To develop a progressive multimetric  configuration  optimisation method for WRF simulations of extreme rainfall events over Egypt

Ying Liu; Yiheng Chen; Chien-Nien Chen; Jiao Wang; Lu Zhuo; Miguel A Rico-Ramirez; Dawei Han

doi:10.1016/j.jhydrol.2021.126237

To develop a progressive multimetric configuration optimisation method for WRF simulations of extreme rainfall events over Egypt

Ying Liu^*
, Yiheng Chen
, Chien-Nien Chen
, Jiao Wang
, Lu Zhuo^*
, Miguel A Rico-Ramirez
, Dawei Han

^*Corresponding author for this work

Research output: Contribution to journal › Article (Academic Journal) › peer-review

25 Citations (Scopus)

107 Downloads (Pure)

Abstract

The Weather Research and Forecasting (WRF) model can help to improve our understanding of analysing and forecasting hydrometeorological disasters. Especially for some regions like the Nile Delta, which faces growing climate hazards but has inadequate in situ rainfall observations. However, identifying an optimal configuration to run the WRF model is often a challenge. In this study, the WRF model was used to simulate extreme rainfall events at high spatial and temporal resolutions centered around Alexandria, in northern Egypt. In particular, a progressive multimetric configuration optimisation (PMCO) method is proposed to identify the possible optimal configurations of WRF in the aspect of domain size, numbers of vertical levels, nesting ratio, spin-up times, and physical parameterization schemes (microphysics, planetary boundary layer, and cumulus), based on 48 specifically designed experiments. The simulation performances are quantified and sorted by the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). All WRF simulations use the ERA5 reanalysis dataset as boundary conditions and the WRF results are verified against the Integrated Multi-satellitE Retrievals for GPM (IMERG). The results show that the rainfall distribution and magnitude are most sensitive to the spin-up time and physical parameterization schemes. It is also observed that the improvement of WRF's reproducibility of rainfall intensity is usually accompanied by a decrease in the reproducibility of rainfall distribution. The best model configuration for the study area comprises of three-level nesting (D01 80x80; D02 112x112; D03 88x88), 58 vertical levels, 1:3:3 downscaling ratio, 48h spin-up time, WRF Single-Moment 6-class microphysics scheme, Mellor-Yamada-Janjic planetary boundary layer scheme, and Grell-Freitas cumulus. The stability of this configuration is also verified with the other three extreme rainfall events over Egypt. The results show that there exists a common WRF configuration set in Egypt that produces the relatively good simulations for extreme rainfall events.

Original language	English
Article number	126237
Number of pages	17
Journal	Journal of Hydrology
Volume	598
Early online date	23 Mar 2021
DOIs	https://doi.org/10.1016/j.jhydrol.2021.126237
Publication status	Published - 1 Jul 2021

Bibliographical note

Funding Information:
This study is supported by the Institutional Links grant [332430681] under the Newton-Mosharafa Fund. The grant is funded by the UK Department of Business, Energy and Industrial Strategy (BEIS) and Egypt Science and Technology Development Fund (STDF) and delivered by the British Council. The first author gratefully acknowledges the China Scholarship Council (CSC No, 201908310086) and University of Bristol joint-funded PhD Scholarship. The authors acknowledge the Advanced Computing Research Centre at the University of Bristol for providing access to the High-Performance Computing system BlueCrystal.

Publisher Copyright:
© 2021 Elsevier B.V.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 13 Climate Action

Research Groups and Themes

Water and Environmental Engineering

Access to Document

10.1016/j.jhydrol.2021.126237

Full-text PDF (author accepted manuscript)
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via Elsevier at https://doi.org/10.1016/j.jhydrol.2021.126237. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 3.46 MBLicence: CC BY-NC-ND

Handle.net

Persistent link

HPC (High Performance Computing) and HTC (High Throughput Computing) Facilities
Alam, S. R. (Manager), Williams, D. A. G. (Manager), Eccleston, P. E. (Manager) & Greene, D. (Manager)
Facility/equipment: Facility

Cite this

@article{22d9f7bc264b4534abb845eac401659c,

title = "To develop a progressive multimetric configuration optimisation method for WRF simulations of extreme rainfall events over Egypt",

abstract = "The Weather Research and Forecasting (WRF) model can help to improve our understanding of analysing and forecasting hydrometeorological disasters. Especially for some regions like the Nile Delta, which faces growing climate hazards but has inadequate in situ rainfall observations. However, identifying an optimal configuration to run the WRF model is often a challenge. In this study, the WRF model was used to simulate extreme rainfall events at high spatial and temporal resolutions centered around Alexandria, in northern Egypt. In particular, a progressive multimetric configuration optimisation (PMCO) method is proposed to identify the possible optimal configurations of WRF in the aspect of domain size, numbers of vertical levels, nesting ratio, spin-up times, and physical parameterization schemes (microphysics, planetary boundary layer, and cumulus), based on 48 specifically designed experiments. The simulation performances are quantified and sorted by the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). All WRF simulations use the ERA5 reanalysis dataset as boundary conditions and the WRF results are verified against the Integrated Multi-satellitE Retrievals for GPM (IMERG). The results show that the rainfall distribution and magnitude are most sensitive to the spin-up time and physical parameterization schemes. It is also observed that the improvement of WRF's reproducibility of rainfall intensity is usually accompanied by a decrease in the reproducibility of rainfall distribution. The best model configuration for the study area comprises of three-level nesting (D01 80x80; D02 112x112; D03 88x88), 58 vertical levels, 1:3:3 downscaling ratio, 48h spin-up time, WRF Single-Moment 6-class microphysics scheme, Mellor-Yamada-Janjic planetary boundary layer scheme, and Grell-Freitas cumulus. The stability of this configuration is also verified with the other three extreme rainfall events over Egypt. The results show that there exists a common WRF configuration set in Egypt that produces the relatively good simulations for extreme rainfall events.",

author = "Ying Liu and Yiheng Chen and Chien-Nien Chen and Jiao Wang and Lu Zhuo and Rico-Ramirez, \{Miguel A\} and Dawei Han",

note = "Funding Information: This study is supported by the Institutional Links grant [332430681] under the Newton-Mosharafa Fund. The grant is funded by the UK Department of Business, Energy and Industrial Strategy (BEIS) and Egypt Science and Technology Development Fund (STDF) and delivered by the British Council. The first author gratefully acknowledges the China Scholarship Council (CSC No, 201908310086) and University of Bristol joint-funded PhD Scholarship. The authors acknowledge the Advanced Computing Research Centre at the University of Bristol for providing access to the High-Performance Computing system BlueCrystal. Publisher Copyright: {\textcopyright} 2021 Elsevier B.V.",

year = "2021",

month = jul,

day = "1",

doi = "10.1016/j.jhydrol.2021.126237",

language = "English",

volume = "598",

journal = "Journal of Hydrology",

issn = "0022-1694",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - To develop a progressive multimetric configuration optimisation method for WRF simulations of extreme rainfall events over Egypt

AU - Liu, Ying

AU - Chen, Yiheng

AU - Chen, Chien-Nien

AU - Wang, Jiao

AU - Zhuo, Lu

AU - Rico-Ramirez, Miguel A

AU - Han, Dawei

N1 - Funding Information: This study is supported by the Institutional Links grant [332430681] under the Newton-Mosharafa Fund. The grant is funded by the UK Department of Business, Energy and Industrial Strategy (BEIS) and Egypt Science and Technology Development Fund (STDF) and delivered by the British Council. The first author gratefully acknowledges the China Scholarship Council (CSC No, 201908310086) and University of Bristol joint-funded PhD Scholarship. The authors acknowledge the Advanced Computing Research Centre at the University of Bristol for providing access to the High-Performance Computing system BlueCrystal. Publisher Copyright: © 2021 Elsevier B.V.

PY - 2021/7/1

Y1 - 2021/7/1

N2 - The Weather Research and Forecasting (WRF) model can help to improve our understanding of analysing and forecasting hydrometeorological disasters. Especially for some regions like the Nile Delta, which faces growing climate hazards but has inadequate in situ rainfall observations. However, identifying an optimal configuration to run the WRF model is often a challenge. In this study, the WRF model was used to simulate extreme rainfall events at high spatial and temporal resolutions centered around Alexandria, in northern Egypt. In particular, a progressive multimetric configuration optimisation (PMCO) method is proposed to identify the possible optimal configurations of WRF in the aspect of domain size, numbers of vertical levels, nesting ratio, spin-up times, and physical parameterization schemes (microphysics, planetary boundary layer, and cumulus), based on 48 specifically designed experiments. The simulation performances are quantified and sorted by the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). All WRF simulations use the ERA5 reanalysis dataset as boundary conditions and the WRF results are verified against the Integrated Multi-satellitE Retrievals for GPM (IMERG). The results show that the rainfall distribution and magnitude are most sensitive to the spin-up time and physical parameterization schemes. It is also observed that the improvement of WRF's reproducibility of rainfall intensity is usually accompanied by a decrease in the reproducibility of rainfall distribution. The best model configuration for the study area comprises of three-level nesting (D01 80x80; D02 112x112; D03 88x88), 58 vertical levels, 1:3:3 downscaling ratio, 48h spin-up time, WRF Single-Moment 6-class microphysics scheme, Mellor-Yamada-Janjic planetary boundary layer scheme, and Grell-Freitas cumulus. The stability of this configuration is also verified with the other three extreme rainfall events over Egypt. The results show that there exists a common WRF configuration set in Egypt that produces the relatively good simulations for extreme rainfall events.

AB - The Weather Research and Forecasting (WRF) model can help to improve our understanding of analysing and forecasting hydrometeorological disasters. Especially for some regions like the Nile Delta, which faces growing climate hazards but has inadequate in situ rainfall observations. However, identifying an optimal configuration to run the WRF model is often a challenge. In this study, the WRF model was used to simulate extreme rainfall events at high spatial and temporal resolutions centered around Alexandria, in northern Egypt. In particular, a progressive multimetric configuration optimisation (PMCO) method is proposed to identify the possible optimal configurations of WRF in the aspect of domain size, numbers of vertical levels, nesting ratio, spin-up times, and physical parameterization schemes (microphysics, planetary boundary layer, and cumulus), based on 48 specifically designed experiments. The simulation performances are quantified and sorted by the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). All WRF simulations use the ERA5 reanalysis dataset as boundary conditions and the WRF results are verified against the Integrated Multi-satellitE Retrievals for GPM (IMERG). The results show that the rainfall distribution and magnitude are most sensitive to the spin-up time and physical parameterization schemes. It is also observed that the improvement of WRF's reproducibility of rainfall intensity is usually accompanied by a decrease in the reproducibility of rainfall distribution. The best model configuration for the study area comprises of three-level nesting (D01 80x80; D02 112x112; D03 88x88), 58 vertical levels, 1:3:3 downscaling ratio, 48h spin-up time, WRF Single-Moment 6-class microphysics scheme, Mellor-Yamada-Janjic planetary boundary layer scheme, and Grell-Freitas cumulus. The stability of this configuration is also verified with the other three extreme rainfall events over Egypt. The results show that there exists a common WRF configuration set in Egypt that produces the relatively good simulations for extreme rainfall events.

U2 - 10.1016/j.jhydrol.2021.126237

DO - 10.1016/j.jhydrol.2021.126237

M3 - Article (Academic Journal)

SN - 0022-1694

VL - 598

JO - Journal of Hydrology

JF - Journal of Hydrology

M1 - 126237

ER -