SBIR Phase I: Real-time Objective Model Analysis Tool and Multi-Model Ensemble Forecast System

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to advance societal goals such as protection of life and property, protection against lost revenue and improved planning by providing a technically innovative product to help operational forecasters create better weather forecasts. The explosive growth in the availability of high-performance computing clusters has significantly increased the amount of forecast data available to operational meteorologists. Making sense of all of the available data in a timely fashion has become difficult and the market exists for products that streamline this process. Any improvement in weather forecasting will have direct and indirect positive effects for businesses and society by increasing safety, helping in daily and weekly planning, helping to mitigate storm related expenses and helping businesses avoid weather-related expenses that can then trickle down as cost-savings for the customer. The technology will allow forecasters to produce better forecasts more quickly and will be valuable in operational settings at national weather centers, to individual users, and in a host of commercial applications such as energy, agriculture, and media. This Small Business Innovation Research (SBIR) Phase I project seeks to demonstrate the technical feasibility of operationally synthesizing multiple streams of global, regional, and ensemble weather model forecast data in real-time to provide a streamlined product of model performance and then use that information to create new global or regional forecast maps. The complexity of accessing and analyzing vast amounts of forecast data creates a significant challenge to the operational forecaster. Analysis of all available forecast data is difficult since forecasters are under pressure to process the data in a short amount of time. Furthermore, the analysis is usually subjective. The proposed technology will ameliorate these issues by creating a model performance and forecast product that consolidates a majority of forecast data into one, easy-to-use interface. The goal of the project is to provide operational forecasters with a tool that gives an objective look at model performance with minimal effort and provides them with a new forecast product that is better than any single model.