CreativeDrought: End of Project Reflections

Mel Rohse (University of Birmingham and Anglia Ruskin University, UK)
Anne Van Loon (University of Birmingham, UK)
Rosie Day (University of Birmingham, UK)
Sally Rangecroft (University of Birmingham, UK)
Lindsey McEwen (University of West of England, UK)
Stephen Birkinshaw (Newcastle University, UK)
Coleen Vogel (University of Witwatersrand, South Africa)
Lyla Mehta (Institute of Development Studies)
Edward Nesamvuni (University of Venda and University of Free State, South Africa)
Eugine Makaya (National University of Science and Technology, Zimbabwe)
Khathu Muthala (Department of Agriculture, Provincial Government of Limpopo, South Africa)

As this is a reflection-based post across our whole team, this is a long read. If you are in a hurry, you can jump to the sections that most interest you using those links:

Origins of the project
The project design
Team reflections
Some context
Thoughts on methods
Successes in working across disciplines

Our project, CreativeDrought: Creative experiments for building resilience to future drought in Africa, came to an end on 28 February 2018. After 14 months of planning and designing research, collaborating across disciplines and across countries, doing fieldwork, modelling and trying to make sense of lots of data, it is time to take stock and reflect on what we achieved and on the challenges encountered along the way. To do so, we invited each team member and partner to contribute a few hundred words reflecting on their experiences with the project. Due to time constraints, not everybody was able to take part but this blogpost collates the contributions of those who could into a team reflection.

Origins of the project

The project was designed as a response to an ‘announcement of opportunity’ called ‘Building Resilience’ and part of the portfolio of research funding known as the ‘Global Challenges Research Fund’ (or GCRF). Much more detail and context can be found here .

The ‘Building Resilience’ call which opened aimed to fund several short projects on “foundation-building activities to stimulate the creation of inter-disciplinary international research communities; to enable broader, deeper and more effective collaborations with beneficiaries and user organisations at the forefront of the development agenda; test new innovative ideas and inter-disciplinary approaches for addressing the issue of developing resilience.” The proposals were expected to be highly interdisciplinary and bring together researchers from environmental sciences, social sciences and the arts and humanities. Further details here.

In our proposal, we stated our research objectives as follows:

– understanding past droughts through discussions with local communities, i.e. narratives of past drought events, indigenous knowledge about water & climate;
– scenario modelling with input of existing information and iterative process with local communities;
– building new narratives of hypothetical future droughts, using creative experimentation based on local cultural expressions and model scenarios; and
– improving local cultural context in decision making by including future drought narratives into integrated water resources management.

The project was due to take place in Zimbabwe and we had written the proposal closely with an NGO working in the Limpopo River Basin. However, we were unable to secure research permits to carry out the fieldwork there, so instead we managed to create a new relationship with a South African partner and move our case study to the South African side of the Limpopo River Basin. We focused on a tributary of the Limpopo, the Nwanedi River and its catchment. Within the catchment, our case study was the village of Folovhodwe (see for example these previous blogposts on the village, water uses, or our summer activities there) .

The project design

In case you’re new to our project and haven’t followed our updates before, here’s a quick overview of the phases we followed. Please feel free to get in touch with us if you want to know more and there’ll be a lot more details in our publications.

The project consisted of four phases.

Phase 1, fieldwork: data collection (past drought narratives with different groups in the village; catchment observations; data gathering on water and drought management)
Phase 2, hydrological modelling (build hydrological model, analyse past droughts, model future scenarios based on local and global information)
Phase 3, community workshops (used future scenarios translated into storylines as starting point for discussion with various groups in the community about impact and adaptation to potential future droughts, filmed those future narratives and edited videos)
Phase 4, engagement with government officials (used videos as starting point for conversations with local and regional water governance actors).

Team reflections

Eight team members contributed to this blog; we have grouped reflections under themes. A recurrent theme is interdisciplinary working, for example on the practicalities of the methodology or communicating and exchanging with colleagues from another discipline.

Some context

We start with a quote from Prof. Lyla Mehta, who usefully contextualised CreativeDrought within work she has been doing over many years: “A lot of my work has been concerned with how the uncertainties associated with climate change, drought and scarcity episodes are understood differently by scientists, policy makers and local people.  As the project ‘Climate Change, Uncertainty and Transformation’, funded by the Research Council of Norway shows, climate change uncertainty is often considered a ‘super wicked problem’ by scientists and policymakers. Still, climate scientists tend to quantify uncertainty in models and risk assessments.  These efforts have rarely taken into account how local people – particularly those living at the margins – make sense of and cope with uncertainties. The Creative Drought project with its focus on local narratives of drought and how and whether these can be brought together with hydrological modelling is also focusing on similar issues. The project has been strong in uncovering rich local narratives about how local people understand and cope with drought in rural South Africa. It has also done very innovative work around modeling different scenarios, building in part on different local narratives.  It was very exciting to be part of such an innovative project and there was a lot of cross-learning.”

Thoughts on methods

In terms of method and bringing together various disciplines, Dr Anne Van Loon and Dr Rosie Day picked up on two aspects that we had not necessarily anticipated when planning the project: the need for compromise and how we talk about the future. Anne wrote about the place of compromise in successfully bringing different methods together: “When we started the CreativeDrought project, I hadn’t realised that by combining natural and social science methods in an interdisciplinary way, both methods could not be applied in the way they are normally used. This meant that we had to improvise and compromise. For example, future scenario modelling is normally used to underpin decision making. For that purpose, the model needs to be set up in the best way possible with accurate input data. The model would need to be calibrated and validated on observations of river discharge and groundwater levels. Also, scenarios need to be a realistic representation of the future, which often means spending a lot of effort in downscaling climate model data and socio-economic scenarios, and then running multi-model ensembles. In the CreativeDrought project, the model was not used for accurate predictions of the future situation, but as a tool in the workshops to give the discussions some numerical basis. Similarly, narrative interviews are normally done to generate qualitative data that can be analysed to provide information, for example about the way that people use water and how they are impacted by and respond to drought. For that purpose the interviews need to be transcribed accurately, translated precisely, analysed with software programmes, and interpreted. In the CreativeDrought project, however, the narrative interviews were done with the aim to get a picture of previous drought events and their impacts, about communication between different groups in the community and between generations, and about how people prepare for an uncertain future. This was then used to test if we could trust the model enough and to design the workshops.”

Still thinking about methods, Rosie considered a more specific aspect of our method around the future. Indeed, a key component of what we proposed was to explore and create stories about the future. “One thing that this project has made me think about is different ways of apprehending the future. The future is a tricky concept – it doesn’t (yet) exist in any concrete sense, it is by nature unknown, and yet, ideas of resilience and preparedness require us to inhabit it and to make it to some extent knowable or at least relatable. Initially, conversations with the villagers brought up narratives of the future being unknowable, or knowable only by God. Natural science in contrast is comfortable in describing possible futures in numerical terms, with the ability to multiply iterations and alternatives made easy by computational power. The status of these future scenarios is quite difficult to grasp, even for scientists – in the team we had quite some discussion about the nature of a prediction vs a projection. Narrative knowledge can inhabit the future in a very different way, by using the imagination to create possible lifeworlds, with actions, reactions, experiences and emotions. This fills out the spaces between the parameters described by the model and allows us to develop a very different kind of relationship with the futures. This is what we were essentially building though the narrative workshops – lifeworlds within parameter space, and inviting others to inhabit them in an exercise of mutual understanding and empathy, and joint enterprise of preparation and lesson learning.”

Professor Lindsey McEwen reflected on the similarities and differences between CreativeDrought and the UK DRY research projects: “Both considered how different forms of knowledge – specialist science and narrative – can come into the same space in exploring different realities and perceptions about past, present and future drought risk.  I found it stimulating to be involved in both projects alongside each other but in very different cultural contexts, as ideas/ learning from one sparked off within the other.  Particularly challenging but important across both cultural settings is the potential to layer drought/water stories and science as new stimuli for place-based thinking about different futures, different adaptations, different roles and responsibilities in drought risk adaptation.  This encourages collective thinking about smaller scale adaptive actions based on traditional coping strategies but also the ‘digging deeper’ to explore what more transformative behaviour changes might look like, and the resources (human capital, technology, funds etc.) needed to implement these.  Thinking about adaptation was challenging territory in Creative Droughts because of the fragility of community that can mitigate against longer term planning and encourages dependency on the State. In both settings, thinking about the creative interplay of low technology, bricolage and civil agency can be particularly useful.  Both interdisciplinary and intercultural working – finding new ways to vision and ask ‘What if’ that combine meaningful conversations and deep listening – seem critical to new engagements for increased shared understanding and action.  Again drawing across projects, creative thinking about how co-production could play out in different research settings is also likely to encourage resilient thinking – in building local capital for drought risk management in a climate change context.”

Successes in working across disciplines

The project was only short, but there were many opportunities for cross-disciplinary learnings, as Dr Mel Rohse explains: “In terms of academic collaborations, I’ve very much enjoyed working with a range of disciplines. There’s something really refreshing about being asked to consider a research problem from a different viewpoint to the one that you’re used to. I felt it was equally useful to have to articulate why my own discipline takes a certain approach or what a specific concept means to colleagues from different backgrounds. That doesn’t mean it didn’t raise challenges, but it was a useful exercise and a good preparation for the fieldwork. In our case study in South Africa, I found that there was a lot of emphasis and interest in modelling and predictions. We were often asked for ‘solutions’ to drought, and understandably so considering the immediacy and everyday difficulties people face. So we had quite a bit of explaining to do about our method and why we were bringing in social sciences and hydrology. Overall, I felt that the project had some successes in using a participatory and mixed-method approach. People who took part in the workshops as much as local government officials seemed to get on board with the potential in looking beyond the physical sciences.” For further details, the processes and outcomes of the project are discussed in an upcoming report on the project.

Dr Sally Rangecroft took a step back to ask: “How best to maximise and enhance interdisciplinary working?” especially in the context of theory and practice informing each other. From her viewpoint, there are two key aspects that helped the interdisciplinary agenda of the project. First, Sally recognised that having hydrologists and social scientists based at the same institution and working together enhanced the opportunities for mutual learning, for shaping ideas and planning the research. Second, she emphasises how key the global North-South institutes’ relationships were. Our contacts in the Limpopo, both academic and governmental, were key to achieve a strong partnership based on sharing vital information and guidance.

Whilst some of the team had been involved in interdisciplinary work before, this was a first for others, like Dr Eugine Makaya. He reflected on what he saw as positives coming from the interdisciplinary approach. “I enjoyed the interdisciplinary methodological approach of the project which married social sciences to physical sciences. The use of the drought narratives, which are imbedded in social sciences, was usefully used to develop hydrological scenarios through workshops. This was very interesting as there was a strong correlation in the results found in both disciplines. The co-creation of future drought narratives was also a very interesting development which I did not believe would give good results in the beginning of the project. Thus, new skills and perceptions were developed following the workshops, especially for me coming from the physical sciences.”


We obviously also encountered some challenges. Anne, Lyla and Rosie picked up on a range of challenges to do with the methodology:

Anne, on justifying the method to ourselves, others and to potential journal editors: “Our alternative use of disciplinary methodologies worked very well and made the workshops successful. But it was also often questioned (both by ourselves and our partners), especially because both the narrative interviews and the modelling contained very interesting information that could have been used in the more classic way. In the year we had for the CreativeDrought project, we did however not have time to additionally perform the modelling and interview data analysis in the classical way. Alternatively, we could have shared the narrative interviews and model data for others to use and analyse them, but there is a large risk to this because the data were collected with a different aim and are therefore not suited to the purpose of drawing conclusions about drought management or predicting future water availability. We noticed this also when writing our scientific papers. Most journals still assume a disciplinary way of applying methods, so we had to explain why we did not use the model and interviews in the classical way. I hope that editors and reviewers will see the surplus value of the alternative use of disciplinary methodologies and understand the compromises that we needed to make.”

Lyla, on the importance of the bigger picture: “It is also important to contextualize the rich local narratives within the larger context of public policy, drought management as well as the political economy of inequality, race, class as well as historical legacies in South Africa. The village studied is strongly embedded in this wider context and the narratives on their own will mean little unless they are placed within this larger context.  Also climate change and drought intersect with other drivers of change that are of wider economic and socio-political trajectories.  While it was intended to capture all these issues, there were challenges concerning time, capacity as well as disciplinary bias.”

Rosie, on dealing with talk about the future: “I think though that the status of these futures remained a difficult issue – how real are they, or how likely to come about? How do/should these imaginary futures affect the present? The community’s lack of familiarity with projecting the future, compared to the academic team, or their different approach to it – mediated by spiritual practice, as opposed to mediated by computational power – is something I hadn’t fully anticipated. In any future work on similar themes I think I’d pay more attention to concepts of time and how different disciplines and cultures mark, create and imagine time.”

Whilst we saw benefits in working across disciplines and generally found it a positive learning experience, it was not always straightforward. Dr Steve Birkinshaw identifies how some things may have been misunderstood along the way and how communicating differently could have helped solve this problem. “A key aspect of the project was the interdisciplinary nature of it, the modellers working together with the social scientists to produce the future drought narratives. There were some misunderstandings about the modelling aspect of the work early on in the project before we started to successfully work with each other. The main misunderstandings were why we were using Shetran (rather than some existing simple large-scale water balance models) and what information the results of the model would provide (possible but not definitive future scenarios). The model needed to have the capability to simulate the upstream reservoir operation, groundwater abstractions, irrigation schemes, land use change and climate change, although there was insufficient high quality data to produce a really good hydrological simulation. These capabilities were necessary so that different future scenarios could be simulated giving realistic future drought narratives for the community-based workshops. These requirements meant that the physically-based distributed Shetran hydrological model was a suitable model and so was used. As a hydrological modeller for over 20 years I appreciated that and I could choose the most fitting models amongst the massive range of hydrological models that are available. The advantages and disadvantages of each model and what results they produce are technical aspects that are not known by non-experts such as the social scientists. What I have learnt is to be more aware of this and in the future, I would explain the modelling aspects in non-technical way early on in the project. A one-page summary document could be most useful.”

Inevitably, questions remain, as Lyla pointed out, it can be difficult to reconcile disciplinary differences around drought and uncertainty. “For example, resilience can be viewed through a systems perspective but it can also be viewed more politically in which case, merely building local resilience to drought lets powerful actors as well as powerful scientific framings off the hook. I also felt that the hydrological dimensions may not always capture the socio-political issues of drought. While most scientists and policy makers recognize that traditional [modelling/quantitative] approaches have limitations, they still tend to persist. While there is a great deal of acknowledgement of the need for local expertise to inform drought scenarios and models and there are emerging hybrid engagements,  there is as yet not much evidence of hydrological science or policy makers embracing or actively incorporating alternative knowledges.”

Professor Coleen Vogel’s thoughts are fitting to end this section.  Her reflections, as a South African with years of experience in climate change policy, relate to “the complicated institutional arrangements for climate and drought risk reduction (DRR). Despite the excellent DRR policy and several pieces of climate policy including a possible Bill on climate change, the country struggles to implement well-intentioned actions across scales and particularly for those living on the margins. The research from this work has confirmed this, that we have several fora and actions areas including varying capacitated and enabled extensions services etc. but the co-framing and co-design of interventions still has a way to go. The issues of mainstreaming and co-designing interventions are much needed and further research needs to be undertaken. This work has also shown me that even if you live in the country, if you are not fully immersed in the daily challenges of the context and space you are engaging in it is often difficult to make sense of a place.




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s