For a financial institution, an infrastructure operator, a manufacturer or simply a company or a local authority with a large spatial footprint, properly assessing a climate risk may seem like an insurmountable challenge: with hundreds or even thousands of assets, locations or suppliers, can you expect to forecast the impacts of global warming in a cost-effective and timely manner?
This case study will show you that the answer is yes. We are going to estimate the frequency of flooding for a real infrastructure operating almost one million sites! Whatever the complexity of your situation, thanks to its experience in big data applied to climate risk assessment, Callendar can help you rigorously calculate your exposure to extreme weather events in a very competitive cost and time frame.
Context: a large decentralized infrastructure exposed to climate risks
The electricity distribution network, operated in France by Enedis, is used to deliver power to end consumers. In most cases, electricity is supplied in low voltage (220 Volts). As a result, a medium voltage/low voltage transformer station (MV/LV transformer or “poste HTA/BT” in French). As these transformers needs to be close to final users, they are extremely numerous and spread over the whole territory.
In this situation, it is inevitable that some of the installations are exposed to climate risks. This is particularly the case with floods: flooding of electrical substations is a situation regularly encountered by Enedis, for example during the Seine flood of 2016 or the Xyntia storm in 2010. Such events cause power outages and represents technical and financial risks, as well as an important image issue for customers.
But with almost one million MV/LV transformer stations in operation in mainland France, how can we assess the extent of this risk? Or, which is even more difficult, how can we identify the vulnerable sites that should be protected in priority?
One in a million: a MV/LV transformer submerged
Methodology and data source
For this case study, we cross-referenced the position of each of the 927,753 MV/LV substations operated by Enedis in mainland France with a detailed mapping of flood-prone areas.
The positions of the installations are available in open data. The flood risk is assessed using a model originally developed by Dottori, et al. This model provides water heights according to the return time with a horizontal resolution of 100 meters.
Our added value: Have you ever tried to open a spreadsheet of one million lines or an image of 250Mb with a general public software? And cross-reference the two? Through Callendar, you have access not only to state-of-the-art climate and weather data. But also to the data processing tools and the computational power required to exploit them.
We used 4 return times: 10 years, 20 years, 50 years and 100 years. We consider that a station is at risk of flooding for a given return time if the corresponding water height exceeds 50 centimetres. The risk of coastal flooding is not taken into account.
If you are interested in the details of the data processing, we have published a tutorial on a similar issue in French that explains step by step the process and the Python code used.
Overview of results
With this definition, we reach the following evaluation:
- 24,161 transformers, or 2.6% of the total, are at risk of flooding with a 10-year return time,
- 28.686 (3.1%) are at risk of flooding with a return time of 20 years,
- 32,947 (3.6%) are at risk of flooding with a 50-year return period,
- 35,947 (3.9%) are at risk of flooding from the 100-year flood.
All this is a bit abstract? Here’s an interactive map that will help you visualize the nearly 36,000 flood-prone facilities:
Our added value: We believe that things are better when everyone does their own job. Ours is the exploitation of climate data, yours is… yours. We can’t tell you how to reduce your risks or what actions to put in your adaptation plan but we can provide you with the data that will allow you to make the right decisions, including in terms of format: databases adapted to your existing tools, static or dynamic graphs and maps, applications…
Added value of large-scale climate risk assessment
Callendar’s mission is to provide useful results that really help our clients reduce their climate risks. How can such a massive climate vulnerability assessment be put to practical use and bring value to an organization? Here are some examples.
Assessing climate risks for external (investors, regulators…) or internal use
The estimation of climate risks on each site allows to calculate reliable indicators on the global exposure of the company. This assessment can help meet the growing demands of investors and regulators.
Internally, it can also help to provision the risk: in this case study, for example, it can be calculated that on average 2758 MV/LV transformers will be flooded every year .
Obtain the best cost/efficiency ratio by concentrating resources on the most vulnerable sites
Thanks to its level of granularity, our approach can also be the key to a well-targeted adaptation strategy: if it is fast and cheap enough, a detailed global assessment is a very useful first step to concentrate the available means on the most at-risk sites.
Let’s imagine, for example, that Enedis is able to invest in protecting 200 substations against the risk of flooding. Our study, possibly supplemented by other data, could make it possible to select 500 serious candidates from nearly one million installations on which to conduct additional studies in order to ensure that investment are directed to the most vulnerable sites.
Beyond energy: applications in finance, infrastructure, industry, administration…
This case study is just one example: we have developed numerous solutions to efficiently cross-reference high-resolution climate projections with large geographical, economic, demographic or social data sets. This expertise, which allows for large-scale climate risk assessment that remains quantitative and localized, can be applied in many areas, including:
- Studies of a portfolio of assets in banking or finance,
- Verification of risk dispersion in insurance,
- Supply chain vulnerability assessment in industry and logistics,
- Quantification of climate change adaptation needs on a territory, for example in a National Adaptation Plan.
Do you have your own needs? Do not hesitate to tell us about it!