Mariana joined ECMWF in 2022 as part of the Destination Earth Project. Her background is in Mathematics and Statistics and her work at ECMWF focusses on using machine learning techniques to improve the post-processing of ensemble forecasts.
- Statistical post-processing
- Probabilistic forecasting
- Assessing uncertainty using machine learning methods
Ph.D., Imperial College London. London, UK (2022). Thesis on “Advanced numerical and statistical techniques to assess erosion and flood risk in coastal zones”. PhD supervisors: Matthew Piggott and Colin Cotter.
MRes in Mathematics of Planet Earth, Imperial College London. London, UK (2018). Final grade: Distinction. Thesis on "Modelling sediment in and around offshore wind farms" (supervised by Matthew Piggott and Colin Cotter).
MMath in Mathematics, University of Oxford. Oxford, UK (2012-2016). Final grade: First. Specialization in numerical methods and fluid mechanics. Thesis on "The fluid mechanics of natural ventilation" (supervised by Ian Hewitt).
Professional Experience
- Since 07/2022: Scientist, Forecast Department, ECMWF
- 12/2021–06/2022: PostDoc, Department of Earth Science and Engineering, Imperial College London, UK
- 11/2021–12/2021: Visiting Research Fellow, LOCEAN, Institut Pierre-Simon Laplace (IPSL), Paris, France
- 10/2020–12/2020: Visiting Research Assistant, UK Meteorological Office, Exeter, UK
- 09/2016–09/2017: Forensic Data Analyst, PwC, London, UK
Visiting Research Assistant at Imperial College London
- 2025
- Zied Ben Bouallegue, Mihai Alexe, Matthew Chantry, Mariana Clare, Jesper Dramsch, Simon Lang, Christian Lessig, Linus Magnusson, Ana Prieto Nemesio, Florian Pinault, Baudouin Raoult, Steffen Tietsche (January 2025) AIFS – ECMWF’s Data-Driven Probabilistic Forecasting . DOI: 10.5194/egusphere-egu24-17158
- Antonio Pérez, Mario Santa Cruz, Javier Diez-Sierra, Matthew Chantry, András Horányi, Mariana Clare, Cornel Soci (January 2025) Testing the use of deep learning techniques for emulating regional reanalysis. DOI: 10.5194/egusphere-egu24-3540
- Giangiacomo Navarra, Curtis Anthony Deutsch, Charlotte Merchant, Mariana C. A. Clare, Maike Sonnewald (January 2025) Predicting Southern Ocean Dissolved Oxygen: Bayesian vs. Deterministic Approach to Forecasting. DOI: 10.22541/essoar.173655319.94315054/v1
- Giangiacomo Navarra, Curtis Anthony Deutsch, Charlotte Merchant, Mariana C. A. Clare, Maike Sonnewald (January 2025) A Bayesian Neural Network approach to study Dissolved Oxygen in Southern Ocean water masses . DOI: 10.22541/essoar.173655319.94315054/v2
- Maike Sonnewald, William Yik, Mariana CA Clare, Redouane Lguensat (January 2025) Discovering Dominant Controls on Southern Ocean Dynamics Under Climate Change: New Knowledge Through Physics-Guided Machine Learning . DOI: 10.5194/egusphere-egu24-21905
- 2024
- Mariana C. A. Clare, Simon C. Warder, Robert Neal, B. Bhaskaran, Matthew D. Piggott (February 2024) An Unsupervised Learning Approach for Predicting Wind Farm Power and Downstream Wakes Using Weather Patterns, Journal of Advances in Modeling Earth Systems. DOI: 10.1029/2023MS003947
- Fatima Pillosu, Mariana Clare, Thomas Haiden, Florian Pappenberger, Christel Prudhomme, Hannah Cloke (August 2024) Can we now predict flash floods globally and up to medium-ranges?. DOI: 10.5194/ems2024-1064
- (October 2024) Data-driven ensemble forecasting with the AIFS, ECMWF Newsletter n. 181, pp. 32-37. DOI: 10.21957/ma3p95hxe2
- 2023
- Sokratis J. Anagnostopoulos, Jens Bauer, Mariana C.A. Clare, Matthew D. Piggott (December 2023) Accelerated wind farm yaw and layout optimisation with multi-fidelity deep transfer learning wake models, Renewable Energy. DOI: 10.1016/j.renene.2023.119293
- Mariana C A Clare, Simon C Warder, Robert Neal, B Bhaskaran, Matthew Piggott (April 2023) An unsupervised learning approach for predicting wind farm power and downstream wakes using weather patterns. DOI: 10.22541/essoar.168121529.93941319/v1
- Mariana Clare, Zied Ben Bouallegue, Matthew Chantry, Martin Leutbecher, Thomas Haiden (May 2023) Combining Bayesian Neural Networks with explainable AI techniques for trustworthy probabilistic post-processing. DOI: 10.5194/egusphere-egu23-946
- Mariana Clare, Thomas Haiden (July 2023) Creating skillful and reliable probabilistic forecasts using machine learning. DOI: 10.5194/ems2023-199
- 2022
- Mariana C.A. Clare, Matthew D. Piggott, Colin J. Cotter (June 2022) Assessing erosion and flood risk in the coastal zone through the application of multilevel Monte Carlo methods, Coastal Engineering, pp. 104118. DOI: 10.1016/j.coastaleng.2022.104118
- Mariana C.A. Clare, Stephan C. Kramer, Colin J. Cotter, Matthew D. Piggott (June 2022) Calibration, inversion and sensitivity analysis for hydro-morphodynamic models through the application of adjoint methods, Computers & Geosciences, pp. 105104. DOI: 10.1016/j.cageo.2022.105104
- Mariana C. A. Clare, Joseph G. Wallwork, Stephan C. Kramer, Hilary Weller, Colin J. Cotter, Matthew D. Piggott (December 2022) Multi-scale hydro-morphodynamic modelling using mesh movement methods, GEM - International Journal on Geomathematics. DOI: 10.1007/s13137-021-00191-1
- Mariana C. A. Clare, Maike Sonnewald, Redouane Lguensat, Julie Deshayes, V. Balaji (November 2022) Explainable Artificial Intelligence for Bayesian Neural Networks: Toward Trustworthy Predictions of Ocean Dynamics, Journal of Advances in Modeling Earth Systems. DOI: 10.1029/2022MS003162
- Mariana Clare, Tim Leijnse, Robert McCall, Ferdinand Diermanse, Colin Cotter, Matthew Piggott (March 2022) Multilevel multifidelity Monte Carlo methods for assessing coastal flood risk. DOI: 10.31223/X5733R
- Mariana C A Clare, Maike Sonnewald, Redouane Lguensat, Julie Deshayes, Venkatramani Balaji (May 2022) Explainable Artificial Intelligence for Bayesian Neural Networks: Towards trustworthy predictions of ocean dynamics. DOI: 10.1002/essoar.10511239.1
- Mariana C. A. Clare, Tim W. B. Leijnse, Robert T. McCall, Ferdinand L. M. Diermanse, Colin J. Cotter, Matthew D. Piggott (August 2022) Multilevel multifidelity Monte Carlo methods for assessing uncertainty in coastal flooding, Natural Hazards and Earth System Sciences. DOI: 10.5194/nhess-22-2491-2022
- Mariana C. A. Clare, Tim W. B. Leijnse, Robert T. McCall, Ferdinand L. M. Diermanse, Colin J. Cotter, Matthew D. Piggott (March 2022) Multilevel multifidelity Monte Carlo methods for assessing coastal flood risk. DOI: 10.5194/nhess-2022-74
- 2021
- Mariana C.A. Clare, Omar Jamil, Cyril J. Morcrette (October 2021) Combining distribution‐based neural networks to predict weather forecast probabilities, Quarterly Journal of the Royal Meteorological Society n. 741, pp. 4337-4357. DOI: 10.1002/qj.4180
- 2020
- Mariana Clare, James Percival, Stephan Kramer, Athanasios Angeloudis, Colin Cotter, Matthew Piggott (March 2020) Hydro-morphodynamics 2D modelling using a discontinuous Galerkin discretisation. DOI: 10.5194/egusphere-egu2020-4990
- Mariana Clare, James Percival, Athanasios Angeloudis, Colin Cotter, Matthew Piggott (January 2020) Hydro-morphodynamics 2D modelling using a discontinuous Galerkin discretisation. DOI: 10.31223/OSF.IO/TPQVY