Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Reference evapotranspiration estimation in hyper-arid regions via D-vine copula based-quantile regression and comparison with empirical approaches and machine learning models

Abdallah, Mohammed ; Mohammadi, Babak LU orcid ; Modathir, Modathir A. ; Omer, Abubaker ; Cheraghalizadeh, Majid ; Eldow, Mohamed E.E. and Duan, Zheng LU (2022) In Journal of Hydrology: Regional Studies 44.
Abstract

Study region: Two hyper-arid regions (Atbara and Kassala stations) in Sudan. Study focus: The study aims to evaluate the potential of the D-vine Copula-based quantile regression (DVQR) model for estimating daily ETo during 2000–2015 based on various input structures. Further, the DVQR model was compared with Multivariate Linear quantile regression (MLQR), Bayesians Model Averaging quantile regression (BMAQR), Empirical Models (EMMs), and Classical Machine Learning (CML). Besides, the CML models including Random Forest (RF), Support Vector Machine (SVM), Extreme Learning Machine (ELM), Extreme Gradient Boosting (XGBoost), and M5 Model Tree (M5Tree) were employed. New hydrological insights for the region: The original EMMs showed poor... (More)

Study region: Two hyper-arid regions (Atbara and Kassala stations) in Sudan. Study focus: The study aims to evaluate the potential of the D-vine Copula-based quantile regression (DVQR) model for estimating daily ETo during 2000–2015 based on various input structures. Further, the DVQR model was compared with Multivariate Linear quantile regression (MLQR), Bayesians Model Averaging quantile regression (BMAQR), Empirical Models (EMMs), and Classical Machine Learning (CML). Besides, the CML models including Random Forest (RF), Support Vector Machine (SVM), Extreme Learning Machine (ELM), Extreme Gradient Boosting (XGBoost), and M5 Model Tree (M5Tree) were employed. New hydrological insights for the region: The original EMMs showed poor performance, which improved after calibration techniques. The DVQR, MLQR, and BMAQR models showed better performance than the calibrated EMMs. However, the DVQR model exhibited the highest accuracy than the MLQR and BMAQR models over two study sites. The M5Tree, SVM, and XGBoost models perfumed better than ELM and RF models at both study sites. The DVQR and XGBoost models showed equivalent performance (R2, NSE, and WIA > 0.99, MAE, and RMSE < 0.2) to the M5Tree and SVM models, but they had significantly more accuracy than the calibrated EMMs, MLQR, BMAQR, ELM, and RF models in two hyper-arid regions. Overall, the high dimensional DVQR model is recommended as a promising alternative technique for estimating daily ETo in hyper-arid climate conditions around the world.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Empirical models, Hyper-arid region, Machine learning, Quantile regression, Reference evapotranspiration, Sudan
in
Journal of Hydrology: Regional Studies
volume
44
article number
101259
publisher
Elsevier
external identifiers
  • scopus:85141253517
ISSN
2214-5818
DOI
10.1016/j.ejrh.2022.101259
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2022 The Authors
id
fa57c6a4-5518-46a6-805e-01e9b52b755e
date added to LUP
2022-11-22 15:02:53
date last changed
2023-12-31 18:49:18
@article{fa57c6a4-5518-46a6-805e-01e9b52b755e,
  abstract     = {{<p>Study region: Two hyper-arid regions (Atbara and Kassala stations) in Sudan. Study focus: The study aims to evaluate the potential of the D-vine Copula-based quantile regression (DVQR) model for estimating daily ETo during 2000–2015 based on various input structures. Further, the DVQR model was compared with Multivariate Linear quantile regression (MLQR), Bayesians Model Averaging quantile regression (BMAQR), Empirical Models (EMMs), and Classical Machine Learning (CML). Besides, the CML models including Random Forest (RF), Support Vector Machine (SVM), Extreme Learning Machine (ELM), Extreme Gradient Boosting (XGBoost), and M5 Model Tree (M5Tree) were employed. New hydrological insights for the region: The original EMMs showed poor performance, which improved after calibration techniques. The DVQR, MLQR, and BMAQR models showed better performance than the calibrated EMMs. However, the DVQR model exhibited the highest accuracy than the MLQR and BMAQR models over two study sites. The M5Tree, SVM, and XGBoost models perfumed better than ELM and RF models at both study sites. The DVQR and XGBoost models showed equivalent performance (R<sup>2</sup>, NSE, and WIA &gt; 0.99, MAE, and RMSE &lt; 0.2) to the M5Tree and SVM models, but they had significantly more accuracy than the calibrated EMMs, MLQR, BMAQR, ELM, and RF models in two hyper-arid regions. Overall, the high dimensional DVQR model is recommended as a promising alternative technique for estimating daily ETo in hyper-arid climate conditions around the world.</p>}},
  author       = {{Abdallah, Mohammed and Mohammadi, Babak and Modathir, Modathir A. and Omer, Abubaker and Cheraghalizadeh, Majid and Eldow, Mohamed E.E. and Duan, Zheng}},
  issn         = {{2214-5818}},
  keywords     = {{Empirical models; Hyper-arid region; Machine learning; Quantile regression; Reference evapotranspiration; Sudan}},
  language     = {{eng}},
  publisher    = {{Elsevier}},
  series       = {{Journal of Hydrology: Regional Studies}},
  title        = {{Reference evapotranspiration estimation in hyper-arid regions via D-vine copula based-quantile regression and comparison with empirical approaches and machine learning models}},
  url          = {{http://dx.doi.org/10.1016/j.ejrh.2022.101259}},
  doi          = {{10.1016/j.ejrh.2022.101259}},
  volume       = {{44}},
  year         = {{2022}},
}