Collecting high accuracy data without Transformations

[MatthewDove]

I am volunteering for a local project and collecting spatial data with a high accuracy GNSS unit. Since it is a large scale type project the site area is very small and I want to get the most accurate data I can. So my unit will be collecting the data in NAD 1983 (2011) and my RTK corrections will also be coming through using the same GCS.

Will my data be as accurate as can be if I am projecting the data to WGS 1984 web aux. sphere OR should I upload a new base map with the same GCS and state plan project to collect the data in. This is so a transformation is not introduced.

The way I understand it is.... As long as my data is being collected in the correct datum I am good when it comes to accuracy right? At the end of the day I just don't want my point data to have a great deal of accuracy errors in terms of distance from each other.

For clarity, I am collecting into ArcGIS Field Maps which requires a Location Profile in order for a transformation to take place between the GPS unit and the maps coordinate system you are collecting into. The things is, I do not want to transform anything if I do not have too.

[Eugene_Adkins]

We have source data that is schema based in GCS NAD1983 and projected to NAD1983 StatePlane. With our GPS unit setup and collection workflow, the features are collected in NAD83(2011) with a RTK correction setup on the unit itself. On Field Maps there is a location profile that transforms WGS to NAD.

This setup (which seems very similar to yours) has worked well for us. In the past, we have been able to walk back to underground GPSed points with a reliable level of accuracy (negative environmental factors not-withstanding).

Hope that helps.

[MatthewDove]

Thanks for the response Eugene_Adkins, as I understand i are collecting in NAD 1983 as well. You are correct it is very similar. My problem however is a visual one. As this project is at a very large scale (zoomed in) and small in terms of site area, I have noticed because of the WGS to NAD datum transformation there is a slight offset issue visually. As the public is viewing this point data overlayed on a vector tile imagery layer while viewing this Storymap on their phones I am trying to collect without a transformation. However it seems as if Field Maps forces to to transform the day to WGS. Is this correct?

[PaulLohr]

I would switch the basemap to use your local state plane coordinate system and avoid the conversion. I would also find a trusted control point to check in and out at. Of course this control point will reference the same coordinate system as your basemap. I like to use NGS control points when reasonably possible. If you use a basemap that references a coordinate system which doesn't match that of your control point, a conversion has to be done. Does NGS or a surveyor stand behind that conversion? 

The conversion may be within a reasonable error tolerance but I don't know how to quantify that. Even if I could quantify the conversion error, I'd rather be following the practices of surveyors.

Points collected between a check in and check out that meet the error tolerance can be assumed to be within that same tolerance (depending on your requirements). As an example, if you space your check in and check out 8 hours apart and collect points in between, a failed check out means you lose ~8 hours worth of work. Make a check at 4 hours and the risk is cut down to losing 4 hours worth of work. You could also watch for issues such as PDOP spikes and ionosphere issues and avoid those times. Surveyors with professional survey software have better ways of limiting error.

Horizontal error of ~ 0.10 feet at control points has been commonly achievable for us without canopy interference. There are times when the R2 / ArcGIS Field Maps / Trimble Mobile Manager simply won't connect or get to an RTK fixed position but tomorrow it will.

Have a great day.

[BillFox]

z-aware too?

[MatthewDove]

Not at the moment but for future collection projects yes.