WunderBlog Archive » Category 6™

Category 6 has moved! See the latest from Dr. Jeff Masters and Bob Henson here.

Super Sunday weather bandwidth

By: Dr. Jeff Masters, 2:26 PM GMT on February 06, 2006

Weather enthusiasts like to watch the Super Bowl--that's my not-so startling conclusion after analyzing the Internet bandwidth curves from the wunderground.com website from last night. In a pattern I've seen during every Super Bowl since wunderground.com opened shop in 1995, our bandwidth plot (Figure 1) shows a clear drop-off in traffic in the 45 minutes prior to the game, a sudden spike in traffic at halftime, a sharp drop when halftime ends, and a return to normal levels at the end of the game. Interestingly, the national championship game for college football has never been apparent on our bandwidth curves. The only other non-weather related event I've seen affect our bandwidth occurred in the aftermath of the 9/11 disaster when our traffic dropped below 50% of normal after 9am and stayed at less than half of normal all day.

Our sharpest spike in bandwidth due to a non-hurricane weather event occurred in 1999 when an F5 tornado ripped through Oklahoma City. Major hurricanes hitting the U.S. regularly cause large bandwidth spikes, particularly in the 30-minute period after NHC issues the 11am advisory. Our busiest day ever was September 22, 2005, as Hurricane Rita approached the Texas/Louisiana coast. Each NHC advisory issued that day created a large bandwidth spike bigger than the Super Bowl halftime spike. Google has a page showing their top searched-for natural disasters of 2005. Hurricanes Katrina and Rita produced roughly equivalent peaks, but Katrina's peak lasted much longer.

Figure 1. Plot of total bandwidth in bits/sec for a portion of the weather imagery sent out by the wunderground.com web site during the Super Bowl on February 5, 2006. The times on the bottom axis are EST.

The views of the author are his/her own and do not necessarily represent the position of The Weather Company or its parent, IBM.

Reader Comments

LOL, well then I am second. The Seahawks were robbed.
So then based on that graph I guess weather geeks are also sports geeks.
Yeah, I was busy watching the officials play the game.
I've got to admit, while I had my normal Sunday gathering of nine folks at my house yesterday, the only time we even talked about the Superbowl was when we got 4 different phone calls telling us the results. The rest of the day was devoted to the further adventures of Captain Jake Riley and the crew of the spacefreighter Croi Lar in the Serenity RPG. O};->>
Guilty as charged on bothcounts. Sports and weather geek right here.
I am a Packers Fan. I am already looking toward next season. Yesterday I was on here because my dad uninvited me the party cause we got into an argument earlyer in the day.
Nope, just a weather geek here. I'd love to see bandwidth for the week preceeding Katrina, Rita, and Wilma.
What would hapen if World Cup Soccer was on? I did not watch the Steelers win. I have watched the last two world cup finals. Guilty on the weather geek. Guilty on certain sports.....Soccer, College Basketball(NCAA comin up in March)
As an IT Director, it is amazing to see the effect different events have on networks. The cellular networks are a perfect example. During Katrina, my cousin at Tulane was trying to get out of town. He couldn't find a cell phone tower with available outbound channels until he reached Jackson, Mississippi (headed from NOLA to Little Rock, Arkansas).

Dr. Masters must have an incredible amount of bandwidth to deal with the spikes from Katrina, Rita, etc.
Not much, haydn. People in the U.S do not watch as much soccer as any other of their sports. I think that is a real shame. By the way, I live for sports as much as weather. I cannot get enough of it. I watch every soccer game that is on TV, every baseball game, a lot of football games, and the entire NCAA Basketball Championships as well as the NBA playoffs. I'm also going to try and fly to Germany to see some of the World Cup games.
In the UK they get similar readings in electical use when there is a spike at half time of soccer matches, when people put the kettle on for a cup of tea

I think the world cup final in 2002 had an estimated 3.3 Billion people watching world wide.
Amazingly the Rugby World cup in 1995 had an estimated 2.2 billion.
And the cricket world cup gets over 1 billion.
The superbowl is estimated at 100million.
Although pulled these figures from the net, so accuracy is probably not great
Considering the world population is around 6.5 billion, that would be truly amazing if 1 in 2 people watched the world cup final!
One final thought before I go on a trip to Myrtle Beach. (There won't be that many people out. The system in the southeast is supposed to drench the costal area of SC and NC in about 24-36 hours.)

The amount of bandwidth required for this site has to be large. Advertising pays for some. The 10 dollar memberships help too. How many of us who use this site pay 10 dollars a year? This is less than one meal at an expensive restaurant. After using this site for a few months, I am going to give my 10 dollars and help support the site. I appreciate the information given. There are many sites on the web that seem pointless. This is not one of them.

Next time there is a world cup I will be watching.
my tunnels will help solve the bandwith problem

sorry, couldn't resist.
June this year is the next one. USA is in the same group as Italy, Czech Rep and Ghana
It's not so surprising spoon, soccer is the most popular and played sport in the world, and has more followers than all other sports combined. In Europe it is amazing: 40 billion out of the 58 billion people in the UK watched the 2002 World Cup quarter-final match between England and Brazil.
That should be *million* on both occassions
That should be *million* on both occassions
Here I see more than a weather~sports connection, I see a general need to watch numbers, compare their statisitics & arcive them. Just as Dr Masters did with the bandwidth (which was way cool & look Aquak the weather geek wants to see more~ bandwidth won't tell you the weather:). I've noticed we have some gamblers around as well. While working in racing I so enjoyed the #'s, stats & records. Before that we'd follow this band, many were down right obsesive about how long since the last (insert any of their song names here) had been played, length of show/song/set), number of show, basically anything to derive & compare, to form a record. Who's here has played D&D or something like it during some part of their life? Guilty there too. Emersing myself in a sort of weather geekdom has really filled that void that appeared after leaving the numbers from racing. Though when I was heavily tied up in the racing, I was still one of those people that were oddly & easily destracted by the weather:)
atmosweather mail for you

oh the I storm update is now out by the nhc
Today while I was at school it rained today! It rained Ash from a near by fire. It lasted for an hour or so.
My tunnels will eliminate the incidence of Florida retirees who never drove a day in New York jumping curbs and taking out a dozen pedestrians at a swipe!
wxfan, check out this link for more info, I think it'll explain it a bit:
this one will take you directly to the article, the other is the blog entry about it, sorry ...

her is some in new to everone to look at
Anything new on that possible tropical system off Africa?

Interestingly our online business is very weather dependant. We have been online since around 1996 and have always seen huge weather related trends in online orders. Cold fronts, snow and foul weather help online sales. Nice weather (especially at first) slows it considerably. We had near record days in sales Saturday and Sunday. Perhaps we should watch the NAO for business planning purposes;) We find your blog riveting, thanks.
I dont want anyone to take this the wrong way, but if Vegas was to make a line for a F-5 tornado hitting the US this year, what would it be?

It is still hard to believe there hasnt been one this millenium when there were many days in the 20th century, with multiple f-5's.
I'm getting excited about the world cup this year, but the US will need help getting out of their group(which is very tough)
a F-5 tornado in Vegas that cool
Stormchaser77~ answer via e`mail. What were the reffs doing in the pile up anyway?
Went out and played Laser Tag with a few friends and husband. Then came on and checked wunderground. hehehe.

Go Full on geek out.
Experts release 2006 forecast

By Tim O'Meilia
Palm Beach Post Staff Writer
Wednesday, Dec. 7, 2005

The hurricane season won't be Greek to us next year, says William Gray's Colorado State University hurricane-forecasting team, and chances of four major hurricanes striking the U.S. again are slim.

Yet, Gray is predicting a double-the-average storm season in 2006 with 17 named storms (reaching only the letter R in hurricane names), nine of which could become hurricanes and five of which are expected to develop into major storms, with winds of 111 mph or more.

While that's far fewer than this year's record-setting marks of 26 tropical storms (forcing the use of the Greek alphabet to name them), 14 of which were hurricanes and seven of which were major hurricanes, it would be more than any other recent year.

"After this year, some people may think, '17 named storms? That's not so bad. We can deal with that,' " said Philip Klotzbach, Gray's top research associate. "But it's still a very active hurricane season."

By Gray's calculations, Palm Beach County has a 10.1 percent chance of a tropical storm strike and a 3.1 percent chance of being hit by a hurricane.

Martin County has a 3.8 percent probability of a tropical storm and 1.2 percent of a hurricane. St. Lucie and Okeechobee counties are even less.

Higher-than-normal Atlantic Ocean temperatures and a weak La Nia in the Pacific make a busy storm season likely, the forecast said.

The historic average is 9.6 named storms, 5.9 hurricanes and 2.3 major hurricanes.

Gray said the increased storm activity of the past few years "is likely to continue in the Atlantic basin for the next 15 to 20 years, but the probability of seeing another two consecutive hurricane seasons with as many land-falling hurricanes as was witnessed in 2004 and 2005 is very low."

Klotzbach said the increased storm activity combined with a high pressure ridge off the U.S. coast worked to steer an unusual number of storms onto the eastern seaboard or into the Gulf Coast.

Of the 13 major storms that formed the past two years, seven struck the U.S. The historic average is one of every three.

"The Bermuda high can't be that strong or in the same place every year," Klotzbach said.

He said the role of the high will be more clear in the spring. Klotzbach will take over as the lead forecaster as Gray concentrates on global warming studies.

Still, the team said there's an 81 percent chance at least one major hurricane will hit somewhere in the United States, compared with the average of 51 percent.

Last December, the Colorado State team forecast 11 tropical storms, which included six hurricanes, three of which were major - less than half of what eventually happened - but bumped their prediction up when hurricane season began in June.

The National Hurricane Center also underestimated the record-breaking season.

"I want the predictions to be spot-on," Klotzbach said, "but the storms to stay out in the middle of the ocean."

How many named tropical storms will there be in 2006?


10 or less 11.43% 667
11-15 9.39% 548
15-19 15.26% 890
20-24 18.46% 1077
25 or more 45.46% 2652

FORT COLLINS - Colorado State University's team of hurricane forecasters is predicting another active season next year.

Dr. William Gray says although 2006 will be another active hurricane season, fewer storms will make landfall compared to recent years.

Dr. Gray's forecast team expects 17 named storms in the Atlantic Basin, with nine of those storms growing to hurricane strength.

Six is the yearly average, 13 hurricanes formed last year.

Dr. Gray says, "We're giving an 81% as being a major landfalling storm in the U.S. in 2006 when the long term average is 52%."

Five of the storms forecast for 2006 are expected to become major hurricanes. There were seven major hurricanes this year. Still, Dr. Gray says it's statistically unlikely that the U.S. will see as many hurricane landfalls next year as compared with 2005 and 2004.

well see!

They predict according to the list of percentages above that there is a 45% chance there will be 25 or more named storms? And that forcasted percentage is 2 times more likely than any other? If thats true thats a bit scary. Yet at the same time, "Dr. Gray's forecast team expects 17 named storms in the Atlantic Basin," I'm banking they stay out to sea and dont meander over in our neck of the woods.

A safe season to everyone

Giants in 06
As for the possible storm off Africa~ Looks like we'll see in around 30 hrs, according to the models. Today they've all switched to a more sub-tropical storm~ dying as it nearly becomes warm core. They all agree as well that the lowest pressure will be in the 990mb range, which could = around 65mph. This is forecasted to occur around the 45hr mark, right before it's forecasted to swing east, than north, before diving south & disapating off the west or into the coast of Africa. The ukmet dissapates it fastest, the Nogaps keeps the storm slightly more to the west, ending the 144hr with a going full tropical depression. The gfs lies in the middle, dissapating at 144 hrs.

Though there is a lot of land around there~ most of these west of Africa storms lately, were forecasted from the get go & continually forecasted to dissapate quickly, though many did not. Hence, now I give the models long term forecast less weight, on this maybe storm.

Due to the fact it is forecasted so close to land & mostly forecasted to not last much more than 4 days, I assume the NWS will be conservative in calling this a STD until a time it maintains for a period, a near or STS strength. On the NWS Atlantic Discussions~ 1:05pm~ Blake stated a cut off low would form, the 7:05pm Formosa (as lastnight) doesn't mention it.

Interesting note~ the whole eastern end of the ITZ has suddenly shifted north into this area with alot of convection.
Just remember - they forecasted 11 storms for '05. Even in July, after all that madness with Dennis and Emily, they said only 17.

"I think the world cup final in 2002 had an estimated 3.3 Billion people watching world wide.
Amazingly the Rugby World cup in 1995 had an estimated 2.2 billion."

Not even half that many people have TVs, your stats are totally wrong.
I remember waking up at 345 in the morning to watch the US get disqualified against Germany. Maybe whole towns and villages gather around one TV to watch those games? Making it at least plausable for that many folks to be watching the game.

Feb. 2, 2006 The NOAA National Weather Service today announced plans to implement the Enhanced Fujita (EF) Scale to rate tornadoes to replace the original Fujita (F) Scale. The EF Scale will continue to rate tornadoes on a scale from zero to five, but ranges in wind speed will be more accurate with the improved rating scale. The NOAA National Weather Service has approved the EF Scale and expects it to be fully implemented by February 2007. (Click NOAA illustration for larger view of tornado. Click here for high resolution version. Please credit NOAA.)

"The EF Scale takes into account additional variables which will provide a more accurate indication of tornado strength," said retired Air Force Brig. Gen. David L. Johnson, director of the NOAA National Weather Service. "The EF Scale will provide more detailed guidelines that will allow the National Weather Service to more accurately rate tornadoes that strike in the United States."

The F Scale was developed in 1971 by T. Theodore Fujita to rate tornadoes and estimate associated wind speed based on the damage they cause. The EF Scale refines and improves the original scale. It was developed by the Texas Tech University Wind Science and Engineering Research Center, along with a forum of wind engineers, universities, private companies, government organizations, private sector meteorologists and NOAA meteorologists from across the country.

Limitations of the original F Scale may have led to inconsistent ratings, including possible overestimates of associated wind speeds. The EF Scale incorporates more damage indicators and degrees of damage than the original F Scale, allowing more detailed analysis and better correlation between damage and wind speed. The original F Scale historical data base will not change. An F5 tornado rated years ago is still an F5, but the wind speed associated with the tornado may have been somewhat less than previously estimated. A correlation between the original F Scale and the EF Scale has been developed. This makes it possible to express ratings in terms of one scale to the other, preserving the historical database.

NOAA, an agency of the U.S. Department of Commerce, is dedicated to enhancing economic security and national safety through the prediction and research of weather and climate-related events and providing environmental stewardship of the nation's coastal and marine resources.

Through the emerging Global Earth Observation System of Systems (GEOSS), NOAA is working with its federal partners and nearly 60 countries to develop a global monitoring network that is as integrated as the planet it observes.

I have to say it's quite odd that Jeff was looking at that info during the events he said.

it's kinda sad and funny at the same time. Although I'd probably do the same thing because I love stats as you guys Know!!

Packers 2007!!!
Forecastercolby- your contributions to this blog indicate open-mindedness. Why would you doubt that half the world's population tune in to the World Cup football final (nobody calls it soccer but the culturally isolated Americans)? No, several billions will indeed watch the last 'soccer' game of the season, and villagers worldwide will share television sets to do it.

We'll never go metric either, dammit.
Glad to see The Bus finally get his ring.

'Skins in SB XLI!!!