Reference Information Specifications for Europe RISE Results

Reference Information Specifications for Europe RISE Results & Lessons Learnt Wyn Cudlip (Qineti. Q) wcudlip@qinetiq. com www. eu-rise. org

Issues • How to satisfy INSPIRE and other EU Directives on data sharing and re-use without completely re-engineering the information databases. • Also applies to cross-border and other cross-sector data harmonisation issues. RISE Results, Porto, 5 th July, 2007. V 1. 0 2

RISE Project goals • Define Methodology & Guidelines Document • for the creation of geospatial Data Product Specifications (DPS), consistent with international and industrial standards. • Test the Methodology • by creating test harmonised DPS for Hydrography, Land. Cover and Elevation; as example input datasets to diffuse nutrient leakage modelling for the Water Framework Directive (WFD). • Test the harmonised Data Product Specifications (h. DPS) • in an on-line “Test Environment” with data from test areas (in Sweden & Norway). • Demonstrate the automatic creation of harmonised Data Products • from existing databases using schema translation RISE Results, Porto, 5 th July, 2007. V 1. 0 3

RISE consortium Public Institutions Source data providers Industry Technology Euro. Geographic s Qineti. Q OGC-Europe Lantmäteriet BKG 2 yr FP 6 Project; started Sept. 05. Working relationship with MOTIIVE. RISE Results, Porto, 5 th July, 2007. V 1. 0 “Harmonisation Advisory Committee & Planning Board” “WFD & GMES” “Key stakeholders” Users and applications 4

Approach Prototype Testing Data Product Specification Requirements Application schema (GML) Conceptual schema (UML) Use Cases Cost benefit Exploitation guidelines RISE Methodology RISE Results, Porto, 5 th July, 2007. V 1. 0 5

Key Steps in the development of “harmonised” Data Product Specification • Analyse Use Cases • To define requirements • Develop Conceptual Schema (in UML) • Including Feature Type Catalogue and Data Dictionary update • Develop GML Application Schema • e. g. , UGAS, UML to GML conversion tool Requirements and Feature Types Identification Data Dictionary Update this can be automated • Use Case Development • Define harmonised Data Product Specification • Data Products created from existing data • DPS and App Schema Development Implementation, testing and validation Goal: to create “on the fly” with suitable server software RISE Results, Porto, 5 th July, 2007. V 1. 0 6

Use Case Development Checklist for data Harmonisation requirements Supporting components: feature data dictionary, CRS registry, etc. Checklist addresses steps to data harmonisation RISE Results, Porto, 5 th July, 2007. V 1. 0 Requirements As-is analysis Gap analysis Requirements and Feature Types Identification Data Product Specification App Schema Development Implementation, testing and validation(using WFS) validation 7

Key assumptions • A Data Specification is based on a data product specification (DPS) according to ISO 19131. Such a DPS includes the following sections: Mandatory: • Overview • Specification scopes • Data product identification • Data content and structure • Reference systems • Data quality • Data product delivery • Metadata Optional: • Data capture • Data maintenance • Portrayal • Additional information • RISE harmonisation is concerned with more than just data content and structure. RISE Results, Porto, 5 th July, 2007. V 1. 0 8

Data Harmonisation Components – Overview 1. INSPIRE Information Model 1. 1 INSPIRE Principles 1. 4 ISO 19100 Profile 1. 7 Object referencing modelling 1. 2 Reference model 1. 5 Multi-lingual text and cultural adaptibility 1. 6 Coordinate referencing and units model 1. 9 Portrayal model 3. 1 Metadata 1. 8 Data translation model/guidelines 1. 3 Application Schemas 3. Guidelines & Best Practice 2. Operational components/registers 2. 1 Identifier Management 2. 3 Feature catalogues 2. 2 Terminology 3. 5 Derived reporting & multiple representations 3. 2 Maintenance 3. 6 Consistency between data 3. 3 Quality 3. 7 Data capturing 2. 4 Dictionaries 2. 5 Conformance 3. 4 Data Transfer (developed by INSPIRE Drafting Team „Data Specifications“ in collaboration with RISE; RISE will focus on the methodology, the DT will focus on generic aspects of the conceptual model) RISE Results, Porto, 5 th July, 2007. V 1. 0 9

Relevant ISO Standards • • • UML ISO 19103 Conceptual Schema ISO 19107/8/9 Application Schema ISO 19109 Feature Type ISO 19109 Data Dictionary ISO 19126 Data Product Specification ISO 19131 Quality ISO 19113/5 Metadata ISO 19115 Multi Lingual ISO 19139 Portrayal ISO 19117 GMLISIO 19136 Translating WFS ISO 19142+43 RISE Results, Porto, 5 th July, 2007. V 1. 0 10

RISE Deliverables Data Harmonisation Requirements Report Methodology & Guidelines for Use Case & Schema Development Use Case Descriptions Conceptual Schema in UML GML Application Schema Data Product Specification Services Architecture Outline Test Environment Report Cost Benefit Report Testing Data Specification Requirements Methodology Conceptual Schema Use Cases Cost Benefit Analysis Architecture Exploitation Guidelines Prototype Exploitation Guidelines Drafts available now; final versions in September 2007 RISE Results, Porto, 5 th July, 2007. V 1. 0 11

Methodology & Guidelines Document • V 1. 0 available. http: //www. eu-rise. org/ • Used by the INSPIRE Drafting Team on Data Specifications for their “Guidelines” Document. • Used by the MOTIIVE Harmonisation Project • Revised version due in September 2007. RISE Results, Porto, 5 th July, 2007. V 1. 0 12

USE Case Architecture Use Case Actors: Users modelling LAN, WWW WFS/WCS harmonised data products Use Case Actors: Data Integrators Processing Services LAN, WWW (WFS/WCS) Source Data Use Case Actors: Data Providers Landcover Clearcuts RISE Results, Porto, 5 th July, 2007. V 1. 0 DEM Hydrography Attributes 13

Testing • Three Themes • Hydrography (i. e. catchment area polygons; river and lake segments) • Landcover (i. e. RISE defined scheme (aggregated CORINE type) • Elevation (i. e. uniform grid of Digital Elervation Model (DEM) • Two Test Areas • Vindan (East Sweden) • Enningsdalsalven (Swedish Norwegian Borders) Enningsdalsalven Vindan RISE Results, Porto, 5 th July, 2007. V 1. 0 14

UML Diagram Elevation Landcover Hydrography (riverbasin and river/lake network) RISE Results, Porto, 5 th July, 2007. V 1. 0 15

Schema Translation • Schema translation definitions created with the use of Matching Tables. • These have to be generated by domain experts. Example of schema translation with Snowflake’s “Go Publisher”. RISE Results, Porto, 5 th July, 2007. V 1. 0 16

RISE Test Environment • Demonstrate the creation of harmonised data products • • “on-the fly” from existing data sets Allow evaluation of the data products (and hence the specifications) RISE Results, Porto, 5 th July, 2007. V 1. 0 17

Conclusion • RISE has produced a Methodology & Guidelines document for the creation of harmonised Data Product Specifications. • Several other documents have also been produced including a number related to schema development. • A RISE Test Environment has shown how h. DP can be produced “on the fly” from existing information through schema translation. • Some types of “missing information” can be created automatically without any database intervention. In practice, the degree of automation will depend on how much information is missing from the input existing datasets. RISE Results, Porto, 5 th July, 2007. V 1. 0 18

Lessons Learnt • Harmonisation must be driven by real requirements • need to avoid lowest common denominator effect. • or compromises which satisfy nobody. • Hence importance of Use Cases • Interface standards need further development • E. g. for access control, handling time series • Enhancement to WCS, improved integration with SOAP etc. • Software tools require further development • E. g. Improved integration of Clients, WMS, WFS and WCS RISE Results, Porto, 5 th July, 2007. V 1. 0 19

Issues for discussion • How to maintain project results after the end of the Project? • How best to influence Standards Bodies? • E. g. Specify requirements; support testbeds? • Commercial versus Open-Source Software? RISE Results, Porto, 5 th July, 2007. V 1. 0 20