2017 Atlantic hurricane season (Reimagined by the GFS)

The Global Forecast System (GFS) is the primary short-to-medium range global synoptic model operated by the National Centers for Environmental Prediction (NCEP). The GFS runs four times daily and forecasts out to 384 hours (16 days). Within this range, the GFS is liable to forecast the formation and development of subtropical and tropical cyclones. While the GFS does show relatively remarkable accuracy for the most part, it too can fall victim to phantom storms and overly bullish forecasting – between 2004 and 2011, the GFS forecast 133 correct cyclogenesis scenarios but produced 250 false alarms.

If we took the worst-case scenario runs from the GFS, what would the 2017 Atlantic hurricane season look like? This page seeks to find out that answer. Certain rules apply for model storms to be included, however – see rules.

Rules
The GFS and its sister models typically produce numerous modelcanes over the course of a season (perhaps 50 or more unique events). Many of these cyclogenesis events, however, are transient, and only appear for one or two runs. To maintain consistency and filter out these errant solutions, the following rules apply for storm inclusion on this page:
 * Tracks and storms are based on the worst-case scenario from the operational GFS, with some liberties taken on the track depictions if the storm remained active at the end of the selected GFS run. That also means no GFS-Parallel or GFS-Ensemble storms or storms from other models.
 * If the strongest run for an actual storm came before the storm formed, then the modelcane is selected over runs while the storm was active.
 * The operational GFS must have depicted the development of the subtropical/tropical cyclone for at least four runs before truncation (240 hours) or eight runs total (before or after truncation, and out to 384 hours); the strongest of these runs is depicted here. What is considered a cyclogenesis event and the classification of model solutions as unique or repeated cyclogenesis events is at the discretion of the author of this page.
 * If the actual storm was stronger than any GFS run (including runs while the storm was still active), or more of a worst case scenario, the strongest GFS run is chosen regardless. This, of course, is rare.
 * More attention was given to the GFS after June 1; thus, activity depicted here prior to June 1 may not be true to the GFS' actual modelling trends.
 * Storm strength is at the discretion of the editor of this page, but is generally based on GFS run data. Occasionally storm winds are based on Brown et. al (2005).
 * Storm names are listed as if the storms did indeed form at the times depicted. Thus a storm's name may differ from what it actually was in reality (e.g. GFS Cindy in the Gulf of Mexico formed prior to GFS Bret in the Central Atlantic, and thus they are shown here as Bret and Cindy, respectively).

Tropical Depression Two

 * In reality: Nothing happened; Run chosen: June 7, 12Z

In mid-June, an area of convection persisted in the Gulf of Panama. In an environment conducive for tropical cyclogenesis, a small area of low pressure developed within the convective envelope and tracked northward across the isthmus of Panama and into the southwestern Caribbean Sea on June 8. The system quickly developed, and on 18:00 UTC June 9 became sufficiently organized to be classified as Tropical Depression Two roughly 70 mi (115 km) northwest of Colón, Panama.

Despite the earlier developmental trends, Two failed to organize much further after cyclogenesis. Tracking slowly westward under the influence of the low-level easterly tradewind flow, Two eventually made landfall near Bluefields, Nicaragua with winds of 30 mph (50 km/h) on June 11. Continuing inland, Two dissipated later that day over Nicaragua.

Hurricane Bret

 * In reality: The eventual storm (Cindy) was broad and disorganized, and eventually struck the Port Arthur area as a tropical storm; Run chosen: June 11, 00Z



Hurricane Cindy

 * In reality: The eventual storm (Bret) was further south and weaker; Run chosen: June 8, 18Z



Tropical Storm Don

 * In reality: ; Run chosen: June 22, 00Z

