We frequently work in a 'local grid' which is a simple orthogonal system typically rotated from north. For all practical purposes this can be considered an orthographic system that can be shifted and rotated to convert to and from UTM or some other national/regional grid. Since these local grids are usually covering fairly small areas distortion issues are generally ignored.
After importing local grid data its quite easy to shift the origin as needed either in the orthographic definition or using Move Horizontal/Vertical. The rotate is a little more tricky. I've been creating an encompassing circle with its centre at the rotation point and then applying Rotate on the local grid data plus circle selection. Since Rotate uses the (approximate) centre of data this seems to work fairly well - although the use of approximate in the documentation is a little concern.
"Most (99% +) of my project involve a layer imported from an AutoCAD file. Surveyors work of arbitrary datums as a rule. They just pick an origin 5 km away to force all of the numbers in the survey area to be positive. The rotation of the survey grid can be anything as surveyors usually rotate the job for best fit on a sheet of paper. Can you image the pain in the arse this causes to GIS users. The standard answer is to import the drawing and use the affine (scale rotate and translate) transform to move the survey data to the map. This system works but has its limitations. So this is what I want.
I want to be able to transform from arbitrary coordinate systems to real world coordinates and the back again to arbitrary coordinates. For example, I would transform a subdivision plan into real world coordinates so that I can do a constraints analysis using Manifold and the town planning layers, vegetation layers, flood layers etc. Then I would draw the no-go areas and translate this data back to the arbitrary coordinate system and export the data in dxf format so that it plugs back into the AutoCAD drawing without any drama.
Being able to import other AutoCAD drawing prepared on the same base as the first drawing. For example one drawing may contain the subdivision layout, another may contain a drainage plan. If there was a saved transform, then they could both be imported and they would register perfectly.
Mine grids. Some of the mines that I have worked on create their own mapping grid. The main axis of the grid is aligned to the main axis of the ore body. Entire GIS databases can be based on mine grid. This is great until you go to find something with a GPS. Again what we need is a saved transform that transforms data to and from mine grid from real world coordinates. "
It's been a while since anyone has added to this post & would have anyone resolved the issue of producing an XML that sets up a local grid? Yes, I am too from a mining/exploration company in Australia and simply can not work effectivly without a local grid projection.
I would like to see a in the Local Grid setup a common point (Lat-Long/XY pair), scale, & rotation method. A Two Common Point projection setup that derives the projection details as I've outlined below would also be particually useful.
What would make Manifold outstanding to other GIS packages is if Manifold was to add the facuility to add a coulmn showing the projection of the Z values to a local grid from a particular datum. For example, for many of the local mining grids that I've encountered generally have added 1,000 or 10,000 Z units (meters) to the Australian Height Datum (AHD). Say, if a height is at 350m AHD then the local grid Reduced Level (RL) is with 10,000m difference in local height 10,350mRL. I understand that adjusted RL's are used with mining & exploration we drill and or mine below the 0m AHD this produces co-ords that are in the minus' which is undesirable, and perhaps also that in the past many mining specific software packages could not handle negative co-ords. It would be very handy to have the calculation of the local grid RL relative to a nominated height datum integrated into the local grid projection as an option. Very handy indeed!
Following is an example of a small local grid around our minesite with projection information .. Even without the reprojn of the Z values, a simple rotation of local grids is essential for everyday geological (incl geochem, geophysical, environmental) work. Sure, this could be done with code. From time to time over the last 2 yrs i tried to point out the attraction of the MapInfo 'model' which allows simple tools to be added as buttons, etc. In part, this is facilitated by the (admittedly limited) ability to program some events in MapBasic - though MapBasic is a very awkward and limited coding system. That's why Encom's Discover add-on system to MapInfo Professional has been so successful. And as I pointed out, it's not cheap.
Since Rotate uses the (approximate) centre of data this seems to work fairly well - although the use of approximate in the documentation is a little concern. The "approximate" bit means that the accuracy of computing the rotation center is given by the location precision parameter of the context component. jharrop Posted:12/01/2004 10:22 AM
Is there a way I'm missing of better controling a rotation? You could try using the following sequence: insert a point at the rotation center, turn on Snap to Points, select the Insert Box on Center tool, click on the created point and increase the box to cover the entire drawing, rotate the entire drawing, remove the box and the point.
Following is an example of a small local grid around our minesite with projection information that I would like to apply to Manifold: ... I have filed a request to add this into our issue tracking system. Shane-O Posted:09/05/2005 4:31 PM
What would make Manifold outstanding to other GIS packages is if Manifold was to add the facuility to add a coulmn showing the projection of the Z values to a local grid from a particular datum. I agree. I have filed requests for the following additions: a) a query function to compute a real height, taking into account the curvature of the context datum, of a given surface location, b) a similar function to convert a given real height back to a "planar" height, c) a similar pair of functions in the Surface - Transform dialog, d) an option to adjust heights when projecting a surface. Anything else?
Having the ability to add user defined local grids would be of great benefit, particually to Geological, Exploration & Mining orientated industries. This should also make it easier to project rotated CAD drawings when they are often drawn on a design plane that isn't placed on a standard projection.
Also, being able to to project Z values would be very useful. I havn't looked into Manifolds GPS connectivity as yet, but to have Z values available in their native format from the geoid and translated to Z value of the defined projection on-the-fly would also be a fantastic addition for fuss-free DGPS data gathering. I have currently real problems with the geoid Z to local projection RL using ArcPad/ArcGIS when connected to a DGPS. ESRI is not helping us with this as ArcPad is essentially unsupported.
7fc3f7cf58