Joanne Robbins
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Joanne Robbins has a total of 97 co-authors. Here are the top co-authors by number of shared publications:
- Michelle D. Spruce
Department of Computer Science, University of Exeter, UK
6 shared publications - Robert Neal
Met Office Exeter UK
6 shared publications Hywel T. P. Williams
College of Engineering, Maths and Physical Sciences, University of Exeter, Exeter EX4 4SB, UK
5 shared publicationsRudy Arthur
College of Engineering, Maths and Physical Sciences, University of Exeter, Exeter EX4 4SB, UK
5 shared publications- Seshagiri Rao Kolusu
UK Met Office, Fitzroy Road, Exeter, EX1 3PB
5 shared publications - Marion Mittermaier
Met Office Exeter Devon UK
4 shared publications
Latest Publications
- Open Access
- 11 September 2026
- Published by Wiley in Meteorological Applications
- Vol. 33 (5) , e70242
Compound Cold Events, Impacts and Weather Patterns: A Case Study of the Scottish Rail Network
Winter weather events disrupt critical national infrastructure, with rail networks particularly vulnerable to blocked tracks, mechanical faults, and frozen infrastructure leading to service delays and cancellations. The impacts of these cold weather events can be amplified by the compounding effect of another meteorological variable, particularly whether heavy precipitation is present or absent, with subsequent impacts dependent on the nature of the cold event. For example, a ‘compounding’ cold‐wet event could incur heavy snowfall, whereas a cold‐dry event could result in extreme low temperatures and icy conditions. Despite their prevalence, such compound cold extremes remain understudied compared to compound heat extremes. In this study, we analyse 10,500 Scottish rail incidents (2006–2023, October to March) using Network Rail impact data, the set of Met Office 30 weather patterns (MO30), and HadUK‐Grid observations to link incidents with compound cold events. We identify high‐impact days based on two classifications: (1) days exceeding the 99th percentile of aggregated daily incidents; (2) days exceeding the 99th percentile of daily customer minutes lost. The majority of high‐impact days were snow‐related, occurring under cold‐dry or cold‐wet conditions with the definition depending on whether snow was falling or lying. These findings highlight the importance of considering co‐occurring cold hazards, as real‐world impacts may diverge from traditional hazard‐based detection. This provides useful insight for compound cold early warning systems.
- 11 September 2026
- Published by Wiley in Meteorological Applications
- Vol. 33 (5) , e70242
Open AccessCompound Cold Events, Impacts and Weather Patterns: A Case Study of the Scottish Rail Network
Winter weather events disrupt critical national infrastructure, with rail networks particularly vulnerable to blocked tracks, mechanical faults, and frozen infrastructure leading to service delays and cancellations. The impacts of these cold weather events can be amplified by the compounding effect of another meteorological variable, particularly whether heavy precipitation is present or absent, with subsequent impacts dependent on the nature of the cold event. For example, a ‘compounding’ cold‐wet event could incur heavy snowfall, whereas a cold‐dry event could result in extreme low temperatures and icy conditions. Despite their prevalence, such compound cold extremes remain understudied compared to compound heat extremes. In this study, we analyse 10,500 Scottish rail incidents (2006–2023, October to March) using Network Rail impact data, the set of Met Office 30 weather patterns (MO30), and HadUK‐Grid observations to link incidents with compound cold events. We identify high‐impact days based on two classifications: (1) days exceeding the 99th percentile of aggregated daily incidents; (2) days exceeding the 99th percentile of daily customer minutes lost. The majority of high‐impact days were snow‐related, occurring under cold‐dry or cold‐wet conditions with the definition depending on whether snow was falling or lying. These findings highlight the importance of considering co‐occurring cold hazards, as real‐world impacts may diverge from traditional hazard‐based detection. This provides useful insight for compound cold early warning systems.
- Open Access
- 13 March 2026
- Published by Copernicus GmbH
ANTICIPATE COST Action: extended-range multi-hazard predictions and early warnings
Operational extreme weather forecasts and early warnings are generally limited to timescales of up to around 10 days and to predicting single events, such as flooding or a heatwave. However, experimental ‘extended-range’ weather predictions that extend up to 46 days have been developed over the last decade by the world’s leading meteorological centres. A key motivation of exploring this prediction timescale is to bridge the gap between timescales, incorporate the latest ‘multi-hazard’ approaches, and improve early warnings and anticipatory actions. Currently, however, the extended-range prediction and the multi-hazard research and operational communities are largely disconnected. To date, there has been no coordinated effort to build a network that connects these disciplines and communities towards the development of operational systems. However, it is essential that these communities come together to explore windows of opportunity and instigate a step-change in the way forecasts are designed, produced and used. To address this challenge, here we present the ANTICIPATE COST Action (CA24144) that has created the first pan-European network focused on extended-range multi-hazard predictions and warnings. Over the next 4 years, ANTICIPATE will bring together existing but largely disconnected disciplines, operational practitioners and stakeholders (including extreme weather forecasting, extended-range prediction and climate dynamics, disaster risk reduction, multi-hazards, and communications) to drive forward advancements in the science, training, communication and application that will support next generation of effective early warnings that enable preparedness and action across hazards and forecasting lead times. In this talk, we will explore upcoming events and activities, and share how ANTICIPATE will provide vital leadership in multi-hazard predictions and warnings, address gaps and challenges, and help educate the next generation of forecasters and communicators for societal benefit. Further details about the ANTICIPATE COST Action are available here: https://www.cost.eu/actions/CA24144/.
- 13 March 2026
- Published by Copernicus GmbH
Open AccessANTICIPATE COST Action: extended-range multi-hazard predictions and early warnings
Operational extreme weather forecasts and early warnings are generally limited to timescales of up to around 10 days and to predicting single events, such as flooding or a heatwave. However, experimental ‘extended-range’ weather predictions that extend up to 46 days have been developed over the last decade by the world’s leading meteorological centres. A key motivation of exploring this prediction timescale is to bridge the gap between timescales, incorporate the latest ‘multi-hazard’ approaches, and improve early warnings and anticipatory actions. Currently, however, the extended-range prediction and the multi-hazard research and operational communities are largely disconnected. To date, there has been no coordinated effort to build a network that connects these disciplines and communities towards the development of operational systems. However, it is essential that these communities come together to explore windows of opportunity and instigate a step-change in the way forecasts are designed, produced and used. To address this challenge, here we present the ANTICIPATE COST Action (CA24144) that has created the first pan-European network focused on extended-range multi-hazard predictions and warnings. Over the next 4 years, ANTICIPATE will bring together existing but largely disconnected disciplines, operational practitioners and stakeholders (including extreme weather forecasting, extended-range prediction and climate dynamics, disaster risk reduction, multi-hazards, and communications) to drive forward advancements in the science, training, communication and application that will support next generation of effective early warnings that enable preparedness and action across hazards and forecasting lead times. In this talk, we will explore upcoming events and activities, and share how ANTICIPATE will provide vital leadership in multi-hazard predictions and warnings, address gaps and challenges, and help educate the next generation of forecasters and communicators for societal benefit. Further details about the ANTICIPATE COST Action are available here: https://www.cost.eu/actions/CA24144/.
- Open Access
- 30 June 2025
- Published by Copernicus GmbH
How different approaches to exposure and vulnerability inclusion affect Impact-based Forecasts and Warnings: Storm Eunice (2022) UK case study
The Met Office National Severe Weather Warning Service (NSWWS) has provided Impact-based Warnings to the public and local responders since 2011. These warnings aim to convey the likelihood of societal impacts resulting from adverse weather. The approach uses a warning impact matrix that categorises impact severity (very low, low, medium and high) and likelihood to assign a warning colour classification (yellow, amber and red). To assess impact severity effectively, consideration of the spatial and temporal variability of exposure and vulnerability is necessary. Different approaches are currently used to integrate this information in risk and impact-based warnings (e.g. Impact tables developed through engagement with the responder communities; expert judgement and engagement with local advisors; quantitative vulnerability and exposure indicators included within automated risk forecasting tools).Using Storm Eunice as a case study, we explore how different methods of quantifying and integrating exposure and vulnerability could influence warning classification. Storm Eunice impacted the UK in February 2022 and led to the issuance of two rare, red wind warnings. Focusing on building damage, which is referred to as structural damage in the Met Office impact tables, and population affected, we review different approaches to warning generation using a: (1) hazard-only approach; (2) exposure-based approach; and (3) a risk-based approach (inclusive of vulnerability) and compare the possible warnings these different methods could produce. Using these results, we then reflect on the influence these differences have on warning recipients and the effect this might have on warning communication and actions taken. We will leverage results from a recent Warning Value Chain Assessment of Storm Eunice to reflect on the uptake and responses to different warnings.
- 30 June 2025
- Published by Copernicus GmbH
Open AccessHow different approaches to exposure and vulnerability inclusion affect Impact-based Forecasts and Warnings: Storm Eunice (2022) UK case study
The Met Office National Severe Weather Warning Service (NSWWS) has provided Impact-based Warnings to the public and local responders since 2011. These warnings aim to convey the likelihood of societal impacts resulting from adverse weather. The approach uses a warning impact matrix that categorises impact severity (very low, low, medium and high) and likelihood to assign a warning colour classification (yellow, amber and red). To assess impact severity effectively, consideration of the spatial and temporal variability of exposure and vulnerability is necessary. Different approaches are currently used to integrate this information in risk and impact-based warnings (e.g. Impact tables developed through engagement with the responder communities; expert judgement and engagement with local advisors; quantitative vulnerability and exposure indicators included within automated risk forecasting tools).Using Storm Eunice as a case study, we explore how different methods of quantifying and integrating exposure and vulnerability could influence warning classification. Storm Eunice impacted the UK in February 2022 and led to the issuance of two rare, red wind warnings. Focusing on building damage, which is referred to as structural damage in the Met Office impact tables, and population affected, we review different approaches to warning generation using a: (1) hazard-only approach; (2) exposure-based approach; and (3) a risk-based approach (inclusive of vulnerability) and compare the possible warnings these different methods could produce. Using these results, we then reflect on the influence these differences have on warning recipients and the effect this might have on warning communication and actions taken. We will leverage results from a recent Warning Value Chain Assessment of Storm Eunice to reflect on the uptake and responses to different warnings.
- Open Access
- 30 June 2025
- Published by Copernicus GmbH
The influence of exposure and vulnerability on designing impact-based warning thresholds: Results of a New Zealand risk modelling experiment
Impact-based warnings are changing the way warnings are designed and communicated. Impact-based thresholds have been proposed to require the inclusion of exposure and vulnerability data for a more ‘risk-based’ approach. However, the influence of this inclusion on assigning warning levels has not been tested, and recent research has identified a concern that exposure may exacerbate the ‘urban-rural bias’. This means that rural areas may be under-warned due to having fewer people exposed, and therefore a ‘lower risk’ compared to an urban area with more people exposed to the same hazard. Thus, how warning levels change based on exposure levels needs to be explored.Our research sought to investigate the difference in the number of people who would receive the highest level ‘red’ weather warning when it was triggered by:the hazard (wind speed thresholds), exposure to the hazard (buildings and people exposed to categories of wind severity), and the overall risk (building damage caused by the wind), in both relative and absolute terms. To do this, we used a high-resolution meteorological model of ex-tropical Cyclone Cook that was a real event and was hypothetically shifted onto Auckland city in New Zealand. Within the RiskScape risk modelling software we were able to calculate the expected impacts to buildings in this scenario. We then experimented with various warning thresholds based on hazard, exposure, and risk. We will describe the methods used in this research and give an overview of the results, which show that there are differences in the number of people warned when different methods are appliedThis research demonstrates how incorporating exposure of buildings and people and risk in the design of a warning system influences the number of people who would receive warnings, in comparison to a traditional hazard-based warning. Our findings will help to inform decisions around the urban-rural bias and reinforces the need to consider who the key audiences are for impact-based warnings. This research has implications on designing impact-based warning systems globally.
- 30 June 2025
- Published by Copernicus GmbH
Open AccessThe influence of exposure and vulnerability on designing impact-based warning thresholds: Results of a New Zealand risk modelling experiment
Impact-based warnings are changing the way warnings are designed and communicated. Impact-based thresholds have been proposed to require the inclusion of exposure and vulnerability data for a more ‘risk-based’ approach. However, the influence of this inclusion on assigning warning levels has not been tested, and recent research has identified a concern that exposure may exacerbate the ‘urban-rural bias’. This means that rural areas may be under-warned due to having fewer people exposed, and therefore a ‘lower risk’ compared to an urban area with more people exposed to the same hazard. Thus, how warning levels change based on exposure levels needs to be explored.Our research sought to investigate the difference in the number of people who would receive the highest level ‘red’ weather warning when it was triggered by:the hazard (wind speed thresholds), exposure to the hazard (buildings and people exposed to categories of wind severity), and the overall risk (building damage caused by the wind), in both relative and absolute terms. To do this, we used a high-resolution meteorological model of ex-tropical Cyclone Cook that was a real event and was hypothetically shifted onto Auckland city in New Zealand. Within the RiskScape risk modelling software we were able to calculate the expected impacts to buildings in this scenario. We then experimented with various warning thresholds based on hazard, exposure, and risk. We will describe the methods used in this research and give an overview of the results, which show that there are differences in the number of people warned when different methods are appliedThis research demonstrates how incorporating exposure of buildings and people and risk in the design of a warning system influences the number of people who would receive warnings, in comparison to a traditional hazard-based warning. Our findings will help to inform decisions around the urban-rural bias and reinforces the need to consider who the key audiences are for impact-based warnings. This research has implications on designing impact-based warning systems globally.
- Open Access
- 18 March 2025
- Published by Copernicus GmbH
Investigating the relationship between compound cold events and impacts on the Scottish rail sector.
Winter weather events can result in costly damages and severe disruption to affected regions. While compound events research has strongly focused on heat-related events, less focus has been placed on extreme cold hydrometeorological hazards. Cold events impact a range of sectors from energy and agriculture to transport and health. The rail sector is particularly sensitive to cold weather hazards resulting in service delays and cancellations. Snowfall can lead to blocked tracks, points failures and issues with electricity supply. Loss of traction, braking issues and frozen infrastructure can arise from ice formation. The impacts of these cold events can be amplified by the compounding effect of another meteorological variable, such as whether heavy precipitation is present or not, with subsequent impacts dependent on the nature of the cold event. For example, a cold-wet event could incur heavy snowfall, whereas a cold-dry event could result in extreme low temperatures and icy conditions. In this study, we analyse the occurrence of 10,000 rail incidents in Scotland over an extended winter period of October to March for 2006-2023 to investigate the relationship between impacts and compound cold events. We use an impact dataset from Network Rail to categorise high-impact days based on two classifications: (1) days with the highest number of aggregated incidents; (2) days with the highest number of accumulated customer minutes lost. Using daily gridded observations from HadUK-Grid at a 5 km resolution we then apply a localised percentile-based methodology to determine the occurrence of cold-dry and cold-wet events on these high-impact days. Initial results show that the majority of high-impact days consisted of incidents caused by severe snow and icing. Analysis reveal that these incidents occurred under hydrometeorological conditions that can be classified as cold-dry and/or cold-wet events. These findings highlight the importance of considering co-occurring hazards rather than single hazards. The results of this study provide a useful insight into compound cold events for rail sector early warning systems, with valuable information on cold weather event hazard characterisation and their associated impacts across varying timescales.
- 18 March 2025
- Published by Copernicus GmbH
Open AccessInvestigating the relationship between compound cold events and impacts on the Scottish rail sector.
Winter weather events can result in costly damages and severe disruption to affected regions. While compound events research has strongly focused on heat-related events, less focus has been placed on extreme cold hydrometeorological hazards. Cold events impact a range of sectors from energy and agriculture to transport and health. The rail sector is particularly sensitive to cold weather hazards resulting in service delays and cancellations. Snowfall can lead to blocked tracks, points failures and issues with electricity supply. Loss of traction, braking issues and frozen infrastructure can arise from ice formation. The impacts of these cold events can be amplified by the compounding effect of another meteorological variable, such as whether heavy precipitation is present or not, with subsequent impacts dependent on the nature of the cold event. For example, a cold-wet event could incur heavy snowfall, whereas a cold-dry event could result in extreme low temperatures and icy conditions. In this study, we analyse the occurrence of 10,000 rail incidents in Scotland over an extended winter period of October to March for 2006-2023 to investigate the relationship between impacts and compound cold events. We use an impact dataset from Network Rail to categorise high-impact days based on two classifications: (1) days with the highest number of aggregated incidents; (2) days with the highest number of accumulated customer minutes lost. Using daily gridded observations from HadUK-Grid at a 5 km resolution we then apply a localised percentile-based methodology to determine the occurrence of cold-dry and cold-wet events on these high-impact days. Initial results show that the majority of high-impact days consisted of incidents caused by severe snow and icing. Analysis reveal that these incidents occurred under hydrometeorological conditions that can be classified as cold-dry and/or cold-wet events. These findings highlight the importance of considering co-occurring hazards rather than single hazards. The results of this study provide a useful insight into compound cold events for rail sector early warning systems, with valuable information on cold weather event hazard characterisation and their associated impacts across varying timescales.
- Open Access
- 5 July 2024
- Published by Copernicus GmbH
Value added by hazard impact modelling within the warning value chain for three impactful surface water flooding events in 2022
The warning value chain is a concept whereby the process of producing a warning is represented by a chain of sources of expertise (components), connected by bridges that convey bidirectional information exchanges. An associated warning value chain questionnaire – produced by the WMO’s High Impact Weather (HIWeather) Warning Value Chain Flagship Project – has been used to carry out a warning value chain assessment for three surface water flooding (SWF) events in the UK. These events occurred during the summer and autumn of 2022 and involved localised convection leading to predominantly minor SWF impacts. Forecast confidence was very low for all events resulting in some short-lead time and responsive warnings. This talk will briefly describe these events before focusing on the value added to the warnings by the Surface Water Flooding Hazard Impact Model (SWFHIM). SWFHIM is an operational forecasting tool used by the Flood Forecasting Centre (a partnership between the Met Office and Environment Agency) to objectively assess the likelihood of SWF impacts across England and Wales, which ultimately feeds into the decision-making process for issuing warnings. The SWFHIM adds value to the weather and hazard outputs by deriving probabilities of impacts using a range of post-processing and forecast presentation methods. However, results from the warning value chain questionnaire helped to identify improvements that may enhance future warnings. For example, SWFHIM forecast presentation changes may improve the pull-through and interpretation of information. Also, recent enhancements to the configuration of the driving convective-scale ensemble are not yet being fully utilized by the hazard-impact modelling. These points and more will be discussed in this talk.
- 5 July 2024
- Published by Copernicus GmbH
Open AccessValue added by hazard impact modelling within the warning value chain for three impactful surface water flooding events in 2022
The warning value chain is a concept whereby the process of producing a warning is represented by a chain of sources of expertise (components), connected by bridges that convey bidirectional information exchanges. An associated warning value chain questionnaire – produced by the WMO’s High Impact Weather (HIWeather) Warning Value Chain Flagship Project – has been used to carry out a warning value chain assessment for three surface water flooding (SWF) events in the UK. These events occurred during the summer and autumn of 2022 and involved localised convection leading to predominantly minor SWF impacts. Forecast confidence was very low for all events resulting in some short-lead time and responsive warnings. This talk will briefly describe these events before focusing on the value added to the warnings by the Surface Water Flooding Hazard Impact Model (SWFHIM). SWFHIM is an operational forecasting tool used by the Flood Forecasting Centre (a partnership between the Met Office and Environment Agency) to objectively assess the likelihood of SWF impacts across England and Wales, which ultimately feeds into the decision-making process for issuing warnings. The SWFHIM adds value to the weather and hazard outputs by deriving probabilities of impacts using a range of post-processing and forecast presentation methods. However, results from the warning value chain questionnaire helped to identify improvements that may enhance future warnings. For example, SWFHIM forecast presentation changes may improve the pull-through and interpretation of information. Also, recent enhancements to the configuration of the driving convective-scale ensemble are not yet being fully utilized by the hazard-impact modelling. These points and more will be discussed in this talk.
- Open Access
- 1 May 2024
- Published by Wiley in Meteorological Applications
- Vol. 31 (3)
Mitigating against the between‐ensemble‐member precipitation bias in a lagged sub‐seasonal ensemble
The Met Office GloSea5‐GC2 sub‐seasonal‐to‐seasonal 40‐member lagged ensemble consists of members who are up to 10 days different in age such that the between‐ensemble‐member bias is not internally consistent. Reforecasts tend to be used to convert these ensemble forecasts into anomalies from a normal state. These anomalies are however not that useful for applications where individual ensemble members are needed to drive downstream applications in the hazard and impact space. Here we explore whether there is a way of correcting for the within‐ensemble bias without using reforecasts. An investigation into the individual daily precipitation distributions from the JJAS 2019 Indian monsoon season, stratified by forecast horizon, highlights how the distribution changes, and shows that the model distribution is markedly different to the observed. Initial results suggest that it could be better to use recent model forecast distribution(s) as the reference for adjusting the model rainfall accumulations as a function of lead day horizon, that is, not attempting to correct the members to a vastly different (observed) distribution shape, but a more subtle shift towards the model's best guess of reality, rather than reality itself, to remove the between‐ensemble‐member bias. A combination of Exponential and Generalized Pareto distributions are used for parametric quantile mapping to remove this internal ensemble bias using computationally efficient pre‐computed lookup tables. Within‐ and out‐of‐sample results for the 2019 and 2020 monsoon seasons show that the method is effective in tightening precipitation gradients, with improvements in spread, accuracy and skill, especially for low accumulations.
- 1 May 2024
- Published by Wiley in Meteorological Applications
- Vol. 31 (3)
Open AccessMitigating against the between‐ensemble‐member precipitation bias in a lagged sub‐seasonal ensemble
The Met Office GloSea5‐GC2 sub‐seasonal‐to‐seasonal 40‐member lagged ensemble consists of members who are up to 10 days different in age such that the between‐ensemble‐member bias is not internally consistent. Reforecasts tend to be used to convert these ensemble forecasts into anomalies from a normal state. These anomalies are however not that useful for applications where individual ensemble members are needed to drive downstream applications in the hazard and impact space. Here we explore whether there is a way of correcting for the within‐ensemble bias without using reforecasts. An investigation into the individual daily precipitation distributions from the JJAS 2019 Indian monsoon season, stratified by forecast horizon, highlights how the distribution changes, and shows that the model distribution is markedly different to the observed. Initial results suggest that it could be better to use recent model forecast distribution(s) as the reference for adjusting the model rainfall accumulations as a function of lead day horizon, that is, not attempting to correct the members to a vastly different (observed) distribution shape, but a more subtle shift towards the model's best guess of reality, rather than reality itself, to remove the between‐ensemble‐member bias. A combination of Exponential and Generalized Pareto distributions are used for parametric quantile mapping to remove this internal ensemble bias using computationally efficient pre‐computed lookup tables. Within‐ and out‐of‐sample results for the 2019 and 2020 monsoon seasons show that the method is effective in tightening precipitation gradients, with improvements in spread, accuracy and skill, especially for low accumulations.
- Open Access
- 13 February 2024
- Published by Wiley in International Journal of Climatology
- Vol. 44 (5) , 1513-1529
Generating weather pattern definitions over South Africa suitable for future use in impact‐orientated medium‐range forecasting
This work aims to define a set of representative weather patterns for South Africa that can be utilized to support impact‐based forecasting of heatwave events. Sets of weather patterns have been generated using k‐means clustering on daily ERA5 reanalysis data between 1979 and 2020. Different pattern sets were generated by varying the clustering atmospheric variable, the spatial domain and the number of weather patterns. These weather patterns are evaluated using the explained variation score to assess their ability to represent the variability of the maximum daily 2m temperature (Tmax,2m). The results indicate that a set of 30 weather patterns generated using mean sea‐level pressure, with a clustering domain in the range 15°–34°E and 21°–36°S, provides a reasonable representation of Tmax,2m variability across South Africa. The implementation of an appropriate weather pattern set into a medium‐range forecasting tool has the potential to extend the prediction of high‐impact weather events in South Africa, such as heatwaves, and also highlight specific impacts on the population, for example, food and water insecurity, heat exhaustion or energy and transport impacts.
- 13 February 2024
- Published by Wiley in International Journal of Climatology
- Vol. 44 (5) , 1513-1529
Open AccessGenerating weather pattern definitions over South Africa suitable for future use in impact‐orientated medium‐range forecasting
This work aims to define a set of representative weather patterns for South Africa that can be utilized to support impact‐based forecasting of heatwave events. Sets of weather patterns have been generated using k‐means clustering on daily ERA5 reanalysis data between 1979 and 2020. Different pattern sets were generated by varying the clustering atmospheric variable, the spatial domain and the number of weather patterns. These weather patterns are evaluated using the explained variation score to assess their ability to represent the variability of the maximum daily 2m temperature (Tmax,2m). The results indicate that a set of 30 weather patterns generated using mean sea‐level pressure, with a clustering domain in the range 15°–34°E and 21°–36°S, provides a reasonable representation of Tmax,2m variability across South Africa. The implementation of an appropriate weather pattern set into a medium‐range forecasting tool has the potential to extend the prediction of high‐impact weather events in South Africa, such as heatwaves, and also highlight specific impacts on the population, for example, food and water insecurity, heat exhaustion or energy and transport impacts.
- Open Access
- 1 January 2024
- Published by Wiley in Meteorological Applications
- Vol. 31 (1)
A seamless blended multi‐model ensemble approach to probabilistic medium‐range weather pattern forecasts over the UK
This paper describes a new seamless blended multi‐model ensemble configuration of an existing probabilistic medium‐ to extended‐range weather pattern forecasting tool (called Decider) run operationally at the Met Office. In its initial configuration, the tool calculated and presented probabilistic weather pattern forecast information for five individual ensemble forecasting systems, which varied in terms of their number of ensemble members, horizontal resolution, update frequencies and forecast lead time. This resulted in multiple forecasts for the same validity time which varied in terms of forecast skill depending on the lead time in question. This presented challenges for end‐users (e.g., operational meteorologists) in terms of knowing which forecast output is best to use and at which lead time, as well as knowing what to do in situations where forecasts varied between ensembles. To get around these challenges, a new seamless blended multi‐model ensemble configuration has been implemented operationally, comprising of output from five separate ensembles, and provides a single best forecast from day one out to day 45. Objective verification for a set of eight weather pattern groups covering forecasts initialized over a 6‐year period (2017–2022) shows that the seamless blended multi‐model ensemble forecasts are at least as good as, if not better than the best performing individual model.
- 1 January 2024
- Published by Wiley in Meteorological Applications
- Vol. 31 (1)
Open AccessA seamless blended multi‐model ensemble approach to probabilistic medium‐range weather pattern forecasts over the UK
This paper describes a new seamless blended multi‐model ensemble configuration of an existing probabilistic medium‐ to extended‐range weather pattern forecasting tool (called Decider) run operationally at the Met Office. In its initial configuration, the tool calculated and presented probabilistic weather pattern forecast information for five individual ensemble forecasting systems, which varied in terms of their number of ensemble members, horizontal resolution, update frequencies and forecast lead time. This resulted in multiple forecasts for the same validity time which varied in terms of forecast skill depending on the lead time in question. This presented challenges for end‐users (e.g., operational meteorologists) in terms of knowing which forecast output is best to use and at which lead time, as well as knowing what to do in situations where forecasts varied between ensembles. To get around these challenges, a new seamless blended multi‐model ensemble configuration has been implemented operationally, comprising of output from five separate ensembles, and provides a single best forecast from day one out to day 45. Objective verification for a set of eight weather pattern groups covering forecasts initialized over a 6‐year period (2017–2022) shows that the seamless blended multi‐model ensemble forecasts are at least as good as, if not better than the best performing individual model.