Difference between revisions of "WCS Wrapper Standards"

From Earth Science Information Partners (ESIP)
(Created page with "Back to [http://aq-ogc-services.sourceforge.net/ aq-ogc-services.sourceforge.net] __toc__ '''Standards used in this project''' * CF-NetCDF for grids * CF-NetCDF for points * W...")
 
Line 5: Line 5:
 
'''Standards used in this project'''
 
'''Standards used in this project'''
  
* CF-NetCDF for grids
+
The OWS project attempts to use existing standards whenever possible,
* CF-NetCDF for points
+
and invent own conventions only when no agreement exists.
 +
 
 +
== NetCDF Standards ==
 +
 
 +
NetCDF is a well established binary standard with cross platform libraries. Unfortunately, NetCDF alone does not guarantee interoperability. NetCDF just describes the syntax, not semantics. Semantic problems include but are not limited to:
 +
 
 +
* How do you define what the projection of the data is?
 +
* How do you calculate lattitude and longitude coordinates to plot it on a map?
 +
* How do you get the time of the measurement?
 +
* What is the physical parameter the data is about?
 +
* What are the units of the data?
 +
* Is the elevation atmosphere of ocean depth?
 +
 
 +
And so on. These problems have been addressed with [http://www.unidata.ucar.edu/software/netcdf/conventions.html NetCDF conventions] which tell people, that if you mark time variable "hours since 1990-01-01" then it's easy to tell that 12 means 1990-01-01T13:00:00 without any help from humans.
 +
 
 +
== CF-Conventions ==
 +
 
 +
The recommended convention is [http://cf-pcmdi.llnl.gov/ NetCDF Climate and Forecast (CF) Metadata Convention] and it's latest version [http://cf-pcmdi.llnl.gov/documents/cf-conventions/1.5/cf-conventions-multi.html CF-1.5]
 +
 
 +
== CF Standard Names ==
 +
 
 +
The standard name table is used to identify real chemical/physical category of the data.
 +
 
 +
Official source: [http://cf-pcmdi.llnl.gov/documents/cf-standard-names/ CF Standard Name Tables]. The latest version is v15, so the document is alive and updated.
 +
 
 +
Datafed has combined some unofficial names from the HTAP project: [http://webapps.datafed.net/table.aspx?database=catalog&table=standard_names datafed standard names table] which may or may not be official one day.
 +
 
 +
=== CF-NetCDF For Cubes ===
 +
 
 +
This is the original data type and it standardizes vast majority of important
 +
metadata so that plug'n play processing is possible.
 +
 
 +
=== CF-NetCDF for Timeseries of Station Data ===
 +
 
 +
The type of data that OWS deals with is [http://cf-pcmdi.llnl.gov/documents/cf-conventions/1.5/cf-conventions.html#id2867470 Timeseries of Station Data]. Currently WCS supports this data type, but source being SQL and output being [http://en.wikipedia.org/wiki/Comma-separated_values CSV, Comma Separated Values] as MS Excel understands them.
 +
 
 +
== OpenGeoSpatial Web Services Standards ===
 +
 
 
* WCS 1.1.0 .. 1.1.2
 
* WCS 1.1.0 .. 1.1.2
 
* CSV to text data
 
* CSV to text data

Revision as of 11:20, December 3, 2010

Back to aq-ogc-services.sourceforge.net

Standards used in this project

The OWS project attempts to use existing standards whenever possible, and invent own conventions only when no agreement exists.

NetCDF Standards

NetCDF is a well established binary standard with cross platform libraries. Unfortunately, NetCDF alone does not guarantee interoperability. NetCDF just describes the syntax, not semantics. Semantic problems include but are not limited to:

  • How do you define what the projection of the data is?
  • How do you calculate lattitude and longitude coordinates to plot it on a map?
  • How do you get the time of the measurement?
  • What is the physical parameter the data is about?
  • What are the units of the data?
  • Is the elevation atmosphere of ocean depth?

And so on. These problems have been addressed with NetCDF conventions which tell people, that if you mark time variable "hours since 1990-01-01" then it's easy to tell that 12 means 1990-01-01T13:00:00 without any help from humans.

CF-Conventions

The recommended convention is NetCDF Climate and Forecast (CF) Metadata Convention and it's latest version CF-1.5

CF Standard Names

The standard name table is used to identify real chemical/physical category of the data.

Official source: CF Standard Name Tables. The latest version is v15, so the document is alive and updated.

Datafed has combined some unofficial names from the HTAP project: datafed standard names table which may or may not be official one day.

CF-NetCDF For Cubes

This is the original data type and it standardizes vast majority of important metadata so that plug'n play processing is possible.

CF-NetCDF for Timeseries of Station Data

The type of data that OWS deals with is Timeseries of Station Data. Currently WCS supports this data type, but source being SQL and output being CSV, Comma Separated Values as MS Excel understands them.

OpenGeoSpatial Web Services Standards =

  • WCS 1.1.0 .. 1.1.2
  • CSV to text data
  • WMS for images
  • WFS for points
  • http-get

Industry Standards for developers

  • python
  • lxml
  • numpy
  • PyNIO
  • webpy