contribute to the destabilization of the atmosphere and the formation of convective clouds. Moreover, the track across equatorial Africa lies atop the November position of the ITCZ, and so the region is one of the cloudiest on the globe at this time of year. Eclipse chasing is a bit of a gamble, but it is no means a lost cause.
http://home.cc.umanitoba.ca/~jander/tot2013/gabon_s.png
Figure 3: The path of the central eclipse across Gabon. Click on the map for a larger version.
Annual precipitation over Africa comes in two rainy seasons that follow the migration of the ITCZ, to the north in March, April, and May and southward from October to December . From Gabon across The Congo and the Democratic Republic of the Congo (DRC), the rainy seasons are approximately equally intense, peaking in March/April and October/November (Figure 4), depending on the location. In the intervening months (December/January and June/July), rainfall usually drops significantly to create a dry, or at least, less wet, season. The spring rainy season is sometimes known as the “long wet” and the fall, the “short wet,” as much for their duration as intensity.
http://home.cc.umanitoba.ca/~jander/tot2013/pcpn1.jpgFigure 4: Annual precipitation patterns along the western half of the African portion of the eclipse track.
Under ordinary circumstances, sites along the coast will derive some benefit from the proximity to the ocean and the cooler waters that inhibit the growth of convective clouds. However the waters off Gabon lie at the confluence of the cold Benguela Current from the south and the warm Guinea Current from the north, creating an extensive cloudiness that is readily apparent in the cloud cover pattern in Figures 1 and 2. Nevertheless, the difference between the cloud amount over land and that over water revealed by the high-resolution contours in Figure 2 does seem to promise at least a 5 percent advantage for a location in Gabon immediately adjacent to the coastline.
The shadow comes ashore in a relatively uninhabited part of Gabon and travel to the centreline will pose a challenge for larger expeditions. The major cities in the area, Port Gentil and Libreville, are a considerable distance from the axis of the eclipse track and overland travellers must skirt rivers and river estuaries to achieve maximum eclipse - essentially a full day’s travel. Another option, though one that results in only a 30-second eclipse, is to go north from Port Gentil to nearby Cape Lopez, a position easily reached by road, as it is only 13 km from the city to the edge of the eclipse track and 16 km to the farthest reaches of the peninsula (which lies 2 km inside the south limit). Fine sand beaches in the area will provide a pleasant eclipse-watching site, if not a long one. For those committed to the centreline, transportation by boat along the coast may be a more convenient option than going overland. From Port Gentil to the point where the centreline first comes ashore is a distance of about 60 km on the water.
In the interior of Africa, there are small regions in Gabon and the Congo where cloud cover is also slightly lower (Figure 2), but in general it could be said that the prospects from the Atlantic coast to the border of Uganda are very similar: pessimistic. To determine more precisely the prospects of seeing the eclipse, we turn to observations from land-based weather stations (Table 1, below). The best statistic for evaluation the eclipse prospectus is the “percent of possible sunshine,” a measurement derived from actual observations of sunny hours at the observing sites. Port-Gentil on the coast and Lambarene in the interior are virtually identical in their measurements (31 versus 32 percent), but Lambarene may be a more convenient choice for an eclipse-watching base, as it lies a mere 14 km from the south limit and 43 km from the centreline along highway N1. Some help in reducing cloudiness will come from the falling temperatures as the shadow approaches. This cooling will inhibit the growth of at least the smaller convective clouds, and so the chances of seeing the eclipse might be increased to about 35 or even 40 percent from the 32 promised by the sunshine statistics in Table 1. Also tempting, but less accessible, is Wonga Wongue Reserve, which sits astride the eclipse path.
Very similar prospects to Gabon are likely in the Congo (Brazzaville). Sunshine measurements are not available there, but the cloud measurements at Makoua (which lies within the track of the shadow) suggest that grey skies are equal to or perhaps a touch less than at Lambarene. In the Democratic Republic of the Congo (DRC), Mbandaka has a slightly lower amount of cloud (based on a limited number of years of observation) according to surface-based observations, but this is not confirmed by the satellite measurements. Collectively the three countries are all very cloudy with similar prospects for a view of the eclipse.
The obvious difficulty with climatological frequencies such as those displayed here, is that the weather is much more variable than the climatology. Invariably, some areas with high average levels of cloudiness will be sunny on eclipse day, and some with good climatological prospects will be cloudy. To reinforce this point, the satellite images in Figure 5 offer an eclipse-time view of central Africa for November 3 in 2010 and 2011. Almost all of the cloudiness is convective with areas where the air is rising to form clouds while compensating adjacent regions have sunnier weather where the air is sinking.
http://home.cc.umanitoba.ca/~jander/tot2013/satpix1.png
http://home.cc.umanitoba.ca/~jander/tot2013/satpix2.png
Figure 5: Satellite images from November 3 2011 (top) and 2010 (bottom) before the time of the eclipse (i.e. 12 UTC) showing characteristic cloud patterns along the track of the lunar shadow across central Africa. Most of the clouds in this image are convective or derived from convective clouds. In particular, a string of growing thunderstorms can be seen along the eastern border of the Democratic Republic of the Congo, beginning just south of the shadow track, in the lower image. Because the time of these images is one to three hours ahead of the eclipse, cloud cover could be expected to be a little heavier than shown in these examples.
Table 1: Surface-Station Observations along the Eclipse Track
http://home.cc.umanitoba.ca/~jander/tot2013/table1.png
Table 1: Climatological statistics along at sites along the eclipse track. The "percent of possible sunshine" is the frequency of sunny hours in percent compared to the total time from sunrise to sunset. Sky condition is defined according to oktas (eights) of cloud cover such that "clear" = no cloud at all, "few" =1-2 oktas; "scattered" = 3-4 oktas; "broken" = 5-7 oktas and "overcast" = 8 oktas (no breaks at all).
Uganda and KenyaThe bimodal precipitation pattern that characterizes the rest of equatorial Africa falls apart in Uganda, with the monthly precipitation graph at Gulu showing a single broad rainfall maximum through the middle of the year followed by a short dry spell that begins in November and ends in March (Figure 6). While the dry spell really gets going in December, the downward trend in November is encouraging for eclipse viewing.
http://home.cc.umanitoba.ca/~jander/tot2013/pcpn2.jpg
Figure 6: Annual precipitation patterns over selected locations along the eclipse track in Uganda, Kenya and Ethiopia. The anomalous pattern at Gulu is due to the proximity of humid air from the Democratic Republic of the Congo and the higher terrain in the area. In Ethiopia, the eclipse track passes north of Mega.
The track across Uganda descends from the high mountains along the border with the DRC, crosses the flat plateau north of Lake Albert, climbs over a lower set of hills to reach Gulu, descends again through the middle of the country and then climbs another set of peaks to reach the Kenya border (Figure 7, below). This up-and-down track makes selection of a site in northeast Uganda complicated but the satellite measurements are pretty clear on the best site - a location north of Lake Albert near the community of Pakwach. The area is protected to the east and west by higher ground and the effects of the nearby lake to the south. The satellite-based cloud measurements in Figure 2 show a distinct minimum in cloudiness for the area. The nearest weather station, Arua, shows an average cloudiness of 71 percent, good for the season, but the centreline north of the lake is almost certainly sunnier than this, perhaps by 5 or 10 percent. The high levels of cloudiness reported at Gulu (Table 1) are out of character with the surrounding stations (though part of this is due to its location on the hills) at with the estimated percent of possible sunshine, and so our trust is probably best placed on the satellite record for site selection.Examination of the daily satellite images over several years centred on November 4 also suggests that the satellite record is to be preferred. Afternoons in northern Uganda, along the eclipse track, are bothered by thundershowers nearly every afternoon, but the region immediately north of Lake Albert seems to have a higher frequency of sunny skies when the landscape elsewhere is blanketed by convective cloudiness. In all likelihood the cool waters of the lake suppress the development of storms and so a small area near the town of Pakwach (Figure 7) enjoys an slightly larger amount of sunshine. The same is true of all of the lakes in Uganda and Kenya, but only Lake Albert and Lake Turkana (in Kenya) are close to or on the eclipse track. The advantage is small, but serves to help make northwest Uganda one of the more favourable spots for eclipse travel.The high frequency of convective clouds - especially thunderstorms - implies that November's rainfall probably comes from a few large storms that build in the afternoon and dissipate by morning. Convective clouds are very sensitive to ground temperature, and so the cooling that comes with the approach of the eclipse shadow will cause cumulus clouds to dissipate. Past experience indicates that any cloud buildup that is not precipitating will fade away about half way between first and second contact, but the late hour of the day for this eclipse will probably mean that the smaller clouds will have begun to fade anyway because of the normal diurnal cycle; the incoming eclipse will hasten the process. The road between Pakwach and Gulu, which approximately follows the eclipse track, will provide a bit of extra insurance by allowing mobile chasers to move along the eclipse track to a more favourable spot at short notice. The road dips out of the shadow path and then back into it, so movements should be made with a GPS receiver and an eye on the time available before first contact. The TravelQuest expedition in 2011 found the Gulu-Arua road (which goes through Pakwach) to be in excellent condition.
http://home.cc.umanitoba.ca/~jander/tot2013/uganda_s.pngFigure 7: Map of the eclipse track across Uganda. Click on the thumbnail for a larger image.
The centreline of the eclipse track north of Pakwach falls across the hamlet of Pokwero (Figure 8), situated on flat agricultural plains that provide no impediment to viewing the eclipse (Figure 9). The eclipse date was gloriously sunny when a TravelQuest scouting team visited the area in November 2011, even though thin high- and mid-level clouds stretched across parts of the sky. Travel is not difficult in the area, but other north-south roads are scarce between Gulu and Pakwatch, and so access to the mid-line of the eclipse is best achieved in the vicinity of Gulu or by heading north from Pakwach along the dirt-and-gravel road to Arua. One of the pleasant surprises during the 2011 scouting trip was the number of sunny days in spite of the contrary climatological record. Some of this could be attributed to the pessimistic nature of cloud observations, which do not clearly distinguish between thick and thin clouds.
http://home.cc.umanitoba.ca/~jander/tot2013/Pokwero.png
Figure 8: The hamlet of Pokwero, Uganda, looking south toward the centreline, in the middle distance.
http://home.cc.umanitoba.ca/~jander/tot2013/Pokero_centre_view_sm.png
Figure 9: Panoramic view to the west toward the location of the eclipsed Sun from a location outside Pokwero. The terrain is flat and largely devoted to agriculture.
As the eclipse track leaves Uganda, it descends abruptly onto the smoother terrain of northern plateau of Kenya, crossing Lake Turkana on its way to the Ethiopian border. The cloud-cover map in Figure 2 shows a region of low cloudiness through most of this region - in fact, the lowest cloud amount along the whole of the eclipse track, with values in the 65-to-70-percent range. Surface observations at Lodwar, only 60 km south of the centreline, are very encouraging, showing a surprising 80 "percent of possible sunshine" for November. The explanation for this sunny refuge lies primarily in the nearby presence of Lake Turkana, which has a dramatic effect on the development and growth of convective clouds in the region. Day-to-day satellite images frequently show a north-Kenya landscape dotted with showers and thundershowers, while the lake and nearby regions stand out, cloud free, in their midst. The region also benefits from the increasing distance from the mean position of the ITCZ, and even though precipitation patterns still show evidence of the double rainy season, the amount of precipitation is considerably lower than other sites along the Moon's path (Figure 6).
http://home.cc.umanitoba.ca/~jander/tot2013/kenya_s.pngFigure 10: Map of the eclipse track across Kenya. Click on the thumbnail for a larger image.
The area surrounding Lake Turkana in Kenya is the best land-based eclipse-viewing site along the track and the centre of activity will be at Lodwar (Figure 10), a rough town of 20,000 people with limited amenities and accommodation. The centre line is reached by a 60-km drive northwestward on the Lodwar-Lokichogio highway and lies in a region of low hills that should not present a viewing problem, as the 13-degree altitude of the eclipsed Sun will be well above the terrain. An alternative approach to the track is to take the right-hand fork in the highway north of Lodwar to the village of Kalikol, which lies just inside the south limit. The road is paved to Kalikol, but turns into a desert track beyond, eventually leading to the shoreline of Lake Turkana where the fishing boats are drawn ashore. To reach the centre line will require an overland trip following the many tracks across the desert terrain; a local guide is essential. Even then, you may be thwarted a few kilometres short of your goal, as the mid-line of the eclipse actually crosses the western shore in a river delta. Our view of the area from the water showed a marshy coastline (Figure 11) that did not look promising. Locals advise that any approach to the eclipse track from Kalokol should stay at least 2 km inland. While the Turkana region around Lodwar has a desert climatology, November does come in a modest rainy season and precipitation can be very heavy, though rare. Even small amounts of moisture can severely affect the ability to travel across the sandy terrain, and the usually dry river and stream beds will fill with water that may present an impossible barrier. Lodwar lies between two rivers, one of which is crossed by a bridge and the other along a spillway. If even a modest amount of rain should fall, the spillway, on the north side of the town, becomes impassable for a day or two, making an approach to the centre line impossible. In short, a site north of Lodwar comes with a small risk of wet weather that would make reaching the centre line nearly impossible.
http://home.cc.umanitoba.ca/~jander/tot2013/LakeTurkanaCentreline.pngFigure 11: View toward the direction of the eclipse from an offshore location north of Kalokol. The land along the shore is marshy in wet weather. The distant hills lie along the Lodwar-Lokichogio highway. The lake is very shallow at this point.
On the east side of the lake, the world-famous Sibiloi National Park provides a not-too-convenient site from which to watch the Moon pass. The centre line lies only a few kilometres south of the Park headquarters, but the overland route to the Park is laborious and accommodation is available for only a very few people - four to six, in fact. The climatology here is very similar to that at Lodwar, with perhaps a slightly less frequency of cloud, as the lake suppresses the formation of convective clouds. The terrain is also flat with a good view across the water toward the eclipsed Sun (Figure 12).
http://home.cc.umanitoba.ca/~jander/tot2013/Parkcentre.png
Figure 12: The view westward from the centreline in Sibiloi National Park. The surface is a coarse hard-packed gravel and sand.
Ethiopia
Ethiopia presents a challenge that extends beyond meteorology, as the eclipse duration becomes very short, the altitude of the Sun at totality sinks below 10 degrees, and transportation becomes more difficult to any location except those along the major highway from Addis Ababa to Moyale (Highway 6). While these might appear to argue against Ethiopia as an eclipse-viewing destination, in fact just the opposite is true. The short duration promises a near-circumferential ring of red chromosphere (the innermost layer of the solar atmosphere) around the dark lunar disk, and dynamic displays of Baily’s beads. The low altitude will deliver a dramatic converging lunar shadow, whose width is barely 10 km, centred on a briefly eclipsed solar disk. The remote location promises a travel adventure into the variegated cultures and terrain of a country that tourists usually overlook.
http://home.cc.umanitoba.ca/~jander/tot2013/ethiopia_s.pngFigure 13: High-resolution map of the track of the eclipse across Ethiopia south of Yabello along Highway 6. Click on the image for a larger version.
I scouted Ethiopia with a team from TravelQuest International in late April and early May of 2011. This is the season of the long rains, but we were early and the weather we encountered was probably not very different than what might be expected during the November short rains. The weather was characterized by prominent afternoon cloud buildups, late day and overnight thunderstorms, and clear or clearing skies in the morning, with the pattern repeating reliably from day to day. The storms were widely scattered, so that in our ten days in the country, we were barely touched by rain, but distant virga provided a common backdrop to the red hills each afternoon and lightning could always be seen somewhere in the distance each night. Rain, when it came, was often very heavy, as evidenced by the flooding we encountered as we drove through the flowing streams after the rain had passed. The combination of heavy short-duration rains makes the climatology look wet, but in fact most of our days were sunny, though with a considerable frequency of small convective clouds and high-level semi-transparent cirrus.The convective character of the daily cloud cycle suggests that eclipse day has a reasonable probability of being sunny. A typical day will have afternoon cumulus clouds, that reach their maximum areal extent around 4 in the afternoon. Some will get big enough to form virga. By 5 p.m., they are largely dissipated. The cooling associated with the approaching lunar shadow will speed up this cycle, and so the “median” eclipse day has a high likelihood of sunshine, though possibly diffused through a thin cirrus layer. From Awassa, the eclipse centre line can be reached by long drive to the south along Highway 6. The road is paved, but speeds are limited by the inevitable cattle, goats, and sheep that drift across the highway, oblivious to all forms of wheeled transportation. The closest community to the track of any size is Yabello (or Yebello), a dusty crossroad town (Figure 13) that provides truckers with a brief respite on the trip from Nairobi to Addis. From Yabello, the eclipse centre line is about an hour away, through a countryside that alternates between flat sandy fields vegetated by the ubiquitous Acacia bush and steep reddish hills. The centreline landscape is inauspicious and open, with moderate hills to the west and southwest that should not interfere with a view to the Sun (Figure 14). About 2 km north of the track lies the village of Madaccho, a community of several hundred that is now aware of the impending spectacle.
http://home.cc.umanitoba.ca/~jander/tot2013/ethiopia_site_s.png
Figure 14: A view looking west from the centreline on Highway 6 south of Yabello. Compare the hills in the background to the map in Figure 8. On eclipse day, the sun will be above the top of the large tree in the right foreground. Click on the image for a larger view.
What is the probability of seeing the eclipse in Ethiopia then?
For the answer, we are obliged to accept the percent of possible sunshine recorded at Moyale, just over the border in Kenya (there is also a Moyale, Ethiopia) and about 90 km south of the eclipse track. In November, 65 percent of the daytime hours are sunny, a figure that probably a little lower than at the centreline south of Yabello.
Summary
The 2013 hybrid eclipse will present challenges in all of its venues, but the best weather prospects are found near the end of the track in Kenya and Ethiopia. In most reachable locations, mobility is limited by the narrow dimension of the track and the lack of a dense road network, but the regional climatology promises a 60- to 80-percent chance of success from Lake Albert onward. In central and western Africa, eclipse viewing will be conducted in a wet and cloudy climate with a sunshine probability around 40 percent. Shipboard observations near the maximum eclipse point provide for easy mobility that will also have the advantage of a longer eclipse with the Sun at a higher elevation.
AcknowledgementsThe details included in this assessment would not be possible without the opportunities to scout the region provided to me by Aram Kaprielian and TravelQuest International. -Jay Anderson Updated: December 2011
原文：http://home.cc.umanitoba.ca/~jander/tot2013/tse13intro.htm 要是是中文就好了。
页:
[1]