The future of Weather Forecasting is Probabilistic. So, Why Isn't Everyone Cheering?
Most of the American public understands the basic odd-making aspects of sports betting and games of chance. An individual who places a $10 bet on a horse that is listed as a 4:1 favorite knows that if this horse wins the race that he/she will receive $40 on a $10 bet. Yet far too often these same individuals who are comfortable with sports betting will throw up their hands when trying to understand what a probability of precipitation forecast of 40% means.
Back in the 1960s, a now-famous meteorology professor at MIT, Dr. Edward Lorenz, demonstrated theoretically that the limit of our ability to predict the occurrence of day-to-day weather forecasts was roughly two weeks. Nothing has changed in the intervening years to suggest that Professor Lorenz was incorrect in his assessment that a limit exists on our skill in forecasting day-to-day weather predictability.
This presentation will discuss why probabilistic weather forecasting is the way of the future, how probabilistic were forecasts are made, how probabilistic weather forecasts are communicated to the public, and how probabilistic forecasts should be interpreted. Mindful of the overall predictability limit of day-to-day weather forecasts established by Professor Lorenz noted above, we will also discuss how certain weather elements can be predicted probabilistically at longer time intervals than others. For example, skillful probabilistic high temperature forecasts might be made upwards of 7–8 days in the future while similar skillful probabilistic forecasts of heavy rainfall for a specific location may only be made 3–4 days in the future.
Back in the 1960s, a now-famous meteorology professor at MIT, Dr. Edward Lorenz, demonstrated theoretically that the limit of our ability to predict the occurrence of day-to-day weather forecasts was roughly two weeks. Nothing has changed in the intervening years to suggest that Professor Lorenz was incorrect in his assessment that a limit exists on our skill in forecasting day-to-day weather predictability.
This presentation will discuss why probabilistic weather forecasting is the way of the future, how probabilistic were forecasts are made, how probabilistic weather forecasts are communicated to the public, and how probabilistic forecasts should be interpreted. Mindful of the overall predictability limit of day-to-day weather forecasts established by Professor Lorenz noted above, we will also discuss how certain weather elements can be predicted probabilistically at longer time intervals than others. For example, skillful probabilistic high temperature forecasts might be made upwards of 7–8 days in the future while similar skillful probabilistic forecasts of heavy rainfall for a specific location may only be made 3–4 days in the future.