Data Services Products: globalstacks_supplement


Files of the global stacks and vespagrams for users wanting to analyze them or generate plots.


This page contains supplementary material including very large .zip files containing all of the files needed should a user want to analyze the stacks or vespagrams or generate plots themselves. Also, some additional figures and animations are available here that likely would interest a more limited audience relative to the images on the main globalstacks page.

Global stacks files

Each zip file has 360 BHZ global stack files, one for each half degree wide distance bin. Zip files are organized by the narrow band-pass filter applied.

Columns within each file are:

  1. time from origin. 0-5400 s for BHZ, 0-10800 s for LH?
  2. length of STA window (s). The STA/LTA functions are delayed by this amount.
  3. number of contributing traces
  4. number of traces rejected by QCing
  5. raw stack of the filtered & QCed traces
  6. raw stack of the filtered & QCed traces using square-root stacking
  7. stack of STA/LTA functions
  8. stack of STA/LTA functions using square-root stacking (preferred for plotting)
  9. stack of envelope functions of the filtered & QCed traces

DIY stacks

Make your own plots: a MATLAB starter script for simple plotting of the stacks PlotGlobalStacksBasic.m. Some published global body wave stacking papers are given below in the Citation section for reference.

Make your own stacks with your own data & processing parameters. This quick and dirty stacking exercise uses Fetch scripts and MATLAB to let users collect & process data and make stacks at a single distance.

Broadband vertical (30MB)

Long period vertical, radial and transverse

Vespagrams (slowness stacks)

Vespagrams are an array analysis tool where traces from stations assumed to be along a common backazimuth and within a narrow distance range are stacked at different slownesses.

Animation of vespagrams made using USArray data in different frequency bands sweeping through different distances

In the frame shown (distance = 60.0 deg) all data from all events with distances between 59.5 – 60.5 degrees were slant stacked along 126 different slownesses to make this one frame. The different panels show vespagrams using different narrow band-pass filters. The animation sweeps through 360 different distance bins, one every half degree.

The vespagrams were generated using all available BHZ data from the BK, CI, TA, US, and UW networks. Data were analyzed in marching 5 degree wide distance bins every 0.5 degrees. Within each distance bin and in different frequency bands, data were culled after calculating basic QC metrics, then each trace was shifted in time according to the average delay from a multi-channel cross-correlation analysis windowed on the first arriving P-wave. Then each vespagram was normalized by the peak amplitude and then further ‘median-normalized’ by iteratively raising the entire trace to the power 0.9 until the median value exceeded a threshold. This final step during event processing ensures a common noise level before stacking vespagrams from all events. Finally, vespagrams from each event within each frequency band are stacked. The dominant phases in the resulting stacks still have a very high signal-to-noise ratio, so a nonlinear color scale is used in the animation in order to highlight weaker phases.

Vespagram files

Each zip file has vespagrams for 0.5 degree wide distance bins. The files are 743400 lines long, 126 slownesses (-10:0.2:15 s/deg), 5900 times (-500:1:5400 s around the origin time), and have four columns: linear stacking, square-root stacking, linear stacking using a higher threshold (higher background noise levels/lower SNR for peak arrivals), square-root stacking using a higher threshold. (1.7 GB)

Comparison of results with those from Astiz et al. 1996.

Rough comparison of the results from this project (left) versus those from the original in 1996 (right). Broadband vertical component STA/LTA stacks are shown. The nonlinear 2013 stacks use data narrowly filtered around 6 sec period and have a 120 sec wide automatic gain control window applied. The stacks on the right are taken from Astiz et al. and are high pass filtered above 6 sec.

Ray paths animation

Animation showing paths of different seismic phases using an earlier global stacks image.


Citations and DOIs

To cite the IRIS DMC Data Products effort:

  • Trabant, C., A. R. Hutko, M. Bahavar, R. Karstens, T. Ahern, and R. Aster (2012), Data Products at the IRIS DMC: Stepping Stones for Research and Other Applications, Seismological Research Letters, 83(5), 846–854, doi:10.1785/0220120032.

To cite the IRIS DMC Global Stacks data product:

  • IRIS DMC (2014), Data Services Products: globalstacks Global stacks of millions of seismograms, doi:10.17611/DP/GS.1.


Astiz, L., P. Earle and P. Shearer, Global stacking of broadband seismograms, Seis. Res. Lett., 67, 8-18, 1996.

Paul S. Earle, Polarization of the Earth’s teleseismic wavefield, Geophys. J. Int. (1999) 139 (1): 1-8 “doi:10.1046/j.1365-246X.1999.00908.x

Rost, S., Thorne, S., and Garnero, E. Imaging global seismic phase arrivals by stacking array processed short-period data, 77(6),697-707, Seismological Research Letters, 2006. 10.1785/gssrl.77.6.697

Shearer, P.M., Imaging global body-wave phases by stacking long-period seismograms, J. Geophys. Res., 96, 20,353-20,364, 1991. doi: 10.1029/91JB00421

Shearer, P. M., Rychert, C. A. and Liu, Q. (2011), On the visibility of the inner-core shear wave phase PKJKP at long periods. Geophysical Journal International, 185: 1379–1383. doi: 10.1111/j.1365-246X.2011.05011.x