Nabro 3D velocity model produced by the FMTOMO code

These files relate to "Seismic tomography of Nabro caldera, Eritrea: insights into the magmatic and hydrothermal systems of a recently erupted volcano" by Gauntlett et al., 2023. 

Original seismic waveforms are from the Nabro Urgency Array (Hammond et al., 2011; https://doi.org/10.7914/SN/4H_2011), which is publicly available through IRIS Data Services (http://service.iris.edu/fdsnws/dataselect/1/). See Hammond et al. (2011) for further details on waveform data access and availability. Sources were originally located by Lapins et al. (2021), and the full catalogue is archived here (Lapins, 2022). 

The FMTOMO package is freely available to download at http://rses.anu.edu.au/~nick/fmtomo.html. 

This repository contains three 3D velocity models:

1. vp_model.out
2. vs_model.out
3. vpvs_model.out

These plain text files comprise the output of the FMTOMO tomography algorithm, which inverts for 3D P-wave velocity and S-wave velocity Vp and Vs) structure and Vp/Vs ratio for a grid centred around Nabro volcano. The velocity is given in km/s for Vp and Vs, and the values are dimensionless for Vp/Vs ratio.

The grid is specified in the first four lines of the file:

• Line 1: for these data, this line will always hold the values `1' and `1'. 
• Line 2: the number of grid nodes in radius (depth), latitude and longitude. 
• Line 3: the radial (depth) node spacing in km, latitude spacing in radians, longitude spacing in radians.
• Line 4: the grid origin radius (km), latitude (radians) and longitude (radians).

Each node of the grid has a P-wave, S-wave and Vp/Vs ratio associated with it, which is specified in lines 5 onward.

• Line 5 - 55229: Value of Vp, Vs or Vp/Vs (given by file name) at each node.

The values loop over the grid, with longitude varying first, and radius last. The first node is at the grid origin. The second node is at the grid origin, plus the grid spacing in longitude. This continues for the `nlon` longitude nodes. The `nlon+1`the point is then at the grid origin, plus the latitude spacing; and so on.

If there are `nr` radial nodes, `nlat` latitude nodes and `nlon` longitude nodes, then the following pseudocode shows how to read lines 5 forwards using the imaginary function `readline`, which reads a single real value from a plain text file and moves to the next line:

```

# Comment: have already read the first four lines

for ir in 1:nr:

for ilat in 1:nlat:

for ilon in 1:nlon:

grid[ir,ilat,ilon] = readline(file_handle)

```

The repository also contains the event catalogue, with hypocenter locations after relocation by the FMTOMO code:

4. event_catalogue.csv

The columns are depth in km (negative values indicate depths below sea level, positive values indicate depths above sea level), latitude in degrees, longitude in degrees, depth error (km), latitude error (km), longitude error (km).