Search Results
You are looking at 1 - 2 of 2 items for
- Author or Editor: Thomas J. Perrone x
- Refine by Access: All Content x
Abstract
We performed a comparative verification of Model Output Statistics (MOS) against Generalized Exponential Markov (GEM), a single station forecasting technique which uses only the surface observation and climatology as input. The verification was performed under three conditions: a “scientific” comparison, where both techniques use the same observation as input; an “operational” comparison, where GEM uses a later observation than does MOS, to simulate the situation where a National Weather Service (NWS) forecaster preparing to make an aviation forecast has a later observation; and a “special operational” comparison, pitting GEM against MOS derived from the previous Limited Area Fine Mesh (LFM) cycle, to simulate the “mid-morning update” operational situation in the NWS where the aviation forecast must be made using previous LFM cycle MOS guidance. Verifications for ceiling, visibility, total cloud amount, temperature, dew-point depression, and wind speed/direction were performed on a full yew of data (a warm season and a cool one) for 21 stations across the United States. GEM demonstrates improvement over MOS for the operational and special operational comparisons, with strongest showing on the major aviation elements—ceiling, visibility, and total cloud amount. For these major aviation elements, a skill crossover between GEM and MOS lies between 5 and 8 hours, and between 3 to 5 hours for the remaining elements. For ceiling and visibility, we also performed experiments blending the GEM and MOS probabilities, and found the resulting categorical, as well as probabilistic, forecasts superior to these produced by GEM or MOS alone.
Abstract
We performed a comparative verification of Model Output Statistics (MOS) against Generalized Exponential Markov (GEM), a single station forecasting technique which uses only the surface observation and climatology as input. The verification was performed under three conditions: a “scientific” comparison, where both techniques use the same observation as input; an “operational” comparison, where GEM uses a later observation than does MOS, to simulate the situation where a National Weather Service (NWS) forecaster preparing to make an aviation forecast has a later observation; and a “special operational” comparison, pitting GEM against MOS derived from the previous Limited Area Fine Mesh (LFM) cycle, to simulate the “mid-morning update” operational situation in the NWS where the aviation forecast must be made using previous LFM cycle MOS guidance. Verifications for ceiling, visibility, total cloud amount, temperature, dew-point depression, and wind speed/direction were performed on a full yew of data (a warm season and a cool one) for 21 stations across the United States. GEM demonstrates improvement over MOS for the operational and special operational comparisons, with strongest showing on the major aviation elements—ceiling, visibility, and total cloud amount. For these major aviation elements, a skill crossover between GEM and MOS lies between 5 and 8 hours, and between 3 to 5 hours for the remaining elements. For ceiling and visibility, we also performed experiments blending the GEM and MOS probabilities, and found the resulting categorical, as well as probabilistic, forecasts superior to these produced by GEM or MOS alone.
Abstract
A statistical forecasting experiment was performed to test the capability of predictors derived from observational data (analysis) fields at 950, 700, 500 and 200 mb to forecast tropical storm formation (genesis). National Oceanographic and Atmospheric Administration tropical mosaic visible satellite images and the Joint (United States Navy and Air Force) Typhoon Warning Center's Post-Season Best Track analyses of tropical storms were used to select a representative collection of tropical cloud clusters, some of which became tropical storms (GO cases), others of which did not (NO GO cases). Navy Fleet Numerical Oceanography Central archived analysis fields of surface pressure, winds, sea surface temperature, and moisture were accessed at locations and times corresponding to cloud cluster positions 24, 48 and 72 hours prior to tropical storm formation/nonformation, and candidate predictors were formed from these analysis data. The number of predictor candidates was increased by also calculating a predictor candidate's Laplacian, and the magnitude of its gradient. A Special local-maximum enhancement technique was also applied to some of the candidate predictors. Stepwise discriminant analysis was applied to these candidate predictors to select subsets with greatest predictive capability for forecasting tropical storm formation at projections of 24,48 and 72 hours. The resulting statistical forecast algorithms were evaluated on independent data, against climatology, and against a basic technique derived solely from latitude and longitude. The results show our forecast technique possesses considerable skill in predicting tropical storm formation with good pre-figurance, post-agreement, threat, and Brier scores.
Abstract
A statistical forecasting experiment was performed to test the capability of predictors derived from observational data (analysis) fields at 950, 700, 500 and 200 mb to forecast tropical storm formation (genesis). National Oceanographic and Atmospheric Administration tropical mosaic visible satellite images and the Joint (United States Navy and Air Force) Typhoon Warning Center's Post-Season Best Track analyses of tropical storms were used to select a representative collection of tropical cloud clusters, some of which became tropical storms (GO cases), others of which did not (NO GO cases). Navy Fleet Numerical Oceanography Central archived analysis fields of surface pressure, winds, sea surface temperature, and moisture were accessed at locations and times corresponding to cloud cluster positions 24, 48 and 72 hours prior to tropical storm formation/nonformation, and candidate predictors were formed from these analysis data. The number of predictor candidates was increased by also calculating a predictor candidate's Laplacian, and the magnitude of its gradient. A Special local-maximum enhancement technique was also applied to some of the candidate predictors. Stepwise discriminant analysis was applied to these candidate predictors to select subsets with greatest predictive capability for forecasting tropical storm formation at projections of 24,48 and 72 hours. The resulting statistical forecast algorithms were evaluated on independent data, against climatology, and against a basic technique derived solely from latitude and longitude. The results show our forecast technique possesses considerable skill in predicting tropical storm formation with good pre-figurance, post-agreement, threat, and Brier scores.
