It's the day before Christmas and all through my house not a model was budging, not even an ounce. UGH! I will get to the late weekend storm in a bit. First, I want to tackle a topic that has come up lately.

Ok, folks, I am trying to stay in a good Christmas mood but am increasingly frustrated with the lack help I am getting from our trusty models in the last few weeks. I am also a little puzzled by some of the comments I am getting and am going to try to see if I correct some of your misconceptions.

Here is a quote from one of the reader comments I received yesterday;

I understand that we make our predictions based on computer models... etc., but once we start reporting information solely based on the 'models', where does the expertise come into play. I mean anybody can read data from a model and report that. The reason why I come to the AccuWeather is because I want to hear what the experts have to say. What is wrong with going out on a limb and making a scientific forecast based on your experience rather than waiting for a consistent model prediction to setup ? (From Paul)

And here was my response;

I have 35+ years of experience and you are right, there are times to make a forecast based on one's experience. However, there are also times, especially more than 3-4 days out, where it makes NO sense going out on a limb with what amounts to a guess. I could do that but what does that get you? Wouldn't you rather KNOW that it is a tricky situation and that by waiting a little longer you would actually get a more accurate forecast?

It is a fine line I and others walk when trying to make forecasts and in dealing with the computer models. Computer models are a guidance to the possible coming weather. It takes professional interpretation of those models and experience with their biases and errors that go into making a forecast. It also takes experience with local variations in a given area to make a complete and accurate forecast. Yes, we do rely on models to help us but that is in the end what it is, one of the tools to help us forecast the weather.

The problem we have had in recent weeks goes to my response to Paul. If I see 5 different models showing no consistency in their solutions even between the same model from run to run, how can I choose the right solution? I am not talking about small differences here. I am talking about the difference between a storm and no storm, not between a storm arriving at 4 p.m. or 7 p.m. Or it's a difference between a storm coming next Monday or waiting until Wednesday or Thursday. No small difference and no small difference that makes in your life. In this case, if I choose one model it amounts to nothing more than a guess. Now that guess may end up being right, but it would not have been based on any scientific process. What does involve a scientific process is to see which model becomes more consistent between runs and see if other models fall in line with it over time. You take this, along with your experience in pattern development, and THEN pick your model of choice.

This is a very small part of what goes on in making up a forecast. It is much more complicated, but I wanted to give a little on the models.

Next Storm:

The ridge protecting the entire West Coast from any storm through Christmas will break down over the weekend. One storm assaults that ridge on Saturday, but most of its energy is wasted on breaking down the ridge and moistening the air mass some. A shower chance does exist north-central and northern California coastal areas late Saturday or Saturday nights. It�s the second storm that has all along caused the problems. Still today we have a big difference in the models with one bringing that storm in on Sunday to central and northern California. The other extreme is that storm going so far west and then south of California it never rains from it. Then there are solutions somewhere in the middle. Right now, I am favoring a possible rain event aimed more at the southern third of California in the time period late Sunday or Sunday night into part of Monday. The most consistent models have had this solution and timing therefore the odds are greater than some semblance of this solution is the best bet.

Now to end things on a good note. I hope all that celebrate Christmas have a wonderful and great holiday tomorrow. Enjoy the people you are with and be thankful for what you have not yearning for what you don't have. I am thankful that my son Brian is here from his job on Mount Washington along with his girlfriend Laura and we can celebrate together. Then our extended family will be getting together on Saturday for a party. Next week we march toward 2010. Will El Nino weather pattern finally get established or will the El Nino pattern we are in continue not to deliver the amount of precipitation it is suppose to? We will walk that walk together for the rest of 2009 and in 2010. Have a great Christmas one and all.

Share this:

The storm that brought wild winds, showers and thunderstorms, and locally heavy snow for a time Monday into Tuesday is now ancient history. In its wake, a ridge aloft in the Eastern Pacific will build some more and move slowly eastward through Christmas Day. This will provide very tranquil weather with no threat of precipitation from Washington and Oregon southward to California, Nevada and Arizona... great news for those taking to the roads to get to the loved ones for Christmas. Notice I said the roads won't have any problems. If you are flying, you may encounter some major problems depending on where you are going. A huge snow and ice storm in the Plains is likely to bring major disruptions to air service into and out of those areas. Already, many flights have been canceled today. This kind of storm can have a rippling effect on other flights leading to the West.

What happens after Christmas? Somehow, my last posting before I took a few days off pre-Christmas was on the flip-flopping on the models. Once again, today the weather in the Sunday/Monday time period is quite uncertain from parts of Oregon on south and east. The models have been going back and forth on when and where the next storm was going to come eastward across the Pacific. The GFS, for example, did a flip and then a flop over the past 36 hours on this storm. The European is pretty much in the same camp as the GFS right now, but both the Canadian and UKMET models are more like last night's run of the GFS.

It does seem likely that a storm moving across the Pacific would bring some rain and mountain snow at a point from Sunday into Monday. But as to when is the biggest chance and where the most amount of precipitation will be is like shooting at a dart board at 50 feet. I think it's better to be a little conservative and wait for another run or two of the models to see some consistency before trying to be specific. Right now, I would think that all models would indicate no big storm is in the offing.

Check back tomorrow and see what the latest models have in store. It's always interesting. And maybe Santa will bring us what we want for Christmas in the way of weather for 2010.

Share this:

It has been a frustrating week trying to make long-range forecasts. Early in the week it looked like by the middle and end of next week the West would go back to a high-level blocking pattern, sending storms farther south again toward the Southwest, but by Wednesday and Thursday that hope had gone by the wayside.

Each day any particular long-range model looked different than the day before and rarely have the models looked similar to one another, except on Monday when they seemingly have forecast a wrong pattern change. I have received a few emails asking why all the changes this week and why does it always seem like the West have more of these flip-flops than other parts of the nation. I will briefly try to answer these questions.

First, the flip-flopping is not just in the West but has had ramifications everywhere in the nation. Case in point is the current East Coast storm that the models have been bringing farther and farther north the last couple of days. The West may see somewhat more of this happening than farther east and the main culprit is the vast Pacific Ocean to the west and less data available to input into the computer models. But another reason for flip-flops in abundance is the kind of weather pattern we are in. High energy (fast jet stream) and high-level blocks are many times very difficult for the computer models to grasp on to.

One thing that looks certain today is that the next storm in the West will move into the Northwest on Sunday with moderate to heavy rain west of the Cascades and snow in the higher Cascades with mostly rain making it into areas east of the Cascades. Some rain can get as far south as northern California Monday and maybe down to areas just south of the Bay Area to northern San Joaquin Valley Monday night and early on Tuesday. The chance of any precipitation in Southern California and Arizona seems unlikely through the middle of next week at least.

On a personal note. I will be away from my desk Monday and Tuesday but be back for Wednesday and Thursday before taking Christmas Day off.

Share this:

Low-level moisture trapped in the Central Valley of California by an inversion can lead to fog and low clouds that never break, or break very little, for days on end in the winter months. This is called Tule fog. It is named after a grass that grows in the wetlands of the Central Valley of California.

Tule fog is the number one leading cause of weather related casualties in California because visibility is reduced to near zero along two heavily traveled freeways, Interstate 5 and Highway 99. Here is a mid-afternoon satellite picture showing low clouds and fog in the Central Valley.

The weather pattern we are in for the next several days will continue to be conducive to widespread low clouds and dense fog that will have a hard time in breaking up, especially from near Sacramento southward to south of Bakersfield. High surface dew points will remain trapped below an upper-level ridge over the area. Night and morning visibility in the fog will drop to as low as a couple of hundred feet, making for hazardous travel. Temperatures where there are no breaks in the afternoon will probably stay mostly in the 50s.

Meanwhile, above the low clouds and fog in the foothills and mountains of the Central Coastal Range and Sierra, it's likely to be partly to mostly sunny and milder. Temperatures at the 5,000-foot level will likely be warmer than in the heart of the Valley by a good 5 to 9 degrees.

Speaking of warmth. Offshore flow Friday and Saturday in Southern California will bring the warmest areas up to near 80 both days, with Friday likely to be a degree or two warmer than Saturday.

Share this:

As I stated yesterday, the weather pattern this week has changed from last with the breakdown of the blocking high over Alaska. Now the storm track is into the Northwest, and several more storms are to come. One of these will be tonight through tomorrow bringing rain west of the Cascades, snow heavy at times in the Cascades, and periods of snow, freezing rain and rain to central and eastern Washington and Oregon. Valley rain and mountain snow is also likely in Idaho. Significant accumulations are likely in the Cascades with 1-2 feet likely. In the higher elevations east of the Cascades, expect 8-18 inches through tomorrow night. The next storm is due in Thursday night and Friday.

But the major long-range models are showing another change in the pattern next week, especially by late next week.

Friday 12Z GFS 500 mb:

Does this map look familiar? It should as it is quite similar to the pattern we had last week. A big blocking high over western Canada and eastern Alaska and a jet stream cutting underneath toward California.

The European model for the same time period has a pretty similar looking pattern.

Now there could be a prelude to all this earlier in the week, let's say about Tuesday, but that is less certain. Though there is no way to be certain how stormy this change in pattern will be, it does seem more likely than not that the upper-level pattern will again be different by the middle and end of next week than it will be the rest of this week.

Share this: