A side note on geospatial data sharing and spatial data infrastructures (SDI)

This blog post is dedicated to provide a general overview over the field of geospatial and environmental data sharing. The term geospatial is actually tautologous: The prefix "geo" implies geography, which always relates things to each other based on their location, where nearer things are stronger related than things further away from each other (Tobler’s 1st law of geography). And the word "spatial" also means having an extent and location in a space. However "geospatial" nowadays is almost exclusively used in the field of digital data with geographical context. Therefore software that delivers, analyses, presents, processes, stores and retrieves is also often called geospatial software (having its origins in the good old GIS – Geographical Information Systems).

Part 1: GIS and GI Science intro

 Research areas that work on the science behind GIS and spatial data, on the analytical methods, processes techniques, on ways of (standardised) spatial data exchange, effective and efficient storage and retrieval, are called GI Science, Geoinformatics, Geomatics, Geocomputation, Spatial Computation or Spatial Science … There is actually quite some discussion, if the necessity, even an entitlement for such a dedicated mixed branch of computer science and geography exists, a similar discussion when geography emerged as an accepted field of research (ref). On the other hand you’ll often find the quote that "80% of data has a spatial component" or something like that. Apparently I can get the source right anymore, therefore you might handle this with care. It might go back to the 1990s when GIS software for PCs wanted to get their feet into the market (ref wikipedia analysis abstract and gis lounge). However  - a lot – of data in the geosciences have a spatial context and – a lot – of those data are needed for governmental agencies to manage land and water resources properly ☺ Agreed?

Part 2: The Opengeospatial Consortium, aka OpenGIS, aka OGC

"The Open Geospatial Consortium (OGC) is an international industry consortium of 479 companies, government agencies and universities participating in a consensus process to develop publicly available interface standards. OGC® Standards support interoperable solutions that ‘geo-enable’ the Web, wireless and location-based services and mainstream IT."
The OGC standards framework provides means to build a spatial Data infrastructure (LINZ - Land Information New Zealand) – which is "the technology, policies, standards, and human resources necessary to acquire, process, store, distribute and improve the usability of geospatial data. (SDI) facilitates the connections between these important sources of information, and allows people to find and access them."
Quite some of the OGC standards and web services are also ISO international standards. There are interface and service descriptions on the one hand and data encodings/formats and conceptual data models on the other side. I will provide a short summary:

WMS – Web Mapping Service (ISO 19128  WMS v1.3.0)

Essentially provides (web) maps (as images like png or jpg) output data Geographiclly correct images, png, jpg, view or portrayal service

Major methods: GetCapabilties and GetMap

WFS – Web Feature Service (ISO 19142 WFS v2.0)

Provides an interface to access, query, store and retrieve vector “features”, aka discrete data – like in ESRI shapefiles. Data is accessible by their data schema, which can be soft-typed and values in schema fields and location queries are utilized. Output GML (ISO 19136), which is in a particular “domain-specific” XML schema. With WFS-T – transactional – there is also support to write back to the WFS server.

GetFeature – get the data
DescribeFeatureType – get schema

WCS – Web Coverage Service

Provides an interface to access, query and retrieve raster imagery and coverages, grids (aka “fields”) as in continuous data. eg NetCDF-CF, GeoTIFF, ArcGRID

GetCapabilities and GetCoverage

CSW – Catalogue Service for Web (ISO 19115 CSW 2.0.2)

Provides an interface to access, query, store and retrieve metadata, aka data about the geospatial data, which is accessible through other geospatial webservices. Output is usually XML ISO 19139 metadata or Dublin core.

GetRecords – find metadata record by search criteria
GetRecordById – get one record by its unique id
DescribeRecord – metadata type
GetDomain – get range of values and/or keywords

SOS – Sensor Observation Service

Provides an interface to access, query, store and retrieve time-series based data that has been measured at locations, eg through sensors or field surveying/sampling. But the focus is to query on temporal and then on spatial or value comparison basis. Standard output formats are O&M, WaterML2.0 time-series and SensorML sensor/procedure metadata.
SOS is part of the sensor web enablement initiative (SWE) which advances to its version 2, where a lot of things become more flexible, but also moe complicated. I will write about that later ☺ SOS explicitly also describes a group of methods to insert sensor and observation data. This is handled through different profiles.


There are quite some commercial and Open Source software packages and tool kits available, for the desktop and server-based for the web that support or where explicitly written for OGC webservices and splendid Open Source resources in the web:

OSGeo Foundation: www.osgeo.org

(Spatially enabled) Databases:

Data Servers (store data in databases):

Web mapping tool kits / frameworks (take data from data servers):

  • openlayers
  • Mapbender
  • MapFish
  • Geomajas
  • Flash
  • Silverlight

Desktop clients supporting (at least partially) OGC standards:

  • QuantumGIS
  • uDig
  • Intergraph Geomedia

Finally a short summary of the eResearch project SOS for CliDB:

  • roughly implemented SOS 1.0.0 Core Profile for the 52North SOS server 4.0.0-beta (bug fixing and testing open)
  • using XMLBeans and substituting SWE 1.0 and O&M 1.0 XML elenents with each other
  • starting with the SOS DAO data backend for the NIWA climate database this week
  • testing a NIWA internal Java library to abstract from actual SQL database queries

And on it goes ... in the next weeks, I will write about New Zealand and international examples of OGC webservices implementations and SDIs for geospatial data publishing (where the data is basically public domain and needs to made accessible) and provide some closer insights to OGC SWE and the next generation sensor networks initiative SWE 2.0 et al. Stay tuned!

Submitted by AlexKmoch on