By Eli Sandberg, Kristian Stenerud Skeie, Christine Hung

As climate change accelerates, it is becoming even more critical for municipalities and insurance actors alike to be prepared for the impact and consequences of severe weather events.

Decision-makers must therefore have access to reliable and high-quality data about which areas and populations are vulnerable to climate effects and what potential costs may arise due to damage from extreme weather events.

SINTEF has collected an overview of what data the SOTERIA partner regions have available that can help describe and assess the climate-related risks they face. We have divided the categories into background data, risk and hazard data and loss data:

✅ Background data describe the assets that are at risk, for example, buildings, infrastructure or agricultural land. Terrain data and vegetation, ecosystems and habitats, cultural heritage, demographic data, societal function and property valuations are also included in this category.

✅ Risk and hazard data identifies and describes the types of natural hazard risks, such as flood zone maps, observed events, and climate data.

✅Finally, loss data reports the cost of restoring damaged assets.

The collected data was evaluated based on various quality criteria aligned with the DCAT-AP standard, a European framework designed to promote interoperability and harmonization of data across sectors and countries. However, we found that the descriptions in regional or national portals often fail to address all aspects covered by DCAT-AP and its geospatial extension.

As a result, we had to manually inspect the various data catalogues, data services, and datasets to understand their structure and scope, including spatial/temporal coverage and resolution, and legal information. Based on the data sources obtained, we have identified best practice examples:

Detailed background data, which can be used to analyse climate-related risks, is generally available to municipalities, as they are often responsible for producing and maintaining information on buildings, land use, and infrastructure. However, the extent to which this information is accessible to other parties varies.

We find that the observed differences between regions correlate with how different countries score in the Open Data Maturity (ODM) assessment, which measures the progress of European countries in promoting and facilitating the availability and reuse of public sector information.

However, we expect both data provisions and portals to improve as the EU Implementing Act on High-value Datasets and other recent policies take effect. A best practice example is how countries, such as Spain, have made building outlines and cadastral parcels available through the INSPIRE geoportal, thereby supporting harmonization efforts like the Open Maps for Europe initiative in developing high-value pan-European datasets and services.

All partner regions or municipalities identified publicly available datasets related to flood risk. However, national or regional datasets for other hazard risk types, such as wildfires, landslides and extreme weather, were not as prevalent. Best practices include incorporating multiple climate scenarios or risk levels (e.g., low, medium and high), as well as considering return intervals and the geographical extent of modelled or recorded events.

Loss data proved to be scarce for the regions, and where available, it is typically aggregated at the municipal level, as opposed to the asset level. The best practice example for insurance loss data comes from insurance companies in Norway, which share loss data with municipalities at the address level. Gabrovo has data on compensation provided by the government, EU and “other sources”, in addition to insurance. For Gabrovo, these data are found online via the National Statistical Institute.

Overall, the data we received provides insights into what information the regions have access to, but also reflects what data the regions are aware of and actively/currently use; in some cases, we know that there is more data available than what was filled in the templates. This may indicate that data has not sufficiently been made available for the regions, which conflicts with the guidelines provided by EU legislation. Our findings may help describe the gap between the data the region knows about and/or actively uses and what data is in fact available.

We have therefore selected the regions of Valencia (Spain) and Trøndelag (Norway) to delve more deeply into the potential best practice for data recording and sharing.

It is, however, important to note that a lack of detailed loss data can be compensated for by robust background and risk data. For example, one way to assess the consequences of climate- related hazards is through satellite data and image recognition techniques. Another example is combining satellite imaging with detailed elevation models and other spatial data as inputs to simulation models. The resulting damage ratios combined with, for example, building characteristics and financial conditions, can then be used to estimate losses. We expect that advancements will further integrate these data types into loss modelling frameworks in the coming years.

To promote data sharing, research, and the development of data-driven services across regions, efforts to harmonize and ensure the interoperability of climate-related risk and loss data must continue. However, our mapping indicates that many promising developments are already underway.

SOTERIA aims to establish regional data hubs in three pilot regions to enable adequate and coordinated measures to reduce the climate and disaster risk, and to be more resilient and better adapted, through better use of data.