Reflexivity on modelling for transboundary water governance

A new special issue explores the social, technical, and political dimensions of modelling in the context of international river basins. One of the editors, Rozemarijn ter Horst, interviews Yasir Mohamed, associate professor at IHE Delft and former Minister of Irrigation and Water of Sudan, who co-authored an article on models for coordinating the management of two dams in the Blue Nile basin.

Collecting, developing, and applying data and models for governing and managing transboundary waters is a challenging balancing act. It requires a consideration of two types of politics: the politics of the region and countries where the waters are located, as well as the politics of knowledge, which is part and results of  the data and data-driven tools. This balancing act calls for reflexivity by those who are commissioning, developing, and applying the tools (Milman & Gerlak, 2020). Models are therefore not a one-size-fits-all tool, but they need to be developed and used in conscious ways. When it comes to transboundary water governance, the interaction between states influences how data- and data-driven tools are used. Yet, there is little written on how this happens in practice.

To foster awareness of the necessity to engage mindfully with the balancing act mentioned above we invited practitioners and scientists to share their experiences and lessons learned. These reflections and analyses were recently published in a special issue on ‘Exploring the use of data and models in transboundary water governance’ in the journal Water International. The special issue contains rich case studies on the interactions between data and models and the politics of transboundary water governance, with examples from four continents.

Here, we celebrate the special issue by highlighting one of the articles that provides a great example on how to balance politics, the social, and technology. Written by Mohamed A. Hassan, Mahmoud F. Hassan, Wail A. Awad (all from the Ministry of Irrigation and Water Resources of Sudan) and tYasir A. Mohamed (IHE Delft), the article “Dam operation using satellite data and hydrological models: the case of Roseires dam and Grand Ethiopian Renaissance Dam in the Blue Nile River” offers an insight into how satellite data and models can be used for the management of transboundary reservoirs. We interviewed one of the co-authors, Dr. Yasir Mohamed, who, as Associate professor of Water Management at IHE Delft and former Minister of Irrigation and Water Resources of Sudan, has a unique and holistic insight on the use of models. On behalf of the co-authors, he reflects on the interactions between science, knowledge and politics.

Rozemarijn: As opposed to many models being offered by foreign consultants, the modelling process you describe in the article is very much homegrown. What is the difference that this makes, in your opinion?

Yasir Mohamed:  It makes a huge difference. First of all, if you bring whatever model from outside, people are always skeptical about it. The models have assumptions. It’s based on physics and equations, but also parameters. I mean people from outside develop a model and they tell nice results. But it is very hard to sell it. So, when a model is homegrown – even if it is a model of less quality – it’s much better than models coming from outside, because the trust and the confidence of the model is so crucial for using its outputs. Sometimes you make a nice model, but it has no roots, then it’s very hard to find a buy in for this model.

Rozemarijn: Did you have a big team working on developing the model that you use?

Yasir Mohamed: It’s not really a big team, but it is interdisciplinary. We have one person working on the hydrology and rainfall runoff model. We have one person working on the remote sensing analysis. And we have one person working on the construction management, who looks into information and photos and then tries to extract the dimensions of the dam [the Grand Ethiopian Renaissance Dam, editor’s note]. How high will the dam be? So those are the three main persons. Maybe you can assume there are other three assistants to them. So, this is the technical team. They present their results to a larger committee for feedback.

Besides them there is also the negotiating team. They bring the political dimension. Those are the people who are really responsible for the negotiations, and they could foresee the political environment in the region and how things will go. As part of the negotiation team there are people from the Defence Department, and from the international water law side. So it’s really a multidisciplinary team. it is also mixed in terms of age: we have people in their twenties, and people of almost 70 who have the institutional memory of the issue itself and the negotiations.

Rozemarijn: How did this come to be? Often a modelling process remains a very technical thing until output is provided, without being contextualized or embedded. But you have automatically then done this.

Yasir Mohamed: There was a need to do it like that. At the Ministry, we needed the information on the water levels in the dam, and then modelling was our only option, in the absence of data sharing.

Rozemarijn: Why did you want to write about this experience? Those kind of articles on how modelling processes are done do not appear often.

Yasir Mohamed: One of the reasons to do this is to document the experience, because the knowledge might be needed in the future. This was the key reason, as the information may become scattered here and there.  And those young people who were part of the team wanted someone to help them with writing this. I mean, they know they have done the work, they have done the computation. But to put it together in a nice story, they needed someone with that particular experience. And there are many good examples of such stories, and I know some of my colleagues, they even have done much better modelling work than we did, but never published.

Rozemarijn: In the editorial we list four lessons based on the case studies from around the world (Table 1). Which one speaks to you, or what do you think is missing?

Yasir Mohamed: For me, it is Engage people first, and model second and Take time for transdisciplinarity. You need to start building a modelling process from the ground. Asking “Do we need a model or not? What kind of model? What kind of input data and how we build it? Who will build it? Who will process the outputs and who will analyse the results?” This kind of things. So I think this one I would rank it number one.

There is an experience I have that supports this, in the Tekeze-Atbara Basin shared between Sudan and Ethiopia. In the past we tried different models at the national level in Sudan, and then bring it to Ethiopia, or vice versa. It was hard to convince decision makers on both sides. So this time we said, okay, we build a joint model by a joint team. The modellers from Sudan went to Ethiopia for three months to build the model. And the modellers from Ethiopia came to Sudan and they lived there for three months. At the end of the day they had developed one joint model. Since it was built by one joint team from both sides, then it was easy to sell the results to both sides.

Rozemarijn: You put it very nicely. It’s really giving the model roots. Thank you for these additional insights, that show how modelling is social, political and technical!

The article, as well as the other case studies that provide insights on how the social, political and technical side of data and models play out in the context of transboundary water governance, are currently available fully open access in the Special Issue on ‘Exploring the use of data and models in transboundary water governance’.

In relation to reflexive modelling, we invite you as well to read and comment on the blog post: Models do not think – The Water Dissensus, in the Forum of the journal Water Alternatives. Here, IHE Delft researcher Jonatan Godinez Madrigal, PhD candidate Bích Trần, affiliate researcher Rozemarijn ter Horst and Humboldt University Researcher Rossella Alba write “Models are useful for computing complex processes that are difficult for humans to calculate. They can help explore the past, present, and future and consolidate hypotheses about the world. However, using them as thinking tools can be challenging because practitioners often confuse them with reality when, in fact, they are only simplifications of reality” .

---

---