1. Seismic Team – 14 May, 2013 – Report to Plenary Session
CEOS Seismic Pilot
Seismic Team – 14 May, 2013 – Report to Plenary Session

2. Thematic Team Members (proposal contributors)
co-leads: Joern Hoffmann (DLR) and Philippe Bally (ESA)
Secretary/Facilitator: Andrew Eddy
Members: Stefano Salvi (INGV), Falk Amelung (U of Miami), Chuck Meertens (UNAVCO) Eric Fielding (NASA/JPL), Tim Wright (U of Leeds), Fabio dell’Acqua (EUCENTRE), Bill Barnhart (USGS NEIC), Steven Hosford (CNES), Tom Cecere (USGS), Francesco Gaetani.

3. Rationale for Pilot
WHY? Most devastating global disaster – large number of events and largest number of total deaths (large events); EO can support increased understanding of nature and extent of risk – cannot predict earthquakes. EO also supports response and other DRM phases.
WHAT’S MISSING? Large data collects over seismic strain belt (15% earth surface) to improve scientific knowledge of seismic hazard; partners to analyse data to generate surface strain model based on interferometric SAR analysis. GSNL provide EO data for improved scientific understanding but only over limited areas. Science products for EQ response (M>5.8 not covered by GSNL, data types for science not covered by Charter).
Precise terrain motion product to support Crustal Block Boundaries analysis over the Cephalonia Island, Greece, Standard Deviation of Velocity Field using ENVISAT Descending (2003 – 2008), Credits: NKUA.

4. Outcome of the session Defined the purpose of the proposal
Defined activities proposed
2.1) List of activities
2.2) Contribution from members/partners
2.3) Steps to be achieved
2.4) Milestones for pilot realization
3) Gathered discussion points
4) Compared of the functions of an exploitation platform
5) EO data requirements
6) High level observation modes concerning seismic hazards

5. Topics discussed but not yet fully addressed:
Detailed review of contributions by partner
Assessment of overlaps/gaps
« Prospective contributions » to address them
However there where good discussions.
Some principles to manage these have been agreed.
Some ‘prospective contributions’ already defined.

6. Objectives
A: Support the generation of globally self-consistent strain rate estimates and the mapping of active faults at the global scale by providing EO InSAR and optical data and processing capacities to existing initiatives, such as the iGSRM.
[Wide extent Satellite observations]
B: Support and continue the GSNL for seismic hazards and volcanoes.
[Satellite observations focused on supersites]
C: Develop and demonstrate advanced science products for rapid earthquake response.
[Observations of earthquakes with M>5.8]

7. Proposed Activities for A) – EO data:
Wide extent & repeat INSAR data to build the global strain model (continuous observations over large areas using SAR data such as Sentinel-1, ALOS-2 and RCM). Coordination and sharing of data acquisition burden among SAR data providers.
Optical ortho-rectified imagery to build regional or global maps of active visible fault (e.g. wide extent mapping using data from the SPOT archive).
InSAR stacks to support study past earthquakes during the satellite era using and, when appropriate, optical data stacks

8. Proposed Activities for A) – derived geo-information
Fault mapping at a regional/global scale
Ground displacement for historical events based on InSAR analysis and optical imagery when appropriate
On going wide extent & repeat ground displacement mapping to contribute to the global strain model (continuous observations over large areas using SAR data such as Sentinel-1, ALOS-2 and RCM).
Access to relevant Digital Elevation Models
Demonstration of EO-based strain rate measurements (representative sites)
Methodologies and tools to produce strain rate estimates
Validation of techniques to measure strain rates (e.g. Western Australia, where there is no strain)

9. Proposed Activities for B)
EO data:
SAR data stacks from several missions.
Optical satellite data, requested by users to support Supersite and natural laboratories (NL) research into Geohazard.
Derived geo-information:
Repeat INSAR measurements over supersites (multi- mission).
Digital Elevation Model

10. Proposed Activities for C):
EO data:
Rapid supply of co-seismic data for event sites (SAR stacks and optical pre and post-event images).
Collection of InSAR data to support fundamental research on earthquake fault mechanics using observations of the early post-seismic phase. These observations (hours to days after the event) are now possible thanks to the multiple sensors available to Event Supersites under the GSNL.
Derived geo-information:
Rapid supply of co-seismic ground displacement from analysis of SAR and optical imagery.
(Semi-) automatic fault modelling, prediction of damage distribution, rapid calculation of Coulomb Stress changes on neighbouring faults (derived from above). (add to list based on INGV contribution)

11. Proposed Activities for A, B & C):
Develop communities: pro-actively develop and manage the community of users & practitioners of Satellite EO for geohazards. Conduct outreach and capacity development concerning EO based techniques within user communities (geoscience centres, observatories, end users)
e-science: accelerate the science use of satellite EO & promote research through new mechanisms using the web.
an exploitation platform: address the goals of the geohazard risks community with dedicated tools (such as atmospheric correction or crustal deformation modelling, etc.) in a cloud-based computing environment. An exploitation platform (e-infrastructure, incl. cloud computing capabilities; provide users with data, software and computation.

12. Underlying principles:
The earthquake response activity associated to Objective C) will be partly covered by the GSNL (Event sites) however C) has a broader scope (earthquakes M>5.8). The pilot sets a target of 4 to 6 moderate-large earthquakes per year. In terms of availability during the time of the pilot this will not be a service, but a demonstrator to show what might be possible.
Data access principles in the pilot: To address objectives B) & C) the CEOS DRM pilot will rely on the GSNL and conduct other activities. While the GSNL are providing freely available EO data for scientific analysis of geohazards, to address objective B) the pilot will coordinate different demonstration activities some following a similar approach (data access under GSNL) and others following different approaches (e.g. data access to a limited user group) in other activities.
Impact on the GSNL: The activities of the CEOS DRM pilot concerning seismic hazards will not affect the policies, rules or access mechanisms of the GSNL.
The added value of CEOS DRM wrt GSNL will be to re-enforce the capacity of the GSNL with the functionalities of an exploitation platform (for instance data storage, etc.)

13. Other agreed principles:
Link to International Charter: the CEOS DRM pilot will contribute to information/promotion actions concerning the International Charter Space & Major Disasters. In some instances users of the pilot with a mandate both in risk assessment and emergency response will access Charter data or results. Objective C) will address information needs of users of the seismological community that will access complementary EO products tailored to their needs and that are not in the scope of the International Charter (e.g. INSAR data).
Note that this draft does not include as an objective supporting the generic need for geo-information to support asset mapping / exposure mapping (in complement to hazard mapping). However some partners have expressed interest in supporting it on a case-by-case basis.

14. Partners & Contributions (& Prospective Contributions)
ESA: data, exploitation platform, scientific animation CSA: data DLR: data, potentially visibility of TanDEM- X DEM through viewer for registered users NASA: data, federated query tool, ARIA rapid imaging and analysis CNES: high res optical imagery (principally archive) ASI: data
INGV: near-real time products
COMET+: large area strain measurements, tools, methodologies, link to Earthquake without frontiers
EUCENTRE hazard exposure layers, links to GEM
University of Miami: validation campaigns
UNAVCO: TBD

15. CEOS Added Value
Large data collects require coordinated approach from leading space agencies
New approaches (e.g. exploitation platform) considered to address very large data volumes & demanding processing required to support users
Close connection to existing CEOS-led initiatives such as GSNL

16. Milestones & Deliverables
Objective A:
Q required data in place
end results of test areas using archived data, initial results for main areas; established beta methodology for processing over large areas; preliminary results for all areas on COSMO Skymed and TerraSAR X data in Eastern Turkey and California
end partial results of validation - Turkey, California
2016? strain rate measurement results
Objective B:
Milestones to be developed (c.f. GSNL plan, c.f. plan for release of E.P.)

17. Milestones & Deliverables
Objective C:
Q Examine the gaps in existing background data over the major cities of the world in areas at high seismic risk (Cosmo SkyMed, RSAT)
Q Example data for past earthquakes put on the SSEP (L'Aquila, Van, Emilia, New Zealand)
Q Implementation of processing algorithms for rapid response products on the SSEP (in case funding is available)
Q Demonstration of the generation of different products (see C2) for 1-2 earthquakes.
Q Comparison of results obtained by different groups/algorithms/approaches; consensus report.
Q Demonstration of the generation of different products (see C2) for 1-2 earthquakes.
Q Product assessment by the final users
Q Demonstration of the generation of different products (see C2) for 1-2 earthquakes.

18. Milestones & Deliverables
Other milestones:
Q Development of the Web site.
Q Development of procedures (for each agency) to ensure optimal data acquisition in case of earthquakes over a certain threshold. (same for volcanoes)
Q Development of a procedure to list EO data acquisitions as they are acquired (on the web site?). Development of a procedure to make data available rapidly.

19. Users Objectives A) & B): Objective C):
Geo-science community incl. Universities (PIs) worldwide
Organisations with a national mandate concerning risk assessment (incl. seismological centres)
Specialists of geohazard risks in other sectors (insurance/re-insurance sector, civil engineering companies, energy, etc.)
Longer term: National & regional civil protection agencies and national & local authorities with a risk management mandate
Objective C):
Geo-science community
Longer term: National & regional civil protection agencies and national & local authorities with a risk management mandate (response)

20. Capacity Building
Capacity building activities will be developed with WGCapDevDD
Strong capacity building under way in different on going activities (e.g. Int. Development activities, United Nations, Institutional Centre for Theoretical Physics (ICTP) in Trieste, etc) – linkages yet to be developed

21. Cross-cutting Elements
Role of GSNL
Part of the planned pilot activities concerns assets that can contribute to Volcanoe theme (e.g. exploitation platform, observation modes for precise terrain deformation monitoring, etc.)
DEM
Exposure data

22. Geography
Global seismic regions, with specific interest over supersites and other test areas, yet to be determined in detail (first ideas: Eastern Turkey, California, Iran, Australia, Dominican Republic).
Stretch goal of mapping strain in Alpine Himalayan zone and Eastern Africa

23. Data Requirements For Objective A
Large data collections for demonstrators of strain rate measurements
Data collections to support the validation of the strain rate methods
Validation with GPS: Southern California - Salton Trough (Terra SAR X, UAVSAR, COSMO Skymed, ALOS-1, ERS, Envisat) insert image from Eric; Eastern Turkey (TerraSAR X - need 150 km not 80km, COSMO-Skymed, large data stack ERS data and PSI products from TerraFirma & DLR WAP), Tim to provide image and define area) ; Greece (TBC)
Japan - Shikoku (Falk to provide image and site definition) Need for L-band and C-band
Validation without GPS data: Dominican Republic
Validation no strain: Australia: Western Australia (use several different satellites to compare results) issues: too dry? not enough topography? need for representativity for atmosheric issues; maybe Eastern Australia works better? 500km? good coherence; Iran (intermediate GPS coverage); Japan - Tokyo?

24. Data Requirements Objective B
 Data over Supersites, Natural Laboratories and Event Sites.
Natural Laboratories are the component that will provide larger amounts of data; GSNL to support A) Objectives for validation.
Objective C
earthquakes of interest above 5.8
Challenges (near-real time data & rapid access to archives; non plannable events)