For many years and still today, statistically post-processed numerical model output has been a primary source of guidance for forecasting surface weather conditions. In particular, Model Output Statistics (MOS) derived from the Nested Grid Model (NGM) and, more recently, the Aviation Model (AVN) provide forecasts of weather parameters at three-hour intervals out to 60 hours. These models are run four times daily, allowing MOS forecasts to become six hours old before an updated product is made available. Therefore, it may be possible to improve conventional guidance for short-term (< 6 h) forecasts of selected parameters if observations taken after the time that the MOS guidance is derived are used to produce updated forecasts.

With this in mind, Vislocky and Fritsch (1997) developed an “observations-based” system for producing short-term forecasts. Specifically, a network of surface observations was used as predictors in a multiple linear regression technique. It was demonstrated that this approach could improve the accuracy of ceiling and visibility forecasts for the hours between the times that the output from the twice-daily operational runs is released.

This technique was applied solely to hourly weather observations. While these observations are commonplace in the weather industry, it is less widely known that additional weather observations are also available 5-minute intervals. However, due to the cost of data transmission, high-frequency observations are usually archived at their respective observing stations (typically a commercial airport) and therefore are not yet available to the general community in real time. Nevertheless, since it is expected that such data will become available in the future, it is of value to explore whether or not this higher frequency data might yield further improvements in the traditional hourly observations-based statistical forecasting systems.

Therefore, a study was undertaken to quantify whether or not increased temporal resolution of surface observing networks yields improvements in one-hour forecasts of ceiling and visibility. The following strategy was employed: 1) a baseline forecasting system was developed for five sites in the New York City area using only the standard hourly observations; 2) an hourly-updating forecasting system was developed using the high-frequency observations from these five sites in addition to the standard hourly observations from other stations in the vicinity, and 3) the results of each system were compared. Hourly forecasts were developed for the top of the hour as well as 15-, 30- and 45-minutes past the hour.

It was found that the use of high-frequency observations to complement the standard hourly observations significantly improved the forecast system’s skill. In particular, it was also found that one-hour forecasts made at 45-minutes past the hour had the greatest improvement when the most recent high-frequency observations were made available to the system. For one-hour forecasts made at the top of the hour, mean square errors were reduced by an additional two to four percent above that produced by the baseline system. This reduction corresponds to an improvement of 15 to 30% compared to the performance of persistence climatology. Meanwhile, for one-hour forecasts made at 45-minutes past the hour, mean square errors were reduced by an additional 12 to 18% above that produced by the baseline system. This reduction corresponds to an improvement of 55 to 110% compared to the performance of persistence climatology.