Weather Monday 2/2/15 (Forecasting Part 1: putting the pieces together)

“Nothing is Ever as it Seems".”

Weather forecasting is much like making a quilt. To make a quilt you have to put all your skills, your vision, and the pieces together. In order to predict future weather, you have to put all your knowledge, skill, and intuition together. Both forecasting and quilt making can be difficult, but I think forecasting the weather is harder by far.

 

 

 

 

Weather is complex

How the atmosphere behaves is extremely complicated. Think about how hard it is to make a quilt with Y-seams compared to making a nine patch. Or how about dealing with getting the center of a mariner’s compass to lay flat and have all the seams meet perfectly. Weather forecasting is definitely not the nine patch.

You might think forecasting something like the high and low temperature would be a simple thing. it is actually not easy to forecast temperature.  The temperature depends on so many things that it becomes something that is hard to predict.

• Temperature is controlled by how many clouds there are. Clouds keep us cooler during the day and warmer at night. So knowing the amount of cloud cover is essential to forecasting the temperature.
• What air mass is moving into the area. Today we might be in a warm air mass, but tomorrow a cold air mass might move in. When will it move in? How cold is it in the air mass that is coming? Will we have clouds or not? How many clouds?
• Do you live near water or inland away from water? If you live along the coast you know that your temperature does not vary much during the day. The high might reach 65 degrees during the day and drop to 57 degrees at night. However, if you live inland away from the coast your temperature during the day might get to 80 degrees on the same day and 65 degrees at night. Coastal cities have a smaller range in temperature than inland cities. And what if you live in a desert? It can get to over 100 during the day and drop to below freezing. Why? Lack of water and clouds.
• Temperature varies with latitude and altitude. So the cities closer to the equator are different than cities closer to the poles. Also temperature decreases the higher up you go in the atmosphere. The temperature on a mountain top is colder than in the valley below. So more factors to take into consideration.
• How many daylight hours is the city having? We get less daylight in the winter and more in the summer.
• Do you have snow covering? Having the ground covered with snow makes it colder than if you did not have the snow covering.

So you can see that temperature is not simple and straight foreword. It is not as easy as making a simple nine patch block. So a weather person has a lot of pieces to put together to figure out what the temperature is going to be.

How accurate are forecasts?

You can’t believe how many times that once someone knows that I am a meteorologist I hear, “Forecasters are never right.” Well for sure they are not right all the time, but their accuracy is a lot better than most people realize. I think we have selective memory when it comes to that. We remember when they are wrong, but forget when they are right.

So how accurate are forecasts and how do you even measure accuracy? If they say the high temperature is going to be 70 degrees and it only reaches 68 is that accurate or not? If they say that all these areas are going to get a lot of snow and only 3/4 of these areas actually get a lot of snow is that accurate or not?  Yes I am talking about last week’s blizzard. NYC did not get the amount of snow forecasted, but Boston certainly did. I don’t thing people realize how hard it is to forecast a winter storm. I doubt we will ever be 100 % accurate.

Here is a website that calculates the accuracy of different forecasting sites like The Weather Channel, Accuweather, Weather Underground, etc.

Forecasting accuracy gets worse the further out in time you try to forecast for. 12 hour and 24 hour forecasts are the most accurate. I usually don’t pay attention to forecasts for more than 2 days out.   I will explain why in the next weather blog. So what this site measures is how accurate the temperature and precipitation forecast is for a 3 day forecast. That means what did the different forecasters say the temperature and precipitation was going to be 3 days later? And how accurate were they? So why just temperature and precip? Well to get temperature and precip  right you have to get lots of other things right. So they are a good test of accuracy.

What you will find if you go to this website is the accuracy varies a lot depending on the location. Some locations are easier to forecast for than others. For example:

• NYC 75% accuracy
• Boston 70% accuracy
• San Francisco 82% accuracy
• Phoenix 90% accuracy

So being a forecaster in Phoenix, Arizona is a lot easier than one in Boston, Mass. Of course you can also see that they are not always wrong. 75% accuracy for a 3 day forecast is pretty good. Certainly a lot better than tossing a coin.

So what about the groundhog?

Since today is Groundhog day I thought it would be a good day to bring Phil up. I guess he saw his shadow today.

His accuracy is only 39%. So I would not rely on Phil. Forecasters have gotten better at doing longer range forecasts for determining what kind of winter we will have or what kind of hurricane season it will be.

Last winter and this winter they forecasted a bad winter for the east coast of the USA. Last winter we had a bad winter and this year is shaping up to be bad as well. Cold and lots of snow! Although generally we will know what the winter will be like after it is over!

So I leave you with this forecast from Accuweather! Enjoy!!!! I will be back next Monday with Part 2 on forecasting.

Meanwhile thanks for reading. If you have any questions or comments just post them.

Chris