Green’s functions databases for Mars

The ETH Zurich stores precomputed Green’s functions databases produced for 16 Martian structural background models. In addition, we distribute 13 models from the literature both with 1 s. and 5 s minimum reliable period. These databases were computed using AxiSEM (Nissen-Meyer et al., 2014), an axi-symmetric spectral element method which runs within a 2D computational domain. The synthetic seismograms for any source/event pair can be calculated using Instaseis (van Driel et al., 2015), a python library built with a fast extraction algorithm for waveforms. Instaseis makes heavy use of Obspy Python package.

To see a list of models for computing the Green’s functions databases, click on the Green’s functions databases tab below. Instructions on extracting waveforms from any of these databases are given under the How to download waveforms tab. The databases are hosted at http://instaseis.ethz.ch/marssynthetics/, and can be accessed using a database name from the table below combined with this URL. For convenience, complete connection URLs are also provided. We also host a pre-calculated set of synthetics

This work was supported by grants from the Swiss National Science Foundation (NSF-ANR project 157133 “Seismology on Mars”) and the Swiss National Supercomputing Center (CSCS) under project ID s682.

Disclaimer: This dataset was prepared as part of the InSight project for the InSight blind test as documented in Ceylan et al (SSR, 2017). If you would like to use any of the material provided within the scope of this test (models, catalogues, waveforms) in a study that may lead to a publication, please contact mqs@erdw.ethz.ch. For any potential users from within the InSight project, note any publications that use this dataset are governed by the InSight 'Rules of the Road'.

4 May 2018: New GF databases are added into our suite for 13 models from the literature (both 1 s and 5 s).

27 July 2017: Non functional GF databases are now active.

29 July 2017: Two GF databases with reliable period of 5 s. are added into our suite. Check out the main page for URLs.

If you use our web services, please cite the study below:

Ceylan, S., M. van Driel, F. Euchner, A. Khan, J. F. Clinton, L. Krischer, M. Boese, S. Staehler, and D. Giardini (2017). From initial models of seismicity, structure and noise to synthetic seismograms for Mars, Space Science Reviews, InSight Special Issue, doi:10.1007/s11214-017-0380-6.

You might also consider citing the ObsPy, AxiSEM, and Instaseis papers:

Krischer, L., T. Megies, R. Barsch, M. Beyreuther, T. Lecocq, C. Caudron, and J. Wassermann, 2015. ObsPy: a bridge for seismology into the scientific Python ecosystem, Computational Science & Discovery, 8(1), 014003, doi: 10.1088/1749-4699/8/1/014003.

Nissen-Meyer, T., van Driel, M., Stähler, S. C., Hosseini, K., Hempel, S., Auer, L., Colombi, A., and Fournier, A., 2014. AxiSEM: broadband 3-D seismic wavefields in axisymmetric media, Solid Earth, 5, 425-445, doi:10.5194/se-5-425-2014.

van Driel, M., Krischer, L., Stähler, S. C., Hosseini, K., and Nissen-Meyer, T., 2015. Instaseis: instant global seismograms based on a broadband waveform database, Solid Earth, 6, 701-717, http://dx.doi.org/10.5194/se-6-701-2015.

Other studies we use:

Bertka, C. M., Fei, Y., 1997. Mineralogy of the Martian interior up to core-mantle boundary pressures. J. Geophys. Res. 102, 5251–5264.

Khan et al., 2016. Single station and single-event marsquake location and inversion for structure using synthetic Martian waveforms, in press, Phys. Earth Planet. Int.

Sanloup, C., A. Jambon, and P. Gillet, 1999. A simple chondritic model of Mars, Phys. Earth Planet. Int., 112, 43.

Taylor, G. J., 2013. The bulk composition of Mars. Chemie der Erde / Geochemistry 73, 401–420.

Blind test Green's functions databases:

These are the candidate models that were included in the recent 2018 MQS blind test. More details about these models and the MQS blind test can be found in Clinton et al.(2017). Each database is created for minimum reliable periods of 1 s. and 5 s. The suffix 'b' at the end of the some model names stands for bedrock. If you are interested in other resources regarding the models or MQS blind test, please visit this web page (Note that the page is not regularly maintained any more).

Database name	Min. period (s)	Record length(min.)	Model data	Complete URL for database	Current status
LFAK_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/LFAK_1s/	Active
LFAK_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/LFAK_5s/	Active
MAAK_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/MAAK_1s/	Active
MAAK_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/MAAK_5s/	Active
SANAK_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/SANAK_1s/	Active
SANAK_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/SANAK_5s/	Active
TAYAK_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/TAYAK_1s/	Active
TAYAK_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/TAYAK_5s/	Active
DWAK_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/DWAK_1s/	Active
DWAK_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/DWAK_5s/	Active
DWThot_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/DWThot_1s/	Active
DWThot_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/DWThot_5s/	Active
DWThotCrust1_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/DWThotCrust1_1s/	Active
DWThotCrust1_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/DWThotCrust1_5s/	Active
DWThotCrust1b_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/DWThotCrust1b_1s/	Active
DWThotCrust1b_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/DWThotCrust1b_5s/	Active
DWThotCrust2_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/DWThotCrust2_1s/	Active
DWThotCrust2_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/DWThotCrust2_5s/	Active
DWThotCrust2b_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/DWThotCrust2b_1s/	Active
DWThotCrust2b_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/DWThotCrust2b_5s/	Active
EH45Tcold_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/EH45Tcold_1s/	Active
EH45Tcold_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/EH45Tcold_5s/	Active
EH45TcoldCrust1_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/EH45TcoldCrust1_1s/	Active
EH45TcoldCrust1_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/EH45TcoldCrust1_5s/	Active
EH45TcoldCrust1b_1s	1	~30	axisem, nd or deck	http://instaseis.ethz.ch/blindtest_1s/EH45TcoldCrust1b_1s/	Active
EH45TcoldCrust1b_5s	5	~300		http://instaseis.ethz.ch/blindtest_5s/EH45TcoldCrust1b_5s/	Active

Summary of the Green's functions databases from Ceylan et al. (2017):

Abbreviations used for mantle geotherm and composition columns are T13: Taylor (2013), SL99: Sanloup et al (1999), BF97: Bertka and Fei (1997), and AK: Khan et al. (2016). Model data are provided in png, nd or deck formats. The columns in the nd files are depth(km), Vp(km/s), Vs(km/s), density(g/cm^3), Qmu and Qkappa. The deck format includes radius(m), density(kg/m^3), Vp(m/s), Vs(m/s), Qkappa, Qmu, V1(m/s), V2(m/s), and iso/aniso columns in order. V1 and V2 are other velocities which we do not use; therefore, we duplicate the values of Vp and Vs.

Crustal thickness (km)	Crustal Vp(km/s)	Mantle geotherm	Mantle mineralogy	Min. period (s)	Record length(min.)	Database name	Model data	Complete URL for database	Current status
30	4.95 (high)	BF97	T13	1	~30	C30VH-BFT13-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C30VH-BFT13-1s	Active
30	4.95 (high)	BF97	SL99	1	~30	C30VH-BFSNL-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C30VH-BFSNL-1s	Active
30	4.95 (high)	AK	T13	1	~30	C30VH-AKT13-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C30VH-AKT13-1s	Active
30	4.95 (high)	AK	SL99	1	~30	C30VH-AKSNL-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C30VH-AKSNL-1s	Active
30	3.60 (low)	BF97	T13	1	~30	C30VL-BFT13-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C30VL-BFT13-1s	Active
30	3.60 (low)	BF97	SL99	1	~30	C30VL-BFSNL-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C30VL-BFSNL-1s	Active
30	3.60 (low)	AK	T13	1	~30	C30VL-AKT13-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C30VL-AKT13-1s	Active
30	3.60 (low)	AK	SL99	1	~30	C30VL-AKSNL-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C30VL-AKSNL-1s	Active
80	4.95 (high)	BF97	T13	1	~30	C80VH-BFT13-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C80VH-BFT13-1s	Active
80	4.95 (high)	BF97	SL99	1	~30	C80VH-BFSNL-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C80VH-BFSNL-1s	Active
80	4.95 (high)	AK	T13	1	~30	C80VH-AKT13-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C80VH-AKT13-1s	Active
80	4.95 (high)	AK	SL99	1	~30	C80VH-AKSNL-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C80VH-AKSNL-1s	Active
80	3.60 (low)	BF97	T13	1	~30	C80VL-BFT13-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C80VL-BFT13-1s	Active
80	3.60 (low)	BF97	SL99	1	~30	C80VL-BFSNL-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C80VL-BFSNL-1s	Active
80	3.60 (low)	AK	T13	1	~30	C80VL-AKT13-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C80VL-AKT13-1s	Active
80	3.60 (low)	AK	SL99	1	~30	C80VL-AKSNL-1s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/C80VL-AKSNL-1s	Active
--	--	AK	DW	5	~300	ModelDW-BFGeoth_5s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/ModelDW-BFGeoth_5s	Active
--	--	AK	DW	5	~300	ModelDW-AKGeoth_5s	png, nd or deck	http://instaseis.ethz.ch/marssynthetics/ModelDW-AKGeoth_5s	Active

Use your own Python scripts [example code]:

We recommend developing your own Python scripts for generating synthetics. If you choose to use your own scripts, simply import Instaseis and connect to one of our databases remotely. Note that our Instaseis servers compute the waveforms in displacement units by default. Most of the default parameters can be overridden in your scripts according to your needs. Consult to Instaseis webpage for more info.

			/usr/bin/python
			>>> import obspy
			>>> import instaseis
			>>> 
			>>> db = instaseis.open_db("http://instaseis.ethz.ch/marssynthetics/C30VH-BFT13-1s")
			>>> print(db)
			RemoteInstaseisDB reciprocal Green's function Database (v7) generated with these parameters:
				components           : vertical and horizontal
				velocity model       : external
				attenuation          : True
				dominant period      : 1.000 s
				dump type            : displ_only
				excitation type      : dipole
				time step            : 0.230 s
				sampling rate        : 4.346 Hz
				number of samples    : 7823
				seismogram length    : 1799.9 s
				source time function : gauss_0
				source shift         : 1.611 s
				spatial order        : 4
				min/max radius       : 3289.0 - 3389.5 km
				Planet radius        : 3389.5 km
				min/max distance     : 0.0 - 180.0 deg
				time stepping scheme : symplec4
				compiler/user        : gfortran 4.9.2 20141030 (Cray Inc.) by vamartin on dora22
				directory/url        : http://instaseis.ethz.ch/marssynthetics/C30VH-BFT13-1s
				size of netCDF files : 564.9 GB
				generated by AxiSEM version v1.2-11-g8461-dirty at 2016-03-11T12:50:33.000000Z 
			>>>

Define a source/receiver pair using the moment tensor solution and location parameters. Use a whole ObsPy.Catalog for this purpose in a loop as you wish.

			>>> receiver = instaseis.Receiver(
			                 latitude=4.5, longitude=136.0, network="7J", station="SYNT1")

			>>> source = instaseis.Source(
			                 latitude=30.0, longitude=30.0, depth_in_m=50000,
			                 m_rr=1.710000e+24,
			                 m_tt=1.810000e+22,
			                 m_pp=1.740000e+24,
			                 m_rt=1.990000e+23,
			                 m_rp=1.050000e+23,
			                 m_tp=1.230000e+24,
			                 origin_time=obspy.UTCDateTime(2020, 1, 2, 0, 0, 0))

Retrieve the waveforms.

			>>> seismogram = db.get_seismograms(source=source, receiver=receiver)

Do some more processing as needed...

			
			>>> print(seismogram)
			>>> seismogram.plot()

Use REST-like API of Instaseis:

Instaseis currently supports nine back-end commands. Check http://www.instaseis.net/routes/seismograms_raw.html and http://www.instaseis.net/routes/seismograms.html webpages for a list of parameters needed to compute waveforms. You can use these commands within your Python scripts, or use wget to download the data.

Use Instaseis GUI:

You also have the option to view synthetics for any source type using the Instaseis GUI. This tool is useful especially for exploring the waveforms before generating a set of them. To start the GUI, simply type the command below in your command screen.

			
			> python -m instaseis.gui