Table of contents
    No headers

    CORMAT is a product derived from the planimetric enhancement of NTDB data sets at the 1:50 000 scale. The correction matrices enable users to enhance the geometric accuracy of the less accurate NTDB.

    Each matrix consists of a set of points arrayed on a regular 100-metre grid. Each point describes the planimetric correction (DX, DY) to be applied at this location. The position of the points is given in UTM (Universal Transverse Mercator) coordinates based on the North American Datum of 1983 (NAD83). Each file constitutes a rectangular area covering the entire corresponding NTDB data set. Its delimitation corresponds more or less to National Topographic System (NTS) divisions at the 1:50 000 scale.

    bulk downoad from

    I've downloaded all the matrices for the Yukon plus a bit (425mb). Call or send an email if you want 'em. So far I haven't been able to use them in ArcMap, there are just too many points I guess. See for a plea for help with this.

    -- MattWilkie - 02 Jun 2005


    no usable responses from support forum or mailng list, initiated a support request today (15 june 2005). ESRI Canada's repsonse: " The research, consensus, and testing here confirms that this format is not usable in our software. You may have better luck contacting the government department that produced this and ask them if there is a way to export this type of data out to a different format - perhaps they have a universal format other than what they currently have this in that may be used in our software." (Incident #66491)


    Response from NRCAN:

    ... Thank you for you comments with regards to the information you have found on our MatCor documentation. I must admit with you, for this paragraph, it will need to be revised and read something similar to this; "Since there are presently no commercial software to accomplish this task, software adaptation is necessary in order to realise the planimetric correction, with the aid of the Bilinear Interpolation Algorithm provided in Appendix A".

    The statement you pointed out in documentation refers to a GIS software capable to adapt the matrix provided in appendix 'A'. At this moment, there no turn key commercial software available that can offer this application. I am told however that FME software can be adapted with the algorithm for planimetric correction, but it requires bench mark application and fine tuning to obtain identical results.

    Most likely you may be obliged to consult with your GIS software provider again such as FME, ArcMap or Arc Info, in order for them to adapt your software with a custom application to run this planimmetric correction matrix. ...

    I invite anyone interested in developing a program to apply the CorMat displacement to existing ArcGIS data to contact me with a proposal to do so with a time & price estimate.

    thank you,

    Matt Wilkie
    Geographic Information,
    Information Management and Technology,
    Yukon Department of Environment
    10 Burns Road * Whitehorse, Yukon * Y1A? 4Y9
    867-667-8133 Tel * 867-393-7003 Fax


    Sent email to gdal-development mailing list last week requesting estimate for custom cormat application development and met with resounding silence. Today tried the same with ESRI forums, this time crossposting in hopes of catching the right eyeballs.

    -- MattWilkie - 14 Dec 2005

    Silence from gdal-dev, so trying again. Frank Warmerdam of FWTools fame says he could do it relatively straightforwardly but is too busy to do it himself. He has too many irons in the fire to do it himself but is confident GDAL/OGR can do it, and that it would be really handy if the cormat data was in the same format as NTv2 datum shift info.

    Basically what I want is a program I can shove a shapefile (or coverage, geodatabase, etc.) at and have it spit back out a corrected one. It needs to be script friendly -- I have over 9 thousand files to throw at it and I'd prefer if it ran on windows.

    I threw up a request on and have had a couple of nibbles; we'll see what develops.

    -- MattWilkie - 19 Apr 2006



        Send feedback