Search

Skillful long-range forecasts of acoustic variables have the potential to be very useful in planning Navy undersea warfare operations. Our study assessed the potential to predict sonic layer depth (SLD) in the western north Pacific at lead times of one to several months. We conducted correlations between SLD and remote climate system variables, and identified a high potential for skillful long-range forecasts of SLD in the western north Pacific using sea surface temperature in equatorial and south Pacific as predictors. We used tercile matching and composite analysis forecast (CAF) methods to develop hindcasts and forecasts of SLD based on SST predictors at lead times of one to four months. Our forecast verification metrics show that the resulting long lead probabilistic forecasts are a clear improvement over presently available long term mean climatology products. We also used conditional compositing techniques to create mean and environmental threshold probability products based on the long lead forecasts. Our results indicate that the support of USW operations by the Navy meteorology and oceanography community could be improved by the use of advanced climate data sets, climate analysis, and long-range forecasting methods., The original document contains color images.

In 1987, an unmanned Atlas-Centaur-67 launched from the Cape triggered a lightning discharge that disabled the on-board guidance system and Range Control destroyed the platform. This incident spurred the review and revision of the natural and triggered lightning launch commit criteria (LCC). The LCC are a set of eleven complex rules that are constantly evaluated by the Launch Weather Team (LWT) of 45th Weather Squadron (45WS). Unfortunately, the 45WS LWT does not have either a descriptive or conditional climatology for many of the LCC. This thesis addresses the lightning and the cumulus LCC. A descriptive climatology for both the lightning and the cumulus LCC is presented for the 1989 to 1998 period Additionally, the climate of the Cape is divided into four seasons, and a conditional climatology is introduced for the cumulus LCC. The conditional climatology procedure uses a season-specific discriminant function to classify the radiosonde observations into either the violation or no violation group for the four seasons Because of the limited number of cumulus violation cases, the statistical significance of the four seasonal discriminant functions could not be verified. Therefore, further refinement of the seasonal discriminant functions is needed to make them a more useful forecasting tool.

This thesis uses statistical analysis to forecast the probability of meeting or exceeding the maximum allowable wind speeds for each of the launch pads at the Kennedy Space Center (KSC) and Cape Canaveral Air Station (CCAS). Wind data were collected from the Weather Information Network Display System (WINDS), a collection of 47 meteorological towers located throughout KSC and CCAS, over a period of five winters. A Fortran program was written to calculate conditional probabilities of meeting or exceeding a given threshold speed during eight consecutive one-hour periods, using the current wind direction and peak wind speed as inputs. Forecast probabilities were displayed in a table according to time period and wind direction. Accuracy was measured by constructing contingency tables and calculating various measures of accuracy. Results were tested for significance by calculating p-values for the chi-square test. This method was found to have very little skill in forecasting maximum wind speeds. It is not recommended for operational use.

This report documents efforts to provide an objective technique for forecasting the onset of fog and visibilities below certain specified thresholds at Zaragoza AB, Spain. The study was in response to a problem with dense fog at Zaragoza. The report addresses the problem in two parts; first, with tables that identify the number of hourly observations of fog at Zaragoza, stratified by certain weather variables, and second, with a fog forecasting model based on discriminant analysis that provides an estimated probability of a specified visibility threshold as a function of time

This report is a continuation of USAFETAC/TN-90/001 (AD-A222574). This report documents the relationship between Goettingen's planetary geomagnetic index (Ap) and various other solar events, such as surges, prominences, filaments, radio bursts and X-ray events. The study is based on solar and Ap data with an 11-year period of record (1975 to 1986--Solar Cycle 21). Solar reports were merged with 3-hour Ap values for 7 days after each event. The resultant datasets were analyzed with respect to type, position on the sun, and size. Contingency tables of means and percent frequency distribution summarize the results.