TY  - JOUR
T1  - Machine Learning Approaches for Prediction of Daily River Flow
A1  - Naser Shiri
A1  - Sepideh Karimi
A1  - Jalal Shiri
JF  - World Journal of Environmental Biosciences
JO  - World J Environ Biosci
SN  - 2277-8047
Y1  - 2023
VL  - 12
IS  - 4
DO  - 10.51847/U72sgqfYRZ
SP  - 33
EP  - 39
N2  - River flow is an important parameter in hydrology, irrigation scheduling, groundwater pollution studies, and hydropower analysis. It depends on various climate and hydrologic factors, e.g. precipitation, temperature, physiography of the river basin, geological characteristics of the basin, etc. Although several factors may affect the quantity and quality of river flow during a certain period, it is difficult to account for all those variables in simulating/predicting river flow values due to the complex relations governing the hydrologic cycle in nature. Therefore, using simpler methods that can be used with fewer required input data would be necessary. A prediction task was implemented in the present study to obtain river flow values based on the previously recorded river flows using three machine learning approaches, namely, multi-variate adaptive regression spline (MARS), boosted regression tree (BT), and random forest (RF). Data from three stations in the state of Iowa (U.S.A.) covering five years of daily records were used for developing the ML models. Based on the results, all three applied models can simulate the river flow values well, when the time lags of two successive days were introduced for model feeding. Also, an analysis was conducted to detect the variations of the applied statistical indicators at the test stage of the k-fold testing data assignment. This analysis showed obvious variations of indicators among the test stages, which revealed the necessity of adopting the k-fold testing in the studied region.
UR  - https://environmentaljournals.org/article/machine-learning-approaches-for-prediction-of-daily-river-flow-853grtfzebbzezo
ER  -