Impact Data Catalogue
Catalogue of impact data
Levels of impact data
From the point of view of adaptation and insurance, not only hazard data are important, but also data on impacts. Natural hazards trigger a cascade of impacts that begin in the geophysical realm, altering hydrologic cycles and soil conditions. These alterations cause ecosystem changes, affecting for example biomass production and essential ecosystem services. Ultimately, the environmental changes impact the man-made environment, for example damaging infrastructure and agriculture, or human beings, for example through health effects.
Any effect following from the original hazard can be termed impact. Different experts and actors apply different distinctions between hazards and impacts. For a climate scientist, a change in river discharge can be understood as an impact, whereas for an insurer, a change in the probability of exceedance of certain water levels in flood prone areas or a change in value at risk is an impact.
As there are differences in the definition of impact, we can distinguish several levels of impact data (Table 1). Table 1 provides an overview starting from a basic level (1) where only damage proneness is indicated, via initial impact probabilities (2) and volume indications of physical impacts on economic assets (3) to more precise physical and monetary damage inventories (4). Data regarding impacts on the insurance sector itself (5) are important for management of the sector and for supervision of financial stability of the sector. Levels 6 and 7 have not been considered in PIISA, as these are less relevant for insurance products. Broader well-being effects and macro-economic effects do play an important role in comprehensive adaptation planning.
| Level | Extent of impact inclusion | Examples |
|---|---|---|
| 1 | Damage proneness of weather conditions |
|
| 2 | Initial impact probabilities (e.g. ~60% likelihood water level > X meters in an area) |
|
| 3 | Observed or modelled physical impacts on production, building stock, public health, etc. |
|
| 4 | Observed and attributed economic impacts |
|
| 5 | Impacts on the insurance sector |
|
| 6 | Broader well-being effects |
|
| 7 | Overall macro-economic impact | Requires reliable, versatile, detailed statistics and validated modelling. Enables attribution to initial impacts, weaknesses, and recovery strategies. |
A catalogue of impact data
Table 2 presents a selection of publicly accessible impact data sets related to the hazards explored in the five PIISA pilots. Some have been used in a PIISA pilot, others are relevant additional options. Some data sets are only offered in the national language, while others also offer an option in English. The emphasis is on datasets that enable representation in maps. Many of the selected datasets allow downloads for the user’s own analysis. The level refers to the impact levels as defined in Table 1.
Of the indices presented in the Risk indices tab, the Yield Loss Drought Index and Standardized Windstorm Index can provide information on (projected) impacts at level 3. The other indices hover between levels 1 and 2.
| Hazard; measured effect | Sector | Name & location of dataset | Brief description |
|---|---|---|---|
| Heat stress — Urban | |||
| Cooling degree days Level 2 |
Urban planning; Public health |
Heat Roadmap Europe – Baseline Scenarios of heating and cooling demand §7.2 Cooling Degree Days projections by EU Member State. | CDD projections, taken from Deliverable D3.3 of Heat Roadmap Europe (“Baseline scenario of the heating and cooling demand in buildings and industry in the 14 MSs until 2050”). Used 2050 projections of CDD for Finland, Netherlands and Italy. |
| Urban Heat Island Level 2 |
Urban planning; Public health |
Climate Impact Atlas — icon ‘Heat’ > Impacts > Urban heat island effect | Zoomable UHI map — current climate / 2050. |
| Warm nights Level 2/3 |
Urban planning; Public health |
Climate Impact Atlas — icon ‘Heat’ > Impacts > ‘Night heat’ > number of nights above X °C. | Zoomable map — number of warm nights/year — current climate / 2050. Netherlands. |
| Clay soil shrinkage (CSS) | |||
| CSS risk of address of choice Level 4 | Real estate | Errial Localized geophysical risk data retrieval service for buildings and infrastructure. | |
| CSS risk (and other) Level 3 | Real estate | GASPAR Identification of affected municipalities, risk zoning, prevention plans and land-use restrictions related to exposed assets. | |
| Soil moisture deficit Level 2 | Real estate | Soil Wetness Index (SWI) from the Météo-France public data portal. | Indicator representing relative soil moisture conditions compared to climatology. Precursor of drought index. |
| Pluvial floods — Urban | |||
| Observed flood damage Level 3/4 | Real estate; infra; etc. | Flood damage statistics (Tulvavahinkotilastot), also other flood types. Microsoft Power BI. | Summary statistics of observed damage over the period 1995–2015 by type of object, by municipality and type of flood. |
| Modelled flood damage Level 3/4 | Real estate; infra; people | Flood damage projections for near future, also other flood types. Microsoft Power BI. | Estimated quantity of affected people, buildings and roads per flood risk area, by return time. |
| Pluvial flood depth Level 3 | Real estate; infra | SYKE storm water map (Beta version 2024). | Zoomable map application for pluvial flood risks in built-up areas in Finland for two levels of extreme downpours. |
| Water depth after downpour Level 3 | Real estate; infra | Climate Impact Atlas – Icon ‘Waterlogging’ > impacts > Large scale extreme rainfall > flood depth 70 mm and 140 mm rain in 2 hours. | Flood depth associated with several short severe precipitation events in the Netherlands. |
| Drought | |||
| Drought Level 1/2 | Agriculture | Combined Drought Indicator (CDI) from the European Drought Observatory. | Developed by the European Drought Observatory, it integrates precipitation anomalies (SPI), soil moisture anomalies, and vegetation stress indicators to classify drought conditions into warning levels. |
| Windthrow | |||
| 5 year wind damage risk in Finland Level 3 | Forests | LUKE forest damage risk monitoring at high resolution. | Zoomable map application showing 5 year wind-induced damage probability (also snow load and bark beetle damage risks can be shown). |
| Damage ranking based windthrow catalogue Level 2 | Forests | FORWIND (article); Dataset Database of wind events in European forests over the period 2000–2018. | |
| Wildfire | |||
| Historically burned area Level 3 | Forests | MODIS burned area. | Historical burned area from MCD64A1 Version 6.1 product at 500 meters resolution over the 2001–2024 period. |
Observed impact data concerning heat stress are rare because of attribution challenges. Furthermore, such data are usually for large spatial scales, e.g. regional. Similarly, harvest losses due to adverse weather have attribution challenges. For the other hazards observed physical impact data do exist, but may have limited access. For forest related damage such data are mostly either an annual summary or a case study. In some countries, e.g. Finland, forest monitoring systems enable high resolution simulation of damage risk for a certain period. For flooding related damage, high resolution observed damage and cost are rarely available, but by combining flood area and flood depth with real estate and population data, reasonable estimates of the number of affected homes and people and approximate direct cost can be made. The same applies to projected flood risks at sufficiently high resolutions.