You are here

Probability Forecasts of Business Impact Variables from CFS2 Ensembles

Description:

Summary:  Many climate-sensitive businesses and activities focus on variables that could be derived from the wide range of meteorological variables produced by the NWS Climate Forecast System v2 (CFS2).  Examples include degree-days in energy and agriculture, potential wind and solar power in renewable energy, and wildfire risk in wildland management.

 

To serve such business users, the CFS2 ensemble probabilities for temperature, precipitation, and other variables must be converted into probabilities about business-impact variables on subseasonal timescales.  This requires the development and validation of algorithms specific to each impact variable.  The business impact probabilities should be presented in graphical and digital forms for further analysis and ingest into business decision support systems.

 

Forecasts of some weather-scale business impact variables are commercially available, but these methods cannot be extended readily to the subseasonal ranges because the model forecasts on these scales must be calibrated in order to produce reliable and skillful probabilities.

 

Project Goals:  The goal of this project is to create a commercially viable system to deliver probabilistic forecasts of business impact variables on subseasonal timescales by suitable transformation of the ensemble forecasts of the CFSv2 and other similar forecast systems or multi-model ensembles.  An important component of the goal is to demonstrate the skill and reliability of the forecasts through appropriate validation studies.

 

Phase I Activities and Expected Deliverables:

Activities

·         Identify a suite of impact variables of primary interest to business and industry

·         Select a small Phase I subset of variables that can be obtained easily from CFSv2 and can be used to develop prototype methods and demonstrate proof of concept  

·         Develop algorithms for computing probability forecasts of the Phase I subset from CFS2 ensembles

·         Develop methods for validating the predicted variables of the
Phase I subset

·         Develop a browser-based system for displaying or delivering the probability forecasts of the Phase I variables

Deliverables

·         A formal report describing or demonstrating

o   the suite of business impact variables and the Phase I subset

o   the algorithms for obtaining Phase I impact variables from model variables

o   the skill and reliability of a sample of forecasts of the Phase I subset variables.

·         A functioning, prototype browser-based system for delivering probability forecasts of the Phase I variables on the subseasonal timescales

 

Phase II Activities and Expected Deliverables:

Activities

·         Extend the Phase I methods to a broader suite of impact variables, including some of the most challenging possibilities

·         Continue development of the browser-based forecast display and delivery system to a commercially viable prototype system

Deliverables

·         Documentation of the algorithms used to create the probabilistic forecasts for the final suite of business impact variables

·         Documentation of the skill and reliability of the probabilistic forecasts for the final suite of business impact variables

·         A browser-based system for displaying and delivering probability forecasts of the final suite of business impact variables that is ready and suitable for initial commercial deployment

US Flag An Official Website of the United States Government