Contents of this page may be automatically translated, we take no responsibility for the accuracy of the translation. Feel free to contact our customer support centre if you have any questions.
Here you can read about harmonization of spatial data.
What is meant by harmonization of spatial data?
Two sets of data can be said to be harmonized with each other if there are no obvious contradictions between them.
Contradictions between two sets of data
Possible contradictions can be, for example, variations in cartographic manner, differences in geometric resolution, differences in the use of concepts, differences in classification systems and more.
How do I reduce the contradictions?
A ways to reduce the contradictions between two data sets are by using common data specifications. The idea here is that these data specifications should be so detailed that any remaining contradictions do not reduce the usefulness of the data sets.
Within the INSPIRE work, several detailed data specifications have been prepared, one for each theme. This means that the concept of "data harmonization" for many has come to be equated with the process of, through a data transformation, ensuring that one's own data set meets the requirements specified in the data specification.
Description of the structure of a data set
A schema is defined here as a formal description of the structure of a data set. This structure can be, for example, tree-oriented (XML), relationship-oriented (GeoPackage), flat (Shape or GeoJSON) or based on triplets (RDF).
What does the term schema transformation mean?
The term schema transformation here means that data is transformed from one structure to another. Often a semantic transformation is also included in this process, ie the naming of object types and attributes is adapted to the rules specified in the data specification.
For many, therefore, the term “schema transformation” has come to be equated with the term “data harmonization ”. But in order for geodata to be considered fully harmonized, and especially in a cross-border perspective, a number of issues need to be considered and included in the entire transformation process.
Issues
Standardized data structure
The structure of a data set can be described by a schema. If data providers use standardized schemes, they can be efficiently processed by different users.
Standardized spatial reference systems
Standardized geodetic reference systems are a basic prerequisite for being able to combine geodata from different sources, and for to enable cross-border co-presentations and analyzes.
Most modern GIS clients today are able to transform in real time between different standardized spatial reference systems. The term "standardized spatial reference system" here means that the reference system must be defined in the EPSG register, where it is indicated by an EPSG code.
Geometric conformity
In order to perform certain calculations, the data sets correspond purely geometrically, for example at national borders. For example, to carry out route planning across a national border, it is required that the roads on both sides of the national border connect to each other.
Similar requirements exist for watercourses and other data themes. The requirement for geometric correspondence applies to all types of boundaries where data from two different sources must be combined, thus also to regional boundaries and municipal boundaries.
The requirement also applies when combining different themes. For example, a road cannot cross a watercourse without being connected to a bridge, tunnel or ferry line.
Geometric conformity can also be important when visually viewing multiple amounts of data. The credibility of the data sets decreases if there are obvious gaps between phenomena that are to be coherent.
Uniform data quality
Information is primarily used to make decisions and the security of these decisions depends to some extent on the quality of the amounts of data that form the basis for the decision. INSPIRE does not specify guidelines regarding accuracy or level of detail for different amounts of spatial data.
However, the European Commission has stated that if a type of spatial data occurs in several different scales (level of detail), then spatial data must be provided in all scales.
In raster data, as a rule, the degree of detail is given in terms of the resolution of the raster, ie how much of the ground is reflected in a pixel in the raster.
Uniform cartographic manner
A uniform cartographic manner makes it possible to present international data in a uniform way. This increases the visual readability and usability of, for example, cross-border display services.
InSPIRE's data specifications, cartographic modes are specified that can be used for presentation of data in display services. Sometimes they are too basic and may need to be modified. This becomes even clearer when several display services are added on top of each other in order to create a self-defined product for a specific area of use.
Uniform use of concepts
By using well-defined and generally accepted terms and concepts, the risk of the meaning of information being misunderstood is reduced. In INSPIRE, the meaning of object classes and attributes has been specified in INSPIRE's data specifications, in so-called object catalogs.
Using these internationally accepted concepts in their data is called semantic harmonization. You can call it using the same semantic reference system.
Information material
Instructional videos for data harmonization are available in Swedish at youtube.com:
Frequently asked questions about data harmonization (new window).
Training materials on INSPIRE and data harmonization can be found on INSPIRE's website:
- Data Harmonization (new window).
- Procedures for Data and Metadata Harmonization (new window).
- Examples of Data Transformation (new window).
To follow these courses, you must first register on the Geospatial Knowledge Base (GKB) Training Platform.