- General Characteristics
The GEOMOD Scenario Model provides simulations of the expected surface deformation pattern due to a wide range of geohazard deformation sources (seismic fault, magmatic cameras, eruptive dykes). It also performs soft computing operations to retrieve the geohazard source parameters through inversion frameworks. It is based on InSAR data and, possibly, on GNSS and other ground data (such as leveling data). Its processing chain exploits the following Geohazard DTC modules: INSAR (producing InSAR datasets), DATAINT (performing integration/fusion of InSAR and GNSS data), DECOMP (retrieving multiple signals - e.g., long- or short- term - from the InSAR or GNSS measurements), and GEOMOD (the actual module performing modelling scenarios, based on the available deformation data).
Fig. 1: Schematic representation of the GEOMOD scenario model.
- Specific Products Description
The products are the parameters of the volcanic or seismic source, for example the position, depth, volume variation or rupture mechanism and intensity. Also, the predicted displacements are datasets of vector type. Other products are graphic images showing a comparison between observed (input) data, modelled (predicted) data, and residuals (observed minus predicted). GEOMOD also accesses the products made available through the EPOS portal in the TCS-Satellite Data framework. Seismic source models are generated by an automated processing chain based on the EPOSAR service, to retrieve the event parameters for significant earthquakes (M 6+) at global scale. The products also include predicted deformation scenarios generated soon after the earthquake. Product ingestion will exploit the web services accessible at https://finitesource.ingv.it/v3/query.
- Products Availability (spatial resolution, spatial coverage, temporal resolution, delay, error etc.)
The products can be delivered a few minutes-hours after the beginning of the run. There is a pre-processing phase in which the modelling is set up, based on the specific characteristics of the studied area. The accuracy depends on the input data and on how well they can be reproduced by a model. A reasonable accuracy should provide residuals within 3-5 times the error associated with the input data. In case of significant earthquakes, the surface displacement scenario is available minutes/hours after the event, and the actual modelling 1-2h after the Sentinel-1 post-event image, according also to the image provider timing.
Fig. 2: Examples of possible scenarios from GEOMOD. Left: slip inversion on the faults responsible for the 2016 Central Italy earthquake. Right: syn-eruptive INSAR results (Sentinel-1 ascending orbit) and modelled eruptive dike (shaded rectangle) related to the December 2018 eruption at Mt Etna.