Skip to Content

Changes for page AusPass Data

Last modified by robert on 2026/02/12 14:01
From version 15.3
edited by robert
on 2026/01/16 11:29
Change comment: There is no comment for this version
To version 4.1
edited by robert
on 2025/06/19 18:15
Change comment: There is no comment for this version

Summary

Details

Page properties
Content
... ... @@ -17,15 +17,16 @@
17 17  
18 18  == External Network data ==
19 19  
20 -We host //some// (a lot, but not all!) complimentary data from the [[Global Seismograph Network (GSN - IU)>>https://www.fdsn.org/networks/detail/IU/]], mainly NWAO, MBWA, and CTAO.
20 +We also host //some// waveform data from GeoScience Australia's national network **AU** and the global network **IU**. This is done as a convenience to our users as much of this data is concurrent and proximal to our networks. Again we reiterate that we only host **SOME** of this data, we do not have the resources to do a full clone. If we do not have the data you're looking for, it's best to try IRIS/Earthscope. If they don't have it, you should get in touch with Geoscience Australia directly.
21 21  
22 -We also archive live streams from the Seismology Research Centre (OZ) as well as the University of Melbourne (VW).
23 -
24 24  (% class="box warningmessage" %)
25 25  (((
26 -Unfortunately, AusPass no longer hosts AU network data~-~- you will now have to get this from Earthscope/IRIS.
24 +AusPass no longer hosts AU network metadata~-~- you will have to get this from Earthscope/IRIS. AU network waveform data is still here.
27 27  )))
28 28  
27 +[[Venn diagram of where AU network data is stored online>>image:GA_iris_auspass.png||data-xwiki-image-style-alignment="center" height="172" width="459"]]
28 +
29 +
29 29  == QA/QC ==
30 30  
31 31  Where possible, we will fix obvious timing errors such as constant time offsets and clock drifts and host the "fixed" version of the waveform data. Sometimes timing corrections cannot be reliably fixed (e.g. clocks jumping around too much intraday) and they are left as-is. Other times, issues were missed or could not be perfectly fixed. If you find any discrepancies, kindly just let us know when and where.
... ... @@ -32,13 +32,13 @@
32 32  
33 33  We do **not** fix north azimuth misalignments with the exception of perfectly inverted N-S alignments, which are extremely rare.
34 34  
35 -We keep copies of the original raw data as well as [[a folder of timing corrections>>http://auspass.edu.au/data/timecorrections]] and are happy to facilitate access to these upon request. If you see any issues, please let us know.
36 +We keep copies of the original raw data as well as [[a folder of timing corrections>>http://auspass.edu.au/timingcorrections]] and are happy to facilitate access to these upon request. If you see any issues, please let us know.
36 36  
37 37  = Station Data =
38 38  
39 39  For all of the networks and stations in our waveform archive, we also host the accompanying STATION metadata (which includes response information). For networks outside of our control (e.g. AU and IU) we download this metadata from IRIS/Earthscope. Any issues with this data should be directed at the respective agencies in charge of keeping it accurate.
40 40  
41 -Response information, particularly for older stations when records were not usually kept very well, might be wrong! Sometimes we just have to take a guess and eyeball appropriate gain values via expected PSD amplitudes. In practice even if the gain is off by 2-4x, the affect on resulting earthquake magnitude estimates are lower than the natural degree of scatter in estimates from station to station. 
42 +Response information, particularly for older stations when records were not usually kept very well, might be wrong! Sometimes we just have to take a guess and eyeball appropriate gain values via expected PSD amplitudes. In practice even if the gain is off by 2-4x, the affect on resulting earthquake magnitude estimates are lower than the natural degree of scatter in estimates from station to station.
42 42  
43 43  = Earthquake Data =
44 44  
... ... @@ -45,33 +45,3 @@
45 45  We host copies of Geoscience Australia's global earthquake catalog which is derived exclusively from the output of the [[Earthquakes@GA>>path:https://earthquakes.ga.gov.au/]] web portal. We do this by manually downloading CSV data and converting it to Seiscomp SCXML, then importing it into our server to be accessed via FDSN as event data. Because we must do this manually, our cloned earthquake catalog is only updated monthly (usually on the 1st). Note that information such as phase arrivals is unfortunately not provided by GA.
46 46  
47 47  Soon we will also begin hosting regionally specific catalogs with full phase picks.
48 -
49 -= SmartSolo Node Polarity Issues =
50 -
51 -There has been a lot of confusion about the polarity of the IGU-16HR (1C and 3C) data, as well as the BD3C-5 data over the years.
52 -
53 -(% class="box errormessage" %)
54 -(((
55 -ALL IGU-16HR channels have negative polarity relative to "standard" FDSNXML convention (positive down = -90 degrees, positive north = 0 degrees, positive east = 90 degrees). Freshly exported data requires a -1 multiplier relative to be "correct" relative to what you're likely expecting.
56 -)))
57 -
58 -(% class="box infomessage" %)
59 -(((
60 -BD3C-5 channels are "normal" and consistent with usual high gain seismometers. These sensors do not require any polarity flips.
61 -)))
62 -
63 -After much consternation was have finally tested this ourselves in a controlled environment, and have also consulted various agencies. We are now certain the above notices to be correct.
64 -
65 -[[Arrivals showing the polarity of various instrument types (courtesy of the China Earthquake Administration). Note that the IGU-HR3C are inverted relative to the BD3C and CMG40T. >>image:Weixin Image_20260115150150_19_1185.png||alt="Arrivals showing various sensors. The HR3C channels are inverted."]]
66 -
67 -In[[ late 2023 it was discovered>>https://epic.earthscope.org/content/smartsolo-archives-flipped-polarity-igu-16hr-3c-5hz]] that the vertical Z channel of **the IGU-16HR short period **SmartSolo nodes is actually inverted relative to the standard expectation of "positive up" (vertical axis dip = -90). EarthScope's solution was to add a -1 multiplier to the //response data.// Because we have a much smaller volume of this data to fix, and far less of it already distributed prior to this announcement, we have opted to do the opposite; e.g. we invert the //raw data// because the data is what is wrong; not the response.
68 -
69 -**In late 2025 we also discovered that the horizontal (N and E) channels are ALSO flipped! Data downloaded before Friday 5 December 2025 is affected- multiply your DH[NE] channels by -1 if you have already downloaded them.**
70 -
71 -Until very recently we also thought that the BD3C-5 "broadband" nodes had inverted data by default, but now understand that their raw exported data has the correct polarity.
72 -
73 -{{warning}}
74 -Do not mix up EarthScope's or the IRIS-NRL node response data with those available at AusPass. Theirs could be inverted relative to our data! In general we have low confidence in the information provided but can confirm that our metadata is at least "pretty close".
75 -{{/warning}}
76 -
77 -
Weixin Image_20260115150150_19_1185.png
Author
... ... @@ -1,1 +1,0 @@
1 -XWiki.robert
Size
... ... @@ -1,1 +1,0 @@
1 -629.1 KB
Content