The traditional method of making a weather forecast is to…
take the best model available and run it until it loses its skill due to the growth of small errors in the initial conditions.
-skill is typically lost after 6 days or so, depending on the season.
What is “ensemble forecasting”?
A method that produces forecasts with skill up to 15 days after the initial forecast.
- uses many model runs with slightly different initial conditions instead of using just one model run.
- an “ensemble mean”, or an average, of the different forecasts is created.
- ensemble mean will likely have more skill because it averages over many possible initial states and essentially smoothes the chaotic nature of the climate.
- -it is now possible to forecast probabilities of different conditions because of the large ensemble of forecasts available.
Every day at 00z, 06z, 12z, and 18z global weather observations are collected, transmitted to major weather centers and, with the aid of a global model “first guess field” to fill in gaps, used to produce a snapshot of the global atmosphere (i.e.) the model analysis or 0-hr forecast. The snapshot includes…
winds, pressure, moisture, temperature, etc. at multiple vertical levels in the atmosphere and at grid intersections of about 100 km.
what do the “observed” global fields provide?
The initial conditions for numerical models that integrate the equations of motion of the atmosphere forward in time to produce a forecast.
What are causes of uncertainty in the ensemble forecast initial conditions?
Instrument error, spatial/temporal sampling error, lack of data, and erroneous influences of the first guess background field.
-bc of these, an individual model run merely “samples” one of many possible current and future states of the atmospheric circulation.
What are perturbations?
slightly altered initial conditions. Somewhere bt 12 and 30 numerical model runs are executed daily using these perturbations.
What are the “ensemble members”?
The 12 to 30 model runs using perturbed initial conditions. Theoretically represent the current best estimate of the distribution of atmospheric states expected in the atmosphere out to forecast times of 15 days.
-one can also use the members to estimate probabilities of certain events, such as much below or much above normal temperatures. (dependent on the spread in the forecasts of individual members)
What is the “ensemble mean”?
an average of all ensemble members
Probability Density Function (PDF)
- the ensemble prediction approach attempts to define the PDF of atmospheric variables.
- for example, the temperature predicted at a specific location would have a most likely value (the ensemble mean) but would also have values that are scattered about the mean.
- this forecast scatter bt ensemble members increases with forecast lead time and should eventually approach the scatter observed in ever day weather.
NWP Process:
(1) Gather Observations
(2) Data Assimilation
(3) Numerical Weather Predictions
(4) Forecast Postprocessing
(5) Issue forecasts, Evaluate
Two Categories of problems with ensemble predictions:
(1) Chaos
(2) Model Error
