Machine learning based parameter sensitivity of regional climate models—a case study of the WRF model for heat extremes over Southeast Australia
Heatwaves and bushfires cause substantial impacts on society and ecosystems across the globe. Accurate information of heat extremes is needed to support the development of actionable mitigation and adaptation strategies. Regional climate models are commonly used to better understand the dynamics of...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2023-01-01
|
| Series: | Environmental Research Letters |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/1748-9326/ad0eb0 |
| _version_ | 1827629995354226688 |
|---|---|
| author | P Jyoteeshkumar Reddy Sandeep Chinta Richard Matear John Taylor Harish Baki Marcus Thatcher Jatin Kala Jason Sharples |
| author_facet | P Jyoteeshkumar Reddy Sandeep Chinta Richard Matear John Taylor Harish Baki Marcus Thatcher Jatin Kala Jason Sharples |
| author_sort | P Jyoteeshkumar Reddy |
| collection | DOAJ |
| description | Heatwaves and bushfires cause substantial impacts on society and ecosystems across the globe. Accurate information of heat extremes is needed to support the development of actionable mitigation and adaptation strategies. Regional climate models are commonly used to better understand the dynamics of these events. These models have very large input parameter sets, and the parameters within the physics schemes substantially influence the model’s performance. However, parameter sensitivity analysis (SA) of regional models for heat extremes is largely unexplored. Here, we focus on the southeast Australian region, one of the global hotspots of heat extremes. In southeast Australia Weather Research and Forecasting (WRF) model is the widely used regional model to simulate extreme weather events across the region. Hence in this study, we focus on the sensitivity of WRF model parameters to surface meteorological variables such as temperature, relative humidity, and wind speed during two extreme heat events over southeast Australia. Due to the presence of multiple parameters and their complex relationship with output variables, a machine learning (ML) surrogate-based global SA method is considered for the SA. The ML surrogate-based Sobol SA is used to identify the sensitivity of 24 adjustable parameters in seven different physics schemes of the WRF model. Results show that out of these 24, only three parameters, namely the scattering tuning parameter, multiplier of saturated soil water content, and profile shape exponent in the momentum diffusivity coefficient, are important for the considered meteorological variables. These SA results are consistent for the two different extreme heat events. Further, we investigated the physical significance of sensitive parameters. This study’s results will help in further optimising WRF parameters to improve model simulation. |
| first_indexed | 2024-03-09T14:00:08Z |
| format | Article |
| id | doaj.art-9d22c5a108764a1cb6d560e9b7986802 |
| institution | Directory Open Access Journal |
| issn | 1748-9326 |
| language | English |
| last_indexed | 2024-03-09T14:00:08Z |
| publishDate | 2023-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Environmental Research Letters |
| spelling | doaj.art-9d22c5a108764a1cb6d560e9b79868022023-11-30T09:15:59ZengIOP PublishingEnvironmental Research Letters1748-93262023-01-0119101401010.1088/1748-9326/ad0eb0Machine learning based parameter sensitivity of regional climate models—a case study of the WRF model for heat extremes over Southeast AustraliaP Jyoteeshkumar Reddy0https://orcid.org/0000-0001-9490-2483Sandeep Chinta1https://orcid.org/0000-0003-0199-1101Richard Matear2https://orcid.org/0000-0002-3225-0800John Taylor3https://orcid.org/0000-0001-9003-4076Harish Baki4https://orcid.org/0000-0003-1956-8280Marcus Thatcher5https://orcid.org/0000-0003-4139-5515Jatin Kala6https://orcid.org/0000-0001-9338-2965Jason Sharples7https://orcid.org/0000-0002-7816-6989Commonwealth Scientific and Industrial Research Organisation Environment , Hobart, TAS, AustraliaCenter for Global Change Science, Massachusetts Institute of Technology , Cambridge, MA, United States of AmericaCommonwealth Scientific and Industrial Research Organisation Environment , Hobart, TAS, AustraliaCommonwealth Scientific and Industrial Research Organisation Data61 , Canberra, ACT, AustraliaGeosciences and Remote Sensing, Civil Engineering and Geosciences, TU Delft , Delft, The NetherlandsCommonwealth Scientific and Industrial Research Organisation Environment , Melbourne, VIC, AustraliaEnvironmental and Conservation Sciences, Centre for Climate-Impacted Terrestrial Ecosystems, Harry Butler Institute, Murdoch University , Murdoch, WA, Australia; Australian Research Council Centre of Excellence for Climate Extremes, University of New South Wales , Canberra, ACT, AustraliaAustralian Research Council Centre of Excellence for Climate Extremes, University of New South Wales , Canberra, ACT, Australia; School of Science, University of New South Wales Canberra , Canberra, ACT, Australia; NSW Bushfire and Natural Hazards Research Centre , Sydney, NSW, AustraliaHeatwaves and bushfires cause substantial impacts on society and ecosystems across the globe. Accurate information of heat extremes is needed to support the development of actionable mitigation and adaptation strategies. Regional climate models are commonly used to better understand the dynamics of these events. These models have very large input parameter sets, and the parameters within the physics schemes substantially influence the model’s performance. However, parameter sensitivity analysis (SA) of regional models for heat extremes is largely unexplored. Here, we focus on the southeast Australian region, one of the global hotspots of heat extremes. In southeast Australia Weather Research and Forecasting (WRF) model is the widely used regional model to simulate extreme weather events across the region. Hence in this study, we focus on the sensitivity of WRF model parameters to surface meteorological variables such as temperature, relative humidity, and wind speed during two extreme heat events over southeast Australia. Due to the presence of multiple parameters and their complex relationship with output variables, a machine learning (ML) surrogate-based global SA method is considered for the SA. The ML surrogate-based Sobol SA is used to identify the sensitivity of 24 adjustable parameters in seven different physics schemes of the WRF model. Results show that out of these 24, only three parameters, namely the scattering tuning parameter, multiplier of saturated soil water content, and profile shape exponent in the momentum diffusivity coefficient, are important for the considered meteorological variables. These SA results are consistent for the two different extreme heat events. Further, we investigated the physical significance of sensitive parameters. This study’s results will help in further optimising WRF parameters to improve model simulation.https://doi.org/10.1088/1748-9326/ad0eb0machine learningsensitivity analysiswrfheat extremesSoutheast Australia |
| spellingShingle | P Jyoteeshkumar Reddy Sandeep Chinta Richard Matear John Taylor Harish Baki Marcus Thatcher Jatin Kala Jason Sharples Machine learning based parameter sensitivity of regional climate models—a case study of the WRF model for heat extremes over Southeast Australia Environmental Research Letters machine learning sensitivity analysis wrf heat extremes Southeast Australia |
| title | Machine learning based parameter sensitivity of regional climate models—a case study of the WRF model for heat extremes over Southeast Australia |
| title_full | Machine learning based parameter sensitivity of regional climate models—a case study of the WRF model for heat extremes over Southeast Australia |
| title_fullStr | Machine learning based parameter sensitivity of regional climate models—a case study of the WRF model for heat extremes over Southeast Australia |
| title_full_unstemmed | Machine learning based parameter sensitivity of regional climate models—a case study of the WRF model for heat extremes over Southeast Australia |
| title_short | Machine learning based parameter sensitivity of regional climate models—a case study of the WRF model for heat extremes over Southeast Australia |
| title_sort | machine learning based parameter sensitivity of regional climate models a case study of the wrf model for heat extremes over southeast australia |
| topic | machine learning sensitivity analysis wrf heat extremes Southeast Australia |
| url | https://doi.org/10.1088/1748-9326/ad0eb0 |
| work_keys_str_mv | AT pjyoteeshkumarreddy machinelearningbasedparametersensitivityofregionalclimatemodelsacasestudyofthewrfmodelforheatextremesoversoutheastaustralia AT sandeepchinta machinelearningbasedparametersensitivityofregionalclimatemodelsacasestudyofthewrfmodelforheatextremesoversoutheastaustralia AT richardmatear machinelearningbasedparametersensitivityofregionalclimatemodelsacasestudyofthewrfmodelforheatextremesoversoutheastaustralia AT johntaylor machinelearningbasedparametersensitivityofregionalclimatemodelsacasestudyofthewrfmodelforheatextremesoversoutheastaustralia AT harishbaki machinelearningbasedparametersensitivityofregionalclimatemodelsacasestudyofthewrfmodelforheatextremesoversoutheastaustralia AT marcusthatcher machinelearningbasedparametersensitivityofregionalclimatemodelsacasestudyofthewrfmodelforheatextremesoversoutheastaustralia AT jatinkala machinelearningbasedparametersensitivityofregionalclimatemodelsacasestudyofthewrfmodelforheatextremesoversoutheastaustralia AT jasonsharples machinelearningbasedparametersensitivityofregionalclimatemodelsacasestudyofthewrfmodelforheatextremesoversoutheastaustralia |