Agencies and individuals responsible for resource management decisions of many types have varying perspectives about hydroclimatic variability and opportunities for using forecasts to improve decision outcomes. Improper interpretation of forecasts is widespread and many individuals find it difficult to place forecasts in an appropriate regional historical context. In addition, the lack of readily accessible quantitative assessments of past forecast performance limits confidence in available forecast products. Further, researchers find it difficult to engage decision makers in detailed discussions about using forecasts to improve risk management when the discussions refer only to abstract notions of natural variability, prediction uncertainty, and forecast quality.

In response, we have developed an interactive forecast assessment tool accessible over the Internet that can be customized by individual users. The tool provides tutorials for guiding forecast interpretation, including quizzes that allow users to test their forecast interpretation skills. Users can efficiently monitor the time evolution of forecasts and subsequent observations. They can also monitor recent and historical observations for selected regions, communicated using terminology consistent with available forecast products. The tool also allows users to evaluate forecast performance for the regions, seasons, forecast lead times, and performance criteria relevant to their specific decision making situations. Using consistent product format, the evaluation component allows individuals to use results at the level they are capable of understanding, while offering opportunity to shift to more sophisticated criteria.

Input from potential users, from a variety of resource management sectors, was instrumental in determining the webtool design. Initial ideas and feedback on the webtool were solicited through workshops, interviews, online surveys, and interaction with a cadre of voluntary webtool beta-testers. In its present form, the webtool can be considered a research ‘product’. However, it is also the basis for an ongoing research ‘process’, whereby new forthcoming forecast products may be more readily understood and used by decision makers, and researchers can engage decision makers in concrete discussions about risk management processes and the value of forecasts.

Our experience in creating the online forecast assessment tool highlights several issues associated with development of nontraditional research products. Design and implementation of commercial quality websites is fundamentally different than producing traditional research products. Websites that incorporate interactive analyses and frequent data updates require highly specialized technical personnel and timeframes that can be difficult to accommodate in many federally funded projects. The long-term sustainability of webtools developed in a university research setting is an open question. Webtool components may not be easily transferable to agencies, and website maintenance requires ongoing commitment. Regardless, from the perspective of decision makers, such products may be more useful than traditional products.