A Large Ensemble Approach to Quantifying Internal Model Variability Within the WRF Numerical Model

The Weather Research and Forecasting (WRF) community model is widely used to explore cross‐scale atmospheric features. Although WRF uncertainty studies exist, these usually involve ensembles where different physics options are selected (e.g., the boundary layer scheme) or adjusting individual parame...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of geophysical research. Atmospheres 2020-04, Vol.125 (7), p.n/a
Hauptverfasser: Bassett, R., Young, P. J., Blair, G. S., Samreen, F., Simm, W.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page n/a
container_issue 7
container_start_page
container_title Journal of geophysical research. Atmospheres
container_volume 125
creator Bassett, R.
Young, P. J.
Blair, G. S.
Samreen, F.
Simm, W.
description The Weather Research and Forecasting (WRF) community model is widely used to explore cross‐scale atmospheric features. Although WRF uncertainty studies exist, these usually involve ensembles where different physics options are selected (e.g., the boundary layer scheme) or adjusting individual parameters. Uncertainty from perturbing initial conditions, which generates internal model variability (IMV), has rarely been considered. Moreover, many off‐line WRF research studies generate conclusions based on a single model run without addressing any form of uncertainty. To demonstrate the importance of IMV, or noise, we present a 4‐month case study of summer 2018 over London, UK, using a 244‐member initial condition ensemble. Simply by changing the model start time, a median 2‐m temperature range or IMV of 1.2 °C was found (occasionally exceeding 8 °C). During our analysis, episodes of high and low IMV were found for all variables explored, explained by a relationship with the boundary condition data. Periods of slower wind speed input contained increased IMV, and vice versa, which we hypothesis is related to how strongly the boundary conditions influence the nested region. We also show the importance of IMV effects for the uncertainty of derived variables like the urban heat island, whose median variation in magnitude is 1 °C. Finally, a realistic ensemble size to capture the majority of WRF IMV is also estimated, essential considering the high computational overheads (244 members equaled 140,000 CPU hours). We envisage that highlighting considerable IMV in this repeatable manner will help advance best practices for the WRF and wider regional climate modeling community. Plain Language Summary The Weather Research and Forecasting model is typically used to dynamically downscale atmospheric conditions for limited geographical regions (e.g., southern United Kingdom), using one or more nested grids of increasing resolution, and with externally provided boundary conditions (e.g., from reanalysis data). Research applications of the model (i.e., separate to forecasting) are wide ranging from hurricanes to urban heat islands. However, despite the model being constrained at its boundaries, subtle modifications to initial conditions (e.g., soil moisture) may propagate and lead to different results. We generate an ensemble of 244 different initial conditions simply by changing the model start time and show large differences in modeled temperature and other meteorological f
doi_str_mv 10.1029/2019JD031286
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2388902481</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2388902481</sourcerecordid><originalsourceid>FETCH-LOGICAL-c3454-e898c8f6623744d24833d3279e519696dca179d7695f4a5f37ecc2c4b3464b4f3</originalsourceid><addsrcrecordid>eNp9kF1LwzAYhYMoOHR3_oCAt1bz1TS5HPtyYyoOdd6VNE23jK6dSYr039sxFa98b8578XB4OABcYXSLEZF3BGE5HyGKieAnoEcwl5GQkp_-_sn7Oeh7v0XdCURZzHogG8CFcmsDx5U3u6w0cLDfu1rpDQw1fG5UFWzR2moNZ1UwrlIlfKhzU8I35azKbGlDC1c2bGwFw8bA1XICH5udcVb_oJfgrFClN_3vvACvk_HL8D5aPE1nw8Ei0geVyAgptCg4JzRhLCdMUJpTkkgTY8klz7XCicwTLuOCqbigidGaaJZRxlnGCnoBro-9nf9HY3xIt3VzMPYpoUJI1FXijro5UtrV3jtTpHtnd8q1KUbpYcj075AdTo_4py1N-y-bzqfLUcwxYfQLQ0Zyjw</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2388902481</pqid></control><display><type>article</type><title>A Large Ensemble Approach to Quantifying Internal Model Variability Within the WRF Numerical Model</title><source>Wiley Journals</source><source>Wiley Online Library (Open Access Collection)</source><source>Alma/SFX Local Collection</source><creator>Bassett, R. ; Young, P. J. ; Blair, G. S. ; Samreen, F. ; Simm, W.</creator><creatorcontrib>Bassett, R. ; Young, P. J. ; Blair, G. S. ; Samreen, F. ; Simm, W.</creatorcontrib><description>The Weather Research and Forecasting (WRF) community model is widely used to explore cross‐scale atmospheric features. Although WRF uncertainty studies exist, these usually involve ensembles where different physics options are selected (e.g., the boundary layer scheme) or adjusting individual parameters. Uncertainty from perturbing initial conditions, which generates internal model variability (IMV), has rarely been considered. Moreover, many off‐line WRF research studies generate conclusions based on a single model run without addressing any form of uncertainty. To demonstrate the importance of IMV, or noise, we present a 4‐month case study of summer 2018 over London, UK, using a 244‐member initial condition ensemble. Simply by changing the model start time, a median 2‐m temperature range or IMV of 1.2 °C was found (occasionally exceeding 8 °C). During our analysis, episodes of high and low IMV were found for all variables explored, explained by a relationship with the boundary condition data. Periods of slower wind speed input contained increased IMV, and vice versa, which we hypothesis is related to how strongly the boundary conditions influence the nested region. We also show the importance of IMV effects for the uncertainty of derived variables like the urban heat island, whose median variation in magnitude is 1 °C. Finally, a realistic ensemble size to capture the majority of WRF IMV is also estimated, essential considering the high computational overheads (244 members equaled 140,000 CPU hours). We envisage that highlighting considerable IMV in this repeatable manner will help advance best practices for the WRF and wider regional climate modeling community. Plain Language Summary The Weather Research and Forecasting model is typically used to dynamically downscale atmospheric conditions for limited geographical regions (e.g., southern United Kingdom), using one or more nested grids of increasing resolution, and with externally provided boundary conditions (e.g., from reanalysis data). Research applications of the model (i.e., separate to forecasting) are wide ranging from hurricanes to urban heat islands. However, despite the model being constrained at its boundaries, subtle modifications to initial conditions (e.g., soil moisture) may propagate and lead to different results. We generate an ensemble of 244 different initial conditions simply by changing the model start time and show large differences in modeled temperature and other meteorological fields across a fixed analysis period. This effect, rarely considered by the Weather Research and Forecasting modeling community (perhaps due to large computational requirements), should be reflected in all future research. Key Points A 244‐member ensemble is used to show internal model variability within the Weather Research and Forecasting model Over 8 °C hourly differences between members were found (median 1.2 °C) simply by changing the model start times This overlooked feature has implications for how users should run one of the most widely used regional climate models</description><identifier>ISSN: 2169-897X</identifier><identifier>EISSN: 2169-8996</identifier><identifier>DOI: 10.1029/2019JD031286</identifier><language>eng</language><publisher>Washington: Blackwell Publishing Ltd</publisher><subject>Atmospheric conditions ; Atmospheric models ; Best practices ; Boundary conditions ; Boundary layers ; Climate models ; Computer applications ; ensemble ; Forecasting ; Geophysics ; Heat islands ; Hurricanes ; Initial conditions ; internal model variability (IMV) ; Mathematical models ; Modelling ; Numerical models ; Parameter uncertainty ; Physics ; regional climate model (RCM) ; Regional climate models ; Regional climates ; Soil ; Soil conditions ; Soil moisture ; Temperature ; Temperature range ; Uncertainty ; Urban heat islands ; Variability ; Weather ; Weather effects ; Weather forecasting ; Weather Research and Forecasting (WRF) ; Wind speed</subject><ispartof>Journal of geophysical research. Atmospheres, 2020-04, Vol.125 (7), p.n/a</ispartof><rights>2020. The Authors.</rights><rights>2020. American Geophysical Union. All Rights Reserved.</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c3454-e898c8f6623744d24833d3279e519696dca179d7695f4a5f37ecc2c4b3464b4f3</citedby><cites>FETCH-LOGICAL-c3454-e898c8f6623744d24833d3279e519696dca179d7695f4a5f37ecc2c4b3464b4f3</cites><orcidid>0000-0001-6212-1906 ; 0000-0002-9522-0713 ; 0000-0002-5334-7951 ; 0000-0002-9741-671X ; 0000-0002-5608-8887</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://onlinelibrary.wiley.com/doi/pdf/10.1029%2F2019JD031286$$EPDF$$P50$$Gwiley$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://onlinelibrary.wiley.com/doi/full/10.1029%2F2019JD031286$$EHTML$$P50$$Gwiley$$Hfree_for_read</linktohtml><link.rule.ids>314,780,784,1417,1433,27924,27925,45574,45575,46409,46833</link.rule.ids></links><search><creatorcontrib>Bassett, R.</creatorcontrib><creatorcontrib>Young, P. J.</creatorcontrib><creatorcontrib>Blair, G. S.</creatorcontrib><creatorcontrib>Samreen, F.</creatorcontrib><creatorcontrib>Simm, W.</creatorcontrib><title>A Large Ensemble Approach to Quantifying Internal Model Variability Within the WRF Numerical Model</title><title>Journal of geophysical research. Atmospheres</title><description>The Weather Research and Forecasting (WRF) community model is widely used to explore cross‐scale atmospheric features. Although WRF uncertainty studies exist, these usually involve ensembles where different physics options are selected (e.g., the boundary layer scheme) or adjusting individual parameters. Uncertainty from perturbing initial conditions, which generates internal model variability (IMV), has rarely been considered. Moreover, many off‐line WRF research studies generate conclusions based on a single model run without addressing any form of uncertainty. To demonstrate the importance of IMV, or noise, we present a 4‐month case study of summer 2018 over London, UK, using a 244‐member initial condition ensemble. Simply by changing the model start time, a median 2‐m temperature range or IMV of 1.2 °C was found (occasionally exceeding 8 °C). During our analysis, episodes of high and low IMV were found for all variables explored, explained by a relationship with the boundary condition data. Periods of slower wind speed input contained increased IMV, and vice versa, which we hypothesis is related to how strongly the boundary conditions influence the nested region. We also show the importance of IMV effects for the uncertainty of derived variables like the urban heat island, whose median variation in magnitude is 1 °C. Finally, a realistic ensemble size to capture the majority of WRF IMV is also estimated, essential considering the high computational overheads (244 members equaled 140,000 CPU hours). We envisage that highlighting considerable IMV in this repeatable manner will help advance best practices for the WRF and wider regional climate modeling community. Plain Language Summary The Weather Research and Forecasting model is typically used to dynamically downscale atmospheric conditions for limited geographical regions (e.g., southern United Kingdom), using one or more nested grids of increasing resolution, and with externally provided boundary conditions (e.g., from reanalysis data). Research applications of the model (i.e., separate to forecasting) are wide ranging from hurricanes to urban heat islands. However, despite the model being constrained at its boundaries, subtle modifications to initial conditions (e.g., soil moisture) may propagate and lead to different results. We generate an ensemble of 244 different initial conditions simply by changing the model start time and show large differences in modeled temperature and other meteorological fields across a fixed analysis period. This effect, rarely considered by the Weather Research and Forecasting modeling community (perhaps due to large computational requirements), should be reflected in all future research. Key Points A 244‐member ensemble is used to show internal model variability within the Weather Research and Forecasting model Over 8 °C hourly differences between members were found (median 1.2 °C) simply by changing the model start times This overlooked feature has implications for how users should run one of the most widely used regional climate models</description><subject>Atmospheric conditions</subject><subject>Atmospheric models</subject><subject>Best practices</subject><subject>Boundary conditions</subject><subject>Boundary layers</subject><subject>Climate models</subject><subject>Computer applications</subject><subject>ensemble</subject><subject>Forecasting</subject><subject>Geophysics</subject><subject>Heat islands</subject><subject>Hurricanes</subject><subject>Initial conditions</subject><subject>internal model variability (IMV)</subject><subject>Mathematical models</subject><subject>Modelling</subject><subject>Numerical models</subject><subject>Parameter uncertainty</subject><subject>Physics</subject><subject>regional climate model (RCM)</subject><subject>Regional climate models</subject><subject>Regional climates</subject><subject>Soil</subject><subject>Soil conditions</subject><subject>Soil moisture</subject><subject>Temperature</subject><subject>Temperature range</subject><subject>Uncertainty</subject><subject>Urban heat islands</subject><subject>Variability</subject><subject>Weather</subject><subject>Weather effects</subject><subject>Weather forecasting</subject><subject>Weather Research and Forecasting (WRF)</subject><subject>Wind speed</subject><issn>2169-897X</issn><issn>2169-8996</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>24P</sourceid><sourceid>WIN</sourceid><recordid>eNp9kF1LwzAYhYMoOHR3_oCAt1bz1TS5HPtyYyoOdd6VNE23jK6dSYr039sxFa98b8578XB4OABcYXSLEZF3BGE5HyGKieAnoEcwl5GQkp_-_sn7Oeh7v0XdCURZzHogG8CFcmsDx5U3u6w0cLDfu1rpDQw1fG5UFWzR2moNZ1UwrlIlfKhzU8I35azKbGlDC1c2bGwFw8bA1XICH5udcVb_oJfgrFClN_3vvACvk_HL8D5aPE1nw8Ei0geVyAgptCg4JzRhLCdMUJpTkkgTY8klz7XCicwTLuOCqbigidGaaJZRxlnGCnoBro-9nf9HY3xIt3VzMPYpoUJI1FXijro5UtrV3jtTpHtnd8q1KUbpYcj075AdTo_4py1N-y-bzqfLUcwxYfQLQ0Zyjw</recordid><startdate>20200416</startdate><enddate>20200416</enddate><creator>Bassett, R.</creator><creator>Young, P. J.</creator><creator>Blair, G. S.</creator><creator>Samreen, F.</creator><creator>Simm, W.</creator><general>Blackwell Publishing Ltd</general><scope>24P</scope><scope>WIN</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7TG</scope><scope>7UA</scope><scope>8FD</scope><scope>C1K</scope><scope>F1W</scope><scope>FR3</scope><scope>H8D</scope><scope>H96</scope><scope>KL.</scope><scope>KR7</scope><scope>L.G</scope><scope>L7M</scope><orcidid>https://orcid.org/0000-0001-6212-1906</orcidid><orcidid>https://orcid.org/0000-0002-9522-0713</orcidid><orcidid>https://orcid.org/0000-0002-5334-7951</orcidid><orcidid>https://orcid.org/0000-0002-9741-671X</orcidid><orcidid>https://orcid.org/0000-0002-5608-8887</orcidid></search><sort><creationdate>20200416</creationdate><title>A Large Ensemble Approach to Quantifying Internal Model Variability Within the WRF Numerical Model</title><author>Bassett, R. ; Young, P. J. ; Blair, G. S. ; Samreen, F. ; Simm, W.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c3454-e898c8f6623744d24833d3279e519696dca179d7695f4a5f37ecc2c4b3464b4f3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Atmospheric conditions</topic><topic>Atmospheric models</topic><topic>Best practices</topic><topic>Boundary conditions</topic><topic>Boundary layers</topic><topic>Climate models</topic><topic>Computer applications</topic><topic>ensemble</topic><topic>Forecasting</topic><topic>Geophysics</topic><topic>Heat islands</topic><topic>Hurricanes</topic><topic>Initial conditions</topic><topic>internal model variability (IMV)</topic><topic>Mathematical models</topic><topic>Modelling</topic><topic>Numerical models</topic><topic>Parameter uncertainty</topic><topic>Physics</topic><topic>regional climate model (RCM)</topic><topic>Regional climate models</topic><topic>Regional climates</topic><topic>Soil</topic><topic>Soil conditions</topic><topic>Soil moisture</topic><topic>Temperature</topic><topic>Temperature range</topic><topic>Uncertainty</topic><topic>Urban heat islands</topic><topic>Variability</topic><topic>Weather</topic><topic>Weather effects</topic><topic>Weather forecasting</topic><topic>Weather Research and Forecasting (WRF)</topic><topic>Wind speed</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Bassett, R.</creatorcontrib><creatorcontrib>Young, P. J.</creatorcontrib><creatorcontrib>Blair, G. S.</creatorcontrib><creatorcontrib>Samreen, F.</creatorcontrib><creatorcontrib>Simm, W.</creatorcontrib><collection>Wiley Online Library (Open Access Collection)</collection><collection>Wiley Online Library (Open Access Collection)</collection><collection>CrossRef</collection><collection>Meteorological &amp; Geoastrophysical Abstracts</collection><collection>Water Resources Abstracts</collection><collection>Technology Research Database</collection><collection>Environmental Sciences and Pollution Management</collection><collection>ASFA: Aquatic Sciences and Fisheries Abstracts</collection><collection>Engineering Research Database</collection><collection>Aerospace Database</collection><collection>Aquatic Science &amp; Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy &amp; Non-Living Resources</collection><collection>Meteorological &amp; Geoastrophysical Abstracts - Academic</collection><collection>Civil Engineering Abstracts</collection><collection>Aquatic Science &amp; Fisheries Abstracts (ASFA) Professional</collection><collection>Advanced Technologies Database with Aerospace</collection><jtitle>Journal of geophysical research. Atmospheres</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Bassett, R.</au><au>Young, P. J.</au><au>Blair, G. S.</au><au>Samreen, F.</au><au>Simm, W.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A Large Ensemble Approach to Quantifying Internal Model Variability Within the WRF Numerical Model</atitle><jtitle>Journal of geophysical research. Atmospheres</jtitle><date>2020-04-16</date><risdate>2020</risdate><volume>125</volume><issue>7</issue><epage>n/a</epage><issn>2169-897X</issn><eissn>2169-8996</eissn><abstract>The Weather Research and Forecasting (WRF) community model is widely used to explore cross‐scale atmospheric features. Although WRF uncertainty studies exist, these usually involve ensembles where different physics options are selected (e.g., the boundary layer scheme) or adjusting individual parameters. Uncertainty from perturbing initial conditions, which generates internal model variability (IMV), has rarely been considered. Moreover, many off‐line WRF research studies generate conclusions based on a single model run without addressing any form of uncertainty. To demonstrate the importance of IMV, or noise, we present a 4‐month case study of summer 2018 over London, UK, using a 244‐member initial condition ensemble. Simply by changing the model start time, a median 2‐m temperature range or IMV of 1.2 °C was found (occasionally exceeding 8 °C). During our analysis, episodes of high and low IMV were found for all variables explored, explained by a relationship with the boundary condition data. Periods of slower wind speed input contained increased IMV, and vice versa, which we hypothesis is related to how strongly the boundary conditions influence the nested region. We also show the importance of IMV effects for the uncertainty of derived variables like the urban heat island, whose median variation in magnitude is 1 °C. Finally, a realistic ensemble size to capture the majority of WRF IMV is also estimated, essential considering the high computational overheads (244 members equaled 140,000 CPU hours). We envisage that highlighting considerable IMV in this repeatable manner will help advance best practices for the WRF and wider regional climate modeling community. Plain Language Summary The Weather Research and Forecasting model is typically used to dynamically downscale atmospheric conditions for limited geographical regions (e.g., southern United Kingdom), using one or more nested grids of increasing resolution, and with externally provided boundary conditions (e.g., from reanalysis data). Research applications of the model (i.e., separate to forecasting) are wide ranging from hurricanes to urban heat islands. However, despite the model being constrained at its boundaries, subtle modifications to initial conditions (e.g., soil moisture) may propagate and lead to different results. We generate an ensemble of 244 different initial conditions simply by changing the model start time and show large differences in modeled temperature and other meteorological fields across a fixed analysis period. This effect, rarely considered by the Weather Research and Forecasting modeling community (perhaps due to large computational requirements), should be reflected in all future research. Key Points A 244‐member ensemble is used to show internal model variability within the Weather Research and Forecasting model Over 8 °C hourly differences between members were found (median 1.2 °C) simply by changing the model start times This overlooked feature has implications for how users should run one of the most widely used regional climate models</abstract><cop>Washington</cop><pub>Blackwell Publishing Ltd</pub><doi>10.1029/2019JD031286</doi><tpages>14</tpages><orcidid>https://orcid.org/0000-0001-6212-1906</orcidid><orcidid>https://orcid.org/0000-0002-9522-0713</orcidid><orcidid>https://orcid.org/0000-0002-5334-7951</orcidid><orcidid>https://orcid.org/0000-0002-9741-671X</orcidid><orcidid>https://orcid.org/0000-0002-5608-8887</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 2169-897X
ispartof Journal of geophysical research. Atmospheres, 2020-04, Vol.125 (7), p.n/a
issn 2169-897X
2169-8996
language eng
recordid cdi_proquest_journals_2388902481
source Wiley Journals; Wiley Online Library (Open Access Collection); Alma/SFX Local Collection
subjects Atmospheric conditions
Atmospheric models
Best practices
Boundary conditions
Boundary layers
Climate models
Computer applications
ensemble
Forecasting
Geophysics
Heat islands
Hurricanes
Initial conditions
internal model variability (IMV)
Mathematical models
Modelling
Numerical models
Parameter uncertainty
Physics
regional climate model (RCM)
Regional climate models
Regional climates
Soil
Soil conditions
Soil moisture
Temperature
Temperature range
Uncertainty
Urban heat islands
Variability
Weather
Weather effects
Weather forecasting
Weather Research and Forecasting (WRF)
Wind speed
title A Large Ensemble Approach to Quantifying Internal Model Variability Within the WRF Numerical Model
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T16%3A09%3A20IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20Large%20Ensemble%20Approach%20to%20Quantifying%20Internal%20Model%20Variability%20Within%20the%20WRF%20Numerical%20Model&rft.jtitle=Journal%20of%20geophysical%20research.%20Atmospheres&rft.au=Bassett,%20R.&rft.date=2020-04-16&rft.volume=125&rft.issue=7&rft.epage=n/a&rft.issn=2169-897X&rft.eissn=2169-8996&rft_id=info:doi/10.1029/2019JD031286&rft_dat=%3Cproquest_cross%3E2388902481%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2388902481&rft_id=info:pmid/&rfr_iscdi=true