Evaluation of dynamic regression and artificial neural networks models for real-time hydrological drought forecasting
- PDF / 1,540,696 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 22 Downloads / 222 Views
ORIGINAL PAPER
Evaluation of dynamic regression and artificial neural networks models for real-time hydrological drought forecasting Majid Dehghani 1 & Bahram Saghafian 2 & Firoozeh Rivaz 3 & Ahmad Khodadadi 3
Received: 5 October 2016 / Accepted: 18 April 2017 # Saudi Society for Geosciences 2017
Abstract In this study, application of a class of stochastic dynamic models and a class of artificial intelligence model is reported for the forecasting of real-time hydrological droughts in the Black River basin in the USA. For this purpose, the Standardized Hydrological Drought Index (SHDI) was adopted in different time scales to represent the hydrological drought index. Six probability distribution functions (PDF) were fitted to the discharge time series to obtain the best fit for SHDI calculation. Then, a dynamic linear spatiotemporal model (DLSTM) and artificial neural network (ANN) were used to forecast SHDI. Although results indicated that both models were able to forecast SHDI in different time scales, the DLSTM was far superior in longer lead times. The DLSTM could forecast SHDI up to 6 months ahead while ANN was only capable of forecasting SHDI up to 2 months ahead appropriately. For short lead times (1–6 months), the DLSTM has performed nearly perfect in test phase and CE * Majid Dehghani [email protected] Bahram Saghafian [email protected] Firoozeh Rivaz [email protected] Ahmad Khodadadi [email protected] 1
Technical and Engineering Department, Faculty of Civil and Environmental Engineering, Vali-e-Asr University of Rafsanjan, Rafsanjan, Islamic Republic of Iran
2
Technical and Engineering Department, Science and Research Branch, Islamic Azad University, Tehran, Islamic Republic of Iran
3
Department of Statistics, Shahid Beheshti University, Tehran, Islamic Republic of Iran
oscillates between 0.97 and 0.86 while for ANN modeling, CE is between 0.72 and 0.07. However, the performance of DLSTM and ANN reduced considerably in medium lead times (7–12 months). Overall, the DLSTM is a powerful tool for appropriately forecasting SHDI at short time scales; a major advantage required for drought early warning systems. Keywords Hydrological drought . DLSTM . Ann . Forecast . SHDI . Drought early warning system
Introduction Drought is a natural dynamic phenomenon whose characteristics vary in space and time. Drought is attributed to the temporal variation of hydro-meteorological time series. There is no universal definition of drought and most of the definitions are study-specific. Generally speaking, drought is eventually sensed after a prolonged precipitation shortage resulting in insufficient water to meet the demands in a period of time. In a way, the difference between water availability and water demand determines the drought severity that humans feel. There are several drought classifications. Wilhite and Glantz (1985) classified droughts into meteorological, hydrological, agricultural, and socioeconomic. Mishra and Singh (2010) proposed to add groundwater drought as another type of drought. La
Data Loading...
