Gaëlle F. Gilson, Hester Jiskoot, Soukeyna Gueye, John H. van Boxel
Abstract This study presents a comprehensive climatology of coastal fog from four synoptic weather stations operated by the Danish Meteorological Institute along the entire East Greenland coast between 1958 and 2016. Elements investigated include fog frequency, daily timing, temperature, wind, visibility and radiosonde profiles during fog. The spatiotemporal patterns in fog from the low‐ to high‐Arctic locations were related to varying regional seasonal temperatures, surface and upper‐air wind and sea ice conditions, and to correlations with the North Atlantic Oscillation (NAO) and the Greenland Blocking Index (GBI). Results indicate that ~70–80% of East Greenland fog occurs in summer (MJJA), and yearly fog onset is near‐coincident with the start of sea ice break‐up. This warm‐season fog has the typical characteristics of advection fog, as shown in the radiosonde profiles and the association with a gentle sea breeze. More than 95% of warm‐season fog is warmer than –10°C, and peaks close to 0°C and, therefore, consists of liquid or supercooled water droplets. In the cold season, mixed‐phase fog prevails in the high‐Arctic locations, accounting for ~70% of observations. Ice fog (T < –30°C) occurs in only 2% of observations and is limited to Northeast Greenland during the cold season. The cold‐season composite radiosonde fog profiles in the high Arctic locations are characterized by deep (~1000 m) and strong (~6°C) surface‐based temperature inversions. Visibility during most fog conditions is lowest during the warm season (< 500 m) and higher during the cold season (< 800 m). In Northeast Greenland, visibility during warm‐season fog has decreased by ~50 m dec ‐1 between 1981 and 2016. In Southeast Greenland, fog visibility is high during low GBI and a positive phase of NAO, but no other correlations with climate indices were found. This article is protected by copyright. All rights reserved.
摘要:本研究利用丹麦气象研究所运营的四个天气气象站在1958年至2016年间沿整个东格陵兰海岸的沿海雾的综合气候学数据。研究的因素包括雾的频率，每日时间，温度，风，能见度和无线电探空仪剖面。从低北极到高北极地区雾的时空格局与区域季节性温度、地表和高空风和海冰状况的变化有关，并与北大西洋涛动(NAO)和格陵兰阻塞指数(GBI)相关。结果表明，约70-80%的东格陵兰岛雾发生在夏季(MJJA)，每年的雾发生时间与海冰破裂的开始时间接近。这种暖季雾具有典型的平流雾的特征，如无线电探空仪剖面图所示，并与温和的海风有关。超过95%的暖季雾温度高于-10°C，峰值接近0°C，因此由液体或过冷的水滴组成。在寒冷季节，混合相雾主要出现在高北极地区，约占观测量的70%。冰雾(T <-30°C)只出现在2%的观测中，并且在寒冷季节仅限于格陵兰岛东北部。高纬度北极地区冷季复合探空雾廓线的特征是深(~1000 m)和强(~6°C)地表温度逆温。在暖季，大多数大雾天气的能见度最低(<500米)和更高在寒冷的季节(<在格陵兰东北部，1981年至2016年暖季浓雾期间的能见度下降了~50 m dec - 1。在格陵兰东南部，低GBI和NAO正相期间雾能见度较高，但与气候指数没有其他相关性。这篇文章受版权保护。版权所有。
{"title":"A climatology of Arctic fog along the coast of East Greenland","authors":"Gaëlle F. Gilson, Hester Jiskoot, Soukeyna Gueye, John H. van Boxel","doi":"10.1002/qj.4617","DOIUrl":"https://doi.org/10.1002/qj.4617","url":null,"abstract":"Abstract This study presents a comprehensive climatology of coastal fog from four synoptic weather stations operated by the Danish Meteorological Institute along the entire East Greenland coast between 1958 and 2016. Elements investigated include fog frequency, daily timing, temperature, wind, visibility and radiosonde profiles during fog. The spatiotemporal patterns in fog from the low‐ to high‐Arctic locations were related to varying regional seasonal temperatures, surface and upper‐air wind and sea ice conditions, and to correlations with the North Atlantic Oscillation (NAO) and the Greenland Blocking Index (GBI). Results indicate that ~70–80% of East Greenland fog occurs in summer (MJJA), and yearly fog onset is near‐coincident with the start of sea ice break‐up. This warm‐season fog has the typical characteristics of advection fog, as shown in the radiosonde profiles and the association with a gentle sea breeze. More than 95% of warm‐season fog is warmer than –10°C, and peaks close to 0°C and, therefore, consists of liquid or supercooled water droplets. In the cold season, mixed‐phase fog prevails in the high‐Arctic locations, accounting for ~70% of observations. Ice fog (T < –30°C) occurs in only 2% of observations and is limited to Northeast Greenland during the cold season. The cold‐season composite radiosonde fog profiles in the high Arctic locations are characterized by deep (~1000 m) and strong (~6°C) surface‐based temperature inversions. Visibility during most fog conditions is lowest during the warm season (< 500 m) and higher during the cold season (< 800 m). In Northeast Greenland, visibility during warm‐season fog has decreased by ~50 m dec ‐1 between 1981 and 2016. In Southeast Greenland, fog visibility is high during low GBI and a positive phase of NAO, but no other correlations with climate indices were found. This article is protected by copyright. All rights reserved.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":" 34","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135241342","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Dhiraj K. Singh, Sebastian W. Hoch, Ismail Gultepe, Eric R. Pardyjak
We present a case study of a coastal‐fog stratus‐cloud lowering event on 13–14 September 2018 during the C‐FOG field campaign conducted along the east coast of Newfoundland, Canada. The goal of this work is to understand the mechanisms governing the life cycle of a four‐hour‐long coastal‐fog event that resulted from the complex interplay of dynamic, thermodynamic, and microphysical processes. In addition to standard meteorological measurements, turbulence, irradiance, droplet‐size spectra, tethered‐balloon wind and thermodynamic profiles, visibility, precipitation and spatial heterogeneity of microphysics measurements are presented to discriminate and interpret the fog formation, development, and dissipation. After sunset, strong radiative cloud‐top cooling induced top‐down convection length scales that can be characterised with the Thorpe scale. Top‐down mixing and turbulent kinetic energy (TKE) generated due to buoyant/shear mixing are characterised using the flux and bulk Richardson number near the surface. Use of these parameters is unique in the analysis of fog events and helped described mixing processes. Downward mixing led to fog droplet formation that precipitated from the cloud base, which in turn cooled the sub‐cloud layers via droplet evaporation and moistened the air beneath the cloud. Once fog formed, it was affected by dry‐air entrainment from its top. As a result, the fog thinned, creating patchy fog that was characterised by remarkable oscillations in visibility near the surface. Dissipation of the fog was driven by strong turbulence above the fog layer and horizontal thermal advection demonstrated using the temperature tendency equation. This work provides novel measurements and analysis techniques that have previously not been used to understand the mechanisms governing stratus‐lowering events. These observations and analyses help highlight processes and explain mechanisms related to the fog life cycle that are inherently challenging to predict in mesoscale models. This article is protected by copyright. All rights reserved.
{"title":"A case study of the life cycle of a stratus‐lowering coastal‐fog event in Newfoundland, CA","authors":"Dhiraj K. Singh, Sebastian W. Hoch, Ismail Gultepe, Eric R. Pardyjak","doi":"10.1002/qj.4615","DOIUrl":"https://doi.org/10.1002/qj.4615","url":null,"abstract":"We present a case study of a coastal‐fog stratus‐cloud lowering event on 13–14 September 2018 during the C‐FOG field campaign conducted along the east coast of Newfoundland, Canada. The goal of this work is to understand the mechanisms governing the life cycle of a four‐hour‐long coastal‐fog event that resulted from the complex interplay of dynamic, thermodynamic, and microphysical processes. In addition to standard meteorological measurements, turbulence, irradiance, droplet‐size spectra, tethered‐balloon wind and thermodynamic profiles, visibility, precipitation and spatial heterogeneity of microphysics measurements are presented to discriminate and interpret the fog formation, development, and dissipation. After sunset, strong radiative cloud‐top cooling induced top‐down convection length scales that can be characterised with the Thorpe scale. Top‐down mixing and turbulent kinetic energy (TKE) generated due to buoyant/shear mixing are characterised using the flux and bulk Richardson number near the surface. Use of these parameters is unique in the analysis of fog events and helped described mixing processes. Downward mixing led to fog droplet formation that precipitated from the cloud base, which in turn cooled the sub‐cloud layers via droplet evaporation and moistened the air beneath the cloud. Once fog formed, it was affected by dry‐air entrainment from its top. As a result, the fog thinned, creating patchy fog that was characterised by remarkable oscillations in visibility near the surface. Dissipation of the fog was driven by strong turbulence above the fog layer and horizontal thermal advection demonstrated using the temperature tendency equation. This work provides novel measurements and analysis techniques that have previously not been used to understand the mechanisms governing stratus‐lowering events. These observations and analyses help highlight processes and explain mechanisms related to the fog life cycle that are inherently challenging to predict in mesoscale models. This article is protected by copyright. All rights reserved.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":"83 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135540097","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract As computer power increases, there is a need to investigate the potential gains of using more than two streams in the radiative transfer calculations of weather and climate models. In this article, seven quadrature schemes for selecting the zenith angles and weights of these streams are evaluated rigorously in terms of the accuracy of thermal‐infrared radiative transfer calculations. In addition, a new method is presented for generating “optimized” angles and weights that minimize the thermal‐infrared irradiance and heating‐rate errors for a set of clear‐sky training profiles. It is found that the standard approach of applying Gauss–Legendre quadrature in each hemisphere is the least accurate of all those tested for two and four streams. For clear‐sky irradiance calculations, “optimized” quadrature is between one and two orders of magnitude more accurate than Gauss–Legendre for any number of streams. For all‐sky calculations in which scattering becomes important, a form of Gauss–Jacobi quadrature is found to be most accurate for between four and eight streams, but with Gauss–Legendre being the most accurate for 10 or more streams. No single quadrature scheme performs best in all situations, because computing irradiances involves two different integrals over angle and the relative importance of each integral depends on the amount of scattering taking place. Additional optimized quadratures for clear‐sky and all‐sky calculations with four to eight streams are presented that constrain the relationships between angles in a way that reduces the number of exponentials that need to be computed in a radiative transfer solver.
{"title":"What are the optimum discrete angles to use in thermal‐infrared radiative transfer calculations?","authors":"Robin J. Hogan","doi":"10.1002/qj.4598","DOIUrl":"https://doi.org/10.1002/qj.4598","url":null,"abstract":"Abstract As computer power increases, there is a need to investigate the potential gains of using more than two streams in the radiative transfer calculations of weather and climate models. In this article, seven quadrature schemes for selecting the zenith angles and weights of these streams are evaluated rigorously in terms of the accuracy of thermal‐infrared radiative transfer calculations. In addition, a new method is presented for generating “optimized” angles and weights that minimize the thermal‐infrared irradiance and heating‐rate errors for a set of clear‐sky training profiles. It is found that the standard approach of applying Gauss–Legendre quadrature in each hemisphere is the least accurate of all those tested for two and four streams. For clear‐sky irradiance calculations, “optimized” quadrature is between one and two orders of magnitude more accurate than Gauss–Legendre for any number of streams. For all‐sky calculations in which scattering becomes important, a form of Gauss–Jacobi quadrature is found to be most accurate for between four and eight streams, but with Gauss–Legendre being the most accurate for 10 or more streams. No single quadrature scheme performs best in all situations, because computing irradiances involves two different integrals over angle and the relative importance of each integral depends on the amount of scattering taking place. Additional optimized quadratures for clear‐sky and all‐sky calculations with four to eight streams are presented that constrain the relationships between angles in a way that reduces the number of exponentials that need to be computed in a radiative transfer solver.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135431841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Leon Hermanson, Nick Dunstone, Rosie Eade, Doug Smith
Abstract Ocean reanalyses covering many decades, including those with few observations, are needed to understand climate variability and to initialize and assess interannual to decadal climate predictions. The Met Office Statistical Ocean Re‐Analysis (MOSORA) exploits long‐range covariances to generate full‐depth reanalyses of monthly ocean temperature and salinity even from sparse observations. We extend MOSORA by generating an ensemble that samples uncertainties in long‐range covariances. Initial covariances are taken from model runs and these are improved with observations using an iterative process. We demonstrate that covariances are improved by iteration, and that this procedure, using very sparse observations typical of the 1960s, captures many features of analyses benefiting from modern observation density. We investigate the ensemble spread and find that salinity trends in the covariances from model runs can introduce unexpected changes in the reanalyses. We nudge the reanalyses into an ensemble of coupled climate models to produce estimates of the Atlantic Meridional Overturning Circulation (AMOC). At 26°N, the AMOC shows decadal variability consistent with observations at this latitude and shows signs of strengthening in recent years. The ensemble spread in AMOC reconstructions increases with time as more observations interact with uncertain covariances. At 45°N, the amount of decadal variability in the AMOC varies between members, but on shorter timescales the variability is similar across the ensemble. At 45°N, the AMOC can be constrained better with more observations on the western boundary, but longer continuous observations are needed to improve covariances and reduce uncertainties in the AMOC.
{"title":"An ensemble reconstruction of ocean temperature, salinity and Atlantic Meridional Overturning Circulation 1960–2021","authors":"Leon Hermanson, Nick Dunstone, Rosie Eade, Doug Smith","doi":"10.1002/qj.4587","DOIUrl":"https://doi.org/10.1002/qj.4587","url":null,"abstract":"Abstract Ocean reanalyses covering many decades, including those with few observations, are needed to understand climate variability and to initialize and assess interannual to decadal climate predictions. The Met Office Statistical Ocean Re‐Analysis (MOSORA) exploits long‐range covariances to generate full‐depth reanalyses of monthly ocean temperature and salinity even from sparse observations. We extend MOSORA by generating an ensemble that samples uncertainties in long‐range covariances. Initial covariances are taken from model runs and these are improved with observations using an iterative process. We demonstrate that covariances are improved by iteration, and that this procedure, using very sparse observations typical of the 1960s, captures many features of analyses benefiting from modern observation density. We investigate the ensemble spread and find that salinity trends in the covariances from model runs can introduce unexpected changes in the reanalyses. We nudge the reanalyses into an ensemble of coupled climate models to produce estimates of the Atlantic Meridional Overturning Circulation (AMOC). At 26°N, the AMOC shows decadal variability consistent with observations at this latitude and shows signs of strengthening in recent years. The ensemble spread in AMOC reconstructions increases with time as more observations interact with uncertain covariances. At 45°N, the amount of decadal variability in the AMOC varies between members, but on shorter timescales the variability is similar across the ensemble. At 45°N, the AMOC can be constrained better with more observations on the western boundary, but longer continuous observations are needed to improve covariances and reduce uncertainties in the AMOC.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":"25 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135432010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jonathan Coney, Leif Denby, Andrew N. Ross, He Wang, Simon Vosper, Annelize van Niekerk, Tom Dunstan, Neil Hindley
Trapped lee waves, and resultant turbulent rotors downstream, present a hazard for aviation and land‐based transport. Though high‐resolution numerical weather prediction models can represent such phenomena, there is currently no simple and reliable automated method for detecting the extent and characteristics of these waves in model output. Spectral transform methods have traditionally been used to detect and characterise regions of wave activity in model and observational data; however, these methods can be slow and have their limitations. Machine‐learning (ML) techniques offer a new and potentially fruitful method of tackling this problem. We demonstrate that a deep‐learning model can be trained to accurately recognise and label coherent regions of lee waves from vertical velocity data on a single level from a high‐resolution numerical weather prediction (NWP) model. Using transfer learning, wave characteristics (wavelength, orientation, and amplitude) can be extracted from the trained segmentation model. The use of synthetic wave fields with prescribed wave characteristics makes this transfer learning possible without the need to characterise real complex wave fields. Addition of noise to the synthetic data makes the models more robust when applied to more complex and noisy NWP data. The collection of trained models produced provides a valuable tool to investigate the prevalence and nature of lee wave activity, as well as a new way for forecasters to detect resolved waves. The deep‐learning model was more capable and quicker at detecting and characterising lee waves than a spectral technique was. This work is just one example of how already established ML techniques can be used to detect and characterise complex weather phenomena from NWP model output and observational data, and how the careful use of synthetic data can reduce the requirements for large volumes of hand‐labelled training data for ML models.
{"title":"Identifying and characterising trapped lee waves using deep learning techniques","authors":"Jonathan Coney, Leif Denby, Andrew N. Ross, He Wang, Simon Vosper, Annelize van Niekerk, Tom Dunstan, Neil Hindley","doi":"10.1002/qj.4592","DOIUrl":"https://doi.org/10.1002/qj.4592","url":null,"abstract":"Trapped lee waves, and resultant turbulent rotors downstream, present a hazard for aviation and land‐based transport. Though high‐resolution numerical weather prediction models can represent such phenomena, there is currently no simple and reliable automated method for detecting the extent and characteristics of these waves in model output. Spectral transform methods have traditionally been used to detect and characterise regions of wave activity in model and observational data; however, these methods can be slow and have their limitations. Machine‐learning (ML) techniques offer a new and potentially fruitful method of tackling this problem. We demonstrate that a deep‐learning model can be trained to accurately recognise and label coherent regions of lee waves from vertical velocity data on a single level from a high‐resolution numerical weather prediction (NWP) model. Using transfer learning, wave characteristics (wavelength, orientation, and amplitude) can be extracted from the trained segmentation model. The use of synthetic wave fields with prescribed wave characteristics makes this transfer learning possible without the need to characterise real complex wave fields. Addition of noise to the synthetic data makes the models more robust when applied to more complex and noisy NWP data. The collection of trained models produced provides a valuable tool to investigate the prevalence and nature of lee wave activity, as well as a new way for forecasters to detect resolved waves. The deep‐learning model was more capable and quicker at detecting and characterising lee waves than a spectral technique was. This work is just one example of how already established ML techniques can be used to detect and characterise complex weather phenomena from NWP model output and observational data, and how the careful use of synthetic data can reduce the requirements for large volumes of hand‐labelled training data for ML models.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":"20 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135431838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Low‐pressure systems (LPSs) are the primary rainbringers of the South Asian monsoon. Yet, their interactions with the large‐scale monsoon circulation, as well as the highly variable land and sea surfaces they pass over, are complex and generally not well understood. In this article, we present a novel, top‐down approach to investigate these relationships and quantify their importance in describing LPS behaviour. We also show that, if the approach is sufficiently well posed, it is productive at hypothesis generation. For each of five predictands (i.e., LPS intensification rate, propagation speed/direction, post‐landfall survival, peak intensity, and precipitation rate) we train an additive decision‐tree ensemble model using the XGBoost algorithm. Shapley value analysis is then applied to the models to determine which variables are important predictors and to establish their relationship with the predictand, with additional analysis following cases of interest. Novel relationships established using this technique include that LPS vorticity intensifies preferentially in the early morning at the same time as the peak in the diurnal cycle of their convection occurs, that vertical wind shear suppresses continued growth of strong LPSs, that large‐scale barotropic instability plays an important role in both the inland penetration and peak intensity of LPSs, and that LPS propagation depends on the depth of its vortex with shallower LPSs advected by low‐level winds and taller LPSs advected by mid‐level winds. We also use this framework to identify and discuss potential new avenues of research for monsoon LPSs.
{"title":"Using interpretable gradient‐boosted decision‐tree ensembles to uncover novel dynamical relationships governing monsoon low‐pressure systems","authors":"Kieran M. R. Hunt, Andrew G. Turner","doi":"10.1002/qj.4582","DOIUrl":"https://doi.org/10.1002/qj.4582","url":null,"abstract":"Abstract Low‐pressure systems (LPSs) are the primary rainbringers of the South Asian monsoon. Yet, their interactions with the large‐scale monsoon circulation, as well as the highly variable land and sea surfaces they pass over, are complex and generally not well understood. In this article, we present a novel, top‐down approach to investigate these relationships and quantify their importance in describing LPS behaviour. We also show that, if the approach is sufficiently well posed, it is productive at hypothesis generation. For each of five predictands (i.e., LPS intensification rate, propagation speed/direction, post‐landfall survival, peak intensity, and precipitation rate) we train an additive decision‐tree ensemble model using the XGBoost algorithm. Shapley value analysis is then applied to the models to determine which variables are important predictors and to establish their relationship with the predictand, with additional analysis following cases of interest. Novel relationships established using this technique include that LPS vorticity intensifies preferentially in the early morning at the same time as the peak in the diurnal cycle of their convection occurs, that vertical wind shear suppresses continued growth of strong LPSs, that large‐scale barotropic instability plays an important role in both the inland penetration and peak intensity of LPSs, and that LPS propagation depends on the depth of its vortex with shallower LPSs advected by low‐level winds and taller LPSs advected by mid‐level winds. We also use this framework to identify and discuss potential new avenues of research for monsoon LPSs.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":"28 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135430958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract With the development of advanced data assimilation and computing techniques, many modern global reanalysis datasets aim to resolve the atmospheric mesoscale spectrum. However, large uncertainties remain with respect to the representation of mesoscale motions in these reanalysis datasets, for which a clear understanding is lacking. The above challenges have served as a strong motivation to reveal and quantify their mesoscale differences. This study presents the first comprehensive global intercomparison of the tropospheric and stratospheric mesoscale kinetic energy and its spectra over two selected periods of summer and winter events among six leading high‐resolution atmospheric reanalysis products, including ERA5, CRA, MERRA2, CFSv2, JRA‐55, and ERA‐I. A state‐of‐the‐art global operational model is adopted as a supplementary reference. Although all reanalysis datasets can reproduce broad distribution characteristics that are grossly consistent with the 9‐km model, there are substantial discrepancies among them in magnitudes. The ability to capture mesoscale signals is closely linked to their resolutions, but it is also impacted by other factors, including but not limited to the selected types of energy, seasons, altitudes, latitudes, model diffusions, parameterization schemes, moist condition, assimilation methods, and observation inputs. Moreover, all datasets illustrate conclusive behaviors for the prevalence of the rotational component in the troposphere, while only very few products fail to exhibit the dominance of the divergent component in the stratosphere. Overall, stratospheric ERA5 and CFSv2 outperform the other reanalysis datasets, and only these two can reproduce the feature of the canonical kinetic energy spectrum with a distinct shift from a steeper slope (~‐3) at lower wavenumbers to a shallower slope (~‐5/3) at higher wavenumbers. In addition, the relative disparities among datasets increase dramatically with height and they are more pronounced in the divergent component. It is also found that the correlations among these datasets are much weaker in the tropics. This article is protected by copyright. All rights reserved.
{"title":"Intercomparison of Tropospheric and Stratospheric Mesoscale Kinetic Energy Resolved by the High‐Resolution Global Reanalysis Datasets","authors":"Ziyi Li, Junhong Wei, Xinghua Bao, Y. Qiang Sun","doi":"10.1002/qj.4605","DOIUrl":"https://doi.org/10.1002/qj.4605","url":null,"abstract":"Abstract With the development of advanced data assimilation and computing techniques, many modern global reanalysis datasets aim to resolve the atmospheric mesoscale spectrum. However, large uncertainties remain with respect to the representation of mesoscale motions in these reanalysis datasets, for which a clear understanding is lacking. The above challenges have served as a strong motivation to reveal and quantify their mesoscale differences. This study presents the first comprehensive global intercomparison of the tropospheric and stratospheric mesoscale kinetic energy and its spectra over two selected periods of summer and winter events among six leading high‐resolution atmospheric reanalysis products, including ERA5, CRA, MERRA2, CFSv2, JRA‐55, and ERA‐I. A state‐of‐the‐art global operational model is adopted as a supplementary reference. Although all reanalysis datasets can reproduce broad distribution characteristics that are grossly consistent with the 9‐km model, there are substantial discrepancies among them in magnitudes. The ability to capture mesoscale signals is closely linked to their resolutions, but it is also impacted by other factors, including but not limited to the selected types of energy, seasons, altitudes, latitudes, model diffusions, parameterization schemes, moist condition, assimilation methods, and observation inputs. Moreover, all datasets illustrate conclusive behaviors for the prevalence of the rotational component in the troposphere, while only very few products fail to exhibit the dominance of the divergent component in the stratosphere. Overall, stratospheric ERA5 and CFSv2 outperform the other reanalysis datasets, and only these two can reproduce the feature of the canonical kinetic energy spectrum with a distinct shift from a steeper slope (~‐3) at lower wavenumbers to a shallower slope (~‐5/3) at higher wavenumbers. In addition, the relative disparities among datasets increase dramatically with height and they are more pronounced in the divergent component. It is also found that the correlations among these datasets are much weaker in the tropics. This article is protected by copyright. All rights reserved.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":"29 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135681997","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Gabriel Stachura, Zbigniew Ustrnul, Piotr Sekuła, Bogdan Bochenek, Marcin Kolonko, Małgorzata Szczęch‐Gajewska
Abstract In the article, a machine learning based tool for calibrating numerical forecasts of near‐surface air temperature is proposed. The study area covers Poland representing a temperate type of climate with transitional features and highly variable weather. A direct output of numerical weather prediction (NWP) models is often biased and needs to be adjusted to observed values. Forecasters have to reconcile forecasts from several NWP models during their operational work. As the proposed method is based on deterministic forecasts from three short‐range limited area models (ALARO, AROME and COSMO), it can support them in their decision‐making process. Predictors include forecasts of weather elements produced by the NWP models at synoptic weather stations across Poland and station‐embedded data on ambient orography. The Random Forests algorithm (RF) has been used to produce bias‐corrected forecasts on a test set spanning one year. Its performance was evaluated against the NWP models, a linear combination of all predictors (multiple linear regression, MLR) as well as a basic Artificial Neural Network (ANN). Detailed evaluation was done to identify potential strengths and weaknesses of the model at the temporal and spatial scale. The value of RMSE of a forecast obtained by the RF model was 11% and 27% lower compared to the MLR model and the best performing NWP model, respectively. The ANN model turned out to be even superior, outperforming RF by around 2.5%. The greatest improvement occurred for warm bias during the nighttime from July to September. The largest difference in forecast accuracy between RF and ANN appeared for temperature drops at April nights. Poor performance of RF for extreme temperature ranges may be suppressed by training the model on forecast error instead of observed values of the variable. This article is protected by copyright. All rights reserved.
{"title":"Machine learning based post‐processing of model‐derived near‐surface air temperature – a multi‐model approach","authors":"Gabriel Stachura, Zbigniew Ustrnul, Piotr Sekuła, Bogdan Bochenek, Marcin Kolonko, Małgorzata Szczęch‐Gajewska","doi":"10.1002/qj.4613","DOIUrl":"https://doi.org/10.1002/qj.4613","url":null,"abstract":"Abstract In the article, a machine learning based tool for calibrating numerical forecasts of near‐surface air temperature is proposed. The study area covers Poland representing a temperate type of climate with transitional features and highly variable weather. A direct output of numerical weather prediction (NWP) models is often biased and needs to be adjusted to observed values. Forecasters have to reconcile forecasts from several NWP models during their operational work. As the proposed method is based on deterministic forecasts from three short‐range limited area models (ALARO, AROME and COSMO), it can support them in their decision‐making process. Predictors include forecasts of weather elements produced by the NWP models at synoptic weather stations across Poland and station‐embedded data on ambient orography. The Random Forests algorithm (RF) has been used to produce bias‐corrected forecasts on a test set spanning one year. Its performance was evaluated against the NWP models, a linear combination of all predictors (multiple linear regression, MLR) as well as a basic Artificial Neural Network (ANN). Detailed evaluation was done to identify potential strengths and weaknesses of the model at the temporal and spatial scale. The value of RMSE of a forecast obtained by the RF model was 11% and 27% lower compared to the MLR model and the best performing NWP model, respectively. The ANN model turned out to be even superior, outperforming RF by around 2.5%. The greatest improvement occurred for warm bias during the nighttime from July to September. The largest difference in forecast accuracy between RF and ANN appeared for temperature drops at April nights. Poor performance of RF for extreme temperature ranges may be suppressed by training the model on forecast error instead of observed values of the variable. This article is protected by copyright. All rights reserved.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":"73 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135725943","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Two methods are widely used to assess the impact of observations in global numerical weather prediction (NWP): data denial experiments (DDEs) and the forecast sensitivity‐based observation impact (FSOI) method. A DDE measures the impact on forecast accuracy of removing an observation type from the system, whereas FSOI estimates the amount by which an observation type reduces the short‐range forecast error within a system containing all observation types. This paper describes the second part of a two‐part study. In the first part, the theory behind DDE and FSOI metrics was presented and then applied to a simple model with two state variables, all in the context of optimal data assimilation (DA), for which the error covariances used in the DA system match reality. The paper showed why and under what conditions the DDE and FSOI metrics give different results, even for an optimal DA system. In this second part, we extend the theory to suboptimal systems, and specifically to systems that are suboptimal in their specification of observation errors, and then apply it to a very simple model, in this case with one state variable. As expected, DDE impacts are reduced when the system is suboptimal. By contrast, relative FSOI impacts (i.e. relative to those of other observation types) increase for an observation type for which the errors are underestimated. This gives the erroneous impression that the change in assumed errors has led to an improvement, whereas the opposite is the case. These results provide some insights into the interpretation of FSOI results from a suboptimal DA system. This article is protected by copyright. All rights reserved.
{"title":"Observation impact metrics in NWP: a theoretical study. Part II: systems with suboptimal observation errors","authors":"J. R. Eyre","doi":"10.1002/qj.4614","DOIUrl":"https://doi.org/10.1002/qj.4614","url":null,"abstract":"Abstract Two methods are widely used to assess the impact of observations in global numerical weather prediction (NWP): data denial experiments (DDEs) and the forecast sensitivity‐based observation impact (FSOI) method. A DDE measures the impact on forecast accuracy of removing an observation type from the system, whereas FSOI estimates the amount by which an observation type reduces the short‐range forecast error within a system containing all observation types. This paper describes the second part of a two‐part study. In the first part, the theory behind DDE and FSOI metrics was presented and then applied to a simple model with two state variables, all in the context of optimal data assimilation (DA), for which the error covariances used in the DA system match reality. The paper showed why and under what conditions the DDE and FSOI metrics give different results, even for an optimal DA system. In this second part, we extend the theory to suboptimal systems, and specifically to systems that are suboptimal in their specification of observation errors, and then apply it to a very simple model, in this case with one state variable. As expected, DDE impacts are reduced when the system is suboptimal. By contrast, relative FSOI impacts (i.e. relative to those of other observation types) increase for an observation type for which the errors are underestimated. This gives the erroneous impression that the change in assumed errors has led to an improvement, whereas the opposite is the case. These results provide some insights into the interpretation of FSOI results from a suboptimal DA system. This article is protected by copyright. All rights reserved.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":"71 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135725812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract The operational Aeolus Level‐2B (L2B) horizontal line‐of‐sight (HLOS) retrieved Rayleigh winds, produced by the European Space Agency (ESA), utilize European Centre for Medium‐Range Weather Forecasts (ECMWF) short‐term forecasts of temperature, pressure, and horizontal winds in the Rayleigh–Brillouin and M1 correction procedures. These model fields or backgrounds can contain ECMWF model‐specific errors, which may propagate to the retrieved Rayleigh winds. This study examines the sensitivity of the retrieved Rayleigh winds to the changes in the model backgrounds, and the potential benefit of using the same system, in this case the National Oceanic and Atmospheric Administration's Finite‐Volume Cubed Sphere Global Forecast System (FV3GFS), for both the corrections and the data assimilation and forecast procedures. It is shown that the differences in the model backgrounds (FV3GFS minus ECMWF) can propagate through the Level‐2B horizontal line‐of‐sight Rayleigh wind retrieval process, mainly the M1 correction, resulting in differences in the retrieved Rayleigh winds with mean and standard deviation of magnitude as large as 0.2 m·s −1 . The differences reach up to 0.4, 0.6, and 0.7 m·s −1 for the 95th, 99th, and 99.5th percentiles of the sample distribution with maxima of ∼1.4 m·s −1 . The numbers of the large differences for the combined lower and upper 5th, 1st, and 0.5th percentile pairs are ∼6,100, 1,220, and 610 between 2.5 and 25 km height globally per day respectively. The ESA‐disseminated Rayleigh wind product (based on the ECMWF corrections) already shows a significant positive impact on the FV3GFS global forecasts. In the observing system experiments performed, compared with the ESA Rayleigh winds, the use of the FV3GFS‐corrected Rayleigh winds lead to ∼0.5% more Rayleigh winds assimilated in the lower troposphere and show enhanced positive impact on FV3GFS forecasts at the day 1–10 range but limited to the Southern Hemisphere.
{"title":"On the Use of Consistent Bias Corrections to Enhance the Impact of Aeolus <scp>Level‐2B</scp> Rayleigh Winds on <scp>NOAA</scp> Global Forecast Skill","authors":"Hui Liu, Kevin Garrett, Kayo Ide, Ross N. Hoffman","doi":"10.1002/qj.4600","DOIUrl":"https://doi.org/10.1002/qj.4600","url":null,"abstract":"Abstract The operational Aeolus Level‐2B (L2B) horizontal line‐of‐sight (HLOS) retrieved Rayleigh winds, produced by the European Space Agency (ESA), utilize European Centre for Medium‐Range Weather Forecasts (ECMWF) short‐term forecasts of temperature, pressure, and horizontal winds in the Rayleigh–Brillouin and M1 correction procedures. These model fields or backgrounds can contain ECMWF model‐specific errors, which may propagate to the retrieved Rayleigh winds. This study examines the sensitivity of the retrieved Rayleigh winds to the changes in the model backgrounds, and the potential benefit of using the same system, in this case the National Oceanic and Atmospheric Administration's Finite‐Volume Cubed Sphere Global Forecast System (FV3GFS), for both the corrections and the data assimilation and forecast procedures. It is shown that the differences in the model backgrounds (FV3GFS minus ECMWF) can propagate through the Level‐2B horizontal line‐of‐sight Rayleigh wind retrieval process, mainly the M1 correction, resulting in differences in the retrieved Rayleigh winds with mean and standard deviation of magnitude as large as 0.2 m·s −1 . The differences reach up to 0.4, 0.6, and 0.7 m·s −1 for the 95th, 99th, and 99.5th percentiles of the sample distribution with maxima of ∼1.4 m·s −1 . The numbers of the large differences for the combined lower and upper 5th, 1st, and 0.5th percentile pairs are ∼6,100, 1,220, and 610 between 2.5 and 25 km height globally per day respectively. The ESA‐disseminated Rayleigh wind product (based on the ECMWF corrections) already shows a significant positive impact on the FV3GFS global forecasts. In the observing system experiments performed, compared with the ESA Rayleigh winds, the use of the FV3GFS‐corrected Rayleigh winds lead to ∼0.5% more Rayleigh winds assimilated in the lower troposphere and show enhanced positive impact on FV3GFS forecasts at the day 1–10 range but limited to the Southern Hemisphere.","PeriodicalId":49646,"journal":{"name":"Quarterly Journal of the Royal Meteorological Society","volume":"176 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135775288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：