SKSarah Kew

Sarah Kew

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Sarah Kew has a total of 146 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Attributing extreme humid heat in South and Southeast Asia in April 2023 to climate change and its implications for vulnerable and disadvantaged communities

    by Mariam Zachariah, Robert Vautard, T. Arulalan, Clemens Schwingshackl, S.T. Chaithra, Joyce J. Kimutai, Clair Barnes, Krishna AchutaRao, Roop Singh, Maja Vahlberg, et al.

  • Open Access

    Automated trigger procedure for operational attribution

    by Sarah Kew, Izidine Pinto, Sjoukje Philip

    The need and interest in operational detection and attribution services is growing. To increase the operational efficiency of relevant and rapid attribution studies, an efficient method for the first step in the process - triggering and selection of extreme events for potential study, is required. It can also be beneficial to ensure that the triggering process is as objective as possible in order to reduce risks of selection bias and ensure any operational service is fully traceable regarding decisions made when initiating a study.The “Heat wave magnitude index daily (HWMId)” of Russo et al. 2015 is adapted for the purpose of triggering heat, cold and wet extremes on a daily basis. Using a clustering algorithm, events are automatically ranked per type across a 2-week period, covering recent and upcoming days, according to an index accounting for their intensity, duration and area. This index, together with related metrics such as DEI, EFI and RX3day, is displayed regionally and globally in the “Hazard Monitoring Tool” - a user interface for operational use.The outcome of the method’s application to heat extremes will be demonstrated and compared to HWMId for well known extreme events and other events of interest. The adaptation of the method for cold and wet extremes will also be presented, along with several other diagnostic tools that can aid the selection of the most suitable extreme events for a rapid attribution study. As this is work in progress, we look forward to discussions on the challenges as well as the promising results of this method.

  • Open Access

    Copernicus Operational Extreme Event Monitoring and Attribution Service

    by Mark McCarthy, Sjoukje Philip, Sarah Kew, Freja Vamborg, Thessa Beck, Zorica Jones

    There are ever increasing demands by society for timely climate attribution information following extreme events, and for inclusion of extreme event attribution within operational climate services. The science supporting routine climate attribution assessments has also progressed substantially in recent years, and there is a landscape of a growing number of providers of such climate intelligence. The Copernicus Climate Change Service (C3S) aims to establish an operational framework for extreme event attribution with a global scope and a European focus. In this presentation the key outcomes from the development phase of this activity will be presented and discussed. The service is building on established methodologies and previous European collaborations, combining scientific, operational and communication expertise from across a number of institutions, to produce a flexible and portable operational extreme event service based on multiple attribution methods and mature science. In this presentation we will present and discuss key outputs and lessons learned during the development phase of this as an operational climate attribution service.The key outputs include:Interactive Web Application – A tool for exploring trends in extreme events and their links to climate change through maps, time series, and statistical analyses.Extreme Event Attribution Office – A distributed office across four European meteorological services, in collaboration with ECMWF, delivering rapid climate attribution assessments.Rapid Attribution Studies – Production of factsheets within five business days of an event, with extended analyses for longer-term events.Counter-Factual Weather Data – Development of alternative climate scenarios using advanced dynamical models.Communication Products – User documentation, visualisation guidance, training materials, and peer-reviewed scientific reports.

  • Open Access

    How Climate and Land Cover Change Shaped Europe's Record Breaking 2025 Fire Season

    by Theodore Keeping, Mariam Zachariah, Clair Barnes, Olivia Haas, Emmanouil Grillakis, Izidine Pinto, Ben Clarke, Joyce Kimutai, Sjoukje Philip, Sarah Kew, et al.

    The 2025 European fire season was characterised by record-breaking burned area and multiple extreme wildfire events across the continent. Increasing wildfire extremes in Europe and globally has intensified interest in how climate and land-use changes are altering European fire regimes. Extreme event attribution of antecedent and concurrent fire weather conditions is used here to assess the changing likelihood and intensity of recent events both relative to preindustrial conditions and under projected climate change.We analyse five regions that experienced extreme wildfire activity in 2025: northwestern Iberia, upland Britain, southwestern Mediterranean France, the eastern Adriatic/Ionian, and northern and western Türkiye. For each region, we attribute changes in the likelihood of short-term fire-weather extremes around peak wildfire activity, using 7-day maxima of vapour pressure deficit (VPD), surface wind speed, and a composite index of VPD and wind, in addition to spring and summer effective precipitation to characterise seasonal drought and fuel accumulation conditions. In addition to weather-related drivers, we assess trends in spring and summer vegetation cover using the leaf area index (LAI) and in land abandonment or reclamation using the changing fraction of managed and unmanaged land.Climate change strongly increased vapour pressure deficit and the composite VPD/wind index for all southern European regions, with the likelihood of drought conditions at least as strong as 2025 also increasing by over a factor of three relative to in the preindustrial climate. Short-term fire weather or summer drought exhibited a weak positive and negative trend with warming respectively, though an increasing likelihood of spring drought conditions, a key driver of 2025’s wildfires, was identified. Spring vegetation significantly increased across Europe, implying higher fuel loads and a potential for more intense wildfires. Land management trends were mixed, with long-term land abandonment in southwestern France and northern and western Türkiye and a recent, rapid land abandonment signal in the eastern Adriatic/Ionian.The record-breaking 2025 European fire season occurred in the context of a climate change driven intensification of the fire weather extremes and drought conditions associated with each of the five wildfire events examined. Combined with increasing growth-season vegetation cover and ongoing land abandonment, these factors suggest increases in European wildfire extremes will continue.

  • Open Access

    Automated trigger procedure for operational attribution

    by Sarah Kew, Izidine Pinto, Sjoukje Philip

    The need and interest in operational detection and attribution services is growing. To increase the operational efficiency of relevant and rapid attribution studies, an efficient method for the first step in the process - triggering and selection of extreme events for potential study, is required.The “Heat wave magnitude index daily (HWMId)” of Russo et al. 2015 is adapted for the purpose of triggering heat, cold and wet extremes on a daily basis, where events are automatically ranked across a 2-week period according to an index accounting for their intensity, duration and area. The method applied to heat extremes will be presented here and results will be compared to HWMId for well known extreme events and other events of interest. As this is work in progress, we look forward to discussions on the challenges as well as the promising results of this method.

  • Open Access

    Multi-method extreme event attribution: Motivation, case study, and implications

    by Shirin Ermis, Vikki Thompson, Marylou Athanase, Lynn Zhou, Ben Clarke, Hylke de Vries, Geert Lenderink, Pandora Hope, Sarah Kew, Sarah Sparrow, et al.

    Since 2004, many methods for event attribution have been developed. Early studies showed that attribution statements are sensitive to the framing of research questions but few large comparisons have been undertaken.Here, we firstly motivate the need for multi-method extreme event attribution, highlighting conceptual differences between methods. In a second part, we present a case study of midlatitude storm Babet (2023) to compare three common storyline attribution methods, alongside a severity-based probabilistic method. We discuss three widely relevant questions which highlight the complementarity and the differences between methods: (1) How has climate change impacted the frequency of the event? (2) How has climate change impacted the event severity? (3) Were the dynamics of the event influenced by climate change and if yes, how?We show that methods differ in the extent to which they reproduce observed weather patterns. This influences attribution statements, and can even change the sign of results for events with uncertain climate signals. We argue that limitations and strengths of methods need to be clearly communicated when presenting event attribution reports to ensure findings can be used reliably by a wide range of stakeholders.

  • Open Access

    Development of the Climate Explorer Circulation Analogues Tool 

    by Vikki Thompson, Izidine Pinto, Sarah Kew, Sjouke Philip

    When an extreme weather event occurs scientists may be asked if similar events have occurred in the past. If the frequency of such events have changed. If such events are becoming more intense or more persistent. And if the impacts of such events are increasing or decreasing. We present a tool that uses atmospheric circulation analogues to provide a framework to answer such questions. When using our analogues tool there are methodological choices that need to be considered. Extreme weather events are, by their very definition, rare, so how can we assess if the analogues are close enough to the observed event to be useful? Analogues can be calculated using spatial correlation or Euclidean distance, from sea level pressure or 500 hPa geopotential height, over different domains, and for different timescales. Investigating how sensitive results are to these choices allows us to provide a set of rules for using our tool for a range of different types of climate extremes, from heatwaves to extreme rainfall. Through a series of case studies, we consider the methodological choices required when assessing analogues, and assess which events are most suited to analogues methods.

  • Open Access

    A comparison of storyline attribution methods for a midlatitude cyclone

    by Shirin Ermis, Vikki Thompson, Nicholas Leach, Hylke de Vries, Geert Lenderink, Lynn Zhou, Pandora Hope, Ben Clarke, Sarah Kew, Sarah Sparrow, et al.

    Since 2004, methods for event attribution have been developed across many groups. Early studies showed that answers to attribution questions are sensitively dependent on the framing of the study used but recently developed methods for storyline attribution have not been compared in detail.Here, we compare three common methods for storyline attribution, alongside the probabilistic method, based on the midlatitude cyclone Babet. This storm caused flooding in the UK and Ireland in October 2023. The three storyline methods are flow analogues, pseudo-global warming, and forecast-based attribution. We discuss four questions that might be asked of attribution studies by the public: (1) Has climate change impacted the event? (2) How has climate change impacted the frequency of the event? (3) How has climate change impacted the event severity? (4) Were the dynamics of the event influenced by climate change and if yes, how?We argue that storyline methods are better suited to answer questions about severity changes in events but that probabilistic methods are needed to determine changes in frequency of the event. Limitations and opportunities of the methods need to be clearly communicated to the public when publishing event attribution studies.Finally, we compare the framing of storyline attribution to that of probabilistic attribution. To the best of our knowledge, this comparison of methods is the first study discussing the differences in the framing and the quantitative results of storyline attribution methods. We hope it represents a basis for future systematic comparisons.

  • Open Access

    Autumn and Winter storms over UK and Ireland about 20% wetter due to human-induced Climate Change

    by Sarah Kew, Mark McCarthy, Emily Wallace, Jennifer Pirret, Oliver Claydon, Fraser Lott, Ciara Ryan, Barry Coonan, James Pope, Ellie Murtagh, et al.

    The 2023/2024 storm season for the UK and Ireland was exceptionally wet as a whole, as well as ‘hosting’ a notably large number of named storms, some of which led to devastating flooding, with cascading impacts on human health, food production and the cost of living.Here we present a attribution study for the Oct-Mar season, quantifying the role of human-induced climate change on the frequency and intensity of strong winds and heavy rainfall on storm days as well as total precipitation for the season as a whole. The storm severity index (SSI) is used to identify stormy days and as an indication of wind intensity. We use probabilistic attribution methods following the world weather attribution protocol, synthesising trends in observations with climate models and communicating uncertainties. The rainfall associated with storms was found to have become about 20% more intense and that the 2023/24 level has become about a factor of 10 more likely. Climate change was also found to have a strong influence on Oct-Mar precipitation totals, in line with expectations. The influence of climate change on storm winds, was less clear however, with average wind (SSI) on stormy days being found to have decreased slightly. Possible reasons for this will be discussed in the light of relevant literature. We also highlight the importance of vulnerability and exposure information in combination with attribution outcomes to provide recommendations for reduced impacts.


  • Open Access

    Attribution using analogues: a case study of the Western European flood event of July 2021 

    by Vikki Thompson, Dim Coumou, Sjoujke Philip, Sarah Kew, Izidine Pinto

    In July 2021 a cut-off low pressure system brought extreme precipitation to Western Europe. Record daily rainfall totals led to flooding that caused loss of life and substantial damage to infrastructure in Germany, Belgium, and the Netherlands. By identifying flow analogues – dynamically similar events - in both reanalysis data and large ensembles of climate model simulations we can investigate how the dynamics involved in this event are changing through time. Analogue methods are increasingly used in event attribution, we highlight considerations that must be made when using such methods. For July 2021, we show that similar low pressure systems are occurring more frequently, and the lows are deepening through time. We find some analogues persist for much longer than was seen in July 2021. These dynamical changes effect surface impacts of such events. Such unprecedented events will become increasingly likely in a warming climate, and society must adapt to reduce future impacts.