CFAN, Climate Forecast Applications Network, LLC.

                                                       © 2019.  All Rights Reserved 

  • Twitter Social Icon
  • LinkedIn Social Icon

 Telephone:  USA  +1.404.803.2012

 Email:  info@cfanclimate.com 

Tropical Cyclones

From the oil rich Gulf of Mexico or the South China Sea, to the

population centers along the Atlantic coast or Bay of Bengal,

CFAN’s forecasts can help you anticipate impacts further in advance

and with greater accuracy and precision. 

 

TropiCast empowers our clients in government and industry to visualize and better understand the overall risk of impact from a hurricane or tropical cyclone in the Atlantic, North Indian Oceans, and West Pacific. TropiCast can support your decision making for: offshore resource operation and safety, utilities operations and recovery, energy demand and trading, insurance and reinsurance needs, emergency management, ocean routing and fleet management, travel and tourism impacts, and supply chain interruption and restoration. 

 

Building upon our innovative tropical cyclone research [link], CFAN has been defining the state-of-the-art in tropical cyclone forecasting since 2007.  

Our extended-range tropical cyclone forecast products out to 15 days are unmatched in the weather services industry. TropiCast’s unique forecasts are produced using CFAN’s proprietary algorithms applied to global model output ensembles in near real time (referred to as ‘calibration’), which provide more accurate forecasts than raw (uncalibrated) model output with features including:

  • calibrated tracks from global model ensembles, which provide up to 3 days of additional lead time relative to raw model output tracks

  • probability forecasts of tropical cyclone formation (genesis) that have skill 3-7 days in advance

  • dynamic cone of uncertainty for tracks forecasts which provides more realistic track uncertainty at extended forecast horizons than raw model output.

  • probability forecasts of hurricane intensity that provide a more realistic assessment of intensity uncertainty

 

For each active system – a named storm or National Hurricane Center 'invest' – the active system page includes:

  • An interactive feature that allows you to overlay track and intensity forecasts from the global models plus NHC’s statistical and regional dynamical models

  • Realtime track and intensity verification for the global and regional dynamical models, for the past 4 days

  • Track probabilities created from synthetic tracks using a proprietary Monte Carlo analysis of the the ECMWF and NOAA global models, providing a dynamic cone of track uncertainty

  • Probability forecasts of maximum and intensity and minimum pressure

  • Forecasts of Integrated Kinetic Energy (IKE) and Cyclone Damage Potential (CDP) to help anticipate relative landfall damage

 

The images below illustrate the active system forecast for Hurricane Dorian (8/30/19 12Z) – at a critical point in the forecast when catastrophic impacts were anticipated for the northern Bahamas and the forecast models showed substantial disagreement on a Florida landfall. ECMWF and NCEP model data depicted in the images have been improved by CFAN’s calibration.

 

Atlantic tropical cyclones

 

CFAN has been providing tropical cyclone forecasts in the North Atlantic since 2007.  CFAN’s probabilistic forecasts of Atlantic tropical cyclone tracks and intensity consistently outperform both government and market forecasts beyond a 2 day time horizon, showing track skill within 300 miles out to 10 days (even before the tropical cyclones actually form).  

CFAN’s superior track forecast is illustrated by comparing CFAN’s forecast for Hurricane Sandy with raw tracks from ECMWF and GEFS, 7 days before landfall.

CFAN Forecast

ECMWF Forecast

GEFS Forecast

This view of the TropiCast dashboard shows forecasts for Hurricane Irma (2017) 6 hours prior to formation and 12 days before landfall on the southwest coast of FL. 

2017 track verification analysis.  At five days lead time, the average track error for the 2017 NHC forecasts was 170 miles.  CFAN’s five day track forecast error was smaller than the official NHC forecast by an average of 39 miles (26%). At 10 days lead-time  the average track error for CFAN’s forecast was 287 miles. CFAN's tracking and calibration improves upon the official ECMWF track forecasts by 7 miles at 5 days and 69 miles at 10 days.  Download the full Report on Track Verification Analysis for the 2017 Atlantic Hurricane Season  [link]

Detailed information on the extent and severity of winds and rainfall in an approaching hurricane assists property owners, facility managers and emergency personnel in preparations and planning aimed at minimizing losses from an eventual impact.  

 

CFAN provides customized high-resolution (5 km) ensemble wind forecasts to support damage estimates and emergency management. CFAN’s predictions of landfall impacts of hurricane winds use an adaptive statistical-dynamical technique to statistically downscale the coarse resolution surface winds from the global model ensembles plus the National Hurricane Center models.  The forecasted wind speeds from the global models are calibrated using CFAN’s calibrated intensity forecasts and initial observations of the radius of hurricane winds.

The movie below portrays CFAN’s prediction of high-resolution landfall winds for Hurricane Hermine (2016), 6 days prior to landfall and before it became a named storm.

 

Landfall impacts

Verification analysis of 2016/2017 FL landfall winds.  CFAN's high-resolution landfall wind forecasts out to 72 hours are evaluated against the operational analyses that are downscaled to the same resolution as the forecasts [link]

CFAN provides an interactive tool that allows the user to dynamically zoom into the region of interest to display the forecasts: 

  • Tracks – overlay CFAN’s calibrated ECMWF and GFS tracks, NHC tracks, other global models, to assess specific landfall scenarios

  • Winds -  calibrated and downscaled winds for each ECMWF and GFS ensemble member; a radial wind model based on forecast parameters from the National Hurricane Center

  • Wind gusts – CFAN’s calibrated ECMWF wind gust forecasts

  • Rainfall – 6 hr, 24hr and 72 hr accumulations, 5 mile resolution.  Based on ECMWF forecasts.

 

CFAN’s landfall impact tool is illustrated for forecasts of Hurricane Dorian (2019):

 
Seasonal hurricane forecasts

Seasonal forecasts of hurricane activity benefit financial decision making and basic risk awareness from tropical storms.

 

North Atlantic hurricane activity varies substantially from year to year, including 5-fold variations in annual basin-scale Accumulated Cyclone Energy (ACE), which integrates the duration and intensity of Named Storms.  The frequencies of U.S. landfalling hurricanes also changes sporadically.  Recent decades have witnessed multiple landfalling Category 3+ hurricanes in consecutive years of 2004 and 2005, two of the most active seasons of the past century, but also an absence of major hurricane U.S. landfalls from 2006 to 2016.

 

CFAN’s seasonal forecasts for Atlantic tropical cyclone activity are based on a breakthrough in understanding of the impact of global climate dynamics on Atlantic hurricane activity.  The basis for our forecasts is a proprietary set of indices based on circulation patterns and trends in the ocean, atmosphere and stratosphere.

 

CFAN issues seasonal forecasts of Atlantic hurricane activity in November, March, May and July. CFAN provides prediction for seasonal ACE and the number of hurricanes and U.S. landfalling hurricanes.

 

[Seasonal hurricane forecasts link]

In addition to the North Atlantic, CFAN provides tropical cyclone forecast and extended range outlooks for the following regions:

 

  • West Pacific (Northern hemisphere)

  • East Pacific (Northern hemisphere)

  • North Indian Ocean

 

 

 
Global Tropical Cyclones
 

Success Stories 

Supporting the petroleum industry

Following the catastrophic 2004/2005 Atlantic hurricane seasons, a major petroleum company perceived the need for extended range, better-than-market tropical cyclone forecast product to anticipate disruptions to energy supply, drilling, refining and transport activities. CFAN responded to this challenge by translating its advanced tropical cyclone research into an innovative forecast product for North Atlantic hurricanes, which became operational in 2007. This product provided the first model-based probabilistic forecasts of tropical cyclone formation and pre-genesis tracks. During 2008, with Hurricanes Ike and Gustav having major impacts on the production region in the Gulf of Mexico, CFAN’s forecasts were also used for operations, emergency management and business continuity.

Predicting hurricane landfall impacts in Florida

CFAN’s early and accurate forecast of Hurricane Sandy in 2012 was presented in a series of  webinars organized by Earth Networks, each of which was attended by over 400 individuals from the energy sector.  Following these webinars, an electric power provider in Florida perceived an opportunity to use CFAN’s tropical cyclone forecasts as part of their strategy to mitigate disruption of the electricity supply from landfalling hurricanes. In response to their needs, CFAN developed a customized high-resolution (5 km) probabilistic hurricane wind forecast product.  CFAN’s forecasts are being used to drive outage models to help manage and recovery from the hurricane impacts, particularly with regards to emergency crew placement. CFAN’s forecasts helped the client to rapidly restore power following Hurricanes Hermine and Matthew in 2016 and Hurricane Irma in 2017.

Research & Publications

 

CFAN's scientific team includes some of the world's leading authorities on tropical meteorology and hurricanes. Of direct relevance to our tropical cyclone forecast product, CFAN's scientists are advancing research on:

 

  • Formation and development of African easterly waves, which seed 85% of major hurricanes in the North Atlantic

  • Advancement of ensemble hurricane prediction methods

  • Development of a new model of hurricane-induced tornadoes

  • Climate dynamics of hurricanes, on seasonal to decadal time scales

 

[Link] to publications by CFAN Scientists

 

 

 

Trial subscription

 

CFAN offers trial subscriptions to TropiCast during the months of May, June, October and November to qualified individuals and organizations. 

To inquire about a trial subscription [link]

 
 

These forecasts are also provided as a data stream to interface with models of damages to client-specific assets and operations.

Supporting the insurance industry

CFAN's clients in the insurance-linked securities market represent 25 percent of ILS assets under management globally.  CFAN's seasonal, monthly, and 15 day forecasts are used to assess and adjust portfolio exposure and contract pricing according to tropical cyclone risk. Recent hurricanes Irma (2017) and Dorian (2019) are examples of how sudden and even slight changes in intensity and track direction near densely populated areas can have a financial impact on ILS funds and the insurance sector generally.  Anticipating changes in tropical cyclone behavior on short notice, as is provided by CFAN's proprietary algorithms, reduces client risk.