July 18th, 2009

July 22 Great Solar Eclipse PFS Bulletin No. 12

Published at 21:25. Filed under Precision Forecast Service; 5 Comments

EDITIONAL NOTICE

The server at NAOC (National Astronomical Observatory of China) was scheduled to be offline for maintenance on July 17 at 0400 UTC to 0800 UTC. However, the server was still offline until 1520 UTC due to some technical problems, led to the cancellation of July 17's update. Meanwhile, my departure delayed from July 18 to 19 for some reasons. Thus, bulletin No. 12 would be my final bulletin in detail for the eclipse. So I include all the information available to manage to make my best guess in this bulletin. During these days, I have experienced a work with hard pressure. I would like to have my most grateful thankness to all of you for your supports. Many people (including me) have work hard for many days for the event, and I wish best of luck on July 22 for every of them.

I have made decision to a 4,100m highland in northern Shangri-La (at the edge of Tibet Plateau) for the eclipse. I will fly to Kunming later tomorrow. Internet connections should be available until July 21, so I may be able to write a few lines when possible and necessary.

OVERVIEW

This is July 22 Great Solar Eclipse PFS Bulletin No. 12, issued in both English and Chinese. The forecast reasoning has not changed since previous bulletin. There is only 4 days away from the event. The uncertainty for this forecast remains very small. The forecast diagrams mentioned below may be found at http://7timer.y234.cn/V3/product.php?language=en&product_id=2 and https://www.fnmoc.navy.mil. Precision numeric forecast for exact locations may be found at http://7timer.y234.cn/V3/product.php?language=en&product_id=1. RSS feed of this service is available at http://7timer.y234.cn/V3/wordpress/?feed=rss2&cat=46.

ANALYSIS

NCEP GFS JUL18/00Z's run depicts the following:

1. Western Pacific subtropical high is rapidly weakening from tau 72 thru tau 96 (JUL21/00Z to JUL22/00Z), with its 500hPa 5880gpm contour drifting eastward from mainland China to around 140E. However, 500hPa streamlines forecast still depicts a weak anticyclone (a splitted high) over South China at JUL22/00Z, with a northward ridge locates at about 110E.
2. A Northeast China vortex locates around 130E 46N as indicated by 500hPa geopotential heights and streamlines forecast for JUL22/00Z, with a shallow trough developing and moving from 124E at tau 72 to 133E at tau 96. At tau 96 the trough is expected to extend to northern Ryukyu Islands.
3. A westerly flow is located along 30N as indicated by 700hPa isotachs forecast around tau 96, an associated 700hPa trough is located around 120E 34N to 126E 40N with a 850hPa shearline. An associated area of positive vorticity will move across East China until tau 108.
4. The main body of Arabian Sea subtropical high remains west of 70E around tau 96, a low pressure is located in eastern India Peninsula.

ECMWF JUL18/00Z's run depicts the following:

1. 500hPa geopotential heights of JUL22/00Z also revealed a rapidly weakening subtropical high over East Asia, the situation has no significant difference with the forecast of GFS. The situations of a low pressure over India Peninsula and the westerly flow over East Asia (revealed by 850hPa wind speed) are also close to the GFS's.
2. 850hPa shearlines at 120E 32N to 127E 34N and 105E 30N to 118E 33N are expected, with a strain of low pressures along the Yangtze River.

FNMOC NOGAPS JUL18/00Z's run depicts the following:

1. Western Pacific subtropical high is rapidly weakening from tau 48 to tau 96, with no area higher than 5880gpm over mainland China at tau 84 (JUL21/12Z; for a comparasion, GFS suggests a splitted high with intensity over 5880gpm is still visible over mainland China). At tau 96, the 5880gpm contour locates around 150E, which is also more eastward comparing with the forecasts by GFS and ECMWF. In all, NOGAPS presents a faster-weaken subtropical high compare to GFS and ECMWF. However, a ridge can be identified from the 500hPa streamlines forecast for JUL22/00Z, extending from 120E 25N to 105E 31N.
2. A Northeast China vortex locates around 128E 52N as indicated by 500hPa geopotential heights and streamlines forecast for JUL22/00Z, with an insignificant trough over northern Korea Peninsula. The intensity of this trough is apparently weaker than the expectation of GFS.
3. A westerly flow is located along 30N as indicated by 700hPa isotachs forecast around tau 96, an associated 700hPa trough is located around 110E 30N to 118E 34N with 700hPa and 850hPa shearlines. An associated area of positive vorticity will lingering around Central China around tau 96.
4. The main body of Arabian Sea subtropical high remains west of 75E around tau 96, a low pressure is located in Gulf of Bengal.

For analyses of three numeric models above, we can now depict:

1. Western Pacific subtropical high will be weakening dramatically around July 21-22, this conclusion is supported by the consistency of all three models as well as the southward behaviour of Qinghai-Xizang anticyclone ridge, the occurence of an El Niño event and the declination of Southern Oscillation Index (SOI). However, the models have also suggested that a weak ridge is possible to remain around the northern part of South China.
2. A significant shearline will be located somewhere along the Yangtze Valley at July 22, with its exact position and intensity being determined; the Southwest China vortex shall be weak.
3. India Peninsula will mostly be controlled by a low pressure instead of a strong Arabian Sea subtropical high at JUL22/00Z.

The uncertainty for this forecast remains very small as the results from all three models are consistent. We shall note that many statistic studies have showed that the ECMWF model is superior than all other models for forecast within 120 hours.

By July 18 at 09Z, Typhoon Molave (07W) is located near Hong Kong and shall be dissipated overland within days. It is unlikely that Molave will affect the weather of July 22.

HUMAN FORECAST

The forecast is remains pessimistic. According to the results of the models and past studies, there seems to be only the luck few who can get the totality. The uncertainty has further reduced since the consistency of three models remains good. Local forecasts as well as advisories of usage are listed below.

Totality area over India:

• Prognostic Reasoning: the Arabian Sea subtropical high remains moderate and is unlikely to bring fine weather for India, a low pressure around the east coast may bring unstable weather.
• Forecast: mainly cloudy over western and eastern totality area, may be mostly clear around Patna.
• Advisory: to weather of India the author is not as familiar as China's. However, due to the lack of an in-control subtropical high and an occurence of a low pressure around the east coast, a clear weather over the entire zone is unlikely. GFS indicates a generally cloudless weather around Patna, while FNMOC EFS predicts a precipitation chance of 20-90% over the totality zone during the day.

Totality area over western Sichuan, China:

• Prognostic Reasoning: the Southwest China vortex is expected to be weak, however the Arabian Sea subtropical high is also moderate, thus, extreme weather over large area is not expected.
• Forecast: mostly cloudy over west, including big Shangri-La (Deqin, Derong, Daocheng), may be suitable for eclipse for some lucky area. More clouds over eastern part (Gongga Shan/Minya Konka, Ya'an, Emei Shan) with moderate precipitation chance at the morning according to FNMOC EFS.
• Advisory: weather forecast for highland is always very challenging, the author has investigated a number of weather archives for Derong and Minya Konka, the relationship between basic meteorological elements and actual weather appeared to be very complex. remarks from experienced local people or guides shall be employed seriously if any. Long term statistic data showed that northern Shangri-La area has a good possibility of being clear even in July. It is reported that the total sunshine time for Derong is 2,700-3,000 hours. For eastern area, July is the rainy season, and the forecast we have does not support a significant chance for an exception.

Totality area over Chengdu-Chongqing Plain, China

• Prognositic Reasoning: the lack of a significant Southwest China vortex suggests the weather shall not be too bad. However, as this area is at the buttom of a trough, plus a weak anti-clockwise stream in-controlled, the weather will not be too opmistic. The weather can be slightly better over east as all three models suggest something like a weak ridge or a weak high.
• Forecast: mainly cloudy with a few showers. There is a weak chance that the weather to be better over east.
• Advisory: there appears to be no significant system in-controlled. It can say with some confidence that rains or clear over large area is unlikely; however, a chance of less clouds and a few more sunshines can not be ruled out, as three models had indicated. The author should point out that minor systems like the weak ridge mentioned above is difficult to predict, so please follow the most recent weather information before you make decision.

Totality area over Hubei Province, China

• Prognositic Reasoning: all three models proposed something like a weak ridge or a weak high over west, but the most area is expected to suffer from affection of a shearline locates over the north.
• Forecast: weak chance for western area (like Enshi) to be better; mainly overcast over the entire province, but clouds may be slightly thinner (may permits a partial eclipse observation) over the south since it locates further from the shearline. Light rain is possible for some area throughout the day.
• Advisory: it can be say with some confidence that decent weather for most area is unlikely. Base on experience, a region of cloudless will locate somewhere around the shearline, however, its size is depending on the intensity of the shearline, which is hard to predict. In all, get prepared to move around and follow the most recent weather information. If possible, check the high resolution satellite image on the early morning of July 22 and make final decision.

Totality area over South Anhui Province, China

• Prognostic Reasoning: the low level shearline is expected to locate over the northern part of Anhui Province, whether or not that its associated clouds would affect sites among South Anhui depends primary on the intensity and exact location of the shearline, which the three models have not yet agree with each other.
• Forecast: since we have noted that ECMWF is more reliable than other two models, we may take ECMWF as consideration. The weather will be generally overcast with rain patches according to the forecast of ECMWF, but may be slightly cloudless over south -- like Huangshan. If NOGAPS or GFS gets it right, it will be cloudy, with much higher chances for the appearance of cloudless regions.
• Advisory: as noted above, the intensity and exact location of the low level shearline play an important role on judging the possibility for South Anhui to get a decent view. However, the southern area of South Anhui might be slightly more promising since it will keep a distance from the shearline no matter which model gets it right. In all, check the high resolution satellite image on the early morning of July 22 to make final decision if you can.

Totality area over pan-Yangtze River Delta, China

• Prognostic Reasoning: the low level shearline is expected to locate around Shanghai. Since the shearline here is close to the front lingering from East China Sea across Japan, the possibility for this shearline to be a weak one is not high.
• Forecast: generally overcast with rain.
• Advisory: weather over this area is less promising than those above. However, weather is varying from day to day, any kinds of situation cannot be ruled out.

Totality area over Ryukyu Islands, Japan

• Prognositic Reasoning: a westerly flow is lifting over this area, however the trough is well clear to the north.
• Forecast: mainly cloudy, with a few or some sunshines.
• Advisory: weather over ocean is quite dynamic. Although this area is not expected to be controlled by subtropical high as usual, cloudless regions are probable.

Conclusion and Remark

• Generally speaking, there is no place to be better from cloud than any other places, except for places around Shanghai seems to be less opmistic according to the forecast we have.
• Weak ridges noted above may bring slightly better weather, however they are hard to be precisely predicted.
• You may better stay at the site you picked and accept whatever happened to you. If you can fly off or can drive around, keep monitoring the weather, preferably weather diagrams prior to July 21 and satellite images in the final hours, and make final decision accordingly. A fine collections of the resources by meteorolgist J. Anderson is available at http://home.cc.umanitoba.ca/%7Ejander/tot2009/fcsts.htm.

12KM MM5/GFS FORECAST

The 12km MM5/GFS model is online and is now under test. Eclipse map produced by Occult 4 has been combined with the cloud cover graph. Please download the KML file and open it with Google Earth (http://earth.google.com/). The forecast time is July 22 from 0000 UTC to 0400 UTC with 1-hour interval, the forecast region includes most of totality area over China. This model shall be updated automatically everyday at about 0200 UTC.

Warning: by using this product you have acknowledge that the product is under very first test and is only posted as an alpha version, 7Timer! is not responsible for any damages or expenses you may incur as a result of any inaccuracy. To help you better assess the products we have, you are highly advised to read the tips below before using.

• Tip 1: 7Timer! has two models now, the original GFS model of 35km and 3-hour interval, which the forecast may be found at APanel section (local forecast for 72 hours) and Weachart section (charts of global for 384 hours) which you can find at the 7Timer! website (http://7timer.y234.cn/V3/product.php?language=en); and the MM5/GFS dynamic numeric model of 12km and 1-hour interval, which uses GFS data for initialization and is under very first test and is only posted here for the eclipse. My experience is that GFS is usually more reliable, despite its resolution is far from as good as MM5/GFS. So, check GFS chart when using MM5/GFS, and assess the reliability of the situation accordingly.
• Tip 2: When there are less than 72 hours from the event, APanel can be used to assess the weather condition for an exact location. However, since APanel also uses the low resolution 35km/3h GFS data, the trick given in Tip 1 still plays -- when the forecast by MM5/GFS (the product you are using) is similar to what GFS given, you may use MM5/GFS for a more precise estimate of the weather as it has a much better resolution. Otherwise, please do not take the result MM5/GFS gives too seriously.

12km MM5/GFS Forecast KML File

NEXT UPDATE

When possible. Might be July 20 at 1600 UTC or earlier.

ACKNOWLEDGEMENT

I would like to thank Dr. Cui Chenzhou from National Astronomical Observatory of China for working with the server problem overnight. I would also like to thank Zhang Linna from Beijing Meteorolical Bureau and President Dickson Fu from Sky Observers' Association (Hong Kong) for their valuable discussion and help. I would also like to thank Cheng-Su Diandian for her support of this work.