A5070920900- Marine and coastal ecosystems
- Scientific Computing and Data Management
- Atmospheric and Environmental Gas Dynamics
- Methane Hydrates and Related Phenomena
- Oceanographic and Atmospheric Processes
- Computational Physics and Python Applications
- Hydrocarbon exploration and reservoir analysis
- Marine Biology and Ecology Research
- Anomaly Detection Techniques and Applications
- Geological Studies and Exploration
- Geochemistry and Geologic Mapping
- Climate variability and models
- Bayesian Modeling and Causal Inference
- Ocean Acidification Effects and Responses
- Geological formations and processes
- Model Reduction and Neural Networks
- Generative Adversarial Networks and Image Synthesis
- Geology and Paleoclimatology Research
- Geological and Geophysical Studies
National Oceanography Centre
2022-2025
University of Oxford
2021-2022
Computer simulations are invaluable tools for scientific discovery. However, accurate often slow to execute, which limits their applicability extensive parameter exploration, large-scale data analysis, and uncertainty quantification. A promising route accelerate by building fast emulators with machine learning requires large training datasets, can be prohibitively expensive obtain simulations. Here we present a method based on neural architecture search build even limited number of data. The...
Abstract The North Atlantic subpolar gyre (SPG) is a key region for the bloom (NAB), phytoplankton foundation of regional food web. NAB depends on nutrients seasonally introduced into surface ocean by deep winter convection. Under climate change, this pattern threatened increasing water column stratification, representing potential “tipping point” in Earth system, and may “collapse” as result. We investigate changes mixing impacts SPG northern using spread future projections from...
Abstract The ocean biological carbon pump (BCP) transports organic matter from the surface to deep ocean. Accurately quantifying efficiency of BCP is essential for understanding potential climate feedbacks and entails measuring flux material in out mesopelagic layer (approximately 100–1,000 m). Observational estimates are often restricted over short timescales. Here we use an biogeochemical model diagnose where, on what timescales, sufficiently steady state that balancing budget may be...
Abstract. The skill of global ocean biogeochemical models, and the earth system models in which they are embedded, can be improved by systematic calibration parameter values against observations. However, such tuning is seldom undertaken as these computationally very expensive. Here we investigate performance DFO-LS, a local, derivative-free optimisation algorithm has been designed for expensive with irregular model–data misfit landscapes typical models. We use DFO-LS to calibrate six...
The Southern Ocean (SO) is a critical component of the global carbon cycle, acting as significant sink for atmospheric dioxide (CO2). Understanding intricate processes governing CO2 uptake in SO paramount comprehending budget and predicting future climate scenarios. Recent observations suggest that changes water masses, driven by climate-induced alterations temperature circulation patterns, can significantly impact uptake. these feedbacks crucial SO's role its broader implications mitigation...
The North Atlantic subpolar gyre is a key region for the phytoplankton bloom (NAB), foundation of regional foodweb. NAB dependent on nutrients seasonally introduced into surface ocean by deep winter convection. Under climate change, this pattern threatened increasing water column stratification, and may “collapse” as result, representing potential “tipping point” in Earth system. We investigate change mixing impacts SPG broader northern using 1. spread future projections from low-resolution...
The North Atlantic subpolar gyre is a key region for the phytoplankton bloom (NAB), foundation of regional foodweb. NAB dependent on nutrients seasonally introduced into surface ocean by deep winter convection. Under climate change, this pattern threatened increasing water column stratification, and may “collapse” as result, representing potential “tipping point” in Earth system. We investigate change mixing impacts SPG broader northern using 1. spread future projections from low-resolution...
Abstract Scarcity of iron and manganese limits the efficiency biological carbon pump over large areas Southern Ocean. The importance hydrothermal vents as a source these micronutrients to euphotic zone Ocean is debated. Here we present full depth profiles dissolved total dissolvable trace metals in remote eastern Pacific sector (55–60° S, 89.1° W), providing evidence enrichment at depths 2000–4000 m. These enhanced micronutrient concentrations were co-located with 3 He enrichment, an...
Abstract The performance of global ocean biogeochemical models can be quantified as the misfit between modeled tracer distributions and observations, which is sought to minimized during parameter optimization. These are computationally expensive due long spin‐up time required reach equilibrium, therefore optimization often laborious. To reduce computational time, we investigate whether a model with shorter spin‐ups provides same optimized parameters one full‐length, equilibrated over several...
Abstract. Biogeochemical model behaviour for micronutrients is typically hard to constrain because of the sparsity observational data, difficulty determining parameters in situ, and uncertainties observations models. Here, we assess influence data distribution, uncertainty, misfit function on objective parameter optimisation a oceanic cycle zinc (Zn), an essential micronutrient marine phytoplankton with long whole-ocean residence time. We aim investigate whether constraints are sufficient...
Abstract. The performance of global ocean biogeochemical models, and the Earth System Models in which they are embedded, can be improved by systematic calibration parameter values against observations. However, such tuning is seldom undertaken as these models computationally very expensive. Here we investigate DFO-LS, a local, derivative-free optimisation algorithm has been designed for expensive with irregular model-data misfit landscapes typical models. We use DFO-LS to calibrate six...
Abstract. Biogeochemical model behaviour for micronutrients is typically hard to constrain because of the sparsity observational data, difficulty determining parameters in situ, and uncertainties observations models. Here, we assess influence data distribution, uncertainty misfit function on objective parameter optimisation a oceanic cycle zinc (Zn), an essential micronutrient marine phytoplankton with long whole-ocean residence time. We aim investigate whether constraints are sufficient...
The uptake of carbon by the Southern Ocean plays a critical role in mitigating atmospheric CO2 increases, but its magnitude and temporal spatial variability are still subject to large uncertainties due scarcity observations, model disagreements. Southeast Pacific is one such region where deep, nutrient carbon-rich circumpolar deep waters upwell, also Subantarctic Mode Water Antarctic Intermediate formed subducted, carrying with them high loadings anthropogenic dioxide into ocean interior....