Ensemble Prediction System
A 100% Chance of Better Aviation Weather Forecasting
Even the most seasoned business aviation flight crews pay heed when they hear they’re flying abroad. Yes, there are support systems, processes, and other resources to help manage the man-made uncertainties of global flight operations—regulations, ground handling, fly over permits, etc. But no one has yet figured out how to successfully manage the most unpredictable part of any flight—weather.
Until now.
In the last 20 years, since the introduction of NEXRAD, there hasn’t been much breaking news in weather forecasting. Most weather predictions were based on Numerical Weather Prediction (NWP) modeling, combining environmental factors, satellite imagery, observations, historical data, and good old-fashioned physics to create a model from which a weather forecast is made.
But here’s how a single weather model has led you astray. Imagine that you were one degree off the heading of your last flight. Not a big deal if you catch it within a few minutes and correct. But, it’s a much bigger deal if you don’t catch it for a few hours. Weather models work the same way. If one variable is just slightly off in a new forecast, then the forecast is less accurate and reliable with each passing minute.
Since weather is not an exact science, each of the variables and algorithms used to make a particular forecast model vary. In the end, flight crews are left to sort out different models making different predictions. But what if there was a way to combine all the models, and all the factors that went into those models, into a single prediction?
A Better Batting Average
Say your favorite baseball player is coming to the plate. How do you know if he’s likely to get a hit? His batting average. A statistician divides the number of his hits by his number of at bats and derives the probability he’ll get a hit. While that number is somewhat accurate, it could be moreso when additional factors are considered. How well does he hit against left-handed pitchers? How well does he hit with men on base? How well does he hit on the road versus his home ballpark? When different factors are considered, and more data is included, a batter’s hitting “forecast” is far more accurate.
In a very oversimplified sense, weather forecasting has moved to the same kind of approach using something called the Ensemble Prediction System (EPS). Used by many forecasting organizations around the world, the EPS approach begins with a traditional forecast model, then slightly “perturbs” (or alters) different factors of that model, running 50 additional models to get a range of possible forecasts. Now, forecasters have 51 different models to analyze to make a final prediction. Imagine the power of running 51 different flight plan scenarios, factoring changes in weather, traffic, or wind speed/direction, before creating your final flight plan. It’s powerful.
Since EPS-based weather forecasts provide more reliable information, forecasters can offer flight crews more:
- Narrow ranges in forecast predictions
- Detailed and specific data
- “Certainty” with each forecast
EPS forecasting will have a ripple effect on your operations, especially for international travel, including:
- Weather Alerting Systems
- Decision Support Systems
- Mission Profile Support
- Optimization Systems
One Last Unanswered Question
Great, you think. I’ll start looking for EPS-based weather forecasts. But if everyone is going to start using EPS, how do I know which forecast is best? Look for forecasts based on the European Center for Medium Range Weather Forecasts (ECMWF)model. Formed in 1975 and based in the United Kingdom, this organization leverages the resources and forecasting intelligence of 34 European member states. Over the years, its global forecasting has proven to be the most reliable. The ECMWF gained public notoriety in 2012 when it predicted the nearly exact path of Hurricane Sandy four to five days before other models. The ECMWF forecast offered the kind of ample time and specificity that helped local and state governments along the Eastern Seaboard a chance to warn its citizens earlier and better prepare for the worst.
Jeppesen believes so strongly in the EPS-based forecasts provided by the ECMWF, that it incorporates their data in the weather forecasting services it makes available to business aviation operators like you. Through Jeppesen Weather Alerts (JWA), Jeppesen offers you a dashboard system, customized for your operation, to help you fully utilize this new level of weather forecasting information. Learn more by clicking here.
A 100% Chance of Better Aviation Weather Forecasting
Even the most seasoned business aviation flight crews pay heed when they hear they’re flying abroad. Yes, there are support systems, processes, and other resources to help manage the man-made uncertainties of global flight operations—regulations, ground handling, fly over permits, etc. But no one has yet figured out how to successfully manage the most unpredictable part of any flight—weather.
Until now.
In the last 20 years, since the introduction of NEXRAD, there hasn’t been much breaking news in weather forecasting. Most weather predictions were based on Numerical Weather Prediction (NWP) modeling, combining environmental factors, satellite imagery, observations, historical data, and good old-fashioned physics to create a model from which a weather forecast is made.
But here’s how a single weather model has led you astray. Imagine that you were one degree off the heading of your last flight. Not a big deal if you catch it within a few minutes and correct. But, it’s a much bigger deal if you don’t catch it for a few hours. Weather models work the same way. If one variable is just slightly off in a new forecast, then the forecast is less accurate and reliable with each passing minute.
Since weather is not an exact science, each of the variables and algorithms used to make a particular forecast model vary. In the end, flight crews are left to sort out different models making different predictions. But what if there was a way to combine all the models, and all the factors that went into those models, into a single prediction?
A Better Batting Average
Say your favorite baseball player is coming to the plate. How do you know if he’s likely to get a hit? His batting average. A statistician divides the number of his hits by his number of at bats and derives the probability he’ll get a hit. While that number is somewhat accurate, it could be moreso when additional factors are considered. How well does he hit against left-handed pitchers? How well does he hit with men on base? How well does he hit on the road versus his home ballpark? When different factors are considered, and more data is included, a batter’s hitting “forecast” is far more accurate.
In a very oversimplified sense, weather forecasting has moved to the same kind of approach using something called the Ensemble Prediction System (EPS). Used by many forecasting organizations around the world, the EPS approach begins with a traditional forecast model, then slightly “perturbs” (or alters) different factors of that model, running 50 additional models to get a range of possible forecasts. Now, forecasters have 51 different models to analyze to make a final prediction. Imagine the power of running 51 different flight plan scenarios, factoring changes in weather, traffic, or wind speed/direction, before creating your final flight plan. It’s powerful.
Since EPS-based weather forecasts provide more reliable information, forecasters can offer flight crews more:
- Narrow ranges in forecast predictions
- Detailed and specific data
- “Certainty” with each forecast
EPS forecasting will have a ripple effect on your operations, especially for international travel, including:
- Weather Alerting Systems
- Decision Support Systems
- Mission Profile Support
- Optimization Systems
One Last Unanswered Question
Great, you think. I’ll start looking for EPS-based weather forecasts. But if everyone is going to start using EPS, how do I know which forecast is best? Look for forecasts based on the European Center for Medium Range Weather Forecasts (ECMWF)model. Formed in 1975 and based in the United Kingdom, this organization leverages the resources and forecasting intelligence of 34 European member states. Over the years, its global forecasting has proven to be the most reliable. The ECMWF gained public notoriety in 2012 when it predicted the nearly exact path of Hurricane Sandy four to five days before other models. The ECMWF forecast offered the kind of ample time and specificity that helped local and state governments along the Eastern Seaboard a chance to warn its citizens earlier and better prepare for the worst.
Jeppesen believes so strongly in the EPS-based forecasts provided by the ECMWF, that it incorporates their data in the weather forecasting services it makes available to business aviation operators like you. Through Jeppesen Weather Alerts (JWA), Jeppesen offers you a dashboard system, customized for your operation, to help you fully utilize this new level of weather forecasting information. Learn more by clicking here.