Hazards by munich re


Hazards by Munich Re is a category you can use in 21RISK, that will fetch Natural Hazards data when a a report is created.


Hazards by Munich Re will only work, if used on a location with an address (and hence Geo coordinates).

A report with this category costs €100 per report. Contact 21RISK to make sure you have the balance available. When used in a report, it looks like this:

Risk Scores

The location in question is given 4 Risk Scores:

Risk Score Description


The Overall Risk Score can be used as a primary identifier of red flags. Financial damage caused by e.g. tropical cyclone or hail events can differ substantially.

Read more

Additionally, the financial damages that can be caused by all individual hazards need to be summed up to represent the total damage potential. Hence, for the Overall Risk Score, all individual hazards are weighted depending on their damage potential (derived from Munich Re`s proprietary collection of financial losses) and the corresponding risk scores are summed up.

The overall risk score includes on all provided NATHAN hazard scores with different weights in combination of an annual loss value for standard industrial business. It has to be taken into account that the wildfire score was not taken into account for the Risk Score split. This could cause small deviations between the overall Risk Score value and the sum of the individual Earthquake, Storm and Flood Risk Score.

Earthquake The Earthquake Risk Score can be used to identify earthquake related risks and includes earthquake, volcano and tsunami risk.
Storm The Storm Risk Score can be used to identify storm related risks includes tropical cyclone, extratropical storm, hail, tornado and lightening risk.
Flood The Flood Risk Score can be used to identify flood related risks includes river flood, flash flood and storm surge risk.


Continue to read more about each individual hazard, and their respective zones. The source for this is Munich Re's Fact Sheet .

Hazard: Active Faults

> The GEM Foundations Global Active Faults is building a comprehensive, global dataset of active fault traces of seismogenic concern. The GEM GAF-DB comprises GIS files hosted here of fault traces and small amount of relevant attributes or metadata (fault geometry, kinematics, slip rate, etc.) useful for seismic hazard modeling and other tectonic applications. The dataset is being assembled primarily as a part of GEMs global Probabilistic Seismic Hazard Modeling efforts, although we hope that the data find wide use in research, education and general interest among many users.

The GEM Global Active Faults project (GEM-GAF) compiles a global dataset of active faults for seismic hazard assessment. The GEM-GAF is building a comprehensive global dataset of active fault traces of seismogenic concern.

The dataset consists of GIS files hosted inhouse, of fault traces and small amounts of relevant attributes or metadata (fault geometry, kinematics, slip rate, etc.) useful for seismic hazard modelling, identifying the distance from a certain point to the nearest fault and other tectonic applications. The dataset currently covers most of the deforming continental regions on earth, with the exception of northeast Asia, the Malay Archipelago, Madagascar, Canada, and a few other regions. These are to be added progressively.

Hazard: Earthquake

Probable maximum intensity (MM: modified Mercalli scale) with an exceedance probability of 10% in 50 years (equivalent to a „return period“ of 475 years) for medium subsoil conditions.

Value Zone
0 Zone 0: MM V and below
1 Zone 1: MM VI
2 Zone 2: MM VII
3 Zone 3: MM VIII
4 Zone 4: MM IX and above

The earthquake map is graded according to the intensity that is to be expected once in a period of 475 years.

Intensity integrates a number of parameters such as ground acceleration and earthquake duration. The return period of 475 years corresponds to a 10% exceedance probability in 50 years, which repre- sents the mean service life of modern buildings. The intensity is expressed in terms of the modified Mercalli scale (MM). The earthquake map is based on an assemblage of existing hazard maps of individual countries. The source maps show:

  • The minimum intensity or peak acceleration to be expected for an exceedance probability of 10% in 50 years
  • The same parameters but for a different reference period
  • The maximum intensity observed
  • Active or potentially active faults
  • Epicentres of earthquakes recorded by instruments and/or historical earthquakes

Merging such heterogeneous sources presents enormous problems, beginning with the process of converting acceleration values into macroseismic intensity, for which various formulas have been proposed (e.g. Trifunac and Brady 1975, Murphy and O’Brien 1977).


Global Earthquake Model (GEM)

The Global Earthquake Model (GEM) Global Seismic Hazard Map (version 2018.1) depicts the geographic distribution of the Peak Ground Acceleration (PGA) with a 10% probability of being exceeded in 50 years, computed for reference rock conditions (shear wave velocity, VS30, of 760–800 m/s).

The map was created by collected maps computed using national and

regional probabilistic seismic hazard models developed by various institu- tions and projects, and by GEM Foundation scientists. The OpenQuake

engine, an open-source seismic hazard and risk calculation software devel- oped principally by the GEM Foundation, was used to calculate the hazard

values. A smoothing methodology was applied to homogenize hazard values along the model borders. The map is based on a database of hazard models described using the

OpenQuake engine data format (NRML); those models originally imple- mented in other software formats were converted into NRML. While trans- lating these models, various checks were performed to test the compatibil- ity between the original results and the new results computed using the

OpenQuake engine. Overall the differences between the original and

translated model results are small, notwithstanding some diversity in mod- elling methodologies implemented in different hazard modelling software.

The hashed areas in the map (e.g. Greenland) are currently not covered by a hazard model. The map and the underlying database of models are a dynamic framework, capable of incorporating newly released open models. Due to possible model limitations, regions portrayed with low hazard may still experience potentially damaging earthquakes.

Hazard: Extratropical storms

Probable maximum intensity with an average exeedance probability of 10% in ten years (equivalent to a „return period“ of 100 years). Areas were examined in which there is a high frequency of extratropical storms (approx. 30°–70° north and south of the equator).

Value Zone
-1 No hazard
0 Zone 0: ≤ 80 km/h
1 Zone 1: 81 - 120 km/h
2 Zone 2: 121 - 160 km/h
3 Zone 3: 161 - 200 km/h
4 Zone 4: > 200 km/h

Extratropical storms are created in the transition region between subtropical and polar climatic zones, i.e. in the latitudes between about 30° and 70°. In these regions, cold polar air masses collide with tropical air masses, forming extensive low-pressure eddies.

The intensity of the storm areas within these eddies is proportional to the difference in temperature between the two air masses, and is therefore at its greatest in late autumn and winter, when the oceans are still warm but the polar atmosphere is already extremely cold. This is why extratropical storms are also referred to as winter storms. Blizzards and ice storms are variants of this type of storm and their potential for damage is often underestimated.

The extratropical storm maps are based on freely available reanalysis data sets which have been downscaled and calibrated by using data from various national weather services, as well as information from global digital terrain models. Gust information from the following centres has been used particularly intensively: the German Weather Service, the Royal Netherlands Meteorological Institute, the UK Met Office, Meteo France, the Bureau of Meteorology (Australia) and the National Oceanic and Atmospheric Administration (USA). An extreme value distribution approach (generalized Pareto distribution includ- ing an upper bound estimation) was used to calculate storm maps with higher return periods. The hazard map is classified into five zones based on peak wind speeds (3 sec gust in km/h). The most exposed areas with respect to extratropical storms are located between 30° and 70° north and south of the equator. The final resolution of the storm maps is 0.01 degrees (roughly 1km).

Hazard: Flash-flood

Frequency and intensity of flash floods.

Value Zone
1 Zone 1: low
2 Zone 2
3 Zone 3
4 Zone 4
5 Zone 5
6 Zone 6: high

Flash floods are short-term events which can be produced by severe convective storms or heavy rain events over one area. Flash floods can be heavily destructive due to the enormous amount of water which often carries rocks, debris and mud.

The hazard is represented by 6 zones, starting from Zone 1 (low hazard) to Zone 6 (high hazard). The flash flood map is based on meteorological data, as well as soil, terrain and hydrographic data (slope and flow accumulation). The meteorological data includes the amount, variability and extreme behaviour of rainfall. Munich Re used soil-sealing maps (detected by looking at impervious surfaces), curvature (from global multi-resolution terrain elevation data with a resolution of 7.5 arcseconds), slope and flow accumulation (from conditioned terrain data based on SRTM elevation with a resolution of 15 arcseconds) as modifiers to generate the final flash flood map. The data is gridded on a 250-metre raster.

Hazard: Hail

Frequency and intensity of hailstorms.

Value Zone
1 Zone 1: low
2 Zone 2
3 Zone 3
4 Zone 4
5 Zone 5
6 Zone 6: High

Hailstorms cause extensive damage to agriculture, as well as to buildings and vehicles. Heavy hailstorms are usually triggered by wide cold fronts. Occasionally, local hot weather thunderstorms – a result of intense insolation over land or mountain slopes – also lead to severe localized hailstorms.

An important precondition for hailstorms is strong instability. This gives rising air at ground level a strong uplift and results in an even higher-reaching upwind zone with powerful cloud formations. In an upwind zone of this kind, hail particles are suspended in the upper section of the cloud so that water droplets and ice crystals are created and the hail seeds grow in layers as the winds successively carry them up. When the weight of the hail seeds becomes too great or the upwind weakens, the ice seeds fall from the cloud and it begins to hail.

The hailstorm map is based on the global distribution of lightning activity (lightning per km2 and year). Data sources of the hailstorm map are OTD/LIS data from NASA, a DEM (interpolated from SRTM data), global temperature data and global precipitation data. Hail as a natural hazard is based on the frequency and intensity of hailstorms. Munich Re does not use statistics on the occurrence of hail events, as such global statistics are not available and/or comparable. Therefore, global standardized records of meteorological data were used. On the basis of this meteorological data it was possible to represent atmospheric conditions which have the potential to create a hailstorm. In fact, the hailstorm map is based on a number of atmospheric conditions with the potential to create a hailstorm.

The following parameters were taken into account for the calculation:

  • Average annual evapotranspiration [mm]
  • Average annual temperature gradient [°C/km] − Average annual potential height of fall of hail [m]

Hazard: Lightning

Global frequency of lightning strokes per km<sup>2</sup> and year. Lightning frequency is determined by counting the total number of lightning flashes independently of whether they strike the ground or not.

Value Zone
1 Zone1: 0,2- 1
2 Zone2: 1- 4
3 Zone3: 4-10
4 Zone4: 10-20
5 Zone5: 20-40
6 Zone6: 40-80

At any given time about 1500 thunderstorms are taking place all over the world, with hardly any region remaining unaffected. Lightning strikes are the main cause of natural fires, which can destroy whole forests and often buildings.

The lightning map shows the global frequency of lightning strikes per km2 and year recorded by satellites and ground-based lightning detection networks. Munich Re classified lightning in 6 catego- ries based on frequency of lightning strikes. It is based on data from NASA: “the product (v2.2) is a 0.5 deg × 0.5 deg gridded composite of total (IC+CG) lightning bulk production, expressed as a flash rate density (fl/km<sup>2</sup>/yr). Climatologies (v2.2) from the 5-yr OTD (4/95-3/00) and 8-yr LIS (1/98-12/05) missions are included, as well as a combined OTD+LIS climatology and supporting base data (flash counts and viewing times). Best-available detection efficiency corrections and instrument cross normal- izations have been applied.”

Hazard: River flood

Areas threatened by extreme floods. JBA flood maps with return periods of 50, 100 and 500. Areas threatened by extreme floods. JBA flood maps with return periods of 50, 100 and 500.


July 2023, new 50 year return period was added.

Read more

Why are the River Flood zones updated? The River Flood hazard zones have been updated with the implementation of the new class for return period 50 to extend the variation of return periods and to derive more granular results.

What changes compared to previously implemented River Flood zones? The implementation of the return period 50 class changes the distribution significantly. In general, ~90% of locations which were located in the return period 100 in a globally representative portfolio (global city database with 854,022 cities) will move to the higher return period 50 zone (applies for ~13% of the total locations in the globally portfolio). This might deviate from the changes within your portfolio due to your specific portfolio structure, in particular since River Flood is only relevant for riverine regions.

How will the updates impact your book of business?

A quantitative analysis on the impact of the implementation of the River Flood return period 50 zone for the globally representative portfolio has been conducted. A comparison of the updated River Flood zoning distribution including return period 50 with the previously implemented River Flood zoning distribution with only return periods 0, 500 and 100 shows the following:

  • Zone 0: No change of city share in zone 0 (previously: 82.4% / updated: 82.4%)
  • Zone 500: No significant change of city share in zone 500 (3.4% / 3.4%)
  • Zone 100: Significant change of city share in zone 100 (14.2% / 1.2%)
  • Zone 50: Significant change of city share in zone 50 (- / 13.0%) Please note that the results for your individual portfolio will likely differ from the described results.
Value Zone
0 Zone 0 minimal flood risk
500 Zone 500 year return period
100 Zone 100 year return period
50 Zone 50 year return period

Munich Re’s river flood hazard data (provided by JBA Risk Management) offer state-of-the-art flood hazard information (with a 30m horizontal resolution), available on a global scale. The global flood maps are constantly improved and are a market standard.

They are based on bare-earth digital terrain data and a consistent worldwide digital surface model. The river flood hazard is represented by three return period zones, ranging from Zone 0 (areas of minimal flood risk) to Zone 100 (100 year return period of river flood). Information on the flood defences’ stand- ard of protection (SoP) is available upon request. River flood projections for the years 2030, 2050 and 2100 are available in Munich Re’s Climate Change Edition.

Flood zone Description
Zone 0 Areas outside the 0.2% annual chance floodplain
Zone 50 2% annual chance flood event (50 year return period)
Zone 100 1% annual chance flood event (100 year return period)
Zone 500 0.2% annual chance flood event (500 year return period)

Hazard: Soil and shaking

Underground conditions influencing earthquake intensity (based on geological, soil and hydrological information).

Value Zone
1 Class 1: low, hard bedrock
2 Class 2: rock
3 Class 3: soft rock/dense soil
4 Class 4: stiff soil
5 Class 5: soft soil
6 Class 6: high, reclaimed land

The soil and shaking hazard layer shows underground conditions that influence earthquake intensity.

There are six different classes from 1 (low risk) to 6 (high risk). The classification is based on geological, soil and hydrological datasets such as: − Geological information: geological map of the world, 1:25m, CGMW/UNESCO 2000 − Soil information: digital soil map of the world and derived soil properties, 1:5m, FAO/UNESCO 1997 − Hydrological information: ArcWorld 1:3m cartographic layer: rivers and water bodies (–>RIV3M), ESRI 1992 − Digital elevation model: provided by Shuttle Radar Topography Mission (SRTM) 30m − World map of sediment thickness: by Gaby Laske

This data complements the interpretation of the earthquake perils by elaborating information about how fast earthquake waves move through the ground based on the soil’s natural composition and its impact on the area of interest.

Hazard: Storm surge

Detailed calculation for coasts and the shores of large lakes. Zones based on 90m MERIT Digital Elevation Model (DEM), taking into account wind speed and bathymetry (underwater depth of lake or ocean floors). Does not consider dykes.

Value Zone
-1 No hazard
1000 Zone 1000 year return period
500 Zone 500 year return period
100 Zone 100 year return period

Storm surges can occur along sea coasts if constant strong wind from one direction causes wind setup on the coast, which can measure up to several metres. Therefore in conjunction with the astronomic tide and high seas, extremely high water levels may occur on certain sections of the coast. The geometry of the coast itself plays an important role regarding the exposure to storm surge. The effects of a rise in sea level also depend on the shape of the coast. The flatter the strip of the coast, the more extreme the effects will be.

Munich Re classified the hazard into three categories; zones 100, 500 and 1000. Coasts in Zone 100 are exposed to a 100 year return period of storm surge (1% annual flood chance), those in Zone 500 a 500 year return period (0.2% annual flood chance) and those in Zone 1000 a 1000 year return period (0.1% annual flood chance). The storm surge map is based on ALOS data (version 1.1.; ©JAXA). The inundation area of these return periods were simulated by applying cost-weighted distance tools. Munich Re simulated multiple wave heights for each coast and calculated the maximum expansion. Wind speeds and bathymetry data were also taken into account.

Hazard: Tornado

Frequency and intensity of tornados.

Value Zone
1 Zone 1: low
2 Zone 2
3 Zone 3
4 Zone 4: High

Tornadoes occur worldwide at latitudes between 20° and 60°, but are undeniably most frequent in the USA. Tornadoes are very localized but extremely intense. The direct damage caused by the high wind speeds is exacerbated by the sharp drop in air pressure (10% or more) at the centre of the funnel.

The tornado zones are based on frequency and intensity interpolated from meteorological data. NOAA data serves as a meteorological parameter. The tornado map is a rough estimate of the global situation and is used to identify risk.

Hazard: Tropical cyclones

Probable maximum intensity with an exceedance probability of 10% in 10 years (equivalent to return period of 100 years).

Value Zone
-1 No hazard
0 Zone 0: 76 - 141 km/h
1 Zone 1: 142 - 184 km/h
2 Zone 2: 185 - 212 km/h
3 Zone 3: 213 - 251 km/h
4 Zone 4: 252 - 299 km/h
5 Zone 5: ≥ 300 km/h

Tropical cyclones are among the most destructive weather phenomena. Coastal regions and islands are particularly exposed as they are affected not only by the direct impact of a storm, but also by the secondary hazards, such as storm surges and pounding waves.

The intensity of a storm rapidly decreases as it moves inland because of the friction increase due to the roughness of the Earth’s surface and reduction in the supply of energy (primarily from water vapour) to the storm system. Orographic effects can also lead to high amounts of rainfall, which in turn can result in severe flooding, producing multi-billion dollar losses in populated regions with high GDP.

Tropical cyclones 2

The Tropical Cyclone zoning system uses forward wind, maximum wind speed, minimum central pressure, radius of maximum wind speeds and track of the centre (“eye”) in 3- to 6-hourly intervals (in exceptional cases, 12-hourly intervals) as main variables for modelling. The wind fields of all historical windstorms were simulated and superimposed in a grid network with a mesh size of 0.1 x 0.1 degrees of geographical longitude and latitude. By means of frequency analysis for each grid coordinate, the maximum wind speed to be expected (probable maximum intensity with an average exceedance probability of 10% in 10 years) was derived for the return period of 100 years chosen for the world map. The hazard zoning is represented by a five-level scale (maximum wind speed that can be expected once in 100 years) based on the Saffir-Simpson scale, multiplied by a gust factor of 1.2. Tropical Cyclone projections for the years 2030, 2050 and 2100 are available in Munich Re’s Climate Change Edition.

Hazard: Tsunami

Zones based on 100m SRTM (Version 4.1) elevation model, taking into account height above sea level and distance from coasts.

Value Zone
-1 No hazard
0 Zone 0 minimal flood risk
1000 Zone 1000 year return period
500 Zone 500 year return period
100 Zone 100 year return period

Tsunamis are seismic sea waves and occur after strong seaquakes or large submarine landslides, often induced by earthquakes or volcanic eruptions in the sea or on the coast.

The greatest risk comes from tsunamis generated by meteorites crashing into the sea. This risk exists throughout the world but, with very low occurrence probabilities, is very difficult to quantify and any dis- cussion of this would go beyond the bounds of this account. Tsunami waves spread out in all directions at a great speed which depends on the depth of water. As the waves can travel 10,000 km or more with- out much attenuation, regions that have not experienced any direct earthquake effects can be affected.

Munich Re classified the hazard into four categories; Zone 0, 100, 500 and 1000. Coasts in Zone 100 are exposed to a 100 year return period of tsunamis (1% annual flood chance), those in Zone 500 a 500 year return period (0.2% annual flood chance) and those in Zone 1000 a 1000 year return period (0.1% annual flood chance). Coasts in Zone 0 (minimal flood risk) have a very low tsunami exposure. The tsunami map is based on SRTM data (version 4.1.). The hazard was calculated with the cost-distance function of ESRI ́s ArcGIS. Munich Re simulated multiple wave heights for each coast and calculated the maximum expansion. Historical tsunami and earthquake data were also taken into account.

Hazard: Volcano

Secondary effects that can occur as a result of the large-scale distribution of volcanic particles (e.g. climate impacts, supraregional ash deposits) are not considered.

Value Zone
-1 No hazard
0 Unclassified
1 Zone 1: minor hazard
2 Zone 2: moderate hazard
3 Zone 3: high hazard

The volcano hazard map is based on the activities of volcanoes. All volcanoes are located and mapped by coordinates. Munich Re calculated the volcanic hazard on the basis of the VEI (volcano explosivity index, US Geological Survey) and its annual return periods given for each VEI index.

As far as technically possible, all volcanoes with known VEI data are classified. 719 volcanoes are therefore classified and the other 830 remain unclassified with no information, due to the fact that those volcanoes have not been investigated or are insufficiently investigated. Each of the 719 volcanoes is given three buffer zones with 10km, 50km and 100km radius. Each buffer zone is assigned with an annual return period of being affected by volcanic hazard. For a 10km buffer, VEI 2–7 are considered for the calculation of the return period, VEI 3–7 are considered for a 50km buffer, and VEI 5–7 for a 100km buffer. This is due to the fact that the area around a volcano affected by an eruption corresponds to the explosion intensity, e.g. a small radius area is affected by small to large eruptions while a large radius area is only affected by large eruptions.

The buffer zones are given their different hazard index depending on the range of the return period. The 830 unclassi-fied volcanoes are given a standard buffer of 50km.

The volcano symbol itself derives its hazard index from the mean of the three buffer zone’s annual return periods. The sources used were the reports from the

University of Bristol:

  • Identifying volcanoes with high hazard and economic exposure
  • Frequency-magnitude relationships for active explosive (ash-producing) volcanoes worldwide

Accordingly, the volcanoes were categorized as follows:

  • Zone 0: Unclassified
  • Zone 1: Minor hazard (> 15,000 years return period)
  • Zone 2: Moderate hazard (200 to 15,000 years return period)
  • Zone 3: High hazard (≤ 200 years return period)

There are several types of hazard associated with volcanoes, the principal hazards being:

  • Ballistic debris av.
  • Shockwaves
  • Lava flows
  • Pyroclastic flows
  • Gases
  • Lahars
  • Lightning
  • Acid rain
  • Tephra fall

It is difficult to assess all the different types of hazard due to volcanism and classify their respective importance for the actual level of risk. As eruptions are typically rare events and systematic investigations on damage-related hazard parameters have just started in the recent past, an absolute measure of volcanic risk is prone to larger uncertainties. However, a relative measure of risk caused by different types of volcanic eruptions, their strengths and return periods seems to be a valid choice for volcanic risk classification for the moment.

Hazard: Wildfire

The effects of wind, arson and fire-prevention measures are not considered.

Value Zone
-1 No hazard
1 Zone 1: low
2 Zone 2
3 Zone 3
4 Zone 4: high

Wildfires are the result of a complex interaction between certain influencing factors, e.g. ignition of the fire, vegetation, meteorological conditions (El Niño/La Niña) and topography.

The wildfire map is based on climatological historical data and GlobeCover (ESA) land cover data:

  • Wildfires are rare in areas where rain is frequent
  • Regions with sparse vegetation are also unlikely to be affected by wildfire
  • Wildfire potential is particularly high when coniferous forests are exposed to dry spells lasting several weeks or even months

The model does not replace a probabilistic model, but it is nevertheless of great value in identifying areas at risk.