Citas bibligráficas
Vega, R., (2023). Modelo Bayesiano para estimación de curvas intensidad-duración-frecuencia basado en lluvia horaria y diaria en la Región Lima [Universidad Nacional Agraria La Molina]. https://hdl.handle.net/20.500.12996/5904
Vega, R., Modelo Bayesiano para estimación de curvas intensidad-duración-frecuencia basado en lluvia horaria y diaria en la Región Lima []. PE: Universidad Nacional Agraria La Molina; 2023. https://hdl.handle.net/20.500.12996/5904
@mastersthesis{renati/248908,
title = "Modelo Bayesiano para estimación de curvas intensidad-duración-frecuencia basado en lluvia horaria y diaria en la Región Lima",
author = "Vega Romero, Ronald Alex",
publisher = "Universidad Nacional Agraria La Molina",
year = "2023"
}
Applying Bayesian approach techniques to the random variable of “maximum annual rainfall intensity” on a sub-daily time scale, lower degrees of uncertainty are achieved for the extreme events forecast. That is true in comparison to classical statistical techniques, when a short observed hourly data is only available. A better decision is achieved to a adequate hydraulic works and storm drainage design in both urban or rural environment. The objective of the study is to model the intensity of rainfall with sub-daily duration, in the Lima region, under a Bayesian approach. The study área is the jurisdiction of the Water Administrative Authority (AAA) Cañete Fortaleza, located on the central coast of the Pacific watershed. Its relevant because in Lima is located the capital country, which is experimenting an uncontrolled urban expansión towards higher altitude sites where occur likely summer storm that trigger debris flows (huaycos) toward vulnerable populations close to streams. Empirically, the potential of these storms to cause damage and casualties has been demonstrated. Rainfall with durations lower than one day are the most difficult to quantify throughout the country, due to the absence of a historical record data at sub-daily scale. Currently in recent years a new generation of automatic weather stations generates historical records at hourly scale. These data still do not exceed 20 years time length time. To date still, the real margin error related on these methodologies has not been quantified, since they are based on indirect methods based on measurements of conventional climatic stations at daily scale. Thus, the proposed bayesian model overcomes this limitation, presenting lower uncertainty degrees despite having a maximum 14 years length of historical record.
Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons