Khosravi Khabat, Attar Nasrin, Bateni Sayed M, Jun Changhyun, Kim Dongkyun, Safari Mir Jafar Sadegh, Heddam Salim, Farooque Aitazaz, Abolfathi Soroush
Canadian Centre for Climate Change and Adaptation, University of Prince Edward Island, Charlottetown, Canada.
Department of Statistical Sciences, University of Padova, Padova, Italy.
Heliyon. 2024 Sep 30;10(20):e37965. doi: 10.1016/j.heliyon.2024.e37965. eCollection 2024 Oct 30.
Accurate prediction of daily river flow ( ) remains a challenging yet essential task in hydrological modeling, particularly crucial for flood mitigation and water resource management. This study introduces an advanced M5 Prime (M5P) predictive model designed to estimate as well as one- and two-day-ahead river flow forecasts (i.e. and ). The predictive performance of M5P ensembles incorporating Bootstrap Aggregation (BA), Disjoint Aggregating (DA), Additive Regression (AR), Vote (V), Iterative classifier optimizer (ICO), Random Subspace (RS), and Rotation Forest (ROF) were comprehensively evaluated. The proposed models were applied to a case study data in Tuolumne County, US, using a dataset comprising measured precipitation ( ), evaporation ( ), and . A wide range of input scenarios were explored for predicting , and . Results indicate that and significantly influence prediction accuracy. Notably, relying solely on the most correlated variable (e.g., ) does not guarantee robust prediction of . However, extending the forecast horizon mitigates the influence of low-correlation input variables on model accuracy. Performance metrics indicate that the DA-M5P model achieves superior results, with Nash-Sutcliff Efficiency of 0.916 and root mean square error of 23 m/s, followed by ROF-M5P, BA-M5P, AR-M5P, AR-M5P, RS-M5P, V-M5P, ICO-M5P, and the standalone M5P model. The ensemble M5P modeling framework enhanced the predictive capability of the stand-alone M5P algorithm by 1.2 %-22.6 %, underscoring its efficacy and potential for advancing hydrological forecasting.
准确预测每日河流量（ ）在水文建模中仍然是一项具有挑战性但至关重要的任务，对于洪水缓解和水资源管理尤为关键。本研究引入了一种先进的M5 Prime（M5P）预测模型，旨在估计 以及提前一天和两天的河流量预测（即 和 ）。对结合了自助聚合（BA）、不相交聚合（DA）、加法回归（AR）、投票（V）、迭代分类器优化器（ICO）、随机子空间（RS）和旋转森林（ROF）的M5P集成模型的预测性能进行了全面评估。所提出的模型应用于美国图奥勒米县的一个案例研究数据，使用了一个包含实测降水量（ ）、蒸发量（ ）和 的数据集。为预测 、 和 探索了广泛的输入场景。结果表明， 和 对预测准确性有显著影响。值得注意的是，仅依靠相关性最高的变量（例如， ）并不能保证对 的稳健预测。然而，延长预测期可以减轻低相关性输入变量对模型准确性的影响。性能指标表明，DA - M5P模型取得了优异的结果，纳什 - 萨特克利夫效率为0.916，均方根误差为23米/秒，其次是ROF - M5P、BA - M5P、AR - M5P、AR - M5P、RS - M5P、V - M5P、ICO - M5P和独立的M5P模型。集成M5P建模框架将独立M5P算法的预测能力提高了1.2% - 22.6%，突出了其在推进水文预报方面的有效性和潜力。
