Global hydrology models (GHMs) play a crucial role in understanding threats to water security. Their global remit means they can be used to assess transboundary impacts from water security crises, e.g. how flooding in East Asia might affect UK supply chains. They also help with quantifying cascading climate change risks, such as how a drought can lead to reductions in crop yields, which in turn lead to rises in food prices, and migration. The UK faces multiple environmental, economic and political threats from around the world. Water is one of these primary risks because it is associated with flooding, droughts and long-term water security, and can affect large land areas and multiple sectors of the economy, from days to years at a time.
Despite the importance of understanding the implications of global hydrological change for the UK and its security, there are only two GHMs in the UK’s environmental modelling national capability. One of these is JULES (the Joint UK Land Environment Simulator), which is a community land surface model that is used both as a standalone model and as the land surface component in the Met Office global climate model. However, whilst JULES includes complex representation of physical land surface environmental processes and feedbacks, its representation of direct human forcings on the hydrological cycle, such as dams, reservoirs, and water abstraction for different water users (e.g. agriculture, industry and domestic) is still being developed. The other model is MacPDM, which was originally developed at UKCEH and has undergone several updates over the past few decades, including several by this project’s supervisor (Gosling & Arnell, 2011; Müller Schmied & Gosling et al. 2024). However, just like JULES, MacPDM does not currently represent direct human forcings. Other national modelling infrastructure capabilities include these important aspects of the global hydrological system, such as WaterGAP (Germany) and H08 (Japan).
This project will provide an important update to the UK’s ability to simulate global hydrological risks and their impacts by using AI methods with global observation datasets to develop, test and optimise the parameterisation of direct human forcings in the MacPDM global-scale hydrological model.
Figure 1. The project will add the missing processes (in grey) to the MacPDM global hydrological model. From Müller Schmied & Gosling et al. (2025).
This project is a CENTA Flagship Project.
This project does not offer a CASE studentship
Each host has a slightly different application process.
Find out how to apply for this studentship.
All applications must include the CENTA application form.
Choose your application route
Model development: AI parameterisation learning methods (e.g. using Claude Code) will develop a new, consensus-based conceptual parametrisation scheme (not black box) informed by laws of hydrology and direct human forcing parametrisations from a large ensemble of GHMs, including WaterGAP and H08.
Optimisation and validation: The new parametrisations will be optimised iteratively for dams and reservoirs, then agriculture, industry and domestic water abstractions. Optimisation involves comparing the new simulations to satellite-derived reservoir storage products like Global Reservoir Storage (GRS), land-based estimates of streamflow (e.g. Global Runoff Data Centre, GRDC) and terrestrial water storage (e.g. GRACE). The parametrisations will be compared bilaterally with those being developed for JULES, in collaboration with the Met Office and UKCEH.
Application: Run the new model with CMIP7 climate projections to assess implications of water security threats to the UK, e.g. identifying future drought hotspots in world regions where the UK relies significantly on supply chains.
DRs will be awarded CENTA Training Credits (CTCs) for participation in CENTA-provided and ‘free choice’ external training. One CTC can be earned per 3 hours training, and DRs must accrue 100 CTCs across the three and a half years of their PhD.
Supervisors will provide training in handling of global environmental datasets, including accessing and downloading data from online repositories, processing using programming languages like Python, Matlab and R, as well as AI tools for research like Claude, and data visualisation. Scientific writing and communication training will be delivered through journal/report writing support, and presentation coaching when preparing for conferences and workshops. The DR will be welcomed into the EU COST action CROSS-CASCADE (co-led by Gosling), opening up a range of international training opportunities on working with climate and socio-economic data, and training targeted at early career researchers.
Partner 1: Dr Andrew Hartley represents the UK Met Office as an External Partner. Dr Hartley has expertise in global-scale hydrological modelling and is Climate Impacts Modelling Team Lead at the UK Met Office, which includes the JULES model. We will apply for CASE funding in 2027 if the project is successful, but Dr Hartley will supervise the project in any case.
Partner 2: Dr Helen Baron represents UKCEH as a CENTA3 Research Centre Partner. Dr Baron is a Hydrological Modeller and has expertise in global-scale modelling, particularly on model development and model representation of human drivers of hydrology.
Year 1: Begin reviewing literature on AI learning methods in environmental modelling. Use AI methods to learn direct human forcing parametrisations from an ensemble of ISIMIP global hydrological models. Update the MacPDM model to include new, consensus-based parametrisations to create a version 1 of the model. Attend an ISIMIP research workshop to present plans and meet other hydrology modellers.
Year 2: Validate and optimise the new parametrisations to create a version 2 (final) of MacPDM. First conference presentation.
Year 3: Run the version 2 MacPDM model with past and future CMIP7 climate data to assess implications of global water security threats to the UK. Start writing-up thesis. Second conference presentation.
Gosling, S.N., Arnell, N.W. (2011) ‘Simulating current global river runoff with a global hydrological model: model revisions, validation and sensitivity analysis’, Hydrological Processes 25, pp. 1129-1145. doi: 10.1002/hyp.7727
Müller Schmied, H., Gosling, S.N. et al. (2024) ‘Graphical representation of global water models’, Geoscientific Model Development, 18, pp. 2409-2425. doi:10.5194/gmd-18-2409-2025
Otta, K., Müller Schmied, H., Gosling, S.N., Hanasaki, N. (2025) ‘Towards the use of satellite remote sensing to validate reservoir storage in global hydrological models: methodology and pilot study in the CONUS’. Environmental Research: Water 1, pp. 015002. doi:10.1088/3033-4942/adc47b
Schewe, J., Gosling, S.N. et al. (2019) ‘State-of-the-art global models underestimate impacts from climate extremes’, Nature Communications 10, pp. 1005. doi:10.1038/s41467-019-08745-6
Zaherpour, J., Gosling, S.N., et al. (2018) ‘Worldwide evaluation of mean and extreme runoff from six global-scale hydrological models that account for human impacts’, Environmental Research Letters 13, pp. 065015 doi: 10.1088/1748-9326/aac547
For any enquiries related to this project please contact Prof Simon Gosling ([email protected]).
To apply to this project:
Applications must be submitted by 23:59 GMT on Wednesday 6th January 2027.