Flood alert system using rainfall data in the Mayagüez Bay drainage basin, Western Puerto Rico

Abstract

Las inundaciones son el tipo de desastres naturales más costoso mundialmente. El presente trabajo es el primer intento para introducir un Sistema de Alerta de Inundaciones en el oeste de Puerto Rico, usando radares con alta resolución temporal y espacial, y desarrollando un modelo de predicción para lluvias convectivas en periodos menores de unas pocas horas. La precisión de estas predicciones generalmente decrece rápidamente durante los primeros 30 min debido a la corta vida de la convectividad en cada uno de los pixeles. Un número de observaciones estudiadas han mostrado que las celdas convectivas tienen un promedio de vida alrededor de 20 min, con el mejor desempeño asociado con un tiempo de espera de 10 minutos. Estudios en simulaciones numéricas han contribuido significantemente al entendimiento de la composición y la duración de las tormentas; este es solo el comienzo de ser reconocido para el uso de sistemas de predicciones en corto tiempo. Una revisión de literatura es proporcionada sobre lo que actualmente se conoce de estudios numéricos y observaciones sobre la organización, tiempo de vida y movimiento de las tormentas.

Floods are one of the most costly types of natural disasters in the world. The current work is an attempt to introduce a Flood Alert System in the western Puerto Rico, using radars with high temporal and spatial resolution and developing a forecasting model for convective precipitation for time periods of a few hours or less (nowcasting). The accuracy of these forecasts generally decreases very rapidly during the first 30 min because of the very short lifetime of individual convective pixels. A number of observational studies have shown that individual convective cells have mean lifetimes of about 20 min, with the best performance associated with a lead-time of 10 min. Numerical simulation studies have contributed significantly to the understanding of storm composition and duration; this is just beginning to be recognized in currents nowcasting systems. A review of the literature is provided related to what is currently known from numerical and observational studies about the organization, lifetime and motion of storms. The nowcasting technique proposed here is a special kind of nonlinear model with stochastic and deterministic components. The rainfall forecasts obtained using the considered method is then routed through a rainfall runoff model Vflo. Thus, implementing a coupled rainfall-runoff forecasting procedure for a watershed in western Puerto Rico. The prediction results with lead-times of 10, 20 and 30 min were analyzed and compared using statistical methods. The forecast result with lead-time of 10 min is the alternative with least percent of error. It was used in the hydrological model Vflo to compare the estimated hydrograph with the observed hydrograph from USGS stations. Furthermore, it was used in the flooding model Inundation Animator to show the extent of flooding superimposed onto a land map.