The Data Deluge: Modern Applications for Bringing Weather to the Decision

This presentation focuses on proven approaches for coupling various disciplines toward a common mission of improving weather-based decisions in the private sector. We will demonstrate methods utilized in the software industry to create user friendly interfaces for analyzing large and unmanageable datasets. We focus on the "agile method" of constructing a team that constantly adapts to customer needs. We will highlight elements of artistic design and consumer usability that are utilized by "web 2.0" companies in order to visualize complex quantitative information.

Decision makers deal with a broad spectrum of weather information ranging from short and medium range weather forecasts to highly specialized fine-scale modeling. Severe weather information on short and long timescales is also a critical input to decision makers across the private sector. Additionally, decision makers are beginning to consider climate change information, regional modeling, downscaling and impacts to the distributions of sensible weather "events" in infrastructure planning multiple decades into the future. The vast amount of data incorporated into the decision making process for an energy user can be overwhelming.

The private weather enterprise and government agencies recognize the complexities in this vast information channel. The National Weather Service recently identified a need to shift its strategic thinking in order to focus on "supporting decisions [through] impact based forecasts, warnings and environmental information" (Berchoff 2010). The vision is to transition warning services to mission specific impact criteria; enable decisions through interpretive collaboration; and improve communication of forecast uncertainty (id). The goal is to improve the decision maker's understanding of risk and deliver data in ways that adapt to user needs (id). The explosion in data sources and the simple volume of data require methods to transform copious amounts of weather data into quick, concise situational awareness. Forecasters must be able to quickly identify what matters, where the uncertainty lies, and where the human can add value. Systems must accelerate insight and delineate patterns, probabilities and impact (id).

In order to address these challenges, the development of decision support software has become a priority for the private weather enterprise and agencies across NOAA. EarthRisk Technologies is a San Diego based software company spun out of the Scripps Institution of Oceanography at UC San Diego. We manage multiple projects that bring together private sector software developers, web designers, private sector meteorologists, energy decision makers and academic researchers. We develop interactive user experiences for analyzing the risk of extreme hot and cold weather around the Northern Hemisphere.