Examples of Gfesuite and D2d Use In Operations During 2004

Abstract

In the past few years, the Graphical Forecast Editor (called the GFESuite or GFE) has become the primary tool that operational forecasters at National Weather Service (NW S) forecast offices use to create and edit their gridded forecast fields (LeFebvre et al., 2003). As the GFESuite continues to evolve, new tools and capabilities are offered for this purpose. Also evolving is the primary AWIPS two-dimensional data display (D2D) that is used by NWS forecasters and offers additional datasets and higher-resolution data. In this study we are evaluating how forecasters are using these tools to forecast the weather and to generate and maintain their gridded forecast fields. An initial evaluation of GFESuite use in 2003 (Roberts, 2004) provided the first baseline for tracking changes in GFESuite use. Several studies have also been conducted over the years evaluating D2D operational use (Kucera, et al., 1997, Roberts et al., 1993, for example). These studies will be used for com parison purposes with these results. In this study, GFE computer logs collected in 2004 are examined to determine which tools and capabilities are being used by the NWS forecasters. The GFE logs record status information, which tools and capabilities are used, and a time stamp indicating exactly when tools are used or when specific actions are performed. Week-long log snapshots are being collected from different forecast offices and during varied weather conditions. With these snapshots, we can examine the range and frequency of product usage by a number of forecasters who have a variety of forecast responsibilities. Additionally, a sample of D2D logs was collected at one of the sites in order to determine how and when forecasters use each of these capabilities. As w ith the GFE logs, the D2D logs record nearly every D2D action taken on the graphics workstation by the forecasters along with a time stamp and workstation identification. The D2D logs were continuously available from this site, so a 4-month sample was collected and analyzed. A summary of these log analyses results, including any common usage patterns that arose are presented here. Comparisons with previous studies are also presented along with a combined GFESuite/D2D analysis to determine when each function is used during the forecast process