Domain Modelling and Implementation Standards context Simon Cox

Domain Modelling and Implementation Standards context Simon Cox Research Scientist Sydney - December, 3 rd 2010

Overview • ISO/TC 211 modeling framework • Standard meta-models – feature, coverage, observation • OGC Services: WMS, WFS, WCS, SOS CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Standards context • ISO/TC 211 – Geographic Information • Main stakeholders: national standards bodies • OGC • Main stakeholders: software vendors; government agencies • INSPIRE • Main stakeholders: European Commission, member states • Base is (profiles of) ISO and OGC standards • ‘Data specifications’ for various domain models CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Feature model CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO/TC 211 basics • Reference model – ISO 19101 • Rules for Application Schema (General feature model) – ISO 19109 • Coverages – ISO 19123 CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO 19101 Reference model CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Application schema CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO 19109 General feature model CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Meta-levels • GF_Feature. Type • all feature types • River • a feature-type • all rivers • Paramatta • a river CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Conceptual object: “feature” • Digital object corresponding with identifiable, typed, object in the real world • mountain, road, specimen, event, tract, catchment, wetland, farm, bore, reach, property, license-area, station • Feature-type is characterised by specific set of properties • May have one or more spatial property CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010 • Specimen • • • ID (name) description mass processing details sampling location sampling time related observation material …

Mapped features CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Fields & coverages CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Objects vs Fields classic geology: “feature” CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010 classic earth-observations: “field” or “coverage”

Spatial function or field: “coverage” • Variation of a property in domain of interest • Sampled discretely, often on a grid (x 1, y 1) (x 2, y 2) • Domain extent is scoped by shape/lifetime of feature of interest • Range-type is scoped by type of feature-of-interest CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Different cross-sections through same dataset Specimen Au (ppm) Cu-a (%) Cu-b (%) As (ppm) Sb (ppm) ABC-123 1. 23 3. 45 4. 23 0. 5 0. 34 • A Column = variation of a single property across a domain (i. e. set of locations) CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010 • A Row summarizes the properties of one feature

Feature may have coverage-valued property Scene + bounds: GM_Envelope + theme: string Weather. Station colour CV_Discrete. Grid. Point. Coverage {n} temperature CV_Discrete. Time. Instant. Coverage CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010 + location: GM_Point + owner: CI_Responsible. Party

Conceptual object model: features • Digital object corresponding with identifiable, typed, object in the real world • mountain, road, specimen, event, tract, catchment, wetland, farm, bore, reach, property, license-area, station • Feature-type is characterised by specific set of properties CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010 • Specimen • • • ID (name) description mass processing details sampling location sampling time related observation material …

Summary of meta-model • Features • Object-centric viewpoint • Artefacts associated with observation campaigns: platform, instrument, site • Detected/interpreted objects in the environment: river, ore-body, plume … • Coverages • Property-centric viewpoint • Variation of a property within a domain • Evidence used in interpretation • Results of model (simulation or interpolation) CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Spatio-temporal elements CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO spatio-temporal utilities • • 19111 – Coordinate reference systems 19107 – Spatial schema (geometry) 19123 – Grid schema 19108 – Temporal schema CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Overview of geometry types CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Temporal primitives CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Temporal reference systems CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO 19111 Coordinate reference systems CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

More information • https: //www. seegrid. csiro. au/twiki/bin/view/App. Schemas • in particular: • • • Geospatial. Standards. Context Iso. Tc 211 Standards OGCInformation. Models Feature. Model Information. Views CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Collecting data CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Assignment of property values • For each property of a feature the value is either i. asserted • E. g. name, owner, price, boundary (cadastral feature types) ii. observed/estimated • E. g. colour, mass, shape (natural feature types) • i. e. error is of interest CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Observations OM_Observation GFI_Feature +feature. Of. Interest 1 0. . * + + + phenomenon. Time result. Time valid. Time [0. . 1] result. Quality [0. . *] parameter [0. . *] GFI_Property. Type 1 +observed. Property 0. . * 1 +procedure OM_Process +result Any An Observation is an action whose result is an estimate of the value of some property of the feature-of-interest, obtained using a specified procedure

Defines the following terms: • • Observation Procedure Observed property Result • • Feature of interest Phenomenon time Result time Valid time – sensor, instrument procedure – e. g. geophysical parameter – e. g. scaled number, vector, image, grid, classification – in-situ, remote, ex-situ (specimen) – time the result applies to – time the result was obtained – time period the result may be used • A neutral terminology to support cross-domain data discovery & fusion

Application to a domain observed property Belongs to feature-of-interest-type OM_Observation GFI_Feature +feature. Of. Interest + + +property. Value. Provider 1 0. . * phenomenon. Time result. Time valid. Time [0. . 1] result. Quality [0. . *] parameter [0. . *] +generated. Observation 0. . * 1 +procedure GFI_Domain. Feature feature of interest Feature-type taken from a domain-model (e. g. Geo. Sci. ML) GFI_Property. Type 1 +observed. Property Range +result OM_Process procedure Standard procedures, per property-type Any result GML, SWE, or pointer to a standard structure (e. g. net. CDF)

Feature with coverage-valued property Scene + + colour CV_Discrete. Grid. Point. Coverage {n} Weather. Station bounds: GM_Envelope theme: string + + location: GM_Point owner: CI_Responsible. Party temperature CV_Discrete. Time. Instant. Coverage Observations with coverage-valued results CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Proximate vs ultimate feature-of-interest The ultimate (“project”) thing of interest may not be directly or fully accessible 1. Proximate feature of interest embodies a sample design • • Rock-specimen samples an ore-body or geologic unit Well samples an aquifer Profile samples an ocean/atmosphere column Cross-section samples a rock-unit 2. Sensed property is a proxy • e. g. want land-cover, but observe colour • Post-processing required • Fo. I may change during processing – e. g. “scene” “tract” Some standard designs are common CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Sampling features Domain feature type +sampled. Feature Sampling. Feature. Complex Intention 0. . * GFI_Feature SF_Sampling. Feature + role +related. Sampling. Feature 0. . * + parameter [0. . *] + lineage [0. . 1] +related. Observation OM_Observation +related. Observation 0. . * SF_Spatial. Sampling. Feature + positional. Accuracy [0. . 2] SF_Specimen SF_Sampling. Point SF_Sampling. Curve Station Traverse Borehole SF_Sampling. Surface Mine. Level Section SF_Sampling. Solid Map. Horizon Mine

Viewpoints CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Oceanographic data processing Stage Viewpoint • Overall : to analyse and predict geophysical properties of ‘the ocean’ • feature • in-situ observations made of various oceanic parameters • coverage + sampling features • satellite observations made of sea surface • coverage • Observations integrated in numerical model gridded analyses and forecasts • coverage + sampling features • individual dynamic features detected (fronts, upwelling zones, etc. ) • feature CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Mineral exploration scenario Stage Viewpoint • identify a prospective region on the basis of general geological information • Feature • obtain geophysical imagery for the region of interest • Coverage • interpret the geophysics in terms of geological features of interest • Feature • construct drill holes for detailed sampling • Sampling features • Map variation of properties along borehole • Coverage • collect specimens along borehole • Sampling features • measure ore-concentration in specimens • Observations • characterize the internal structure of the ore body (concentration of commodity) • Coverage • Interpret geological features implied by property variation along borehole • feature CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Three viewpoints: Feature, Coverage, Observation • A Cell reflects the result of a single observation Specimen Au (ppm) Cu-a (%) Cu-b (%) As (ppm) Sb (ppm) ABC-123 1. 23 3. 45 4. 23 0. 5 0. 34 • A Column = variation of a single property across a set of locations CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010 • A Row gives properties of one feature

OGC Service stack • Different information-types accessed using different interfaces • • Maps – WMS Features – WFS Coverages – WCS Observations – SOS • Each interface is composed of a “set of operations” CSIRO Cox + Woolf, NDG Data and Information Modelling 2009

Serialization CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Data transfer • Data instances must be serialized for transfer CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

UML → XML/GML <gsml: Borehole gml: id="gc. DD 0214"> <gml: name code. Space="urn: cgi: authority: CGI: GSV“ >gc. DD 0214</gml: name> <sa: length uom="m">48. 3</sa: length> <sa: shape> <gml: Line. String gml: id="gc. DD 0214 survey“ >. . . </gml: Line. String> </sa: shape> <gsml: collar. Location> <gsml: Borehole. Collar gml: id="gc. DD 0214 c"> <gsml: location> <gml: Point gml: id="gc. DD 0214 p"> <gml: pos srs. Name="urn: ogc: crs: EPSG: : 4326“ >-31. 939 115. 832</gml: pos> </gml: Point> </gsml: location> <gsml: elevation srs. Name=“urn: ogc: crs: EPSG: : 5711 ” >34. 5</gsml: elevation> </gsml: Borehole. Collar> </gsml: collar. Location> <gsml: index. Data> <gsml: Borehole. Details>. . . </gsml: Borehole. Details> </gsml: index. Data>. . . </gsml: Borehole> CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

More information • https: //www. seegrid. csiro. au/twiki/bin/view/App. Schemas • in particular: • • OGCInformation. Models Feature. Model Observations. And. Sampling Information. Views CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Earth Science & Resource Engineering Simon Cox Research Scientist Phone: 08 6436 8639 Email: simon. cox@csiro. au Web: www. csiro. au Thank you Contact Us Phone: 1300 363 400 or +61 3 9545 2176 Email: enquiries@csiro. au Web: www. csiro. au CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Formalization CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO 19103 Conceptual Schema Language (UML Profile) • UML patterns • Avoid multiple inheritance • Put role-names on association-ends • Meaningful nouns – not ‘uses’ or ‘has’ • Guidelines for terms and labels • Lexical forms, nouns not verbs, singular not plural, short but not code • Packaging • Utility classes • Primitive types – Character. String, Boolean, Integer … • Names • Measure • Are float/decimal/double ever correct? • Record & record-type CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

19123 Coverage geometry and functions • Coverage = map from domain range CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Alt. view – collection of geometry-value pairs CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO 19107 Spatial schema CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

GM_Point vs. GM_Position CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO 19123 Grid types CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO 19108 Temporal schema CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO Documents CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

How to read ISO 19100 documents • IS vs TS • Standard vs technical specification • e. g. ISO 19115 Metadata model vs. ISO TS 19139 Metadata XML Implementation • ‘Normative’ vs. ‘Informative’ content • • • Chapter = “Clause” Clause is normative unless explicitly labelled “informative” Normative statements use SHALL, SHOULD, MAY, NOT Externally imposed requirements use MUST NOTE: • Informative consequences or corollary of a normative statement • EXAMPLE: • Non-normative illustration CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

What should you read? • Substance appears in different places in difference documents • ISO 19107 spatial schema • model is in the main clause • ISO 19115 metadata • model in Annex A; data dictionary in Annex B • Standard document structure • • Clause 1 – Scope statement Clause 2 – Conformance classes Clause 3 – Normative references Clause 4 – Terms and definitions Clause 5 – Symbols, notation, formalization Clause 6–N – text Annex A usually contains test-suite CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Metadata CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

ISO 19115 Metadata • ISO/TC 211 ‘poster child’ CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

MD_Metadata class CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

MD_Identification CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

SV_Service. Identification (ISO 19119) CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

LI_Lineage CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

DQ_Data. Quality, DQ_Element CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

EX_Extent CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Citations, contacts CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Can use the pieces without MD_Metadata CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Scientific data and metadata • OGC Sensor Web Enablement (SWE) initiative • producer-oriented viewpoint - Sensor. ML • Focuses on data production processes • processing chain, expressed in an ‘executable’ form • Optimised for satellite applications • consumer-oriented viewpoint – O&M (Observations and Measurements) • Focuses on the ‘subject’ of the observation • Feature of interest • Observed property • Observation procedure is ‘metadata’ for the result production event CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Specialization of Observations 1. Domain specific feature-type catalogue for the Feature of Interest 2. Controlled vocabulary of observed property-types 3. Set of instruments, procedures, protocols 4. Specialized observation types CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Applications define (and provide URIs) for: • • Feature-types Observed properties Procedures Result formats • Specialized types support intra-domain processing and analysis

Examples of observed property assignment • The 7 th banana weighed 270 gm on the kitchen scales this morning • The attitude of the foliation at outcrop 321 of the Leederville Formation was 63/085, measured using a Brunton on 2006 -08 -08 • Specimen H 69 was identified on 1999 -01 -14 by Amy Bachrach as Eucalyptus Caesia • The image of Camp Iota was obtained by Aster in 2003 • Sample WMC 997 t collected at Empire Dam on 1996 -03 -30 was found to have 5. 6 g/T Au as measured by ICPMS at ABC Labs on 1996 -05 -31 • The X-Z Geobarometer determined that the ore-body was at depth 3. 5 km at 1. 75 Ga • The GCM simulation run today using CMIP 3 indicated that the pressure field in the atmosphere tomorrow will be as given in pf 999_20081020_1 CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

In “pictures” CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Sampling & Observations OM_Observation GFI_Feature +feature. Of. Interest + + +property. Value. Provider 1 0. . * phenomenon. Time result. Time valid. Time [0. . 1] result. Quality [0. . *] parameter [0. . *] +generated. Observation 0. . * 1 +procedure GFI_Domain. Feature 1. . * Intention SF_Sampling. Feature +sampled. Feature protocol OM_Process GFI_Property. Type 1 +observed. Property Range +result Any

Compare with other models • • result ~ ‘observation’ Result ~ value Observed-property ~ parameter, measurand, observable Specimen ~ sample

CSML Sampling. Feature + sampling. Time [0. . 1] + parameter [0. . *] Spatially. Extensive. Sampling. Feature Specimen + + + material. Class sampling. Method [0. . 1] sampling. Location [0. . 1] size [0. . 1] current. Location [0. . 1] Sampling. Point Sampling. Curve Sampling. Surface Sampling. Solid + length [0. . 1] + area [0. . 1] + volume [0. . 1] Traverse Trajectory Profile CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010 Swath Ragged. Section

Discretization of sampling domain • Shape of extensive sampling feature does not describe the discretization • Different sampling regimes for different properties, within same manifold • E. g. sampling down a borehole, along a shipstrack or flightline • Add diagram here CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Value-adding chain • Observation/result • estimate of value of a property for a single specimen/station/location • data-capture, with metadata concerning procedure, operator, etc • Coverage • compilation of values of a single property across the domain of interest • data prepared for analysis/pattern detection • Feature • object having geometry & values of several different properties • 1. classified object, snapshot for transport • geological map elements • 2. object created by human activity, artefact of investigation • borehole, mine, specimen CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Sensor. ML – Process taxonomy CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010

Classification of processes CSIRO. Masterclass: Domain Modelling and Implementation - Sydney 03/12/2010