The goal of this website is to provide forecasters with a novel set of visual tools to evaluate the full ensemble spread of forecast solutions for a variety of forecast variables. This website is developed at Stony Brook University and motivated by a project with Consolidated Edison, Inc. of New York and the Stony Brook Collaborative Science, Technology, and Applied Research (CSTAR) program.

This site uses data from four ensmbles: The Global Ensemble Forecast System (GEFS), the Canada Meteorological Center (CMC) Ensemble, the Fleet Numerical Meteorology and Oceanography Center (FNMOC) Ensemble, and the Short Range Ensemble Forecast System (SREF). All images can be organized into two basic categories: 1) Ensemble cyclone track images and 2) Ensemble non-track images. Below we outline a brief description of the overall website. For more information on the individual images, please visit the help files associated with that type of plot.

1) Ensemble cyclone track images: Displaying all forecast cyclone tracks would result in an overwhelming and confusing bunch of images. For this reason, we employ methods to reduce the clutter. Two basic methods are used here; A) Storm Separated Tracks and B) Moving Window Tracks. The first method displays cyclone tracks that move through a pre-determined region and can be used to determine the ensemble spread of cyclone intensity and position. The second method plots cyclone tracks within a 24 hour window around that forecast time and is a useful tool for elucidating ensemble spread with regard to intensity and timing. When applicable, both tropical and extra-tropical cyclones are plotted for different regions in the Northern Hemispehre.

2) Ensemble non-track images: To compliment the cyclone track plots, additional plots of ensemble probability/mean data are created: A) 32/90 F Isotherm and accumulated 6-hour precipitation probability, B) Previous 24-hour minimum/maximum 32/90 F Isotherm and accumulated 24-hour precipitation probability, and C) 32/90 F Isotherm and 10-m wind speed probability. These plots are operationally being generated for the eastern United States and Alaska region.