OpenSHA

Open-Source Seismic Hazard Analysis (OpenSHA)

Modular Fault System Solution

PREVIEW: This is a preview of the proposed file format. UCERF3 files are still stored in the Legacy Fault System Solution format, but branch averaged solutions in this format are available to download for testing: FM 3.1, FM 3.2, FMs Combined.

The UCERF3 model introduced Fault System Rupture Sets and Solutions as data containers for earthquake rupture forecasts. A Rupture Set defines all of the on-fault supra-seismogenic ruptures in a fault system, and their properties (magnitude, rake, etc). A Solution defines the annual rate of occurrence of each rupture, and may also supply information about gridded seismicity.

Data are stored in a zip file consisting primarily of JSON, GeoJSON, and CSV files for ease of use. Rupture sets and solutions are stored in the ruptures and solution subdirectories of the zip file, respectively. OpenSHA codes will include a modules.json file at the top level of the zip file that includes information used by OpenSHA to load in the Rupture Set and Solution, but this can be omitted for externally-created files.

Example files presented here are often shortened for brevity, indicated by ....

A small but complete Fault System Solution (that, by definition, also contains a Rupture Set) can be downloaded here for testing.

Files in this format can be loaded in nightly builds of OpenSHA application, or via code in OpenSHA via the load(File) method on org.opensha.sha.earthquake.faultSysSolution.FaultSystemSolution and org.opensha.sha.earthquake.faultSysSolution.FaultSystemRupSet.

Table of Contents

Fault System Rupture Set

(return to top)

Fault System Rupture Sets define a set of fault sections and supra-seismogenic ruptures (and their properties) on those sections. Their data are stored in the ruptures subdirectory of a zip file.

Here is a summary of files likely to be in a rupture set zip file:

File Name Required? Format Description
ruptures/fault_sections.geojson YES GeoJSON Fault section geometries
ruptures/indices.csv YES CSV Lists of section indices that comprise each rupture
ruptures/properties.csv YES CSV Rupture properties (mag, rake, length, area)
ruptures/average_slips.csv (no) CSV Average slip information for each rupture
ruptures/modules.json (no) JSON Manifest of Rupture Set modules, used by OpenSHA

Fault Section Data

(return to top)

Rupture sets usually contain a large number of fault subsections, which are small equal-length subdivisions of each parent fault section (typically with length approximately equal to half the siesmogenic thickness of the fault). Those subsections are stored in the Fault Section GeoJSON format with the added requirement that each section be listed in order of their ids, starting with id=0 and ending with id=(numSections-1). The GeoJSON will be stored in ruptures/fault_sections.geojson.

Here is an example fault section data file with 9 subsections spanning two faults:

{
  "type": "FeatureCollection",
  "features": [
    {
      "type": "Feature",
      "id": 0,
      "properties": {
        "FaultID": 0,
        "FaultName": "Demo S-S Fault, Subsection 0",
        "DipDeg": 90.0,
        "Rake": 180.0,
        "LowDepth": 12.0,
        "UpDepth": 0.0,
        "DipDir": 0.0,
        "AseismicSlipFactor": 0.0,
        "CouplingCoeff": 1.0,
        "SlipRate": 10.0,
        "ParentID": 11,
        "ParentName": "Demo S-S Fault",
        "SlipRateStdDev": 1.0
      },
      "geometry": {
        "type": "LineString",
        "coordinates": [
          [
            -118.0,
            34.7
          ],
          [
            -118.00000000000001,
            34.75
          ]
        ]
      }
    },
...
    {
      "type": "Feature",
      "id": 8,
      "properties": {
        "FaultID": 8,
        "FaultName": "Demo Reverse Fault, Subsection 2",
        "DipDeg": 45.0,
        "Rake": 90.0,
        "LowDepth": 12.0,
        "UpDepth": 0.0,
        "DipDir": 0.0,
        "AseismicSlipFactor": 0.0,
        "CouplingCoeff": 1.0,
        "SlipRate": 3.0,
        "ParentID": 25,
        "ParentName": "Demo Reverse Fault",
        "SlipRateStdDev": 0.5
      },
      "geometry": {
        "type": "LineString",
        "coordinates": [
          [
            -118.29993824145802,
            35.300020582168905
          ],
          [
            -118.35,
            35.35
          ]
        ]
      }
    }
  ]
}

Rupture Section Indices

(return to top)

The ruptures/indices.csv file lists the participating subsections for each rupture. It is stored in a CSV file. The first row of the CSV file shall contain column headings, but the content of the header is not checked and need not exactly match the example given below (for example, the ‘# 1’, ‘# 2’, etc., columns are included here for readability but can be omitted). Ruptures should then be listed in order, and the first rupture shall be index 0.

The participating subsections are indicated by their (0-based) index, so for the example below, rupture 0 consists of subsections 0 and 1, and rupture 1 consists of subsections 0, 1, and 2. The total number of columns in the CSV file is equal to the number of sections in the largest rupture plus 2, and each line may have different column counts.

An example is given below with 28 ruptures on the 9 previously defined fault subsections:

Rupture Index Num Sections # 1 # 2 # 3 # 4 # 5 # 6 # 7 # 8 # 9
0 2 0 1              
1 3 0 1 2            
2 4 0 1 2 3          
3 5 0 1 2 3 4        
4 6 0 1 2 3 4 5      
5 8 0 1 2 3 4 5 6 7  
6 9 0 1 2 3 4 5 6 7 8
7 2 1 2              
8 3 1 2 3            
9 4 1 2 3 4          
10 5 1 2 3 4 5        
… … … … … … … … … … …
27 2 7 8              

Rupture Properties

(return to top)

Rupture properties are stored in ruptures/properties.csv which gives the magnitude, rake, area, and length of each rupture. It is stored in a CSV file. The first row of the CSV file shall contain column headings, but the content of the header is not checked and need not exactly match the example given below. Ruptures should then be listed in order, and the first rupture shall be index 0.

An example is given below with 28 ruptures on the 9 previously defined fault subsections:

Rupture Index Magnitude Average Rake (degrees) Area (m^2) Length (m)
0 6.105 180.0 1.33434E8 11119.5
1 6.329 180.0 2.00151E8 16679.3
2 6.496 180.0 2.66868E8 22239.0
3 6.625 180.0 3.33585E8 27798.8
4 6.73 180.0 4.00302E8 33358.5
5 6.972 148.7 6.43906E8 47713.0
6 7.062 137.6 7.65707E8 54890.2
7 6.105 180.0 1.33434E8 11119.5
8 6.329 180.0 2.00151E8 16679.3
9 6.496 180.0 2.66868E8 22239.0
10 6.625 180.0 3.33585E8 27798.8
… … … … …
27 6.367 90.0 2.43603E8 14354.5

Fault System Solution

(return to top)

Fault System Solutions define the rate of each rupture from a Rupture Set (called a ‘solution’ because those rates are usually the result of an inversion). They may optionally also include gridded seismicity data. Their data are stored in the solution subdirectory of a zip file.

A solution must also contain a rupture set (in the ruptures top-level subdirectory). In addition to the standard rupture set files, here is a summary of files likely to be in a solution zip file:

File Name Required? Format Description
solution/rates.csv YES CSV Annual rates for each rupture
solution/grid_mech_weights.csv (no) CSV Focal mechanism weights for each gridded seismicity location
solution/grid_region.geojson (no) GeoJSON Gridded seismicity region
solution/grid_sub_seis_mfds.csv (no) CSV Sub-seismogenic MFDs for gridded seismicity
solution/grid_unassociated_mfds.csv (no) CSV Gridded seismicity MFDs that are not associated with any fault
solution/modules.json (no) JSON Manifest of Solution modules, used by OpenSHA

Rate Data

(return to top)

Solution annual rate data for each rupture is stored in a simple 2-column CSV file, solution/rates.csv. The first row of the CSV file shall contain column headings, but the content of the header is not checked and need not exactly match the example given below. Ruptures should then be listed in order, and the first rupture shall be index 0. An example is below:

Rupture Index Annual Rate
0 0.005320239194616782
1 0.005298660169966246
2 7.834210706407277E-4
3 1.8703779639406976E-7
4 0.0026397968587955787
5 1.4234970401212632E-8
6 5.675850061305903E-4
7 9.987701252958843E-9
8 3.866169289703722E-7
9 3.636288131991794E-7
10 4.219820930141324E-7
… …
27 3.7916393801626976E-7

Gridded Seismicity Data

(return to top)

Solutions may optionally provided gridded seismicity information. For a fault system solution, gridded seismicity can refer to either off-fault earthquakes (those ‘unassociated’ with any fault) or sub-seismogenic ruptures on a fault (see UCERF3 reports for additional information). This data is stored in 4 files, each of which is summarized below.

Gridded Seismicity Region

(return to top)

The gridded region used to define the set of gridded seismicity locations is stored in solution/grid_region.geojson. It follows the OpenSHA Gridded Region File Format, and is omitted here for brevity, but the region used for the examples below has 81 grid nodes.

This file is optional. If omitted, the region will be inferred from grid nodes supplied in the focal mechanisms CSV file, but note that the grid nodes must be evenly spaced in latitude and longitude and not contain any holes (though it can be irregularly shaped)

Gridded Seismicity Focal Mechanism Rates

(return to top)

Each gridded seismicity node can have ruptures of various focal mechanisms: strike-slip, normal, and reverse. This file gives the fraction of seismicity associated with each node that corresponds to each of those focal mechanisms. This data is stored in solution/grid_mech_weights.csv and the format is given below. The first row of the CSV file shall contain column headings, but the content of the header is not checked and need not exactly match the example given below. Grid nodes should then be listed in order, and the first node shall be index 0.

Node Index Latitude Longitude Fraction Strike-Slip Fraction Reverse Fraction Normal
0 34.0 -120.0 0.5 0.25 0.25
1 34.0 -119.75 0.5 0.25 0.25
2 34.0 -119.5 0.5 0.25 0.25
3 34.0 -119.25 0.5 0.25 0.25
4 34.0 -119.0 0.5 0.25 0.25
5 34.0 -118.75 0.5 0.25 0.25
6 34.0 -118.5 0.5 0.25 0.25
7 34.0 -118.25 0.5 0.25 0.25
8 34.0 -118.0 0.5 0.25 0.25
9 34.25 -120.0 0.5 0.25 0.25
10 34.25 -119.75 0.5 0.25 0.25
… … … … … …
80 36.0 -118.0 0.5 0.25 0.25

Gridded Seismicity MFDs

(return to top)

Magnitude-Frequency distributions (MFDs) for each grid node are stored in a CSV file format. Each grid node can have 2 MFDs, one for sub-seismogenic ruptures associated with a fault (solution/grid_sub_seis_mfds.csv), and another for ruptures unassociated with any fault (solution/grid_unassociated_mfds.csv), but many grid nodes will only have 1 of those types.

The format for each file is identical, and the first row of each CSV file shall contain column headings which are used to define the X-values of the MFD (this may vary from model to model). Grid nodes need not be listed in order and can be skipped if no MFD exists for a particular node (or, alternatively, rows can be left blank after the 3-column row header to indicate that no MFD exists for a node).

Here is an example file that contains MFDs for some nodes and omits them for others:

Node Index Latitude Longitude 5.05 5.15 5.25 5.35 … 8.4
8 34 -118 0.0381873 0.0303332 0.0240946 0.019139 … 0.0
17 34.25 -118 0.0311113 0.0247126 0.0196299 0.0155926 … 0.0
26 34.5 -118 0.0381873 0.0303332 0.0240946 0.019139 … 0.0
35 34.75 -118 0.0381873 0.0303332 0.0240946 0.019139 … 0.0
44 35 -118 0.0381873 0.0303332 0.0240946 0.019139 … 0.0
52 35.25 -118.25 0.00806943 0.00640978 0.00509147 0.0040443 … 0.0
53 35.25 -118 0.0381873 0.0303332 0.0240946 0.019139 … 0.0
59 35.5 -118.75 0.00806943 0.00640978 0.00509147 0.0040443 … 0.0
60 35.5 -118.5 0.00807135 0.0064113 0.00509268 0.00404526 … 0.0
62 35.5 -118 0.0381873 0.0303332 0.0240946 0.019139 … 0.0
67 35.75 -119 0.00806943 0.00640978 0.00509147 0.0040443 … 0.0
68 35.75 -118.75 0.00807135 0.0064113 0.00509268 0.00404526 … 0.0
71 35.75 -118 0.0381873 0.0303332 0.0240946 0.019139 … 0.0
75 36 -119.25 0.00806943 0.00640978 0.00509147 0.0040443 … 0.0
76 36 -119 0.00807135 0.0064113 0.00509268 0.00404526 … 0.0
80 36 -118 0.0381873 0.0303332 0.0240946 0.019139 … 0.0

Copyright ©2022 University of Southern California. All rights reserved. License—Disclaimer

This website is generated automatically from the OpenSHA wiki on GitHub, and is powered by GitHub Pages, Jekyll, and this GitHub Action.