Short-Term River Flow Forecasting Framework and Its Application in Cold Climatic Regions

Catchments located in cold weather regions are highly influenced by the natural seasonality that dictates all hydrological processes. This represents a challenge in the development of river flow forecasting models, which often require complex software that use multiple explanatory variables and a large amount of data to forecast such seasonality. The Athabasca River Basin (ARB) in Alberta, Canada, receives no or very little rainfall and snowmelt during the winter and an abundant rainfall–runoff and snowmelt during the spring/summer. Using the ARB as a case study, this paper proposes a novel simplistic method for short-term (i.e., 6 days) river flow forecasting in cold regions and compares existing hydrological modelling techniques to demonstrate that it is possible to achieve a good level of accuracy using simple modelling. In particular, the performance of a regression model (RM), base difference model (BDM), and the newly developed flow difference model (FDM) were evaluated and compared. The results showed that the FDM could accurately forecast river flow (ENS = 0.95) using limited data inputs and calibration parameters. Moreover, the newly proposed FDM had similar performance to artificial intelligence (AI) techniques, demonstrating the capability of simplistic methods to forecast river flow while bypassing the fundamental processes that govern the natural annual river cycle.

Training and Testing Data Division Influence on Hybrid Machine Learning Model Process: Application of River Flow Forecasting

The hydrological process has a dynamic nature characterised by randomness and complex phenomena. The application of machine learning (ML) models in forecasting river flow has grown rapidly. This is owing to their capacity to simulate the complex phenomena associated with hydrological and environmental processes. Four different ML models were developed for river flow forecasting located in semiarid region, Iraq. The effectiveness of data division influence on the ML models process was investigated. Three data division modeling scenarios were inspected including 70%–30%, 80%–20, and 90%–10%. Several statistical indicators are computed to verify the performance of the models. The results revealed the potential of the hybridized support vector regression model with a genetic algorithm (SVR-GA) over the other ML forecasting models for monthly river flow forecasting using 90%–10% data division. In addition, it was found to improve the accuracy in forecasting high flow events. The unique architecture of developed SVR-GA due to the ability of the GA optimizer to tune the internal parameters of the SVR model provides a robust learning process. This has made it more efficient in forecasting stochastic river flow behaviour compared to the other developed hybrid models.