IWRI - Postdoctoral Position in Hybrid Machine Learning Physics Modeling

About UM6P:

Mohammed VI Polytechnic University is an institution dedicated to research and innovation in Africa and aims to position itself among world-renowned universities in its fields

The University is engaged in economic and human development and puts research and innovation at the forefront of African development. A mechanism that enables it to consolidate Morocco's frontline position in these fields, in a unique partnership-based approach and boosting skills training relevant for the future of Africa.

Located in the municipality of Benguerir, in the very heart of the Green City, Mohammed VI Polytechnic University aspires to leave its mark nationally, continentally, and globally.

About IWRI:

The International Water Research Institute (IWRI) provides research, education, and innovation in the fields of water and climate. In order to face the next challenges. IWRI seeks to rethink and adapt research development, innovation, and training to new paradigms. It is an institute that outlines forward-thinking pathways to address water issues in a systemic manner in Africa.

IWRI's vision is to Lead and develop integrated research and processes for Water & Climate concerns in Africa. We act as an African water hub through strategic cooperation and partnerships. We share comprehensive and diverse expertise to provide scientific information by focusing on four pillars: Integrated water resources management, Climate change and adaptation, Hydroinformatics, and Water technologies.

Position Summary:

The International Water Research Institute (IWRI) of the College of Agriculture and Environmental Sciences (CAES) at UM6P invites applications for a postdoctoral researcher position in hybrid machine learning-physics based modeling of hydroclimatic extremes in data scarce regions.

The postdoctoral researcher will develop innovative hybrid machine learning-physics based modeling approaches to improve the understanding and prediction of hydroclimatic extremes in data-scarce regions, with a primary focus on Morocco. The position aims to advance methods that integrate data-driven algorithms with physically based hydrological and climate models under conditions of limited observations, while ensuring that the resulting tools are both scientifically robust and operationally relevant. A key component of the role is the translation of model outputs into actionable information for stakeholders, particularly water resource managers and planning authorities, and the evaluation of long-term water management strategies aimed at mitigating water scarcity. The project also includes assessing the transferability of the developed approaches to other regions facing similar constraints.

Scientific context:

Hydroclimatic extremes such as droughts and floods are intensifying under climate variability and change, particularly in regions where observational data are sparse and monitoring systems are limited. In such context, traditional physically based models are often constrained by insufficient data, while purely data-driven approaches may lack physical consistency and generalizability. Hybrid modeling frameworks that combine machine learning with process-based understanding offer a promising pathway to overcome these limitations. Morocco provides a compelling case study, given its pronounced hydroclimatic variability, increasing water stress, and ongoing investments in large-scale water management strategies. However, significant challenges remain in reliably predicting extreme events and assessing long-term adaptation options under deep uncertainty and data scarcity.

Research Scope:

The project will focus on the development of a new generation of hybrid modeling approaches that couple advanced machine learning methods with hydrological and/or climate models. Emphasis will be placed on designing approaches that are robust to data scarcity through the integration of heterogeneous data sources, including remote sensing products, reanalysis datasets, and sparse in-situ observations, as well as techniques such as transfer learning and uncertainty quantification. The research will address the detection, attribution and prediction of hydroclimatic extremes under non-stationary conditions and evalaute model performance over Morocco as a primary testbed. A central components of the work will invovle translating model outputs into decision-relevant indicators and assessin the feasibility and robustness of long-term water maangemnt strategies currently under consideration. The transferability of the developed methods to other regions will also be systematically investigated.

Candidate Profile:

• PhD in hydrology, climate science, geosciences, applied mathematics, computer science or related field
• Expertise in mahcine learning, physically based modeling or hybrid approaches combining boths
• Strong programming skills (Python, MATLAB, Fortran, or similar) and familiarity with ML frameworks.
• Experience with hydroclimatic data, including remote sensing products, reanalysis datasets and/or in-situ observations Knowledge of uncertainty quantification, data assimilation, or transfer learning is a strong advantage.
• Proven ability to handle HPC environements is an added asset
• Demonstrated ability to develop and evaluate models under data-scarce conditions.
• Ability to communicate results to both scientific and non-technical stakeholders, including water managers and policy actors.