Q&A: Two Junior Researchers Help to Lead Research on Optimizing the Costs and Benefits of Hydropower in Africa

A June 4th study in Nature Sustainability showed how to reduce the environmental impacts and maintain high levels of power generation for proposed hydropower projects across all of Africa. The contributions of two junior researchers, Anna Clark and Angelo Carlino, were integral to this work. We talked with them about what they learned from their experience as co-authors of the paper.

Hydropower, often perceived as a cheap, reliable, and low-carbon option, has attracted the interest of governments and investors worldwide. This is particularly salient in Africa, where hydropower accounts for 17% of electricity generation. Dams are often a necessary component of hydropower, but they have the potential to damage ecosystems. They can fragment rivers, diminishing biodiversity. They may also shift floodplains, inundating previously dry land and catalyzing the release of methane – a potent greenhouse gas which contributes to climate change. In addition, locating dams on smaller rivers reduces biodiversity impacts, but also lowers energy production and cost–effectiveness. 

The goal of the study was to support energy system planning in a way that is both beneficial to humans and less detrimental to nature: selecting portfolios of dams to meet energy demand while minimizing costs, river fragmentation, and greenhouse gas emissions from hydropower. Few hydropower analyses have worked at the continental scale to balance these multiple goals – an analytical technique known as multi-objective optimization. Carlino and Clark worked together at every stage, with the support of Stanford Natural Capital Project Lead Scientist Rafael Schmitt and Andrea Castelletti, a professor at Politecnico di Milano in Italy. 

This interview has been edited for clarity and brevity.

A CONVERSATION WITH ANNA CLARK AND ANGELO CARLINO 

Q: Tell me about your academic and professional background, and how you came to this project?

Clark: I came to the project as a senior at Stanford. I was majoring in Engineering Physics in the Renewable Energy track, so I had a more first-principle, physics-based understanding of electricity and renewable energy systems. There is also engineering in the degree, so it had a fair amount of computation, but I was definitely lacking experience with some of the computer-science-heavy tasks. This project truly helped me grow as a student and researcher: I now have such a better understanding of hydropower – specifically its challenges, drawbacks, and benefits as a renewable energy resource. But moreover, I learned how to organize a research project as big as this— how to split it up into the different stages, as well as how to organize the directories and files we needed to be running. By the time the project had finished I was completing my “co-term”— a fifth year I took to complete my Master’s degree—in Management Science and Engineering, in the Energy and Environment track.

Carlino: I started on this project a few months after finishing my PhD in Milano. My PhD work focused on energy systems, hydropower, and optimization, with some work on climate policy as well. My background was more in environmental system management and planning and multi-objective optimization, all of which were a part of this project. Specifically, I was trying to bring the multi-objective component into the energy system planning field. I was a visiting postdoc at NatCap for ten months, which is when Rafael and I started with the initial exploration of African hydropower. 

Q: What was novel about this work?

Carlino: Usually energy system planning looks at cost alone, leaving out other considerations that could be examined via multi-objective optimization. It’s hard to do multi-objective optimization when the objectives— things like methane emissions reduction— are functions of other decisions that impact multiple sectors. For example, hydropower contributes to both the water and energy sectors. In this case, it’s crucial to balance the benefits and drawbacks of many variables. The method we adopted is not the only way to do this, but it is well-suited to the specifics of the problem we wanted to research.  

Clark: That is one of the novelties of this work: it’s a new method of balancing those objectives (power production, cost, methane emissions, river fragmentation)  and embedding everything in a framework for power and energy systems planning. Such an integration across economic, environmental, and energy domains is novel. 

Q:  Tell me about your role and each other’s roles in this project. What were each of your key responsibilities? 

Carlino: As Rafael and I started narrowing the scope of the project, we onboarded Anna and she took on more and more. I had supervised Master’s thesis students before but this collaboration with Anna was definitely closer. She feels more like a colleague than someone I had to supervise. 

Clark: The biggest thing I was contributing to was some of the GIS (geographic information system) analysis methods and some of the data-science aspects. I prepped a lot of the data, such as those representing the river network, by sorting and simplifying it, as well as correcting inconsistencies and errors. As the project progressed, I was also involved in some data visualizations. 

Carlino: There was a lot of data involved with this project and Anna kept it highly organized. She helped a lot with formulating the final data package and cleaning up the code. She also checked that we were making consistent assumptions and selecting the river flow accurately. When you have a river network, you have both creeks and big rivers. When specifying hydropower project locations, it is important to make sure you’re actually picking the big river when you want to pick the big river, and the smaller one when you want to pick the smaller one. Anna was integral in ensuring the river network was mapped correctly. But one of her most important contributions was shaping the research itself, simply by asking questions and trying to understand. Rafael and I— we have worked in multi-objective optimization longer than Anna, and at times we were not fully understanding each other because we were talking in our own terminology. By having Anna there asking questions, it forced us to flesh out our ideas more fully, to explain in simple terms what we meant. We ended up understanding each other better and also helping her understand. By prompting these discussions, Anna also helped illuminate discrepancies between what I wanted to do and what Rafael wanted to do, so we were then able to resolve these in a way that made sense for all three of us. And then you also presented at the conference, Anna!

Q: Tell me about that! 

Clark: This conference (C3E Women in Clean Energy Symposium) was for women researchers in clean energy, with a competition component: several finalists from across the country were selected to go to MIT and present their posters for the final round of this poster competition, and I was lucky enough to be one of them!  I had presented a poster at a much larger conference prior to this, but it was entirely virtual. This experience was far more interactive and felt like my first “real” experience. It was amazing to get some advance publicity and interest for our research at this conference. An added bonus was that some of the judges were Stanford professors! It felt very full-circle to share our paper with the wider world while simultaneously having some Stanford and NatCap connections there at the conference. 

Q: What did you learn from this research and publishing experience? 

Clark: I knew the challenge science communication posed was significant, but one big challenge I learned about was the difficulty of communicating all these concepts in a paper, and then making sure that the paper is satisfactorily clear to everyone. We want a reader with a reasonable amount of background knowledge to be able to understand why we did what we did and why we couldn’t have done something else that may have been computationally easier, for example. The added complexity is that there are multiple objectives we are focusing on, and the energy system we were considering spans several decades. There were so many iterations of the paper, abstract, and poster in order to get all of them to the point where they were digestible, but also very clear and accurate. There was so much time invested in communicating these concepts to ensure the paper was suitable for publication.

Carlino: Among the things that Anna contributed to, science communication was a big one. Neither Rafa nor I are native English speakers, so having Anna— who knew what we did very well— was so helpful in translating the work into more understandable terms. I also personally learned how to delegate and trust my team. I think so many of us have the tendency to rush through meetings, trying to get to the point really quickly. But so often, the implications of some prior steps or ideas haven’t been fully considered. So a big takeaway for me was to take the time to go through the meeting and discuss: What was the idea? Why are you doing this? At every step, remind yourself of the rationale behind the whole project.