Publication:
Development of drought index for Puerto Rico's hydrologic conditions

Thumbnail Image
Authors
Malavé-Pérez, Luis G.
Embargoed Until
Advisor
Rivera-Santos, Jorge
College
College of Engineering
Department
Department of Civil Engineering
Degree Level
M.S.
Publisher
Date
2020-07-24
Abstract
This study intends to develop a new drought index for Puerto Rico (PRDI) that combines relevant hydrological parameters necessary to obtain an appropriate drought indication value with a focus on hydrological conditions and water amount. The PRDI was developed to measure the drought for Carraizo watershed delimitation. The PRDI demonstrated an acceptable representation during the two biggest drought events (1994 and 2015) with exceptional and extreme drought, respectively. The second objective was to adjust a stochastic forecast model. This new PRDI was analyzed to determine the best stochastics ARIMA model to forecast the PRDI to the future. The ARIMA(2,0,1)(2,1,0) was selected for 3, 9-, and 12- while ARIMA(1,0,1)(3,1,0) was selected for 6 months. The PRDI forecast models can predict the drought values with totals errors between 20.3% to 45%. This index could significantly improve the decision-making process before, during and after considerable drought events occur.

Este estudio desarrolla un nuevo índice de sequía para Puerto Rico (PRDI) que combine parámetros hidrológicos relevantes para obtener un valor de sequía apropiado enfocado en condiciones hidrológicas y cantidad de agua. El PRDI se desarrolló para medir la sequía en la cuenca del embalse Carraizo. Este demostró una representación aceptable durante los dos eventos de sequía históricos (1994 y 2015) con sequía excepcional y extrema, respectivamente. El segundo objetivo era desarrollar un modelo de pronóstico estocástico ARIMA para pronosticar el PRDI a futuro. El ARIMA (2,0,1)(2,1,0) fue seleccionado para pronósticos de 3, 9 y 12 meses, mientras que el ARIMA (1,0,1)(3,1,0) fue seleccionado para pronósticos a 6 meses. Los modelos de pronóstico de PRDI pueden predecir los valores de sequía con errores totales entre 20.3% y 45%. Este índice podría mejorar significativamente el proceso de toma de decisiones antes, durante y después de que ocurran eventos de sequía considerables.
Keywords
Drought index,
Índice de sequía
Usage Rights
Except where otherwise noted, this item’s license is described as CC0 1.0 Universal
Cite
Malavé-Pérez, L. G. (2020). Development of drought index for Puerto Rico’s hydrologic conditions [Thesis]. Retrieved from https://hdl.handle.net/20.500.11801/2702