KEM-35 Public seismic hazard and risk model Groningen version 6

In 2019, a public version of the Groningen seismic hazard and risk model (TNO HRA model) was delivered by TNO in the KEM-03b and KEM-10 project. This version is functionally a 1:1 replication of the NAM HRA version 5, including all assumptions and choices made, and is able to reproduce the outcome of the NAM HRA, while performing faster due to a slightly different statistical approach.

For the hazard and risk assessment of 2020 NAM uses updated models for the hazard and risk calculation. For the seismological model version 6, for the ground motion model version 6 en for the damage module version 7 (version 6 was applied in 2019).
As the model chain of TNO should be used in calculations and evaluations of hazard and risk, the models need to be updated to the latest models that are being used.

Specific research questions are:
1. What is the effect of the new updated models on hazard and risk assessment in Groningen?
2. Is the hazard and risk calculated with the updated public model of TNO comparable to the NAM results? And is not, how can the differences be explained?
3. Which improvements (as recommended by NAM or various KEM projects) should to be included in a next version of the TNO HRA model?
4. What is the effect off the outcomes of KEM-04/KEM-02 projects on the hazard and risk assessment?
This project is commissioned in view of having a public independent probabilistic seismic HRA for Groningen.

RESEARCH REQUEST FULL TEXT

KEM 35 was appointed to upgrade the hazard and risk assessment (HRA) made by TNO for Groningen based on the latest available models developed by NAM regarding: the seismological source model, the ground motion model, the damage and consequence model. The methods employed to produce the results, that is, numerical integration rather than simulation, appear superior with respect to those of NAM. The project team updated the seismic hazard and risk assessment (HRA) for Groningen with the latest available (version 6) models from NAM.

RESEARCH PROJECT REPORT

As expected, the results of the risk assessment are comparable to (i.e., match) those of NAM because TNO employs the same models. The research questions are fully addressed for what concerns the reproduction of the NAM results. In fact, TNO was able to spot some issues that could be better addressed, computationally, by the NAM procedure. Differences in the results are likely attributable to numerical approximations. The results are relevant as the testify public competency building on the topic on seismic hazard and risk assessment for Groningen. The results from this project provide the basis for the next step which will be the ownership, by TNO, of further developments of the probabilistic seismic HRA, which is more complicated.

The fourth RQ “What is the effect off the outcomes of KEM-04/02 on the hazard and risk assessment?” has intentionally not been addressed in this project.

The projects was evaluated by the KEM scientific expeert panel.

RESEARCH PROJECT EVALUATION

The quality of the work made by TNO is excellent and it proves its continuously growing confidence in managing the hazard and risk assessment for Groningen. It appears that, from the risk evaluation point of view, TNO is now mature to let the HRA go public. TNO is now ready to perform autonomously HRA, although based on models already made available from others (i.e., NAM).
A proposal by the Ministry to evaluate future developments on the TNO PSHRA with an extended KEM expert team (KEM subpanel) is fully supported by KEM.

On November 19 a KEM DeepNL presentation on the KEM-35 project was given by TNO during the DeepNL annual meeting.