Modelo estadístico de pronóstico de convección para la zona norte de la Provincia de Mendoza

• Autores: Simonelli, Silvia Carmen
• Año de publicación: 2000
• Idioma: español castellano
• Tipo de recurso: tesis doctoral
• Colaborador/a o director/a de tesis: Nicolini, Matilde
• Materia: CONVECCION; PRECIPITACION CONVECTIVA; ANALISIS DISCRIMINANTE; PRONOSTICO OBJETIVO; CONVECTION; CONVECTIVE PRECIPITATION; DISCRIMINANT ANALYSIS; OBJECTIVE FORECAST
• Institución: Universidad Nacional de Buenos Aires. Facultad de Ciencias Exactas y Naturales

El modelo de pronóstico de ocurrencia de convección y de precipitación convectiva obtenido para la zona norte de la provincia de Mendoza desarrollado en este Trabajo de Tesis representa un avance respecto a la metodología actual de pronóstico de convección al proveer una nueva herramienta de fácil uso computacional que sintetiza la información meteorológica disponible y aplica una metodología estadística de evaluación de los pronósticos de convección en la región.
Mesoscale phenomena are relevant in the province of Mendoza, Argentina. Among them, severe weather related to convection represents a major economical and social hazard and has been addressed in the past by different hail suppression programs. This region has singular characteristics in regards to initiation, development and evolution of convection owing to its location leeward of the abrupt Andes range and of the Precordillera. These topographical features modify the airflow in a complex way and affect the westerlies in a deep tropospheric layer. The period of storm occurrence extends from October to March, although during some years the first storm events happen in September and the last ones in April. These events are related to frontal systems, air mass instability or cooling at middle levels. The use of a statistical technique, like the stepwise discriminant analysis, has permitted to separate convective and non-convective events from the sample, and to finally obtain a probability of convection occurrence from surface and upper air soundings data sets. The importance of thermodinamical conditions are stressed by the selection of instability indexes that include low-level relative humidity in their definition as the best predictors both for convection ocurrence in the time scale and for convective precipitation in the spatial scale. Model validation shows a positive association between predicted and observed occurrence with high probability of detection and low false alarm ratio during forecast periods closer to the observational time. The forecast model developed in this Tesis for the northern zone of Mendoza province represents an improvement respect to current operative routine and provides a statistical methodology to evaluate the forecast in the region.
Fil: Simonelli, Silvia Carmen. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina.