New Datum For GCS in Australia GDA2020 is comming

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Fri, 13 Mar 2020 06:11:24 GMT

I used old or original MGA94 set in in V8i placed two points from old land gate data and labelled the coords with Civtools ( 3 decimals rounded off)

then I opened in CE 13 which already had MGA 94 ( but I didn't use it why later ) jumped at it...corrected error in post from here on

" use the GSB not the 7p ,to reset the MGA94 datum and then the reproject to mga2020 for realistic project areas you get very good results"

Now I've been told by cadastral surveyors that legally they only report to + or - 0.050m or 50mm for establishment and title survey.

But my data was accurate to the mm... maybe just lucky and only used small sample for WA and 3 grid pages from Melways for victoria melbourne for my trails but I'm pretty confident its the way to go with most jobs we do for road and rail projects...

RE: New Datum For GCS in Australia GDA2020 is comming

Christopher Albany — Fri, 13 Mar 2020 05:16:40 GMT

Hi Lorys,

I used the GSB (NTv2 grid file). When I tried the 7 parameter, I get differences in the 1m range.

Harping back to the old AMG84 to MGA94 days, I was always told the grid file option gave better results than the 7 parameter conversion. This concurs with the advice provided by Landgate - see above post.

I am very curious how you are getting exact conversion results with Landgate published values. Can you please breakdown how you achieved this?

Thanks,

Chris.

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Fri, 13 Mar 2020 03:20:02 GMT

[quote userid="935076" url="~/products/microstation/f/microstation-forum/168350/new-datum-for-gcs-in-australia-gda2020-is-comming/573365"]<10mm range, but I'm puzzled as to why there is any differenc[/quote]

" CORRECTION OF THIS POST "

SHOULD READ-

use the GDA94- GSB to CORRECT THE datum before reproject to gda2020?

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Fri, 13 Mar 2020 03:16:43 GMT

our IT dept would NOT like that and make the program folder pretty much locked down...

But I had success with the CE doing 2 steps if data is old v8i MGA94 an then rename the default model name to MGA2020 so its always obvious in v8i if the datum is MGA 2020.. but for now we suggest only use on datum in your project like pre v8i olden days...

RE: New Datum For GCS in Australia GDA2020 is comming

Mark Shamoun — Tue, 10 Mar 2020 08:46:39 GMT

No probs, I think we need better accuracy than that.

Maybe better responses on this in the Map forum?

RE: New Datum For GCS in Australia GDA2020 is comming

Christopher Albany — Tue, 10 Mar 2020 07:59:06 GMT

Thanks Mark.

Following this, the data re-projects to MGA2020. However, the results vary to those published by Landgate. It's in the <10mm range, but I'm puzzled as to why there is any difference?

Also, thanks too for the MGA2020 Custom GCS for V8i - I get the same results in CE & V8i.

Cheers,

Chris.

RE: New Datum For GCS in Australia GDA2020 is comming

Mark Shamoun — Sat, 07 Mar 2020 17:17:20 GMT

Hi Lorys,

Have you tried getting the MGA2020 Custom GCS in v8i working as I did here to see if MGA2020 GCS applied in CE was still visible in v8i?

https://communities.bentley.com/products/geospatial/desktop/f/bentley-map-forum/173593/can-anybody-using-bentley-map-select-series-4-help-with-datum-files/502358#502358

I know some guys in RMS and VicRoads were trying this out in the interim before they migrate to CE in the future.

RE: New Datum For GCS in Australia GDA2020 is comming

Mark Shamoun — Sat, 07 Mar 2020 17:08:40 GMT

Hi Chris,

Here is another explanation I got in the Maps forum when I was testing this out:

https://communities.bentley.com/products/geospatial/desktop/f/bentley-map-forum/174192/new-gda2020-geocoordinate-systems-in-ce---how-does-it-work

I probably should have shared that here to save a lot of wasted trial and error time...

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Wed, 04 Mar 2020 05:34:24 GMT

its worse if you change to MGA2020 via connect edition and then return to v8i, then the datum is none but the coordiantes are correct for MGA2020..

We need to go back to the old way we did things before we could do datum transforms in microstation ,,

I put some suggestions at the end of my tutorial on MGA2020 CE and V8i see ABOVE

a few were that you add the datum to the end of the file name, or rename the default model to MGA2020 this is a very easy fix

After you do the transform the file open in v8i press Alt and enter keys together and add into the title properties> summary>TITLE> TEST DRAWING_MGA2020

Then in the Microstation Manager you can see the title tab ( you have to enable it right click see right)

then repeat to enable this in win explorer

This title method works very well in both V8i and windows Explorer and once you've enabled this t stays visible

you can add it in CE via alt enter and edit the properties >extended>title >MGA2020_55 and it will be visible in Windows Explorer

but it seems in CE only visible after properties from Alt Enter, but the GCS tool in CE will tell you the datum is MGA2020 but not in V8i as that datum no recognised..

Alternative was we put nonprintable text " THIS FILE HAS BEEN REPROJECTEDTO MGA2020-55"

I'm not using any worksets in CE so I'm not sure if you can see the title tag attribute in CE start screen any using connect can you confirm how to set this like we had it in V8i.

RE: New Datum For GCS in Australia GDA2020 is comming

Christopher Albany — Wed, 04 Mar 2020 03:35:12 GMT

A big issue is once the data has been transformed to MGA2020, how do we know how the data was transformed?

Without this information, any transformation back to MGA94 may be incorrect. You need to be 100% sure that the data transforms back to the exact same values.

Perhaps instead of having to change the MGA94 files, the MGA2020 one changes i.e. MGA2020/GSB-50 or MGA2020-7P-50, then any backwards transformations will need to adhere to the chosen parameters.

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Wed, 04 Mar 2020 03:12:56 GMT

[quote userid="935076" url="~/products/microstation/f/microstation-forum/168350/new-datum-for-gcs-in-australia-gda2020-is-comming/572111"]I don't follow why Alain is saying that a change to the MGA94 [/quote]

What Chris & I are trying to say is if you created data in v8i and applied the MGA94 datum thru V8i, then open in CE if you have to change it to the other MGA94 datums the one I recommended is the ( corrected & updated this post 31/08/20) GSB and correct the datum without reprojection, then set the datum to MGA2020 and use reproject.

---------------------------------------------------------------------------------------

Further to this

However if your in CE and create / open a new file if no datum preset then just use the add datum MGA2020 thru the GCS tool

Note that in CE if you try to use the MGA94 datum is has a note to tell you don't use this use GSB version and later if you try to reproject this to MGA2020 it will transform and change your coordinatess I got better results in Victoria using that one for realistic size project areas...

RE: New Datum For GCS in Australia GDA2020 is comming

Christopher Albany — Wed, 04 Mar 2020 01:51:13 GMT

A good source of info can be found here;

https://www0.landgate.wa.gov.au/business-and-government/specialist-services/geodetic/gda2020

Simply, WGS84 is a dynamic datum, coinciding with GDA94 in 1994 and again with GDA2020 in 2020. This creates a well know issue for the web-mapping environment where GDA94 and GDA2020 data are misaligned when using null transformations [EPSG: 1150 and 8450].

It also states;

Unless the benefits of adopting another transformation method [e.g. EPSG: 8446] have been expressly identified, the conformal + distortion NTv2 transformation grid [EPSG: 8447] is the recommended transformation method for spatial data in WA. Use the 7-parameter similarity transformation method or NTv2 conformal only grids to transform data outside the conformal + distortion grid. Horizontal datum transformation should not affect orthometric height (AHD71).

I don't follow why Alain is saying that a change to the MGA94 datum would have been required. The transformation is to re-project data from one know datum to another. This is as we do with data from AMG84 to MGA94. This also uses grid files, just no distortion grid.

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Mon, 02 Mar 2020 01:45:53 GMT

here is the link to a tutorial I have put together to assist going forward with GDA2020 datum

communities.bentley.com/.../MGA2020-Datum-transform-for-Microstation-CE.pdf

if anyone has feedback to improve the document please PM me below..

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Wed, 26 Feb 2020 23:43:47 GMT

[quote userid="15894" url="~/products/microstation/f/microstation-forum/168350/new-datum-for-gcs-in-australia-gda2020-is-comming/571030"]To perform the conversion of MGA94 data to MGA2020 you must select one of the MGA94-7p or MGA94/GSB datum based GCS (set not reproject) then reproject to the MGA2020 based GCS. This will apply the 7paramter transform or the grid shift file based transf[/quote]

Well that's exactly what my research experiments has demonstrated, and you have confirmed it, thank you..

The old AMG 66 to MGA94 in V8i transformed beautifully which lead us to use the then new datum MGA94 as our seed file for new works ...

So we will need to be careful moving forward and try not to mix old data with new and set it all in the current datum or in the old datum its just using existing projects done in of MGA94 under CE cant use new datum without changing the datum via the method I out lined and you confirmed... I can foresee a lot of users are going to get confused here in Australia.

However now that I have figured it out I'll be sharing it with our Bentley User Group Victoria Australia and put it on our website to share with the rest of the user groups Australia wide.. hopefully with this post in the forum and our website most users here in Oz will be aware of how to deal with the new datum transformation.

RE: New Datum For GCS in Australia GDA2020 is comming

Alain Robert — Wed, 26 Feb 2020 15:22:54 GMT

Lorys,

MGA94 used to be the reference datum in Australia and GPS coordinates and control points would ultimately be expressed as MGA94 latitude/longitude (we could concider MGA94 to be coincident with WGS84 over Australia). This is why the MGA94 datum defined a null transformation to WGS84

With the introduction of MGA2020 this was changed and now coordinates considered coincident with WGS84 is MGA2020 and MGA2020 now is the one having a null transformation to WGS84. This implies that MGA94 is NOT coincident with WGS84 and is not the reference datum in Australia anymore. We cannot change the MGA94 datum without introducing difficulties in the management of older data. What we did instead was introcudes 2 new versions of MGA94 (MGA94-7P and MGA94/GSB) that define transformation to MGA2020.

To perform the conversion of MGA94 data to MGA2020 you must select one of the MGA94-7p or MGA94/GSB datum based GCS (set not reproject) then reproject to the MGA2020 based GCS. This will apply the 7paramter transform or the grid shift file based transform,

I understand this approach requires the additional step or 'correcting' previous data based on MGA94 but doing otherwise could have led to difficulties notably in the selection of the transformation method (7param of grid files).

These definitions are not available in v8i, unfortunately.

Alain

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Wed, 26 Feb 2020 06:07:33 GMT

ok I have done some more tests in V8i the datum for us in Austraila was MGA94 and Zone 55 as I'm in Victoria the zone 50 was to just check against the provided SSM HZ878 as known set of coordinates in both old and new datums...

we have set this datum MGA94 -55 in our v8i seed files

in CE when we attach the MGA94-55 datum and reproject to mga2020-55 nothing happens

But if I change the datum to MGA94/gsb-55( not avail in v8i) no reprojection (not avail in V8i) no reprojection

Then change datum to MGA2020-55 and reproject data it works and cords and objects moved to correct locations

Very strange as we would need to rebuild old files before using old existing exrefs and old data in CE..I can foresee this is going to give us some real headaches moving into future working with CE13.....

GSB IS RECOMMEDED and in v8i the GSB was only called up by EPSG 28355 not MGA94-55

THIS WAS REVISED 31/08/20 DO NOT USE THE 7P TRANSFORMATION......

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Wed, 26 Feb 2020 03:51:21 GMT

Mark,

I'm having a lot of difficulty with this..in CE 13 Mstn

I used the cords for SSM HZ 878 provided by Christopher Albyany... in MGA2020-50 to place a circle

I tested the cords in the freeware GdAy to see if the listed cords in the pdf from the web site match when converted to MGA94-50 and give or take a few mm the coordinates of MGA94-50 agree with the listed..

ok now starts my problem I created new file with MGA94-50 and place the circle at the coordiantes for H878 in that GCS datum

when I tried to ref attach my MGA2020-50 file on top of the active MGA94-50 no mater what setting I use reproject or transform

It doesn't land in the same place as the MGA2020-50 circle... it just comes in at its coordiantes .. this should not happen it should reproject and move the ref file circle to match up centres with the active file circle..

undo xref I tried the reverse open the mga2020-50 file and ref attached the MGA94-50 and it doesn't reproject so the ref file circle is not copied the MGA94-50 s another file and told it to reproject it data by change the GCS datum to MGA2020-50 and the circle didn't move it kept its old coordinates...it shouldn't work like this ..should it

I have attached here my files

here is the data for the circles you can copy paste xy = to draw the circles at the coordinates not transcription errors that way

HZ878 MGA2020-50 xy=386477.299,6468404.023

HZ878 MGA94-50 xy=386476.282,648402.528

( FYI checked with GDAY mga9450->mga2020-50 xy=386476.286,648402.522 so matches website cords near enough)

could somebody repeat these steps and can anyone check my 2 files.. don't know whats wrong here it should work same as when we worked in v8i and converted AMG66 to MGA94 by both methods reprojection changed GCS and by ref reprojection via xref

the result should be data moves and takes on new coordiantes, which is what used to happen..

communities.bentley.com/.../MGA94_2D00_50.zip

RE: New Datum For GCS in Australia GDA2020 is comming

Christopher Albany — Tue, 25 Feb 2020 05:56:32 GMT

Hi Mark,

Yes, specifically comparisons against published results, such as Landgate (WA - Department of Land Administration) Standard Survey Mark summary sheets.

We have found something odd happening, it almost appears as if the software is computing the transformation backwards. Essentially instead of having the MGA2020 values transformed in a NE direction, they seem to be transformed in a SW direction instead.

Have you found anything like this? I've attached the summary pages that we've used.

communities.bentley.com/.../Landgate-SSM-Summary-Pages.pdf

Thanks,

Chris.

RE: New Datum For GCS in Australia GDA2020 is comming

Mark Shamoun — Tue, 25 Feb 2020 05:31:25 GMT

Hi Chris, i've done a little bit of testing of this - anything specific?

RE: New Datum For GCS in Australia GDA2020 is comming

Christopher Albany — Tue, 25 Feb 2020 04:15:03 GMT

Just wondering if anyone in Australia has done any testing with MGA94 to MGA2020 using any of the Connect products?

RE: New Datum For GCS in Australia GDA2020 is comming

Mark Shamoun — Tue, 22 Jan 2019 13:16:41 GMT

GDA2020 Geographic Coordinate System files for v8i converted from CE U11 and discussed here:

https://communities.bentley.com/products/geospatial/desktop/f/bentley-map-forum/173593/can-anybody-using-bentley-map-select-series-4-help-with-datum-files

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Wed, 26 Sep 2018 01:44:48 GMT

Thanks Andrew but will copying the gcs library from CE to replace the V8i cgs library work? I some how doubt the compatabliity one being 64 bit and the other being 32 bit.

RE: New Datum For GCS in Australia GDA2020 is comming

Andrew Bell — Wed, 26 Sep 2018 00:48:49 GMT

No further development of MicroStation V8i is planned and as Alain mentioned the...

[quote userid="15894" url="~/products/microstation/f/microstation-forum/168350/new-datum-for-gcs-in-australia-gda2020-is-comming/482624"]GDA2000 geographic coordinate systems will be present in MicroStation CONNECT edition Update 12[/quote]

Also, as Mark mentioned

[quote userid="334014" url="~/products/microstation/f/microstation-forum/168350/new-datum-for-gcs-in-australia-gda2020-is-comming/482738"]replace your gcs library in or add a cusom one in v8i...in theory.[/quote]

https://communities.bentley.com/products/geospatial/desktop/w/geospatial_desktop__wiki/6548/--defining-a-custom-coordinate-system

but this would require Bentley Map, as you have also noted, to add a custom GCS.

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Tue, 25 Sep 2018 12:02:21 GMT

Andrew as ss4 will continue a while yet any plans for the new datum in the gcs library to have an update for GDA2020 or will that have to wait until 2020? ...

We used the AMG66 for many years before we actually took up the GDA94 ( around 2001-2005 ) as industry std for survey in Australia so I assume we'll keep using the GDA 94 for a few years more before we adopt GDA2020 as the industry survey std datum here in Oz.

RE: New Datum For GCS in Australia GDA2020 is comming

Lorys — Tue, 25 Sep 2018 11:53:16 GMT

thanks Andrew its good to know the actual time table....too bad I wont have retired before then... drat I needed an extra 3 years ... oh well by 2021 CE should be working as intended and a few extras beside....