Data Archiving and IT Issues for PBO

1
Data Archiving and IT Issues for PBO

• Douglas Neuhauser
• UC Berkeley Seismological Lab
• Northern California Earthquake Data Center

2
Plate Boundary Observatory

• Measure broadband deformation across the western
  US.
• Measure strain rate in 4D.
• Determine forces responsible for plate boundary
  deformation.

3
PBO Instrumentation

• Continuous GPS.
• Borehole strainmeters and related sensors.
• Surface strainmeters and related sensors.
• Borehole seismometers.

4
Borehole Installations

• 3 component tensor strainmeter
• Pore pressure sensor.
• Barometric pressure sensor.
• Borehole seismometers.

5
PBO Borehole Installation

• Borehole contains strainmeters, seismometers, and
  other geophysical sensors.
• GPS attached to borehole casing.

6
Surface Strainmeter Installations

• Laser interferometer for horizontal distance
  measurement.
• Laser interferometers for pier anchors.
• State-of-health sensors for data correction
  (vacuum pressure, laser frequency, etc.)

7
Data Archive Issues for EarthScope

• 1. What standard data processing should be
  performed on EarthScope data to create standard
  products?

8
Seismic Data Processing

• Minimal processing required (clock timing
  corrections).
• Sensors directly measure ground motion (velocity
  or acceleration).
• Sensors have well defined instrument responses.
• SEED standard for data archiving and data
  distribution include data and response.

9
GPS Data Processing
10
Strain Data Processing

• Multiple stages involved in processing strain
  data.

11
Data Archive Issues for EarthScope

• 2. How do we establish procedures for different
  stages of data processing?
• Correct strain from pore pressure and atmospheric
  pressure channels?
• Remove tides from strain data?

12
Strain Data Processing Procedure
13
Data Archive Issues for EarthScope

• 3. How do we organize the data so that users
  know which data channels for an experiment are
  associated with each other, where there are
  several raw channels of the same type?
• Tensor strain
• Pore pressure
• Barometric pressure
• Temperature channels
• Seismic channels
• Various voltage / State-of-Health channels

14
USGS Low Frequency Geophysical Network (UL)

• Diverse set of geophysical experiments
• Strain (tensor, volumetric, wire)
• Creep
• Tilt (lake level, long baseline)
• Well water levels
• Electric potential
• Each experiment has primary and auxiliary data
  channels.

15
USGS Low Frequency Geophysical Network (UL)

• Data used by and distributed to other researchers
  by the PI.
• Data available in raw and processed versions.
• Instrument responses often not well known.

16
UL Data Archiving issues

• 1. What format(s) to use for archive and
  distribution for raw and processed data?
• How many versions of the data (raw and
  processed) should be archived?
• How to make the data readily available to the
  community for use with other data sets?

17
Data Archive Issues for EarthScope

• 4. What data format(s) do we need for the
  various types of data?
• How do we indicate what processing has been done
  to data channels?
• SEED naming convention of NET, STATION, CHANNEL,
  and LOCATION may not provide sufficient diversity.

18
UL Data Archive at NCEDC

• Raw data time series archived in SEED format at
  NCEDC.
• NETWORK name is UL.
• STATION name is geographic site.
• CHANNEL name designates type of measurement
  (instrument, sample rate, instrument bandwidth).
• LOCATION name signifies experiment type and
  instance at the site.

19
UL Data Archive at USGS

• Automated cleaned data available from USGS in
  simple ASCII format.
• Glitch and offset removal.
• Cleaned primary data channels may be corrected
  by data from auxiliary data channels.
• Minimal info about cleaning process provided
  with cleaned data.
• Data distributed with USGS 4 character channel
  name.

20
Data Archive Issues for EarthScope

• 5. What data archive and distribution models are
  needed to provide users with integrated access to
  the complementary data sets? For example
• Will people want to integrate GPS data and
  strainmeter data?
• What are data format and metadata requirements to
  accomplish this?

21
NetDCArchiveModel
22
NetDCArchiveModel
23
GSAC Archive Model
24
FISSURESArchiveModel
25
Data Archive Issues for EarthScope

• 6. How do we address these issues in a framework
  that allows both people and inference or
  processing systems to make intelligent use of
  EarthScope data?