Super ensemble learning for daily streamflow forecasting: Large-scale demonstration and comparison with multiple machine learning algorithms

Abstract

Daily streamflow forecasting through data-driven approaches is traditionally performed using a single machine learning algorithm. Existing applications are mostly restricted to examination of few case studies, not allowing accurate assessment of the predictive performance of the algorithms involved. Here we propose super learning (a type of ensemble learning) by combining 10 machine learning algorithms. We apply the proposed algorithm in one-step ahead forecasting mode. For the application, we exploit a big dataset consisting of 10-year long time series of daily streamflow, precipitation and temperature from 511 basins. The super learner improves over the performance of the linear regression algorithm by 20.06%, outperforming the "hard to beat in practice" equal weight combiner. The latter improves over the performance of the linear regression algorithm by 19.21%. The best performing individual machine learning algorithm is neural networks, which improves over the performance of the linear regression algorithm by 16.73%, followed by extremely randomized trees (16.40%), XGBoost (15.92%), loess (15.36%), random forests (12.75%), polyMARS (12.36%), MARS (4.74%), lasso (0.11%) and support vector regression (-0.45%). Based on the obtained large-scale results, we propose super learning for daily streamflow forecasting.

References in corpus (4)

• XGBoost: A Scalable Tree Boosting System
• ranger: A Fast Implementation of Random Forests for High Dimensional Data in C++ and R
• To Explain or to Predict?
• The Evolution of Boosting Algorithms - From Machine Learning to Statistical Modelling

Cited by in corpus (9)

• Merging satellite and gauge-measured precipitation using LightGBM with an emphasis on extreme quantiles
• Comparison of tree-based ensemble algorithms for merging satellite and earth-observed precipitation data at the daily time scale
• Massive feature extraction for explaining and foretelling hydroclimatic time series forecastability at the global scale
• Comparison of machine learning algorithms for merging gridded satellite and earth-observed precipitation data
• Probabilistic water demand forecasting using quantile regression algorithms
• Ensemble learning for blending gridded satellite and gauge-measured precipitation data
• Hydroclimatic time series features at multiple time scales
• A review of machine learning concepts and methods for addressing challenges in probabilistic hydrological post-processing and forecasting
• Time series features for supporting hydrometeorological explorations and predictions in ungauged locations using large datasets