skip to content

The Godwin Laboratory for Palaeoclimate Research

 

Wed 28 May 13:00: Title to be confirmed

http://talks.cam.ac.uk/show/rss/15125 - Mon, 12/05/2025 - 16:28
Title to be confirmed

Abstract not available

Add to your calendar or Include in your list

Wed 28 May 13:00: Title to be confirmed

Departmental Talks - Mon, 12/05/2025 - 16:28
Title to be confirmed

Abstract not available

Add to your calendar or Include in your list

Wed 14 May 14:00: Explainability can foster trust in artificial intelligence in geoscience and disaster risk management

http://talks.cam.ac.uk/show/rss/15125 - Thu, 08/05/2025 - 14:40
Explainability can foster trust in artificial intelligence in geoscience and disaster risk management

Artificial Intelligence (AI) is transforming many fields, including geosciences and disaster risk management, by offering powerful tools for analysing complex systems and supporting critical decision-making process. However, as the complexity and potentially the predictive skill of an AI model increases, its interpretability — the ability to understand the model and its predictions from a physical perspective — may decrease. In critical situations, such as scenarios caused by natural hazards, the resulting lack of understanding of how a model works and consequent lack of trust in its results can become a barrier to its implementation.

This talk focuses on the emerging field of Explainable AI (XAI), which enhances the human-comprehensible understanding and interpretation of opaque ‘black-box’ AI models, can build trust in AI model results, and encourage greater adoption of AI methods in geoscience and disaster risk management. Drawing on recent research, it highlights how XAI can enhance the adoption of AI in this field and outlines key challenges, current trends, and promising directions for future integration of XAI into geoscience and disaster risk management applications.

Add to your calendar or Include in your list

Wed 14 May 14:00: Explainability can foster trust in artificial intelligence in geoscience and disaster risk management

Departmental Talks - Thu, 08/05/2025 - 14:40
Explainability can foster trust in artificial intelligence in geoscience and disaster risk management

Artificial Intelligence (AI) is transforming many fields, including geosciences and disaster risk management, by offering powerful tools for analysing complex systems and supporting critical decision-making process. However, as the complexity and potentially the predictive skill of an AI model increases, its interpretability — the ability to understand the model and its predictions from a physical perspective — may decrease. In critical situations, such as scenarios caused by natural hazards, the resulting lack of understanding of how a model works and consequent lack of trust in its results can become a barrier to its implementation.

This talk focuses on the emerging field of Explainable AI (XAI), which enhances the human-comprehensible understanding and interpretation of opaque ‘black-box’ AI models, can build trust in AI model results, and encourage greater adoption of AI methods in geoscience and disaster risk management. Drawing on recent research, it highlights how XAI can enhance the adoption of AI in this field and outlines key challenges, current trends, and promising directions for future integration of XAI into geoscience and disaster risk management applications.

Add to your calendar or Include in your list

Wed 14 May 14:00: Title to be confirmed

Departmental Talks - Wed, 07/05/2025 - 17:20
Title to be confirmed

Abstract coming soon

Add to your calendar or Include in your list

Mon 12 May 13:00: Title to be confirmed

http://talks.cam.ac.uk/show/rss/15125 - Wed, 07/05/2025 - 12:48
Title to be confirmed

Abstract not available

Add to your calendar or Include in your list

Mon 12 May 13:00: Title to be confirmed

Departmental Talks - Wed, 07/05/2025 - 12:48
Title to be confirmed

Abstract not available

Add to your calendar or Include in your list

Wed 28 May 14:00: Seismic Constraints on the Temperature and Composition of Earth's Interior: Some Open Questions

http://talks.cam.ac.uk/show/rss/15125 - Wed, 07/05/2025 - 12:03
Seismic Constraints on the Temperature and Composition of Earth's Interior: Some Open Questions

Despite major advances in the quantity and quality of seismic data, several first-order questions about the thermal and compositional structure of the Earth’s interior remain unresolved. This is due in part to the uncertain relationship between seismic velocities and physical properties such as temperature and composition, but more critically, to the limited resolving power of seismic observations and the lack of a clear physical meaning of seismic models.

In this talk, I will review several efforts made over the years to address two fundamental questions through an interdisciplinary approach: a) Is the Earth’s mantle well-mixed, and to what extent? b) What do we know about the bulk composition, thermal state (and thickness) of the continental lithosphere?

The first question will involve a review of seismic constraints on the spherically averaged structure of the Earth’s mantle, focusing on current insights into thermal and compositional variations with depth. Given the complexity of this topic, I will highlight select studies and findings from my past research that help illuminate these issues, while leaving open areas of ongoing debate and uncertainty.

For addressing the second question, I will focus on recent advancements in imaging the continental lithosphere, with a particular focus on the crustal portion. I will discuss techniques such as ambient noise analysis to measure the dispersion and attenuation of short-period surface waves, along with receiver function analysis, to refine our understanding of the composition and thermal state of the lithosphere. Finally, I will explore efforts to develop thermochemical models of the Earth’s crust that align with these seismic observations.

Add to your calendar or Include in your list

Wed 28 May 14:00: Seismic Constraints on the Temperature and Composition of Earth's Interior: Some Open Questions

Departmental Talks - Wed, 07/05/2025 - 12:03
Seismic Constraints on the Temperature and Composition of Earth's Interior: Some Open Questions

Despite major advances in the quantity and quality of seismic data, several first-order questions about the thermal and compositional structure of the Earth’s interior remain unresolved. This is due in part to the uncertain relationship between seismic velocities and physical properties such as temperature and composition, but more critically, to the limited resolving power of seismic observations and the lack of a clear physical meaning of seismic models.

In this talk, I will review several efforts made over the years to address two fundamental questions through an interdisciplinary approach: a) Is the Earth’s mantle well-mixed, and to what extent? b) What do we know about the bulk composition, thermal state (and thickness) of the continental lithosphere?

The first question will involve a review of seismic constraints on the spherically averaged structure of the Earth’s mantle, focusing on current insights into thermal and compositional variations with depth. Given the complexity of this topic, I will highlight select studies and findings from my past research that help illuminate these issues, while leaving open areas of ongoing debate and uncertainty.

For addressing the second question, I will focus on recent advancements in imaging the continental lithosphere, with a particular focus on the crustal portion. I will discuss techniques such as ambient noise analysis to measure the dispersion and attenuation of short-period surface waves, along with receiver function analysis, to refine our understanding of the composition and thermal state of the lithosphere. Finally, I will explore efforts to develop thermochemical models of the Earth’s crust that align with these seismic observations.

Add to your calendar or Include in your list

Thu 15 May 11:30: Vertical mixing and associated biogeochemical fluxes via radium isotopes

http://talks.cam.ac.uk/show/rss/15125 - Wed, 07/05/2025 - 10:37
Vertical mixing and associated biogeochemical fluxes via radium isotopes

The coastal ocean is a highly dynamic and vital biogeochemical mediator between land and sea. Coastal waters frequently experience poor water quality derived from land-based anthropogenic pressure, which is often attributed to surface water sources, such as rivers. Subsurface sources (e.g., submarine groundwater discharge, benthic fluxes), however, often rival or exceed river contributions to coastal water and chemical budgets. Yet subsurface sources are understudied because they are challenging to quantify.

In this talk, I will first give a broad overview on quantifying subsurface flows using U-Th series geochemical tracers (radium, radon). Then I will present an ecosystem-scale study across the Baltic Sea, where we used 224Ra to quantify vertical mixing across a largely hypoxic deep water column. We collected radium and solute (e.g., dissolved silicate) bottom water profiles from 50 stations along a ~5000 km cruise track in the Baltic Sea. 224Ra-derived vertical mixing rates were on the order of 10-4 m2/s, well-within range of previous local-scale estimates based on modeling and sediment core incubations. Diffusive solute fluxes were also similar in magnitude to previous studies. Overall, this talk will highlight an innovative method for quantifying diffusive fluxes and contextualize findings in terms of broader biogeochemical significance.

Add to your calendar or Include in your list

Thu 15 May 11:30: Vertical mixing and associated biogeochemical fluxes via radium isotopes

Departmental Talks - Wed, 07/05/2025 - 10:37
Vertical mixing and associated biogeochemical fluxes via radium isotopes

The coastal ocean is a highly dynamic and vital biogeochemical mediator between land and sea. Coastal waters frequently experience poor water quality derived from land-based anthropogenic pressure, which is often attributed to surface water sources, such as rivers. Subsurface sources (e.g., submarine groundwater discharge, benthic fluxes), however, often rival or exceed river contributions to coastal water and chemical budgets. Yet subsurface sources are understudied because they are challenging to quantify.

In this talk, I will first give a broad overview on quantifying subsurface flows using U-Th series geochemical tracers (radium, radon). Then I will present an ecosystem-scale study across the Baltic Sea, where we used 224Ra to quantify vertical mixing across a largely hypoxic deep water column. We collected radium and solute (e.g., dissolved silicate) bottom water profiles from 50 stations along a ~5000 km cruise track in the Baltic Sea. 224Ra-derived vertical mixing rates were on the order of 10-4 m2/s, well-within range of previous local-scale estimates based on modeling and sediment core incubations. Diffusive solute fluxes were also similar in magnitude to previous studies. Overall, this talk will highlight an innovative method for quantifying diffusive fluxes and contextualize findings in terms of broader biogeochemical significance.

Add to your calendar or Include in your list