locked
is there an elevation dimension to the new spatial data type? RRS feed

  • Question

  • just started reading up on this, not a "geographer" by any stretch but am noticing that height or "elevation" is missing from the 2 or 3 threads I've done my best to read thru thus far.   From what I'm gathering, the new data type is still essentially 2 dimensional (eventhough curvature is accounted for in geography flavor).  It looks like you feed (in a ceratin order) the "corners" of your earth "chunk", and you have a data type representing that chunk, upon which intersection functions etc can be run.

     

    Our app may need to record location of assets in 3 dimensions, probably a lat and long but also a height.  Can the new data type do this?   We may need to know the distance between two points at different heights, will the new data type functions support this?    

    Wednesday, April 2, 2008 9:20 PM

Answers

  • All points has both X/Long, Y/Lat, Z and M values, so you should be fine.

    However, all spatial operations is only performed in 2D.

     

    Wednesday, April 2, 2008 10:41 PM
    Answerer
  • Z is regarded as the height, whether is the height above mean sea level, above terrain or above the geoid (it's really up to you or your vertical reference system). M is for Measure and is a little more tricky. Here's how ESRI defines it:
    Measures are values assigned to each coordinate. They are used for linear referencing and dynamic segmentation applications. For example, milepost locations along a highway can contain measures that indicate their position. The value represents anything that can be stored as a double-precision number.
    Microsoft's own definition is:
    The semantics of the measure value are user-defined but generally describe the distance along a linestring. For example, the measure value could be used to keep track of mileposts along a road.
    You can use the Measure value for anything you like. For instance it could also be accuracy of each vertex, or time represented as a double to give you a 4th dimension to your data.
    Thursday, April 3, 2008 4:53 PM
    Answerer

All replies

  • All points has both X/Long, Y/Lat, Z and M values, so you should be fine.

    However, all spatial operations is only performed in 2D.

     

    Wednesday, April 2, 2008 10:41 PM
    Answerer
  • thanks Morten, can I assume Z is 3rd dimension/coordinate?  What is M? 

     

    Thursday, April 3, 2008 12:57 PM
  • I think I get it, looked at BOL and it looks like Z is akin to the Z axis we used in high school, M looks like a measurement (eg meters).  I'm guessing the measurement is most important in the geography data type's elevation, I dont believe long and lat are ever stated in terms of a Measurement.

     

    But M also seems to be an ingredient in categorizing "like" spatial columns, so much so that they cannot participate in the same geo function together otherwise, eventhough they are in the same table and column.

     

    Measurement and apparently some angular stuff (sounds like there are different preferences/needs out there for the angular stuff) that helps to project a "chunk" of earth onto a flat surface for spatial recording and functionality, define the well known category or SRID (Spatial Reference Identifier)  that a particular spacial value (instance)  belongs to.

     

    Apparently this SRID is stored with the other spatial "pieces" of info associated with a particular spatial value, and your table can contain spatial values with different SRIDS.  SRIDS of zero are allowed, it looks like this is a generic category most appropriate for simple/straight geometrical manipulation.      

     

    Thursday, April 3, 2008 2:46 PM
  • Z is regarded as the height, whether is the height above mean sea level, above terrain or above the geoid (it's really up to you or your vertical reference system). M is for Measure and is a little more tricky. Here's how ESRI defines it:
    Measures are values assigned to each coordinate. They are used for linear referencing and dynamic segmentation applications. For example, milepost locations along a highway can contain measures that indicate their position. The value represents anything that can be stored as a double-precision number.
    Microsoft's own definition is:
    The semantics of the measure value are user-defined but generally describe the distance along a linestring. For example, the measure value could be used to keep track of mileposts along a road.
    You can use the Measure value for anything you like. For instance it could also be accuracy of each vertex, or time represented as a double to give you a 4th dimension to your data.
    Thursday, April 3, 2008 4:53 PM
    Answerer