Climate Service for the Chief Snowmaker – Opportunities and Challenges for Winter Tourism Management
About:
Carlo Carmagnola is a post-doc at Centre National de Recherches Météorologiques in France. He works as a snow scientist, focusing on prediction capabilities in projects PROSNOW and CLIMSNOW.
Martin Coath is a computer scientist and science communicator. He currently works as a senior researcher at the University of Lapland, Finland, on the Blue-Action project.
In this blog post, Carlo Carmagnola (PROSNOW) and Martin Coath (Blue-Action) discuss the differences and shared challenges between two projects looking at climate services for winter tourism*.
The YOPP-endorsed Blue-Action project is an EU Horizon 2020 project looking at the impact of a changing Arctic on northern hemisphere weather and climate. One of the case studies from the project is demonstrating the possibility of a climate service around snowmaking, co-designed with a ski resort in Northern Finland.
PROSNOW is a Horizon 2020 research project that aimed to build that has succeeded in building a seamless meteorological and climate prediction system applied to snow management, tailored to the needs of the industry through a co-design approach.
Carlo and Martin, you’re both involved in projects which design tools dedicated to optimizing snow management. Could you tell us a little about your respective tools and what they are meant to do?
Carlo: I’m a snow scientist working at the French Snow Research Centre. My main field of interest is in snow modelling and management, focusing on the physical properties of the snowpack. I usually spend half the time modelling, and then half the time in the field working with the stakeholders.
In PROSNOW, I was responsible for the French pilot ski resorts. The idea behind PROSNOW was to combine several elements – weather predictions, local observations and snow modelling – to come up with predictions for different variables such as wet bulb temperatures and snow depths. We then translated this information into simple terms for the resort partners and pushed the results to a web interface demonstrator. The users can apply this information to answer questions, such as “will there be enough snow on this slope for Christmas, do I need to make more snow, and how much water do I need to make the snow?”
Martin: I’m a data scientist and programmer working on the Blue-Action winter tourism case study. Blue-Action is a big project, so this is just a small part of it, making data available on seasonal to decadal predictions that are useful to a ski resort in Finland.
The size of PROSNOW is really interesting. Blue-Action is a tiny case study in comparison. We only considered one ski resort, one industrial partner, who are dependent upon manufactured snow. Every year, they open on 8th October, which they manage by storing snow made the previous season. Our role was to concentrate on snowmaking conditions, and boil that down into an app that we could deliver to mobile phones to predict snowmaking conditions, and an indication of how uncertain these predictions were. The aim was to give them aid to decision-making.
Developing such innovative, high-tech tools aimed at a niche market, there isn’t a roadmap to success. What particular challenges did you face?
Carlo: In our product, we use a real-time modelling chain, where we insert observations from the ski resorts and data such as the water consumption by each snow gun. Then we retrieve the forecast and feed it into three different snow models, one each for France, Austria and Switzerland. Each country runs its own snow model. A challenge was finding common parameterizations that could be applied to all our resorts and all our models.
Martin: That’s really interesting and very different from our challenge in Blue-Action. Our design process was about pruning out the complexity and finding the information that we could present to the chief snowmaker of this single resort that could affect his decision. We thought that every resort has a similar central job position, an expert who made the decision based on various inputs, but this doesn’t seem to be the case. So the product is applicable to other resorts certainly, but in its current state is very tailored and based on the co-design process with this partner.
Carlo: Of course the co-design process was also a key benefit and challenge for PROSNOW. The main point for us was to make it simple, but it was a difficult process to simplify so much complexity. In our first version of the prototype, for example, the amount of water was expressed in kilograms per square metre. This is a scientific metric, which is easy to understand in a scientific lab, but in the field, this is not easily understood by a stakeholder. Another challenge was the representation of uncertainty. Most stakeholders just look at the average value but don’t really look at the probability. We needed training and consulting process on a weekly basis to explain how the model works. We tried to develop something different from the graphs that could be useable and easily understood, and to put it into the demonstrator. A lot of effort was put into user-friendliness and ergonomy, but training people to correctly use the interface was a real challenge.
These projects provide valuable scientific advancements. But on top of that, your experience highlights the hurdles in transforming scientific modelling results into operational information for ski resorts. What factors do you see influencing user uptake?
Martin: This is a philosophical problem – if you had someone brought up one hundred years ago, a weather forecast would be nonsensical. But now everyone looks at it and understands because it is part of our cognitive skillset. This is a barrier that we can break down. People may have had to be trained to use your product initially, but then these skills are available to them, and this is the way it is going. The next generation will have the skills to interpret these things.
Carlo: I totally agree. And it really depends on the motivation of the end-user. We had nine ski resorts. Some people were ready to use PROSNOW, and others weren’t keen on having some scientist coming in and telling them what to do. We had to explain that it was a decision-making tool – it is simply there to help them!
Martin: That really resonates with me. The people involved in this process have so much information already, which can leave us in a difficult position because if we essentially say we have one more screen for you to look at, they understandably wouldn’t be that keen. We need to make it clear that it will add value, and that can be hard when you are piloting these processes. But in some ways, this is the beauty of it – it is important to start the pipeline and get buy-in from the users, but as the modelling improves, the quality will go up and up.
All these things considered, have your users perceived an added value in the deployment of these services?
Carlo: That’s a really important point for all climate services. We are taking new science to people who need something simple and are working hard, they don’t have time to explore the complexities. What we did for PROSNOW was focusing on the forecast one or two days ahead, then moving little by little to the long-term forecast. If we’d gone straight into the long-term forecast, where there is so much more uncertainty, it would have been harder to show the added value of the service, and they wouldn’t have necessarily trusted it so much.
Martin: Demonstrating value is key. We said from the outset if we gave you three days of extra planning, what would this be worth to you? We asked them to make that calculation. Together we were able to show there was economic value in the forecasts over the first four weeks, and in ten years’ time, this could extend to eight to twelve weeks of quality information. But we need to push forward now, to show people that there is value in climate services, to help businesses and people start to adapt. And from our perspective, we need to find out how we can best translate the huge gigabytes of data produced by science to these end-users, so it is useful. We need to fill in the gaps until we have a great communication line from beginning to end, then as the science improves, we already have a way to provide it directly to the people who need it.
PROSNOW has now ended its H2020 project phase. The service is now entering a new phase while being commercialized, and it will be available in French ski resorts as early as next winter season. The Blue-Action case study has also completed. The team are exploring commercialisation for other Finnish ski resorts over the coming season.
Watch the videos:
Blue Action Winter Tourism in Finland http://blue-action.eu/climate-services/1-winter-tourism-northern-finland
PROSNOW Snow Management Service: https://www.youtube.com/watch?v=ckVccQbnM44
The interview was conducted by Hannah Grist, communications lead of Blue Action.
*We acknowledge that there are debates and research around the impacts and ethics of snow-making, including the long-term sustainability, benefits to local economies, ecosystem changes and water use. It is beyond the scope of this blog to comment on this debate: for further information, please look at e.g., Rixen et al., (2003), Rolando et al. (2007), and Smith (2019). The aim of this piece is to highlight the potential of scientific developments in predictive capacity to improve efficiency of current snow-making, and ultimately decrease the requirement for and water consumption of such activities.
Rixen, C., Stoeckli, V., Ammann, W. (2003): Does artificial snow production affect soil and vegetation of ski pistes? A review. Perspectives in Plant Ecology, Evolution and Systematics, Volume 5, Issue 4.
Rolando, A., Caprio, E., Rinaldi, E., & Ellena, I. (2007). The Impact of High-Altitude Ski-Runs on Alpine Grassland Bird Communities. Journal of Applied Ecology, 44(1), 210–219.
Smith (2019): Luxury, Tourism and Harm: A Deviant Leisure Perspective. In: Raymen T., Smith O. (eds) Deviant Leisure. Palgrave Studies in Crime, Media and Culture. Palgrave Macmillan, Cham.
