Post-Graduate Student Showcase 2025
LOCATION
Augustine United Church
41 George IV Bridge
Edinburgh
EH1 1EL
This meeting will take place as a hybrid meeting so both in person or virtual attendance is available.
Each talk will last for 30 minutes. The first speaker will be Toby Jones followed by Lucy Seabourne and her talk will start soon after 6:30 pm. There will be questions afterwards. The venue is the Augustine Church, 41 George IV Bridge, Edinburgh EH1 1EL. Tea and biscuits will be served from 5:30 pm. Those viewing the lectures remotely can join from 5:50 pm using Zoom and the event will start promptly at 6 pm. Would all those attending, whether in-person or remotely, register for the event on www.rmets.org. Please note that joining instructions will not be provided unless you have registered.
Speakers
Toby Jones
Correlation of Wind and Precipitation AnnualAggregate Severity of European Cyclones
The yearly cost of wind and flooding from European cyclones frequently reaches billions of euros. This study has investigated the correlation between wind gust and precipitation annual aggregate severity arising from extra-tropical cyclones over the Europe-Atlantic region using ERA5 reanalysis from 1980-2020. Simple annual aggregate severity indices have been constructed by aggregating exceedances above chosen damage thresholds of wind gust speed maxima and precipitation totals for all storms in a year. At low thresholds, there is a strong positive correlation between wind and precipitation aggregate severity, likely induced by the common dependence on the total number of storms. However, at higher thresholds, where the severity indices are expected to be a better reflection of wind and flood losses, negative correlations start to appear especially over western Europe. A correlation of -0.22 is observed between aggregate severity indices over France at thresholds of 20 m/s and 20mm. Furthermore, regions that experience positive correlation at these thresholds exhibit negative correlation values for sufficiently high thresholds. This suggests that aggregate wind and flood losses in Europe should not be assumed to be either independent or positively correlated, and that there is a potential for risk diversification. A framework has been developed to better understand the negative correlation occurring at high thresholds. Only the mean and variance of the severities and counts of storms are used to estimate the correlation between aggregate severities. A reduction in error occurs once the framework is modified to reflect how atmospheric conditions can lead to stormier or calmer years. The overall trend of the correlation being more positive over the ocean and closer to zero (and negative) over land is well captured by the framework.
Lucy Seabourne
Automated Weather Impact Data Collection
Impact-based Forecasts and Warnings aim to help stakeholders better understand the potential consequences of hazardous weather events, rather than simply forecasting the weather itself. This increased awareness improves the actionability of warnings produced and highlights the importance of collecting weather impact data to improve situational awareness and verify impact-based warnings. However, collecting this data manually is a laborious task and inevitably susceptible to temporal bias with fewer impact reports being recorded overnight and at weekends. For this reason, it is advantageous to collate an automated weather impact database. At the Met Office our automated workflow involves scraping websites that reliably report global socio-economic weather impact data such as deaths, injuries and buildings damaged. Where websites report impacts from events that are not always relevant, a machine learning model is used to filter out those irrelevant reports. Tabulated statistics are extracted directly, and natural language processing is applied to extract numerical information from free text descriptions. This data is then stored in our impact database and visualized on a map displaying weather impacts globally over the past seven days. We are currently working on producing a similar weather impact database focusing specifically on UK impacts. This database will allow us to identify affected areas on a more detailed scale, thereby enhancing the verification of UK Impact-based Forecasts and Warnings. This talk will provide an overview of our automated impact data collection methodology and the steps we are taking to adapt this approach for collecting UK-specific impacts.
Registration
REGISTRATION IS NOW CLOSED
Registration for this event is closed.
If you have any queries with regards to this event or require any further information please contact us at meetings@rmets.org.
We take data privacy seriously. Please read the RMetS privacy policy to find out more.
The Royal Meteorological Society has a number of local centres across the UK, where meetings are held throughout the year. The Scotland Local Centre are currently hosting monthly meetings which can be attended in person or virtually. For further information on the local centre, upcoming events or to be added to the mailing list, please contact scotland@rmets.org.
This meeting will take place as a hybrid meeting so both in person or virtual attendance is available.
Each talk will last for 30 minutes. The first speaker will be Toby Jones followed by Lucy Seabourne and her talk will start soon after 6:30 pm. There will be questions afterwards. The venue is the Augustine Church, 41 George IV Bridge, Edinburgh EH1 1EL. Tea and biscuits will be served from 5:30 pm. Those viewing the lectures remotely can join from 5:50 pm using Zoom and the event will start promptly at 6 pm. Would all those attending, whether in-person or remotely, register for the event on www.rmets.org. Please note that joining instructions will not be provided unless you have registered.
Speakers
Toby Jones
Correlation of Wind and Precipitation AnnualAggregate Severity of European Cyclones
The yearly cost of wind and flooding from European cyclones frequently reaches billions of euros. This study has investigated the correlation between wind gust and precipitation annual aggregate severity arising from extra-tropical cyclones over the Europe-Atlantic region using ERA5 reanalysis from 1980-2020. Simple annual aggregate severity indices have been constructed by aggregating exceedances above chosen damage thresholds of wind gust speed maxima and precipitation totals for all storms in a year. At low thresholds, there is a strong positive correlation between wind and precipitation aggregate severity, likely induced by the common dependence on the total number of storms. However, at higher thresholds, where the severity indices are expected to be a better reflection of wind and flood losses, negative correlations start to appear especially over western Europe. A correlation of -0.22 is observed between aggregate severity indices over France at thresholds of 20 m/s and 20mm. Furthermore, regions that experience positive correlation at these thresholds exhibit negative correlation values for sufficiently high thresholds. This suggests that aggregate wind and flood losses in Europe should not be assumed to be either independent or positively correlated, and that there is a potential for risk diversification. A framework has been developed to better understand the negative correlation occurring at high thresholds. Only the mean and variance of the severities and counts of storms are used to estimate the correlation between aggregate severities. A reduction in error occurs once the framework is modified to reflect how atmospheric conditions can lead to stormier or calmer years. The overall trend of the correlation being more positive over the ocean and closer to zero (and negative) over land is well captured by the framework.
Lucy Seabourne
Automated Weather Impact Data Collection
Impact-based Forecasts and Warnings aim to help stakeholders better understand the potential consequences of hazardous weather events, rather than simply forecasting the weather itself. This increased awareness improves the actionability of warnings produced and highlights the importance of collecting weather impact data to improve situational awareness and verify impact-based warnings. However, collecting this data manually is a laborious task and inevitably susceptible to temporal bias with fewer impact reports being recorded overnight and at weekends. For this reason, it is advantageous to collate an automated weather impact database. At the Met Office our automated workflow involves scraping websites that reliably report global socio-economic weather impact data such as deaths, injuries and buildings damaged. Where websites report impacts from events that are not always relevant, a machine learning model is used to filter out those irrelevant reports. Tabulated statistics are extracted directly, and natural language processing is applied to extract numerical information from free text descriptions. This data is then stored in our impact database and visualized on a map displaying weather impacts globally over the past seven days. We are currently working on producing a similar weather impact database focusing specifically on UK impacts. This database will allow us to identify affected areas on a more detailed scale, thereby enhancing the verification of UK Impact-based Forecasts and Warnings. This talk will provide an overview of our automated impact data collection methodology and the steps we are taking to adapt this approach for collecting UK-specific impacts.
Registration
REGISTRATION IS NOW CLOSED
Registration for this event is closed.
If you have any queries with regards to this event or require any further information please contact us at meetings@rmets.org.
We take data privacy seriously. Please read the RMetS privacy policy to find out more.
The Royal Meteorological Society has a number of local centres across the UK, where meetings are held throughout the year. The Scotland Local Centre are currently hosting monthly meetings which can be attended in person or virtually. For further information on the local centre, upcoming events or to be added to the mailing list, please contact scotland@rmets.org.
