The Importance of a GNSS Site Calibration

Free  SX10 1-day Rental*

*Must have liability insurance and a valid California business. Limited to new customers only.

At CSDS, we strive to ensure that our network is providing the utmost in RTK precision and repeatability. However, real-world accuracy and assurance can only be achieved by checking into control and performing a GNSS site calibration aka “localization.” This transforms our network reference frame and corrections onto your own network of control points.

There are four important questions you should ask when using an RTN:

  1. What is the datum or reference frame that is required for your particular surveying project?
  2. Does your job require only relative RTK precision, or precision and accuracy to a predefined coordinate system?
  3. How well am I checking into control on my current project?
  4. How will I be able to transform onto control on an older project if the RTN I’m using has undergone an adjustment?

The CSDS RTN was established using a very precise adjustment over a large area using the NAD83 (2011) reference frame. Although this degree of precision and the reference frame may be acceptable for some applications, the only way to ensure the highest degree of accuracy from any RTN is by means of a GNSS site calibration onto known survey control.

Furthermore, as minor adjustments will be performed on our network in the future, locating or establishing your own project site control will be vital in order to ensure that you have a method of checking your precision and accuracy after future adjustments. If necessary, you may need to perform a new GNSS site calibration to compensate for any significant differences.

GNSS, Precision, Accuracy & Grid Coordinates

In the most generic terms: GNSS receivers speak natively in Site Calibration“WGS 84 Latitude and Longitude” coordinates. There are other refined, or “local” datums built for specific regions, such as the NAD27 & NAD83 (NAD meaning “North American Datum”). NAD27 is (based on the Clarke 1866 ellipsoid) or the more modern and regularly readjusted NAD83 datum (based on the GRS 80 ellipsoid).

One fundamental difference between WGS 84 and NAD83 is that NAD83 is defined to remain essentially constant over time for points on the North American Plate, whereas WGS 84 is defined with respect to the average of reference locations all over the world. But still, the raw coordinate frame/format for WGS 84 and NADXX are expressed in latitude, longitude and ellipsoidal height (not mean sea-level, or “geoid” elevations).

On a project site that is to be surveyed, engineered and constructed upon, a grid system Northing, Easting and Elevations (YXZ) coordinates allow us to calculate ground distances and elevations a bit more easily than trying to calculate the differences between raw observed latitude, longitudes and ellipsoid heights.

Published coordinate systems such as US State Plane or UTM grids have been created in order for us to transform these latitude and longitude values from WGS 84 or NAD83 ellipsoids into relative grid systems and positions expressed as northings, eastings and elevations. After all, who wants to try to calculate distances or angles on a construction site using degrees, minutes and seconds rather than grid coordinates?

Would you like an experienced land surveyor/geodesy expert to teach you about coordinate systems in much more depth? Then join us for our Basic Geospatial Concepts class: Click here to sign up for Training.

RTK GNSS systems typically yield +/- 1-2 cm relative precision. Traditional base/reference receivers can be set up on top of tripods over known points within published coordinate systems, thus allowing the rover receivers to yield both relative RTK precision and accuracy in relationship to other points within these published grid coordinate systems. But a few questions to ask yourself:

  • What if I’m unable to physically setup a base receiver over a known, published coordinate due to obstruction or risk of the base receiver being left alone in an unsafe location?
  • Now that I’m leveraging the value of an RTN and don’t need an additional base receiver on site anymore – how can I transform my GNSS project onto a coordinate system other than that of the CSDS RTN?
  • What about arbitrary or assumed coordinate systems (i.e. 5000/5000/100) that were derived by a total station or autonomous GNSS survey? How can I correlate the GNSS systems native latitude/longitude/height with these unpublished coordinate systems?
  • What if I create a network of control points based on a known coordinate system, perform a least-squares adjustment, therefore slightly changing the geometry of my control points from their known, published origins? How can I tie my GNSS system into them as precisely and accurately as possible?
  • How do I tie into property or section corners calculated from assumed coordinates and record distances/bearings?

All good questions, and the answer to all is a GNSS site calibration

Here in California, we are fortunate to live and/or work in a region with beautifully amazing, diverse topography. Unfortunately, we are sitting above (and within reach) of a few tectonic plates that are moving in different directions – and fairly rapidly in terms of tectonic velocities:

  • The Pacific Plate is moving to the northwest at a speed of between 7 and 11 centimeters or ~3-4 inches a year.
  • The North American plate is moving to the west-southwest at about 2.3 cm per year.

Such rapid movement made the recent split in our RTN necessary in order to achieve the highest amount of repeatable precision possible within each plate/subnet (Click here to visit our GPS Network Services page then go to the Network Updates tab).

The CSDS RTN reference frame will remain the same (based on NAD83 (2011)), but in order to compensate for subsidence and subtle movement upon each plate, we will continue to perform adjustments within each subnet a few times a year, hence the highly recommended practice of locating or establishing your own site control and performing a site calibration after each adjustment.

Understanding GNSS site calibrations (what exactly happens under the hood?)

The mathematical transformations that are applied in order to convert WGS 84 positions to grid coordinates are:

    • A datum transformation to convert the WGS-84 latitude, longitude, and ellipsoidal height coordinates to latitude, longitude, and ellipsoidal height coordinates relative to the ellipsoid of the local map grid.
    • A map projection to convert the local ellipsoid latitude and longitude coordinates into local map grid northing and easting coordinates (the height value is not altered in this process).
    • A geoid model to WGS-84 height to get approximate elevation above sea level.
    • A horizontal adjustment of the transformed grid coordinates to best fit local control data. This adjustment allows for any local variations in the projection system that cannot be accommodated in the overall datum transformation.
    • A height adjustment to convert the heights above the local ellipsoid or elevations derived from the geoid to local control elevations above sea level.
    • The horizontal and vertical adjustment are stored as part of the coordinate system definition for the project. All GNSS points in the database are updated using the calibration parameters, resulting in more accurate local grid coordinate values.

For more information on how to perform a GNSS site calibration using Trimble hardware and software, please visit our Support page.

If you have any additional questions about the CSDS RTN or GNSS site calibrations, please contact Tom Cardenas at:


To learn more about Site Calibration Setup Using Trimble Access v2015.10, view the link below:

Site Calibration Instructions

8 thoughts on “The Importance of a GNSS Site Calibration

    • Hi Pravesh,

      In most applications, the user doesn’t need to calibrate every time they’re on site, it would only be necessary if checking into a point yields high residuals.

    • Hi Pravesh,

      In a monitoring scenario, you’d want to calibrate one time and then specifically not calibrate again since the whole point is to catch movement of the points, not mathematically flatten the residuals.

  1. What is the difference between doing a site calibration after setting reference station at a known point within the local system and setting reference at an unknown and assuming a “herefix”. Will the Site Calibration parameters be different in these two cases?

    • Hi, the difference is that a site calibration is going to calculate a transformation based on the points you measure and may involve adjusting the position of the base station. If the base station is located on a known coordinate, these two things will conflict.

      The recommended workflow for a site calibration is to use the “Here” point you mentioned and calibrated to at least 4 points that surround the project. Hope that helps!

  2. What is the difference between performing a site calibration such as an initial go using calculated search points derived in the office, vs setting on one found point and calling it, say 10000,10000, and then setting a bearing to hold between that and a second found point and ultimately just moving from there to measure subsequent points?

Have a question or comment?

Your email address will not be published. Required fields are marked *