N 2401/N 2402 New Work Item Proposal - Observation and Sampling schema Simon Cox Research Scientist...

N 2401/N 2402 New Work Item Proposal - Observation and Sampling schema

Simon Cox

Research Scientist

28 May 2008

ISO/TC 211 Copenhagen 2008 CSIRO

Outline

• Background and history• Observation schema• Sampling features• Applications• Issues• Next steps


Background & history


Origins and motivation

• XMML• Boreholes

• Geophysics, gravity obs.

• Geochemistry/Assay data

• OCG Web Services/Sensor Web Enablement• Remote sensing

• Environmental monitoring

• Wildfire detection

• Geology field & lab observations, boreholes• Climate/Atmospheres/Oceans• Ecosystems/Taxonomic obs.• Water quality, groundwater• …


Motivation for developing a common model

• Cross-domain data discovery and fusion• Re-usable service interfaces


OGC Sensor Web Enablement

• OGC Web Services testbeds• OWS-1 2001 – OWS-5 2007

• Core elements of OGC SWE suite• SensorML – provider-centric information viewpoint

• O&M – consumer-centric information viewpoint

• SOS, SAS – http interface to observations

• SPS – tasking interface

• sweCommon – data-types & encodings, including coverage encoding

• TML – low-level sensor streams


O&M states

• OGC IPR 2002 (OWS-1.2)• OGC “Recommendation” 2003 (OWS-2)• OGC “Best Practice” 2006 (OWS-4)• OGC “Standard” 2007 • ISO/OGC NWIP 2008


Acknowledgements

• XMML Consortium• OGC testbed sponsors• Geoscience Australia• eWater CRC• CSIRO


Observation schema


Science relies on observations

• Provides evidence & validation• Involves sampling


What is “an Observation”

• Observation act involves a procedure applied at a specific time (Fowler & Odell, 1997ish)

• The result of an observation is an estimate of some property• The observation domain is a feature of interest at or over some

time


Examples

• The 7th banana weighed 270gm 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 H69 was identified on 1999-01-14 by Amy Bachrach as Eucalyptus Caesia

• The image of Camp Iota was obtained by Aster in 2003

• Sample WMC997t 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 simulation run on 2004-09-09 indicated that the pressure in the hanging-wall at 618 Ma was reduced 4 MPa


In “pictures”


Generic pattern for observation metadata

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

Where’s the “location”?

In the feature-of-interest, reconciling remote and in-situ observations

Many complexities are encapsulated in the Process (sensor), PropertyType, Result

Observ ation

+ metadata: MD_Metadata [0..1]+ samplingTime: TM_Object+ resultTime: TM_Object [0..1]+ resultQuality: DQ_Element [0..1]+ parameter: Any [0..*]

ProcessAnyFeature

Any{n}

PropertyType

generatedObservation

0..*

procedure1

propertyValueProvider

0..*

featureOfInterest

1

result

observedProperty1


Is O&M just an application schema?

• Yes – but domain neutral• Fundamental data-estimation pattern


«FeatureType»Observation


constraints{observedProperty must be member or component of member of featureOfInterest}{procedure must be suitable for observedProperty}{result type must be suitable for observedProperty}

«FeatureType»Process

«FeatureType»AnyFeature

«type»Any

{n}

PropertyType

«metaclass»GF_FeatureType

{n}

+ definition: CharacterString+ isAbstract: Boolean = false+ typeName: LocalName [0..1]

«metaclass»GF_PropertyType

{n}

+ definition: CharacterString+ memberName: LocalName

«FeatureType»ObservationCollection

member

1..*


0..*

procedure1


0..*

featureOfInterest1

result

observedProperty

1

«instanceOf»

1

carrierOfCharacteristics 0..*

«instanceOf»

Observation model is corollary of GFM

Observation role wrt target is propertyValueProvider

result type (and encoding) must be consistent with PropertyType

observedProperty is bound to target type


Procedure vs. observedProperty

• observedProperty supports discovery by observation users• “show me all the observations of temperature and wind-speed”

• procedure provides strict definition• “how was that value obtained?”

• …or provider-centric discovery• “show me all the data collected by instrument X”


Some properties vary within a feature

• colour of a Scene or Swath varies with position• shape of a Glacier varies with time• temperature at a Station varies with time• rock density varies along a Borehole

• Variable values may be described as a Function on some axis of the feature

• Corresponding Observation/result is a Function • If domain is spatio-temporal, also known as coverage or map


Variable property coverage valued result

«FeatureType»PointCov erageObserv ation

«FeatureType»Observ ation

«FeatureType»DiscreteCov erageObserv ation

CV_DiscreteElementCov erage

«FeatureType»ElementCov erageObserv ation

CV_DiscreteTimeInstantCov erageCV_DiscretePointCov erage{n}

CV_Coverage

CV_DiscreteCov erage{n}

«FeatureType»TimeSeriesObserv ation

result

result result result


Other specializations by result-type


«FeatureType»Measurement

«FeatureType»CategoryObserv ation

«FeatureType»CountObserv ation

«FeatureType»TruthObserv ation

«FeatureType»ComplexObserv ation

GenericName

«type»ScopedName

{n}

«type»Record

Number

«type»Integer

{n}

«enumeratio...Boolean

{n}

Measure{n}

Type

«type»RecordType

{n}

«FeatureType»GeometryObserv ation

«FeatureType»TemporalObserv ation

«type»GM_Object

{n}

TM_Object{n}

result

record 0..*

recordType 0..1

result

result

result

result

result

result


Assignment of property values

• For each property of a feature, the value is eitheri. asserted

• name, owner, price, boundary (cadastral feature types)

ii.estimated• colour, mass, shape (natural feature types) • i.e. error in the value is of interest


When is the Observation viewpoint interesting?

• If the data-acquisition metadata is of concern


Sampling features


Proximate vs ultimate feature-of-interest

The ultimate (“project”) thing of interest may not be directly or fully accessible

1.Sensed property is a proxy• e.g. want land-cover, but observe colour• Post-processing required• FoI may change during processing – e.g. “scene” “tract”

2.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

Some standard designs are common


«FeatureType»SamplingPoint

+ position: GM_Point

«FeatureTyp...SamplingFeature

«FeatureTyp...Trav erse

«FeatureType»SamplingCurv e

+ length: Measure [0..1]+ shape: GM_Curve

«FeatureType»SamplingSurface

+ area: Measure [0..1]+ shape: GM_Surface

«FeatureTyp...Interv al

«FeatureType»SamplingSolid

+ volume: Measure [0..1]+ shape: GM_Solid

«FeatureTyp...Swath

«FeatureTyp...Section

«FeatureTyp...LidarCloud

«FeatureTyp...Scene

«FeatureTyp...Observ ationWell

«FeatureTyp...Flightline

«FeatureTyp...Station

«FeatureTyp...Trajectory

«FeatureTyp...ShipsTrack

«FeatureType»SpatiallyExtensiveSamplingFeature

«FeatureTyp...Profile

«FeatureTyp...MapSurface

«FeatureTyp...Flitch

Name:Package:Version:Author:

SamplingManifoldpedagogy1.0Simon CoxSampling Manifolds

• N.B. SF provide bounds-for sub-sampling, but not details of their spatial decomposition


Basic sampling features

SamplingPoint


SamplingFeature

SamplingCurv e


SamplingSurface

+ area: Measure [0..1]+ shape: GM_Surface

SamplingSolid

+ volume: Measure [0..1]+ shape: GM_Solid

SpatiallyExtensiveSamplingFeature

Observ ation

AnyFeature

Spec imen

SamplingFeatureRelation

relatedOb servation

0..*

Inten tion

sampledFeature

1. .*

relatedSamp lingFeature 0..*

source

0..*

target

Nam e:Package:Version:Author:

Summary diagram: External classes«Leaf» Sampling Features1.0Simon Cox

Domain feature type


«FeatureType»Specimen

+ materialClass: GenericName+ currentLocation: Location [0..1]+ samplingMethod: ProcessModel [0..1]

«estimatedProperty»+ samplingTime: TM_GeometricPrimitive+ size: Measure [0..1]

«FeatureType»SamplingFeature

«FeatureType»LocatedSpecimen

«estimatedProperty»+ samplingLocation: GM_Object


«DataType»SamplingFeatureRelation

+ role: GenericName


processingDetails

0..*

0..*

target

relatedSamplingFeature0..*

source

Intention

sampledFeature 1..*

Name:Package:Version:Author:

Specimenpedagogy1.0Simon Cox

Specimen

• Specimens are SamplingFeatures used for ex-situ observation and analysis


Applications


Boreholes and logs

CV_Coverage

CV_DiscreteCov erage{n}


Observ ation

SamplingFeature

SamplingCurv e


DiscreteCov era geObserv ation

Bore hole

relatedOb servation

0..*

result


Borehole and logExternally governed packages1.0Simon Cox


Water quality measured along a ferry track

FerryTrack 20071113 :SamplingCurv e

notesshape=Curve987:GM_LineString

WQ200711113-1 :Observ ation

WQ/C-9 (Water Quality) :PropertyType

Albemarle-Pamlico Sound :AnyFeature

Fooglemeter 2000 :Observ ationProcess

WQ20071113/r1 :Rec ord





Station1 :Sa mplingPoint

notesposition=Point1:GM_Point





FT/S1 :SamplingFeatureRelation

notesrole="host track"





These stationsmust l ie on this track

relatedOb servation

target

sampledFeature

target

sampledFeature

featureOfInterest

relatedSamp lingFeature

target

observedProperty

featureOfInterest

proce dure

result

result

relatedOb servation

sampledFeature


sampledFeature

featureOfInterest

proce dure

observedProperty

result

proce dure

observedProperty


relatedOb servation


FerrySam pling-PExamples1.0Simon Cox

FerryTrack 20071113 :SamplingCurv e

notesshape=Curve987:GM_LineString

WQ/C-9 (Water Quality) :PropertyType

Albemarle -Pamlico Sound :AnyFeature


WQ200711113 :Observ ation

WQ20071113/r :CV_DiscretePointCov erage

sampledFeature

relatedOb servationproce dure

featureOfInterest

result

observedProperty


FerrySam pling-CExamples1.0Simon Cox


Water quality of aquifers observed in wells

Ev ertsWell

Leederv illeFormation

CottesloeWedge

GnangaraMound

Interv a lEW2 :SamplingCurv e

WQ3/1 :Observ ation

WQ2/1 :Observ ation

WQ2/2 :Observ ation

WQ3/1 :Observ ation

EW/EW1 :SamplingFeatureRelation

notesrole="intervalHost"

EW/EW2 :SamplingFeatureRelation

notesrole="intervalHost"


Farkleme ter XP :Observ ationProcess

NWC200 7/WQ :Observ ationCollection

RobsWe ll

Interv a lEW1 :SamplingCurv e

WQ3/1/r :CV_DiscreteTime InstantCov erage

WaterQuality :PropertyType

sampledFeature


sampledFeature

featureOfInterest

relatedOb servation

target

target

proce dure

observedPropertyobservedProperty

proce dure

observedProperty

proce dure

observedProperty

featureOfInterestrelatedOb servation

result

member

relatedOb servation

relatedOb servation

sampledFeature

relatedOb servation

sampledFeature

featureOfInterest

relatedOb servation

sampledFeature


featureOfInterestrelatedOb servationmember

member

memberproce dure


Well&IntervalsExamples1.0Simon Cox


Specimens and “outcrops”


Specimen processing chains


Domain profile

• feature of interest

• Feature-type is taken from a domain-model(e.g. Geology)


+ procedureParameter: TypedValue [0..*]+ procedureTime: TM_Object [0..1]+ quality: DQ_Element [0..1]+ observationMetadata: MD_Metadata [0..1]+ resultDefinition: CharacterString [0..1]+ samplingTime: TM_Object



«type»Any

{n}

PropertyType

result


0..*

featureOfInterest

1


0..*

procedure1

observedProperty

1

• observed property

• Belongs to feature-of-interest-type

• procedure

• Standard procedures, suitable for the property-type


In GeoSciML

SamplingPoint



Spec imen

SamplingFeature

SamplingCurv e


Bore hole Outc rop

AnyIdentifiableObject

AnyFeature Observ ation

Inten tionsampledFeature 1. .* relatedOb servation 0..*


Borehole a nd outcropExternally governed packages1.0Simon Cox


Deployments

• GeoSciML (IUGS Testbed)• Climate Science Markup Language v2 (UK Research Councils)• Water Observations Markup Language (Australia/US)• Various database schemas have been designed to match

• GA, NVCL


Issues


Document format & structure

• ISO format (not OGC)• Project leader will need some help here …

• (e.g. where is the ISO template?

• One-part or two-parts?• Clarify scope statements

• Title• O&M, vs O&S

• Abstract spec?• Include XML implementation? Status?

• Use of stereotype «FeatureType» in UML


Model

• Some properties are promoted to top level – are these justified?

Observ ation


ProcessAnyFeature

Any{n}

PropertyType


0..*

procedure1


0..*

featureOfInterest

1

result

observedProperty1


Relationship to ISO 19109 & 19123

• IS or TR• UK suggests incorporate content in revision of ISO 19109

(Observations) and 19123 (Sampling)

• O&S are domain-independent application schemas• AnyFeature-->GF_FeatureType

• PropertyType-->GF_PropertyType

• Sampling features provide context for coverage results• Manifold, but not decomposition details – i.e. domain-extent & CRS

• Clarification required in ISO 19109 re. coverages• feature-properties may be non-constant

GF_AssociationType may link a feature to a coverage

• New stereotype for properties?• «estimatedProperty» for those feature-properties whose values are

obtained through observation, and hence have a finite error


Relationship to GML

• gml:Observation – permissible for observation target not to be a feature?


Dependencies

• Feature & Coverage models - OK• Property-types & registers

• ISO 19110 revision deadlock

• Process (SensorML)• SWE Common – coverages & records


Next Steps?

Thank you

Exploration & MiningSimon CoxResearch Scientist

Phone: +61 8 6436 8639Email: [email protected]: www.csiro.au/em

Contact UsPhone: 1300 363 400 or +61 3 9545 2176

Email: [email protected] Web: www.csiro.au


Observed property

• Observed property• Length, mass, temperature, shape

• location, event-time, orientation

• colour, chemical concentration

• count/frequency, presence

• species or kind (classification)

• Expressed using a reference system or scale• Scale may also be ordinal or categorical

• May require a complex structure

• “Sensible”, but not necessarily physical …


Feature-of-interest

• The observed property is associated with something• Location does not have properties,

the substance or object at a location does

• Observed property must be logically consistent with the feature-of-interest

• E.g. rock-density, pixel-colour, city-population, ocean-surface-temperature

• … i.e. the Observation “target”


Procedure

• Instruments & Sensors • Respond to a stimulus from local physics or chemistry

• Intention may concern local or remote source

• Sample may be in situ or re-located


Procedures are usually process chains

• Procedure often includes data processing, to transform “raw” data to semantically meaningful values

• Voltage orientation

• count radiance NDVI

• Position + orientation scene-location

• Mercury meniscus level temperature

• Shape/colour/behaviour species assignment

• This requires consideration of “sensor”-models and calibrations


Advanced procedures

• Modelling, simulation, classification are procedures• “raw” data == modeling constraints

(sensor-outputs=process-inputs)

• “processed” data == simulation results (outputs)

• “interpreted” data == classification results (outputs)

• SensorML provides a model and syntax for describing process-chains


RockSample-A :Specimen

DensityItA :Observ ation

Density :Phenomenon

Densitometry :Observ ationProcedure

2610 kg/T :Measure

2006-11-23 :TM_Instant

Leederv ille, WA :Location

RockSample-B :Specimen

DensityItB :Observ ation

2580 kg/T :Measure


West Leederv ille, WA :Location

+time+result

+procedure+observedProperty

+featureOfInterest

+sampl ingLocation

+density

+sampl ingLocation

+time


+featureOfInterest

+result

+density

ProbeItA :Observ ation

Material :Phenomenon

Microprobe :Observ ationProcedure

MineralDistribution :CV_Cov erage

2006-11-24/2006-11-26 :TM_Period



+observedProperty +procedure

+result+time

+material

+featureOfInterest

+sampl ingLocation RockSample-A :Specimen

2610 kg/T :Measure


+density



Density :Phenomenon


2610 kg/T :Measure



+featureOfInterest


+result

+density

+time


2610 kg/T :Measure


RockSample-B :Specimen

2580 kg/T :Measure

West Leederv ille, WA :Location

+density

+sampl ingLocation

+density

+sampl ingLocation

ProbeItA :Observ ation

Material :Phenomenon

Microprobe :Observ ationProcedure


2006-11-24/2006-11-26 :TM_Period



Density :Phenomenon


2610 kg/T :Measure




+result+time

+featureOfInterest

+material

+featureOfInterest


+result

+density

+time

+sampl ingLocation



2610 kg/T :Measure


+material

+density

+sampl ingLocation

Observations, features and coverages

Feature summary

Property-valueevidence

Multiple observations one feature, different properties:feature summary evidence

A property-valuemay be a coverage

Same property onmultiple samplesis a another kindof coverage

Multiple observations different features, one property:coverage evidence


“Cross-sections” through collections

Specimen Au (ppm) Cu-a (%) Cu-b (%) As (ppm) Sb (ppm)

ABC-123 1.23 3.45 4.23 0.5 0.34 • A Row gives properties of one feature

• A Column = variation of a single property across a domain (i.e. set of locations)


SOS

getObservation

getResult

describeSensor

getFeatureOfInterest

Accessing data using the “Observation” viewpoint

WFS/Obs

getFeature, type=Observation

WCS

getCoverage

getCoverage(result)

Sensor Register

getRecordById

WFSgetFeature

e.g. SOS::getResult == “convenience” interface for WCS


WFS/SFS

Accessing data using the “Sampling Feature Service” viewpoint

WFSgetFeature

WCSgetCoverage

getCoverage(property value)

SOSgetObservation

Commondata

source

getFeature(sampling Feature)

getFeature(coverage property value)

getFeature(relatedObservation)

getCoverage(result)

SensorRegister

getRecordById (procedure)

getFeature(featureOfInterest)

getObservation(relatedObs)

getResult(property value)


WFS

Accessing data using the “Domain Feature” viewpoint

WCSgetCoverage(property value)

getFeatureSOS

getResult(property value)

The “George Percivall preferred™” viewpoint #1– observations are property-value-providers for features

??


WCS

Accessing data using the “just the data” viewpoint

WFSgetFeature/geometry(domain exent)

getCoverageSOS

getResult (lots of ‘em)(range values)

The “George Percivall preferred™” viewpoint #2 – observations are range-value-providers for coverages


Conclusions

• Different viewpoints of same information for different purposes• Summary vs. analysis

• Some values are determined by observation• Sometimes the description of the estimation process is necessary

• Transformation between views important• Management of observation evidence can be integrated

• (Bryan Lawrence issues) • For rich data processing, rich data models are needed

• Explicit or implicit

• Data models (types, features) are important constraints on service specification


Development and validation of O&M model

• Developed in the context of • XMML Geochemistry/Assay data

• OGC Sensor Web Enablement – environmental and remote sensing

• Subsequently applied in• Water resources/water quality (AWDIP, WRON)

• Oceans & Atmospheres (UK CLRC, UK Met Office)

• Natural resources (NRML)

• Taxonomic data (TDWG)

• Geology field data (GeoSciML)

• I could have put dozens of logos down here


O&M Status

• OGC Standard December 2007

• ISO/TC 211 New Work Item June 2008

• Core piece of OGC Sensor Web Enablement suite• See also: SensorML, Sensor Observation Service, SAS, SPS

• Basis for Water Observations ML• Key aspect of GeoSciML

N 2401/N 2402 New Work Item Proposal - Observation and Sampling schema Simon Cox Research Scientist...

Documents

Transcript of N 2401/N 2402 New Work Item Proposal - Observation and Sampling schema Simon Cox Research Scientist...