Making SENSES — a Project Report
Reflections on communicating and visualising climate change scenarios
The core aim of the SENSES project was the visualisation and communication of climate change scenarios. We started the project in late 2017 — and submitted the last modules in late 2020. It was a long and intense project — and since we just published an extensive Perspective in the journal One Earth, it seems like a good time to reflect on the project in a more informal way. So, here are some impressions from the design process and the project development.
This project report was written by the SENSES team at the UCLAB in Potsdam: Francesca Morini, Boris Müller, Jonas Parnow, Fidel Thomet and Nadia Zeissig.
What is SENSES all about?
The overall aim of the SENSES Toolkit — available under climatescenarios.org — is to make climate change scenarios more understandable, accessible, and usable. The results of the project are available for the general public. The SENSES project did, however, collaborate specifically with users from three different communities: policy, finance, and regional scenario users.
What are climate change scenarios?
Climate change scenarios are a powerful tool for describing how global climate will change in the next decades. They allow us to explore possible climate futures, chart response strategies, and inform climate policy making.
Scenarios do not predict the future. Rather, they are projections of what can happen and describe pathways of what should happen.
In short, climate change scenarios are plausible, coherent and internally consistent descriptions of possible climate change futures. They can also describe pathways towards certain goals. In a way, climate change scenarios try to answer two questions: ‘What can happen?’ (projections) and ‘What should happen?’ (goal-oriented pathways).
How did we start the design process?
You probably noticed: climate research in general and climate change scenarios in particular are not trivial. And since no one at the UCLAB is a climate scientist, the first thing we had to do was to establish an understanding of the subject matter. This meant reading a lot of scientific papers. Furthermore, we talked a lot about our understanding and our insights — both internally and with climate scientists. During this process, we started to sketch little explainers that helped us to establish a common ground.
At one point we realised that our notes and sketches might be super-helpful for explaining climate change scenarios to a wider audience. So we came up with the idea to create a Primer on Climate Change Scenarios. Reflecting on what we had learned, we were able to turn these sketches and insights into a concise and interactive introduction to climate change scenarios. Everyone in the SENSES consortium was quite enthusiastic about it — so it was the first module we launched during the project. The Primer was much more work than initially anticipated, but it was extremely useful for communicating the basic concepts behind scenarios. Furthermore, it was a great foundation for subsequent collaborations with scientists and stakeholders. And it was very successful — until the end of the SENSES project, it was the most frequented and most popular of all modules in the toolkit.
In the Primer, we often use simple graphics that explain a lot. The following graphic encapsulates the process that drives global warming.
If you haven’t come across the terms ‘mitigation’ and ‘adaptation’: they are the two main strategies for engaging climate change. In this graphic you can immediately grasp where they can come into effect.
Whom were we designing for?
The SENSES Toolkit is a web site that is available for the general public. However, actively using climate change scenarios requires a certain level of expertise. While we wanted to make the results publicly available, we understood that there was a real need to make climate change scenarios better accessible to decision-makers and stakeholders. So, in the design and development process, we collaborated with four distinct stakeholder groups: policy, finance and regional scenario users. Our partners in the consortium had excellent connections to these groups, so we were able to engage directly with our stakeholders.
Our aim was to provide current and potential climate scenario users with practical tools and visualisations that would allow them to productively include current scenarios in their own work. We wanted to enable our stakeholders to literally see the impact and implications of the climate change scenarios and to generate actionable insights. Especially the finance community is used to working with scenarios and showed a great interest in using climate change scenarios.
An interesting user group were regional scenario users. We had to address the needs of regional decision makers and to consider local implications of global socio-economic change, climate change and climate policies. The challenge was to downscale global scenarios and match existing national and global scenarios to local requirements.
So on the one hand we had direct access to clearly defined user groups — but on the other hand each user group had very distinct interests and demands.
How did we engage the stakeholders?
Together with the colleagues from the SENSES consortium, we conducted several co-design workshop with the stakeholder groups. The main objective of the first workshops was to understand how our users actually work with climate change scenarios. We wanted to find out what kind of topics they are interested in and how scenarios could be used to address specific questions like sustainable investments or mitigation strategies.
The output of the workshop showed that we were dealing with quite a number of diverse topics and questions. We realised that a single tool or visualisation for all issues would simply not work. In order to meet the different questions and topics, we needed to provide our stakeholders with a more modular structure that would enable them to access the scenario space from their own distinct perspective. Furthermore, we understood that we needed to provide more context for the scenarios. So, we developed a conceptual framework that guided us in the development of the SENSES Toolkit.
The conceptual framework consisted of three different things we wanted our user to do:
- Learn: acquire a general understanding of the fundamental mechanics of climate change scenarios as well as more specific narratives that provide context to specific scenarios.
- Explore: facilitating self-directed as well as guided exploration of the climate change scenario space.
- Share: using, reusing, and sharing visualisations and climate scenario data.
Based on this conceptual framework, we created a modular structure for the SENSES Toolkit. In this sense, each module is either a Learn or an Explore Module — and each module provides the users with the functionality to share the data and the visualisation.
This modular structure had several advantages. We were able to address a diverse set of requirements from our stakeholders and we could distribute the responsibilities in the team. Every module had a lead designer who oversaw the design, the communication and the implementation — even if he or she got support from the consortium and the rest of the design team.
How did we collaborate with climate scientist and stakeholder?
After establishing a modular structure for the SENSES toolkit, we obviously had to fill it with both content (Learn Modules) and interfaces that would allow our users to access the scenario database (Explore Modules).
Explore Modules were not trivial — but fairly straight forward. Since they did not have an underlying narrative, it was more about understanding the data structures and finding visual ways for accessing and representing them.
The SENSES Scenario Finder is a good example for this approach. It is a versatile tool for exploring the scenario space in several ways. The Finder acts as an advanced table and allows the user to filter each column through various facets. All facets are dynamic, meaning they instantly react to the selection and visualise options that are no longer selectable or how the current distribution of values is compared to the unfiltered list.
Another very effective way to introduce our stakeholders to scenario data are a special kind of Explore Modules — the Guided Explore Modules (GEMs). To bridge the gap between the explanatory Learn Modules and the IIASA Scenario Explorer aimed at experts, we equipped several Learn Modules with Guided Explore Modules. Here, readers can deepen their knowledge and understanding by accessing and downloading the underlying data and by exploring different perspectives.
All GEMs have a consistent layout that features an introduction, guiding questions, the actual data, display options, and metadata. GEMs are driven by config files which are standardised and documented, making it fairly easy to extend the set of available GEMs in the future. All data are directly retrieved from the IIASA Scenario Explorer API and are thereby ensured to stay up to date.
The workflow for creating the Learn Modules was more complex. Since each Learn Module has a topic and a narrative, we collaborated closely with domain experts — i.e. climate scientists. We realised that the best way to define and fix a topic was to find an existing academic paper and translate it into a more concise and comprehensible Learn Module. This approach proved to be very effective. It allowed us to confine a subject and have reliable academic references as well as generate a narrative and develop visualisations. Thus, we were able to create a high number of Learn Modules in a relative short time span.
We needed a workflow that was both focused and effective, so we came up with a procedure that we called ‘Tandems’. A Tandem consists of one domain expert and one visualisation designer, who are closely collaborating and balancing their different perspectives. Tandems have key characteristics that enabled us to deal with mentioned challenges in a very effective way.
First, tandems have very focused ambitions. They address a single and specific issue in the context of a larger project framework. Secondly, while co-design processes usually have a variable number of people in different roles, tandems are very rigorous: only one domain expert and one visualisation designer are involved in the core design phase. (The domain expert was often the author of the academic paper that was the foundation of the Learn Module.) These two initial conditions allow for a great degree of freedom in terms of both workflow and output.
As shown in the figure above, only the Tandem itself is active throughout the entire process. The work of the Tandem is supported by the consortium and by selected stakeholders — but only at specifically defined points.
Although Tandems are fairly flexible, we can summarise four typical phases: Ideation, Design, Evaluation and Publication. Those phases are drawn from existing design and co-design frameworks, such as the Design Activity Framework proposed by McKenna et al.
The ideation phase involves the entire consortium: after a brainstorming session, possible topics are narrowed down to one specific question that the Tandem will address. In the SENSES project, the specific question to be addressed in the Learn Module was defined in this phase.
Right after comes the design phase: a visualisation designer and a domain expert work autonomously on content, format, and design. In this phase, the domain expert and the visualisation designer need to identify data that can answer the previously formulated question. The domain expert is usually providing knowledge, datasets, and relevant literature. The visualisation designer organises this knowledge and starts prototyping, using the material provided by the domain expert. As in traditional prototyping, this phase requires various iterations to define, realise, and refine the final product.
During the evaluation phase, experts and designers present their work to the research group and to potential users. Feedback can be given both in unstructured and informal settings (e.g. by navigating through the module and expressing opinions on the content), and in defined, formal workshops (e.g. by compiling written questionnaires on specific aspects, such as user experience, accessibility of contents, and so on). After a successful evaluation, the work is ready for the publication phase.
From a technical point of view, the Tandems developed the modules on individual microsites and git-repositories. The decision to split the toolkit into several modules and work in Tandems made it necessary for us to think of ways to use synergies and ensure consistency. We built a dedicated SENSES Library as a living style guide and component library. Styles, functionalities and components (e.g. input elements like radio buttons and dropdown menus) developed by the Tandems and approved by the consortium were integrated into the library. This made them available for reuse, which in turn simplified the overall development process.
What can our stakeholders do with the toolkit?
As discussed above, the SENSES Toolkit has a modular structure. There is, however, an underlying course of action that guides the users in the Toolkit:
- Acquire general knowledge on climate change scenarios in the Primer
- Find out about specific topics in the Learn Modules
- Get acquainted with selected data points in the Guided Explore Modules (GEMs)
- Autonomously explore the scenario space
- Download data and visualisations for personal and professional use
The overall aim of the project was to empower our users. We wanted them to be able to use climate change scenarios for their own projects. For example, climate change scenarios can be used to demonstrate the most cost-effective ways to reach the 1.5° goal, reveal how global warming will lead to more extreme events, or identify financial investment opportunities. By providing extensive background knowledge and a simple access to scenario data, the SENSES toolkit enables our users to identify many more applications and use scenarios for their own work.
We created 20 modules for the SENSES Toolkit — some of them like the Primer have already been referred to. If you are interested in climate change, it is definitely well worth to go through the entire toolkit!
In any case, we would like to discuss four modules in more detail as we want to address different design approaches in the toolkit.
(Design lead: Fidel Thomet)
Climate change increases both frequency and intensity of extreme events such as heatwaves and tropical cyclones. SENSES Earth explores the spatial distribution of exposure to extreme events by projecting them on a 3D globe. The visualisation displays data for six extreme events (crop failure, droughts, heatwaves, river floods, tropical cyclones, and wildfires) at various degrees of global warming and for individual climate and impact models as well as for multi-model medians.
The data, which are derived from climate and impact models, differs from other datasets we use in the toolkit because of its high resolution spatial dimension. The three-dimensional representation of the data and its overall visual appearance give the module a unique and distinct look and feel.
Additionally, the globe is designed to support reusability and allowed us to embed it in the closely related Learn Module on extreme events.
Senses Earth also serves as a bridge from the Senses Toolkit to the extreme events study in ISIpedia, which offers country-level reports for the same set of events.
Fossil Fuel Risks
(Design lead: Francesca Morini)
Our society is currently relying heavily on fossil fuels to provide energy for our industries, for keeping our homes warm, and for moving goods and persons around the globe. However, mitigating climate change will require us to reduce the emission of greenhouse gasses by reducing the use of fossil fuels. As a result, the demand for fossil fuels will evolve differently in the future than what we have seen in the past. This has consequences for the risk of investing in the fossil fuel sector.
In the module Fossil Fuel Risks, we show the risks associated with investing in the fossil fuel sector and explain why investors should consider climate policy in their decision making process. Essential for this is understanding the range of possibilities — so we addressed uncertainty in a dedicated visualisation. The main concept is to go beyond uncertainty as a ‘side parameter’ and treat it as a central element that needs to be understood when dealing with climate change scenarios. The reader is guided through different types of risk, including primary energy quantity and carbon taxes occurrence. The module itself is part of a wider collection of chapters dedicated to financial instability and opportunities under different scenarios.
Power Sector Transformation
(Design lead: Nadia Zeissig)
The module Power Sector Transformation examines the different types of risks the electricity sector is facing resulting from the decarbonisation of global electricity generation. Each of the risks covered in the module relates to one of the Risk Factor Pathways developed by the UNEP Finance Initiative. The module is addressed towards finance experts and explains why climate policy should be considered in their decision-making.
On the basis of three transition scenarios the module analyses the changes in electricity production from fossil fuel electricity production to renewable and nuclear-based electricity driven by a carbon price. This dynamic results in a shift of revenues from fossil utilities to renewable and nuclear based utilities.
Interactive data visualisation and storytelling are used in the module to convey the information. At the beginning, the annually produced amount of electricity is displayed in the form of circles of corresponding sizes. This row of circles is carried along throughout the module as a constant visual element and as a comparative figure. Further information is then added or contrasted to this variable in different graphs presenting the risk types.
Dutch Case Study
(Design lead: Jonas Parnow)
The Learn Module on the Dutch Case Study was specifically designed for the regional scenario user community. It is a best-practice example for the co-production of robust adaptation pathways. An essential tool for our stakeholders is the co-creation technique “Fuzzy Cognitive Map”. It consists of a network of drivers (e.g. population decline and technology investment) and concepts (e.g. biodiversity, climate-smart land use and local economy) that influence each other positively or negatively. It is possible to activate and deactivate each driver individually to see how they influence the process of the concepts. The design team created an interactive visualisation of the “Fuzzy Cognitive Map” that enabled the stakeholders to actively use it in workshops and explore regional implications of climate change.
Global warming is one of the greatest challenges humanity is facing right now. The way we deal with global warming today will determine how future generations will live on this planet. Climate change scenarios are a way to plan this future. They can be used to make both qualified and sustainable decisions today.
Throughout the entire SENSES Toolkit, visualisation design played an important role. We use visualisation techniques to explain how models work, we use them for comparing scenarios, and we use visualisation interfaces for designing access to the scenario database. The visual design creates a contextual bridge between general scenario knowledge and the complex scenario space of the database.
For us, it was a remarkable experience to work on such a relevant topic and in such a notable consortium! We hope this project report provides some insights and inspirations for other design projects that deal with similar issues.
By making climate change scenarios more understandable, more accessible, and more useful, we hope to contribute to the ongoing societal effort for reaching the 1.5° target.
Fachhochschule Potsdam / UCLAB: Boris Müller, Fidel Thomet, Francesca Morini, Nadia Zeissig, Jonas Parnow
Potsdam Institute for Climate Impact Research: Elmar Kriegler, Christoph Bertram, Cornelia Auer, Jan Volkholz
International Institute for Applied System Analysis: Volker Krey, Daniel Huppmann, Michael Pimmer
Stockholm Environment Institute: Henrik Carlsen, Sara Talebian, Simona Pedde
Wageningen University & Research: Kasper Kok, Lotte de Jong
SENSES is part of the European Research Area for Climate Services (ERA4CS), an ERA-NET initiated by JPI Climate. It is funded by BMBF (DE), BMWFW (AT), NWO (NL), FORMAS (SE) with co-funding by the European Union (Grant 690462).