HR: 1340h
AN: PA33A-1021    [Abstracts]
TI: Application of Conjunctive Nonlinear Model Based on Wavelet Transforms and Artificial Neural Networks to Drought Forecasting
AU: * Abrishamchi, A
EM: abrisham@sharif.edu
AF: Dept. of Civil Engineering, Sharif Univ. of Technology, Azadi Ave., Tehran, Teh 1458889, Iran (Islamic Republic of)
AU: Mehdikhani, H
AF: Dept. of Civil Engineering, Sharif Univ. of Technology, Azadi Ave., Tehran, Teh 1458889, Iran (Islamic Republic of)
AU: Tajrishy, M
EM: tajrishy@sharif.edu
AF: Dept. of Civil Engineering, Sharif Univ. of Technology, Azadi Ave., Tehran, Teh 1458889, Iran (Islamic Republic of)
AU: Marino, M A
EM: mamarino@ucdavis.edu
AF: LAWR Dept. and Dept. of Civil and Environmental Engineering, UC Davis, Davis, Ca 95616, United States
AU: Abrishamchi, A
EM: abrisham@ucdavis.edu
AF: Dept. of Civil and Environmental Engineering, UC Davis, Davis, Ca 95616, United States
AB: Drought forecasting plays an important role in mitigation of economic, environmental and social impacts of drought. Traditional statistical time series methods have a limited ability to capture non-stationarities and nonlinearities in data. Artificial Neural Network (ANN) because of highly flexible function estimator that has self- learning and self-adaptive feature has shown great ability in forecasting nonlinear and nonstationary time series in hydrology. Recently wavelet transforms have become a common tool for analyzing local variation in time series. Wavelet transforms provide a useful decomposition of a signal, or time series; therefore, hybrid models have been proposed for forecasting a time series based on a wavelet transform preprocessing. Wavelet-transformed data aids in improving the ability of forecasting models by diagnosing signal's main frequency component and abstract local information of the original time series on various resolution levels. This paper presents a conjunctive nonlinear model using Wavelet Transforms and Artificial Neural Network. Application of the model in Zayandeh-Rood River basin (Iran) shows that the conjunctive model significantly improves the ability of artificial neural networks for 1, 3, 6 and 9 months ahead forecasting of EDI (effective drought indices) time series. Improved forecasts allow water resources decision makers to develop drought preparedness plans far in advance.
DE: 1812 Drought
DE: 6620 Science policy (0485)
SC: Public Affairs [PA]
MN: 2007 Fall Meeting