Long-term Streamflow Forecasting and Uncertainty Analysis for Hanjiang River using XGB Model

Huaping Huang, Gaoyang Jin, Kaixia Yin, Ling Yi, Dong Wang, Yujie Li

2021

Abstract

In this study, we proposed a hybrid modelling processor to generate highly performed streamflow forecasts. As a demonstrated case, the extreme gradient boosting (XGB) algorithm was firstly employed to forecast monthly streamflow series of the Huangzhuang hydrological station located in Hanjiang River Basin, China. To further improve the forecast accuracy and quantify the uncertainty, model conditional processor (MCP) approach was then used to postprocess the forecasts produced by the XGB model. The findings reveal that: (1) the XGB algorithm performed well for simulating and forecasting monthly streamflow series, (2) The median forecasts generated by the MCP approach exhibited smaller errors than the deterministic results of XGB model. (3) The 90% confidence interval was reasonable and reliable as most of observations lied within the prediction bounds.

Paper Citation

in Harvard Style

Huang H., Jin G., Yin K., Yi L., Wang D. and Li Y. (2021). Long-term Streamflow Forecasting and Uncertainty Analysis for Hanjiang River using XGB Model. In Proceedings of the 7th International Conference on Water Resource and Environment - Volume 1: WRE, ISBN 978-989-758-560-9, pages 5-11

in Bibtex Style

@conference{wre21,
author={Huaping Huang and Gaoyang Jin and Kaixia Yin and Ling Yi and Dong Wang and Yujie Li},
title={Long-term Streamflow Forecasting and Uncertainty Analysis for Hanjiang River using XGB Model},
booktitle={Proceedings of the 7th International Conference on Water Resource and Environment - Volume 1: WRE,},
year={2021},
pages={5-11},
publisher={SciTePress},
organization={INSTICC},
doi={},
isbn={978-989-758-560-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 7th International Conference on Water Resource and Environment - Volume 1: WRE,
TI - Long-term Streamflow Forecasting and Uncertainty Analysis for Hanjiang River using XGB Model
SN - 978-989-758-560-9
AU - Huang H.
AU - Jin G.
AU - Yin K.
AU - Yi L.
AU - Wang D.
AU - Li Y.
PY - 2021
SP - 5
EP - 11
DO -