Using Ensembles for Short-Range Forecasting

Abstract

Numerical forecasts from a pilot program on short-range ensemble forecasting at the National Centers for Environmental Prediction are examined. The ensemble consists of 10 forecasts made using the 80-km Eta Model and 5 forecasts from the regional spectral model. Results indicate that the accuracy of the ensemble mean is comparable to that from the 29-km Meso Eta Model for both mandatory level data and the 36-h forecast cyclone position. Calculations of spread indicate that at 36 and 48 h the spread from initial conditions created using the breeding of growing modes technique is larger than the spread from initial conditions created using different analyses. However, the accuracy of the forecast cyclone position from these two initialization techniques is nearly identical. Results further indicate that using two different numerical models assists in increasing the ensemble spread significantly. There is little correlation between the spread in the ensemble members and the accuracy of the ensemble mean for the prediction of cyclone location. Since information on forecast uncertainty is needed in many applications, and is one of the reasons to use an ensemble approach, the lack of a correlation between spread and forecast uncertainty presents a challenge to the production of short-range ensemble forecasts. Even though the ensemble dispersion is not found to be an indication of forecast uncertainty, significant spread can occur within the forecasts over a relatively short time period. Examples are shown to illustrate how small uncertainties in the model initial conditions can lead to large differences in numerical forecasts from an identical numerical model.

Experimental ensemble forecasts of precipitation based on a convection‐resolving model

“…The relative influence of these two types of uncertainty depends on the properties of, for example, the synoptic situation, the boundary layer, the soil moisture, and the orography (Anthes et al, 1989;Stensrud et al, 1999;Bright and Mullen, 2002;Sattler and Feddersen, 2003).…”

A Comparison of ETKF and Downscaling in a Regional Ensemble Prediction System

“…Some other studies try to generate IC perturbations for REPS by using a regional version of traditional IC perturbation methods, such as the Breeding Growing Mode (BGM), Singular Vectors (SVs), Ensemble Transform Kalman Filter (ETKF), etc. It is proved that these methods can also trigger limited ensemble spread and benefit forecast skill for REPS [13][14][15][16].…”

Development of decision support products based on ensemble forecasts in the European flood alert system

“…It has been gradually expanded to short-range highresolution probabilistic forecasting in view of improving warnings for extreme weather events on local scales (Stensrud et al, 1999;Molteni et al, 2001;Nicolau, 2002). In hydrology, the main approaches developed for probabilistic streamflow prediction are based on: (1) generating ensemble runs with different calibrated hydrological models, (2) using analogbased techniques to statistically assess the probability of a future event based on observed past situations, or (3) nesting single or combined sources of uncertainty from model structure, parameters, input and/or measurements in rainfall-runoff simulations (Georgakakos et al, 2004;Bartholmes and Todini, 2005;Pappenberger et al, 2005;Siccardi et al, 2005;Carpenter and Georgakakos, 2006;Diomede et al, 2006;Gourley and Vieux, 2006;Vrugt et al, 2006, and references therein).…”