Saturday, December 15, 2012

An Early Look At Weather For Christmas 2012

The holidays are right around the corner, so let's start to take a look at who could be seeing good holiday travel weather, who may see some stormy conditions, and even who could see a White Christmas.
This is the GFS model forecast for Christmas Day. Pictured above are precipitation values in colors, high and low pressure systems denoted by H and L signals, respectively, and 850mb temperature lines in light reds and light blues. The dark blue line is the rain/snow line. The GFS forecast calls for some light snow in the Great Lakes, with some flakes flying in the immediate coastal areas in the Northeast. 850mb temperatures are as low as -18 (Celsius) in the North Plains, so a chilly Christmas is not out of the question. An emerging storm system in the Plains could cause New Year's troubles, something I'll be addressing in coming days. Also, a rain event could be ongoing for the West US. But before you take this and run with it, the GFS is typically unreliable beyond the 5 day mark, and this forecast is more than 10 days out.


The two images above are of the CFS v2 weekly forecasts. The top image shows temperature anomalies for the timeframe of December 22-27, a frame that includes Christmas. The forecast shows warmth prevailing across the Central US, with a frigid West Canada in the forecast. The bottom image shows maximum snow depth from 4 CFS v2 ensemble members from the Dec. 22-27 period, the same timeframe as the temperature anomaly forecast. The ensemble members hint at some good snows along the Northeast and North Plains, something I can't say won't happen. But in all honesty, the CFS v2 weekly forecasts aren't exactly the most reliable either. However, I think you should watch that snow chart if you're in the Northeast- all 4 members are showing it, so snow on the ground for Christmas doesn't seem all that far-fetched.

Below is my forecast for a White Christmas potential I issued a while back. I will have a new forecast out in coming days.



Andrew

December 26-28 Potential Winter Storm

The GFS model has now been showing a big Northeast snowstorm for at least 4 model runs, meaning it's time to take a look at this potential.

We'll start off with the oldest model forecast, the 0z model run from last night. This forecast of the GFS shows a wide swath of snow ongoing in the Plains and Midwest, with heavy rain in the Southeast and Mid Atlantic. The system in question is in the southern Midwest, and the way the precipitation and freezing line is angled tells me that, for many in the Northeast, such a solution would be a solid rain-to-snow event. Frankly, I'm going to discount the 0z GFS because of how different it is from the 6z GFS.

This is the second-oldest GFS model run forecast, the 6z forecast, which was made early this morning. For roughly the same timeframe, the morning of December 27th, we see a fairly strong storm system in New England with a wide swath of precipitation. The solid blue line is the rain/snow line, and its take on this forecast gives inland locations of New York, Pennsylvania and similar states a good batch of sticking snow. Coastal cities receive a nice dumping of rain per this forecast. In the wake of the storm, a  big blast of Arctic air flows into much of the nation, no doubt bringing the lake effect snow machines to  maximum power.

We now turn our attention to the second-newest forecast, the 12z GFS, which was made in the morning hours, about 6 hours later than the 6z. The 12z GFS isn't all that different from the 6z, with a system in the vicinity of the Northeast and a wide precipitation shield. Two main differences include a much weaker precipitation shield and the storm system isn't even onshore with this forecast. Coastal locations now see chances of snow, and Arctic air still flows south into the Plains and Midwest, cranking up the lake effect snow machines. Note the heavy snow ongoing in the Virginias.

And now the newest, and most startling GFS model forecast- the 18z. The 18z GFS, valid for the evening of December 26th, brings an epic storm system up the coast. The storm system bombs out to a minimum central pressure of 968 millibars- not your everyday low pressure system strength. The precipitation shield is wild, with up to 1.5 inches of rain hitting New Jersey. You weather enthusiasts know that average snow-to-rain ratios is 10:1 (inches), so this could mean 15 inches of snow for New Jersey! Heavy snow falls in the immediate coastal regions, with relatively lighter snow farther inland. Again, the arctic air continues to come on down into the Plains and Midwest, Ohio Valley and Mid-Atlantic behind this system to provide a base for lake effect snow and enhanced chances for snow in the Northeast.

It's obvious that we are dealing with something on our hands, the question is exactly how strong could this system be, where could it go, what type of precipitation could it bring, and the most important question: Will it even happen??

We still have a lot to learn on this potential, but it looks like the potential could be there.

Andrew

December 20-22 Potential Blizzard Event (Updated 12/15)

(Originally posted earlier today) I have upgraded my thinking of this winter event from a 'Significant Storm' to a 'Blizzard' event. I'll now explain why.

This is the ECMWF forecast for Hour 144, a.k.a. the evening of December 20th. We see a very strong storm system in northeast Illinois at this time with a minimum central pressure of 986 millibars- pretty strong for an onshore system. Let me first start off discussion of the model by saying I trust the ECMWF more than other models at the moment, as although all models have been having troubles, the ECMWF has historically outperformed most models in the past. The ECMWF shows a high pressure system in the Southeast, off to the east of Florida.

Moving quickly along, we now visit the ECMWF Ensemble prediction system, or the ECMWF EPS. This forecast is for the same timeframe as above- the evening of December 20th. We see our storm system is on the Michigan/Indiana/Ohio border, with a central minimum pressure of 998 millibars. Considering the ECMWF EPS is composed of 51 separate forecasts, and we are 144 hours away, the strength of this system that is being projected is astounding. It's not that common to see an average of 51 forecasts, 6 days out projecting such a strong storm system. The ECMWF EPS believes the system will be submerged in cold air, as shown in 850mb temperatures (colors). This, combined with tightening of isobars, makes me believe that a blizzard-like solution could evolve in the Upper Midwest and South Canada into Michigan.

The GFS model has not been mentioned, but for those wondering, it is similar to the ECMWF EPS solution above as far as placement of the system goes.

Going back to the ECMWF model, we can see that it is projecting very heavy snow to fall over some Midwestern states. Snow is shown as precipitation north of the dark blue 32 degree line:

ECMWF Precip/Temp forecast for the Morning of Dec. 20

ECMWF Precip/Temp forecast for the evening of Dec. 20

ECMWF Precip/Temp forecast for the morning of Dec. 21
I don't have access to snow depth forecasts from the ECMWF, but if I were to give a rough estimate of snow amounts from the above few images, I would predict the following accumulations for the following locations:

-Northeast IL: 6 inches+
-South WI: 5-8 inches
-Michigan: 5-8 inches+


(Begin updated data) Now, I bring to you the updated 12z model suite and some information you'll find pretty darn interesting.

The ECMWF Ensembles are farther south than their 0z counterparts this morning, as shown in the below images:



As I explained, to have the ensemble system so confident on such a strong storm system this far out is very encouraging to those wanting snow. Now, in this scenario, Wisconsin, Iowa and maybe north Illinois would get the heaviest snow before the system moves off to the east. When that happens, we see a massive cold air dump from the heart of Canada into the Northeast. If this happens, heavy lake effect snow is likely. Not possible, LIKELY. This would be air that doesn't force you to wear a heavy jacket, it chills you to the bone- and then some. The warm Great Lakes will take this and, if such a solution happens, vigorously expel potentially high amounts of lake effect snow. The fact that this is the ECMWF EPS showing this happening is great news for the states I mentioned above for snowfall, because when you have the ECMWF Ensembles on your side, you're looking pretty good as far as probability of a forecast happening goes.

Now onto the interesting part. As I previously stated, the ECMWF is my trusted model at the moment. I will show you why I trust it's 0z solution just below:


These are North America model track errors. Basically, the higher the model is on this chart, the worse it has verified. If we look in the long range, we see there is one blue line that actually goes down in the long range, meaning it gets more accurate as time goes on (not a common sight at all). This blue line is none other than the ECMWF model. The Hour 120 forecast (just about when this storm gets going) is looking very well according to this chart above- all the more reason to trust the ECMWF. If you're looking for the GFS, it's that red line that is too high up to see at hour 96. (For those of you wondering about the 12z models, the GFS is about the same as the red line above, and the ECMWF has not come in yet.)

Forecast Preference: 12z ECMWF EPS / 0z ECMWF
Confidence: 45%

I'll have my first call out tomorrow morning. Until then, updates on this system may be found on the Facebook page at www.facebook.com/TheWeatherCentre

Andrew

December 20-22 Potential Blizzard Event

See the updated version HERE

I have upgraded my thinking of this winter event from a 'Significant Storm' to a 'Blizzard' event. I'll now explain why.

This is the ECMWF forecast for Hour 144, a.k.a. the evening of December 20th. We see a very strong storm system in northeast Illinois at this time with a minimum central pressure of 986 millibars- pretty strong for an onshore system. Let me first start off discussion of the model by saying I trust the ECMWF more than other models at the moment, as although all models have been having troubles, the ECMWF has historically outperformed most models in the past. The ECMWF shows a high pressure system in the Southeast, off to the east of Florida.

Moving quickly along, we now visit the ECMWF Ensemble prediction system, or the ECMWF EPS. This forecast is for the same timeframe as above- the evening of December 20th. We see our storm system is on the Michigan/Indiana/Ohio border, with a central minimum pressure of 998 millibars. Considering the ECMWF EPS is composed of 51 separate forecasts, and we are 144 hours away, the strength of this system that is being projected is astounding. It's not that common to see an average of 51 forecasts, 6 days out projecting such a strong storm system. The ECMWF EPS believes the system will be submerged in cold air, as shown in 850mb temperatures (colors). This, combined with tightening of isobars, makes me believe that a blizzard-like solution could evolve in the Upper Midwest and South Canada into Michigan.

The GFS model has not been mentioned, but for those wondering, it is similar to the ECMWF EPS solution above as far as placement of the system goes.

Going back to the ECMWF model, we can see that it is projecting very heavy snow to fall over some Midwestern states. Snow is shown as precipitation north of the dark blue 32 degree line:

ECMWF Precip/Temp forecast for the Morning of Dec. 20

ECMWF Precip/Temp forecast for the evening of Dec. 20

ECMWF Precip/Temp forecast for the morning of Dec. 21
I don't have access to snow depth forecasts from the ECMWF, but if I were to give a rough estimate of snow amounts from the above few images, I would predict the following accumulations for the following locations:

-Northeast IL: 6 inches+
-South WI: 5-8 inches
-Michigan: 5-8 inches+

Forecast Preference: ECMWF
Confidence: 55%

Andrew

Are We Living In A Computer Simulation?

Recently, the idea that the universe as we know it is a computer simulation has emerged. What do you mean 'computer simulation', you ask? Well, let's see what philosopher Nick Bostrom says on this topic.


In 2003, Swedish philosopher Nick Bostrom proposed the idea that the world we live in, and indeed we ourselves, may be a computer simulation run by people in the far, far future. Now, a team of physicists from the University of Washington report that they may have a way to test that idea.
Bostrom's 'Simulation Hypothesis" proposes that one of the three following possibilities is true:
1) The human race will go extinct before it ever reaches a "post-human" stage of evolution that possesses a technological level capable of running complex computer simulations of our universe (what he calls "ancestor simulations"),
2) The human race evolves to a "post-human" level and is capable of running ancestor simulations, but does not do so — possibly due to the energy requirements needed, or perhaps due to ethical concerns about simulating self-aware beings,
3) We (in the here and now) are most certainly living in a computer simulation.
Furthermore (and this is where it gets a bit 'trippy'), unless we are currently living in a computer simulation, it is highly unlikely that the human race will evolve to a post-human level that runs ancestor simulations. (text from Yahoo! Voices)

Basically, Bostrom believes that one of three things is true in life as we know it: Humans go extinct before we reach a stage of intelligence where we can simulate the universe (called the post-human stage), humans evolve to this post-human stage, but, for ethical reasons and the like, we choose not to run these simulation. The final (and most intriguing possibility) is that we do live in a computer simulation.

Researchers from the University of Washington believe they can find out whether such an idea is true. By using computers of immense power (called lattice quantum chromodynamics, or QCD), the researchers could find out if our world could be simulated. The basic idea of QCD computing is that one starts with the very basic physics components that exist today. Now, all computers have limitations. For this particular computer scenario, researchers are forced to use a cube-shaped grid to make simulations on these computers. So, by using the basic components of physics laid out on a cube-shaped grid, researchers could run a simulation. If the simulation from the researchers prints out a situation similar to ours (we don't know just what that could be, but in physics, the possibilities are endless), then we've just found evidence that everything we know is being controlled by more advanced life forms, and we're in a computer simulation somewhat like The Sims.

If such evidence exists that we are actually in a computer simulation, what do we do next? Well, the logical thing to do would be to attempt to make contact with these advanced life forms, which would be fairly difficult. Because we would be a simulation and our 'creators' would be more advanced, the technology we are using could be far outdated and thus unknown (a.k.a. forgotten, like the Spanish Conquistador's mediocre guns) to the advanced lifeforms. The awkward-ness then arises- we are essentially under the control of our creators. We start to wonder, should we notify our potential creators that we are now essentially self-aware? Much like The Sims, our creators could have the power to obliterate us in a flash. But that hasn't happened in the last several million years, so I don't believe it will happen soon.

That's what could happen if evidence is found we are in a computer simulation. There is no way of knowing if we are in a simulation, and the chances of ever finding out we are 100% sure that we are in a computer simulation is next to nothing.

It's pretty difficult to even comprehend what living in a computer simulation could mean for us, but it's one of the exciting things facing our world today.

Andrew