The Use of Large-Scale Climate Indices in Monthly Reservoir Inflow Forecasting and Its Application on Time Series and Artificial Intelligence Models

Hanbeen Kim,Taereem Kim,Jun-Haeng Heo,Ju-Young Shin,Sunghun Kim

doi:10.3390/w11020374

Abstract

Climate variability is strongly influencing hydrological processes under complex weather conditions, and it should be considered to forecast reservoir inflow for efficient dam operation strategies. Large-scale climate indices can provide potential information about climate variability, as they usually have a direct or indirect correlation with hydrologic variables. This study aims to use large-scale climate indices in monthly reservoir inflow forecasting for considering climate variability. For this purpose, time series and artificial intelligence models, such as Seasonal AutoRegressive Integrated Moving Average (SARIMA), SARIMA with eXogenous variables (SARIMAX), Artificial Neural Network (ANN), Adaptive Neural-based Fuzzy Inference System (ANFIS), and Random Forest (RF) models were employed with two types of input variables, autoregressive variables (AR-) and a combination of autoregressive and exogenous variables (ARX-). Several statistical methods, including ensemble empirical mode decomposition (EEMD), were used to select the lagged climate indices. Finally, monthly reservoir inflow was forecasted by SARIMA, SARIMAX, AR-ANN, ARX-ANN, AR-ANFIS, ARX-ANFIS, AR-RF, and ARX-RF models. As a result, the use of climate indices in artificial intelligence models showed a potential to improve the model performance, and the ARX-ANN and AR-RF models generally showed the best performance among the employed models.

Highlights

Reservoir inflow forecasting is an essential task in dam operation and is strongly linked to water resource planning and management
To determine the exogenous variables for the second type of input variables, the monthly reservoir inflow series was decomposed by ensemble empirical mode decomposition (EEMD) to extract the intrinsic mode functions (IMFs)
The reservoir inflow of the SY dam was decomposed into eight IMFs and a residue, and the reservoir inflow of the CJ and GS dams were decomposed into seven IMFs and a residue

Summary

Introduction

Reservoir inflow forecasting is an essential task in dam operation and is strongly linked to water resource planning and management. Reservoir inflow forecasting has become increasingly complex and important due to changes in the frequency and magnitude of water-related disasters under climate change. To better understand the responses to climate change, a large number of models have been developed for more accurate and reliable inflow forecasting [1,2,3,4,5,6,7,8,9]. ARIMA with eXogenous variables (SARIMAX) models have been widely applied to model hydrological time series considering seasonality [11,12,13,14,15]. Previous studies have successfully proved the applicability of the SARIMA model following the Box and Jenkins procedures, because of the simple mathematical structure, ideal representation of the statistical and correlation structures, and relatively small number

Objectives

Results

Discussion

Conclusion