Probability Forecasts of Business Impact Variables From CFS2 Ensembles

Period of Performance: 01/01/2015 - 12/31/2015

$94.7K

Phase 1 SBIR

Recipient Firm

Prescient Weather Ltd
200 Innovation Blvd Suite 229
State College, PA 16803
Firm POC
Principal Investigator

Abstract

The value of the subseasonal and seasonal probability forecasts of the National Weather Service (NWS) will be enhanced when the information about standard meteorological variables is converted into information about business impact variables such as degree days, wind and solar power potential, and end-of-season crop yield. Toward that goal, in Phase I we will identify a suite of such action variables for a broad range of industries and activities, develop algorithms to obtain probabilities about a representative sample of such variables from the NWS Climate Forecast System Version 2 (CSF2) forecasts, and develop methods for verifying forecasts about action variables. Computing the forecasts and verification over a historical period will provide an estimate of the skill of the forecasts of the impact variables. The ongoing CFS2 forecasts can then be combined with forecast skill to show decision makers the expected consequences of acting at various predicted probabilities to seize opportunity or mitigate advedse events. This will lead in Phase II to web-based interactive decision advisory systems tailored to industries such as energy, agriculture, transportation, and insurance and finance that will allow their decision makers to assess alternative actions, reduce climate variability risk, and increase profits.