ÃÛÌÒÁµÈË

Note: This is Part II of a two-part series. In Part I, we looked at How Conventional ÃÛÌÒÁµÈË Forecasting Works.

There are several different methodologies and phenomena that can come into play in generating long-range forecasts, including:

1. Climate models: These are long-range computer-generated forecasts that are similar in concept to the short-range computer-generated forecasts that are a foundation of forecasts of weather over the upcoming days to weeks. They use simplifications different from those of the short-range forecast models and are designed to forecast monthly and seasonal trends rather than day-to-day and hour-by-hour weather.
2. Teleconnections: These are patterns in the atmosphere that typically persist from weeks to years and influence the temperatures, precipitation, storm tracks, and jet stream over large areas. Think of a teleconnection as being like construction on an urban highway: It may impact the traffic flow for months (much as a teleconnection impacts the flow of the atmosphere), controlling the traffic flow completely and causing it to back up for miles during rush hour (much as a teleconnection can lead to a series of snowstorms), while at night causing traffic to slow down only through the construction zone (much as a teleconnection may not control the weather continuously, with dry periods occurring between the storms that it brings).
3. Analogs: The concept behind analogs is that if we can find past weather patterns similar to the current one, the weather that follows may also be similar.  Of course, these patterns are only similar and not exactly the same—if weather patterns were that simple, everyone would always have accurate long-range forecasts.
4. Solar cycles: The most controversial of the four methodologies, the use of solar cycles is based on the concept that output from the Sun controls our weather. Most meteorologists and climatologists have scoffed at this idea, because changes in solar energy are tiny and they did not believe that these were large enough to affect the weather. But recent research has shown physical mechanisms by which these small changes in solar output can be magnified in the upper atmosphere and funneled downward to affect and even control the weather. You might ask which of these methodologies we use here at The Old Farmer’s ÃÛÌÒÁµÈË to determine our long-range weather forecasts.

Photo credit: BrianAJackson/GettyImages

The answer is that we use all of them to some extent.

• We look at the climate model forecasts and consider them, although we count them less than the other inputs and, on average, our forecasts have been more accurate than those of the climate models.
• We strongly consider the expected teleconnections, but our experience has been that the published forecasts of teleconnections made 6 months to 2 years in advance, which is what we need for our weather forecasts, are often incorrect. Instead, we believe that changes in solar activity strongly influence the phase and strength of the various teleconnections, and we use the solar cycle to forecast teleconnections.
• The solar cycle is the predominant input that we consider in making our long-range forecasts. Some studies have shown correlations between the length of a cycle and temperatures, or between the magnitude of a cycle and temperatures.  But our methodology goes far beyond any simple correlations, finding analogs to each phase of the current cycle, determining the weather that prevailed in those analog periods, and making adjustments for differences between the analogs and for changes in the climate.

Our methodology has been developed over two centuries of use, with enhancements and refinements added as we have learned more about the physics of the atmosphere, interactions between the ocean–atmosphere system, teleconnections, space weather, and climate change and have had the ability to add computer technology and power to our research and methodology.

We are not always right, but we believe that verification studies show that our long-range forecasts are better than those from any other source.

In all of this, there is one thing of which you can be absolutely certain: We are always striving to refine our forecast methodology to make it even more accurate and useful than ever before.