OR/14/064 EO information services assessment and feedback

Kessler, H, and Dearden, R. 2014. Scoping study for a Pan-European geological data infrastructure: D3.4: technical requirements for serving 3D geological models. British Geological Survey, OR/14/072.

Scope

With respect to the four attributes of a sustainable service (available, useful, reliable, affordable) it is useful to determine the success criteria for Service acceptance by WB and by Local Users. The primary aim is to respond to the requirements. These requirements are not fully identified at present (due primarily to a lack of direct contact with Users), however it is understood that ITC have completed a questionnaire survey and once the results of these are known, and the CHARIM workshop has been held, BGS can incorporate the information into the service design and the service assessment.

Available. As noted previously in this report, one of the strengths of EO is its global availability. Archive data have been obtained and additional very high resolution stereo and triplet imagery has been tasked. The short timescale of this project means that the tasking has not been aligned with the optimum time of year (in terms of climate, cloud cover) and so there is a risk that weather conditions will limit the availability of the service in terms of its comprehensiveness. The availability of the service into the future is less prone to time restrictions, therefore the service(s) can be readily updated as soon as suitable imagery is acquired. Long term sustainability of the service is dependent on local experts applying the methodology and providing the service to agreed standards. This project does not have the remit or the resources to provide training, although BGS is contributing to the WB Caribbean Handbook for Risk Information Management which provides use cases and training for future service provision.

Useful. It is not expected that the EO-RISC services will be stand-alone. For example, the landslide inventory is a valuable product in its own right that will identify current areas undergoing ground motion — however its value will be increased when additional work is undertaken (in conjunction with ITC) to address risk levels associated with the hazard. This stepped usefulness will need to be accounted for in the service assessment.

Reliable. Each service has appropriate validation methodologies to gauge the reliability of the result. In brief these are:

Service 1 — Existing land use classifications will be used as a baseline to define training areas for supervised classification of the optical imagery. The thematic accuracy will be determined using a standard confusion matrix approach supplemented by limited field verification at point sample locations.

Service 2 — A supervised classification of exposed ground (i.e. vegetation scars due to landslides) using the optical imagery will be assessed using a standard confusion matrix approach. However, the classification results will not be integrated directly into the service; each potential landslide will undergo manual visual assessment using the EO data by experienced landslide and geoengineering specialists at BGS. The inventory will subsequently evaluated against existing inventories (bearing in mind different scales and sources of the inventories) augmented by ground truthing.

Service 3 — The DEMS generated will be validated against ground control data, where available. The ground control data should be in the form of authoritative survey points such as trig stations.

The reliability of the services into the future is determined by the quality of the EO data, and the expertise of the image analysts.

Affordable. The affordability of any service is defined by what is believed to be within ones financial means. It does not mean that something is necessary low cost, but that it is value for money. In this case ESA has provided a modest budget ~€120 000 to provide all three services over several territories. To define if each service is affordable by one or more users (or a consortium of users) it would be necessary to obtain feedback from the users regarding their abilities to pay for one or more service.

Schedule

Planned schedule for all Service assessment activities is outlined in this section. At the highest level, the service assessment plan was outlined in the SoW as follows:

• KO+1: Service Readiness Review meeting with Users, ESA and WB in Washington DC. This was to provide the first opportunity to meet the WB and Local Users and to define the services and discuss their mode(s) of assessment. This meeting was cancelled.
• KO+2: Delivery of D1, Service Readiness Document (SRD) i.e. this document. At the time of writing BGS has had extremely limited to contact with the Local Users via WB and their CHARIM project. We have had access to the CHARIM Preliminary Assessment Report
• KO+7: Delivery of D3, Operational Documentation (D3) and visit to the project User with the WB TTL.
• KO+9: Service Utility Review (SUR) at the WB in Washington
• KO+10: Delivery of D4, Service Utility Document.

At this point it is clear that increased contact with the Local Users is mandatory. BGS will participate in the CHARIM workshop in the Caribbean at the end of September and we intend to use this brief meeting to discuss the services with the Local Users along with ways in which they can be assessed.