This thesis looked at using new relative humidity (RH) climatologies to improve the Air Force Global Weather Center's (AFGWC) contrail forecasts. To study the effect of the new RH climatolgies, the currently used empirical relative humidity (RH) profile is replaced with a more accurate climatological one, the Stratospheric And Gaseous Experiment II (SAGE II). The study begins by examining accuracy and bias of forecast contrail bases generated by the empirical and SAGE II RH profiles on 42 days. Both sets of forecast bases are shown to be statistically similar with a series of hypothesis tests. Additional RH profiles, from 0% to 100%, are then tested to gage their affect on forecast base. Again, little statistical difference in forecast bases are noted between the additional pro files. In general, a high forecast base bias is shown for contrail algorithms derived from the Appleman theory. This thesis also reveals the dependence of forecast bases on RH and lapse rate. Lapse rates ranging from 20C/km to 90C/km and forecast bases generated by RH values of 0% and 100% are used to show how RH variations of more than 30% may only vary forecasts by less than 1,000 feet. The thesis demonstrated the AFGWC can not improve its contrail forecasts by using a more accurate climatological RH profile, the AFGWC contrail forecast algorithm has an inherent high forecast base bias, and the degree to which forecast bases are affected by RH greatly depends on the atmospheric lapse rate.

A contrail tracking system was developed to help in the assessment of the effect of commercial jet contrails on the Earth's radiative budget. The tracking system was built by combining meteorological data from the Rapid Update Cycle (RUC) numerical weather prediction model with commercial air traffic flight track data and satellite imagery. A statistical contrail-forecasting model was created a combination of surface-based contrail observations and numerical weather analyses and forecasts. This model allows predictions of widespread contrail occurrences for contrail research on either a real-time basis or for long-term time scales. Satellite-derived cirrus cloud properties in polluted and unpolluted regions were compared to determine the impact of air traffic on cirrus. Duda, David P. and Russell, James, III Langley Research Center

Accurate forecasts of contrail occurrence are essential to military aircrews. Although classical forecast methods have been reasonably successful predicting contrails, there is need for improvement at low ambient relative humidity. This thesis examines the performance of the Hanson method, which was developed to provide better contrail forecasts under drier atmospheric conditions. As a secondary objective, the forecast methods of Schumann and Hanson are compared to the algorithm currently in use by the Air Force Global Weather Central. Data used to validate the algorithms were collected at Wright- Patterson AFB, OH and Edwards AFB, CA. Theoretical contrail forecasts were made for each observation, using the flight level pressure, ambient temperature, and relative humidity. Comparisons were then made between the forecast and actual observation of contrail conditions. Forecast and occurrence data were then statistically analyzed to gauge each method's performance. All methods detected roughly 75 percent of observed contrails under moist atmospheric conditions. However, the Hanson method's performance decreased when drier atmospheric observations were tested. Schumann's method performed as well as the AFGWC algorithm under all atmospheric conditions. Based on this research the Hanson method is not recommended for operational use.

Forecasting for some basic models; The forecasting process.