6 long range forecasting questions you had but were afraid to ask


What exactly is “long range” forecasting anyway?

Weather forecasting has several different ranges. A one to 5-day forecast is a “short range” forecast. 6 to 10 days is considered a “medium range” forecast. Forecasts for any period greater than about 14 days is called a “long range” forecast. Read that again twice. It may just help next time you check the weather on your favorite app WeatherPro.

As weather forecasting technology improves, forecasts of all time periods have also improved. But long-range forecasts, which were not that reliable not long ago, have become much better and continue to improve.

You will probably get perplexed, perturbed or puzzled by the next question, but we've got your back. (And your umbrella.)


How do you even spell “probabilistic”?

Weather forecasting for a few days in the future can be very specific. For example, the “Maximum temperature of 15 degrees Celsius with rain starting in the afternoon” is typical of a short-term forecast. Long term forecasts cannot be as specific. No doubt you are asking yourself, why? Well the information that drives the skill of the short-term forecasts gets less important the longer the time from when the forecast is made. The long-term forecasts become more uncertain. But, there is still enough information to know the chances of ranges of weather and climate conditions occurring. A typical long-range forecast might state that weekly temperatures three weeks from now have a 60% chance of being greater than normal temperatures. This is a probabilistic forecast. Using probabilities for long range forecasts allows the users to understand the chances of certain conditions from occurring. You see clearer skies now, right?



ECMWF, CFS, SCAND. Why is long lead forecasting such an alphabet soup?

Like hail, traffic jams and bad coffee, acronyms are an unfortunate part of our lives. They probably won't go away anytime soon. Acronyms are used to quickly provide a lot of information quickly when a particular phrase is used repeatedly. It turns out that long range forecasting is no different because there are many sources of information to inform a well-made forecast. Who knew? Now you do.

The ECMWF is the European Center for Medium Range Weather Forecasting. They provide an important weather model, a mathematical representation of our atmosphere, that is used as a critical component for long term forecasts. In fact, the ECMWF creates a 46-day weather forecast from their model twice per week. The US National Atmospheric and Oceanic Administration Climate Forecast System Version 2 (that’s a mouthful!) is another important model for long range forecasting.

And finally, meteorologists monitor and forecast many climate indices that include the AO (arctic oscillation), PNA (Pacific North America Pattern), and the SCAND (the Scandinavian pattern) are all modes of the atmosphere that can impact temperatures and precipitation over very large regions. Monitoring these indices help to create a long lead forecast.    

You are probably wondering what the other acronyms mean? Well, you'll have to wait for our next long range forecasting blog. Stay tuned.

What’s trendy now in the world of long-range forecasts?

Many long-range forecasts present information relative to normal. No doubt you are asking yourself again, well,  what is normal? Especially with regards to the weather anno 2019.  

Climatologists define long periods of time. Let's say, for example, 30 years. They do this to measure the average climate for a region.  An important question for long range forecasts for monthly and seasonal forecasts is “what is normal?” because of temperature trends measured over many years.   Good long-range forecasters consider the climate trends in their forecasts because what is “normal” today may not have been the same normal 30 years ago.

Think back about the increase in acronyms. The trends in long range forecasts continue to evolve. We will elaborate on those another time.


Analog forecasting. Is that about knowing which LPs are in that corner record store?

Like in music. Analog is back. Actually, it never left the world of forecasting. Long range forecasts sometimes look back to see forward. Say what?

A frequently used technique for long range forecasting is to find time periods in the past 70 years that the weather and climate were very similar to today. No, that is not a type meant to be 20. That's correct. 70!

Finding the weather and climate conditions that occurred after those similar periods in the past helps to forecast today’s future. This is called analog forecasting. Analog forecasting is a useful tool because sometimes previous weather patterns can be very similar to what we are experiencing today.     

Next time you are in a music store, ask that hipster mister know it all if he knows anything about analog forecasting. Now you do.


Honey, don’t forget to turn off the teleconnection.

Remember the acronyms? The patterns these acronyms identify are called “teleconnection”. This has nothing to do with the teleprompters used to aid people giving speeches.

This phenomena describes a process in which conditions at locations that are very far apart can be related to each other. See that image at the top of this blog? That shows the impact of a strong Arctic Oscillation and the impact on temperatures during winter times.The map shows much greater chances of warmer than normal conditions over Scandanavia, colder than normal over eastern Canada, and warmer than normal temperatures over the Eastern United States.This pattern is typical of the Arctic Oscillation teleconnection.


Interested in knowing more long range forecasting? Need further clarification? We are happy to help.

Get in touch with us 




A one to 5-day forecast is a “short range” forecast. 6 to 10 days is considered a “medium range” forecast. Forecasts for any period greater than about 14 days is called a “long range” forecast.

Jan Dutton
Prescient Weather