Nature Based Solutions are techniques for urban sustainability, adaptation, mitigation and improvement of quality of life, that are “synergistic with nature” and have multiple outcomes or services. Plants, animals, soil, water, and other natural elements can either be solutions, or components of solutions, in themselves, or may inspire those solutions (i.e. biomimesis). NBS may also be good for non-human species (although that isn’t the goal), and thus may sometimes be cross-categorized as Nature-Inclusive Design (http://www.bioveins.eu/blog/making-urban-nature-book-review).

NBS thus involve a very wide array of things, from plastic “cement” that absorbs water and allows it to evaporate away, rather than creating runoff, to green walls or roofs, to trees. The wide range of solutions, corresponding to a wide range of urban problems and urban planning goals, apparently causes a lot of difficulty for urban planners and decision makers. How to find out what solutions exist, how to implement them, how effective they are, or how to trade them off or combined them synergistically with other NBS?

Today I attended the Nature4Cities workshop in Brussels to find out. To be honest, I signed up with the quaint idea that we would be sitting around a table role-playing implementing specific NBS tools and techniques. I was wrong. NBS implementation is all about data integration and decision-support algorithms.

Nature4Cities is an H2020 (EU-funded) project with 27 institutional partners from 9 European cities (https://www.nature4cities.eu). It aims to create a “knowledge and decision platform for renaturing cities”. It brings together databases of assessed case studies of NBS implementation (https://www.nature4cities.eu/platform), as well as a step-by-step automated process that helps the user to build an NBS implementation scenario, assess it, and then plan an implementation strategy with citizen participation.

Their planning and assessment tool, which was the focus of the workshop, is rather complex, with about ten separate steps. At the core of it, if I understood correctly, are three kinds of assessments: a Simplified Urban Assessment Tool, and Environmental Impact Assessment, and a Socio-Economic Impact Assessment. Each of these is composed of several separate data analysis tools. The Simplified Urban Assessment Tool, for example, is composed of four tools called Greenpass, Colouree, Expert Box and Noise Box. Greenpass and Colouree are born of public-private collaborations. They are both simplified versions of commercial products that have been modified to serve NBS assessment ends, and integrated into the Nature4Cities process. Greenpass provides thermal data related to the human microclimate in and around buildings. Colouree provides data about how NBS affects quality of life and neighbourhood desirability assessments. Expert Box estimates carbon sequestration. I think the Noise Box measured urban noise levels.

When I asked why these four kinds of data analysis were chosen to make up the Simplified Urban Assessment Tool, I was told it was a pragmatic choice because the opportunities to integrate the assessment tools into Nature4Cities was available. The project doesn’t claim that they accurately represent all issues you would ever want to know about related to the impact of NBS on urban quality. The researcher I was talking to also pointed out that they could not integrate the four tools into a single analysis, or create a single output for the Simplified Urban Assessment Tool, because the input data for each analysis is at different scales and measures incomparable things. She argued that users don’t want a single evaluation score anyway, they want to see individual outputs of each component.

She may be right, but so far the tool, which is not yet finished, has something of a data visualization problem. The output of each assessment tool is a set of indices, ranging incomprehensibly from, as far as I could see, around 10 x -6 to 100. These are not helpful to anyone unless a visual method for contextualizing and interpreting them can be found.

The case study cities for the project are Milan, Alcalá de Henares (Spain), Szeged (Hungary) and çankaya (Turkey). You can’t currently do the full analysis for any other city because some of the input data is in public databases (e.g. municipal datasets about land uses), some comes from tailor-made analyses of existing datasets (e.g. number of green roofs analysed from satellite data), and some is generated through specific studies carried out in each city (e.g. citizen perceptions). Supposedly, all these datasets will be produced for all the major European cities at some point, so that those municipalities will have access to the tool.

I was surprised by how data-intensive it would be to implement NBS using the Nature4Cities platform. I am sure that the database of assessed existing solutions and programmes will be inspiring. Will the scenario building and assessment tools really be that valuable though? I must have been hungry (despite the apple fritters I ate earlier at the Brussels Christmas market) because I started to think about the data issues in terms of food. Suppose that you identify that you are hungry (your problem or question), and decide that you want to eat a hamburger (the data you need to answer your question). To satisfy you, your hamburger should include a sourdough bun, pickles, a leaf of lettuce, a small amount of mayonnaise, two slices of tomato, and a medium-rare beef patty. You go into the kitchen to make your hamburger (data analysis). Unfortunately, all you have available are two Wassas (large dry Scandinavian crackers), a meatball from the Italian grocery, and an egg (not exactly the data you were hoping for). Those ingredients can be arranged into a form resembling a hamburger (data analysis). But is it really the same? Will it actually satisfy your hunger, or even taste good? Should you just go food shopping for the ingredients you want, or to a restaurant (create your own data through tailored studies or expert analysis)? Almost all data analysis has some problems of this kind. Why, then, are complex data-heavy automated decision-making tools so popular? I think that the real value here is the production of scenarios and the decision-making support. Will users always accept the assessment of each scenario and decision? I am interested in how municipalities (or other actors) will combine this tool with other ones.