Adaptive delta management (ADM) deals with making flexible and robust plans in the uncertain environment of deltas. ADM receives increasing attention to be applied globally, including in Argentina. In this research, I have studied how modelling can assist ADM in Argentina through action research by applying various participatory methods like workshops, group model building and interviews. In the exploratory phase, I could make several observations regarding the application of ADM in Argentina. I discovered something unexpected: Multiple co-existing ontologies of the Paraná Delta were shaped in practice, influencing the ADM process. The Paraná Delta seemed to be continually changing per actor, time and space in the Delta. In order to introduce the reader to this multiplicity, I offer an approach to formulate these ontologies and provide several examples of ontologies in the Delta. Next, I propose a structured approach to connect ontologies to modelling. The added value of using ontologies for the modelling of the Paraná Delta was discussed with experts. The results indicate that by connecting ontologies and modelling, the ADM process (in the Paraná Delta) will be strengthened, since ontologies (1) illustrate modellers assumptions, (2) improve stakeholders and politicians understanding of the Delta, (3) guide research and data collection to verify assumptions on the Delta, and (4) strengthen the understanding of one's own ontology. Therefore, the results show that incorporating ontologies in ADM research is a valuable approach. Future research should focus on formulating models to support the claims made in this article. Further, a theoretical investigation is needed to incorporate the use of ontologies in the ADM process. In order to guide future researchers to incorporate ontologies in their modelling, a framework is constructed based on my research experience: ‘the Kite framework’.

In this thesis, the usage of scenarios for the participatory application of ADM is investigated in the context of the Paraná Delta, in Argentina. The focus is two-fold: (1) the research focuses on how a participatory design for ADM can be made, and (2) it focuses on which moment of the framework scenarios should be introduced in the participatory ADM

Conceptual models in hydrology are widely used, allow for easy interpretation and require little data. Machine learning models in hydrology often outperform conceptual models but lack the ease of interpretability, require large amounts of data and and do not obey physical laws. Hybrid approaches aiming to combine the advantages of both approaches are becoming more popular. A Neural Ordinary Differential Equations approach is introduced to combine a differential equation-based conceptual model with a neural network. Additionally, conceptual models and LSTM models are used as benchmarks. The models are tested using the LamaH-CE dataset as well as the E-OBS dataset. In many cases the hybrid models outperform the conceptual model. However, to further improve the performance of hybrid models more research is needed to make the models more computationally efficient and optimized training strategies are required to explore the full potential of the approach.