Active control stations are reference points on which GNSS receivers are permanently deployed to continuously track all navigation satellites in view.  They are usually connected to a communication network providing remote access to the observation data. Active stations may also stream satellite range measurements in real-time, enabling precise differential positioning and navigation.

Passive control points are ground markers or pillars on which users install their survey instruments to connect to the reference frame and integrate their surveys.  Geodetic control points with stable monuments can be re-observed periodically to estimate displacements caused by crustal dynamics.  Lower quality survey control points are usually suitable to validate low resolution geospatial datasets or loosely connect a survey to the reference frame.

The hierarchy of active and passive networks is presented in Figure 1.  More details about the Canadian active and passive control networks can be found on the Natural Resources Canada (NRCan) Canadian spatial reference system information page (accessed Oct. 21, 2021).

Figure 1. Hierarchy of the active and passive networks in Canada.

The separation between horizontal and vertical control networks originated from the techniques used to conduct conventional land surveys.  The network trilateration and spirit-levelling methods also referenced different surfaces, the geometric ellipsoid and geoid model. Horizontal networks consisted mainly of points separated by tens of kilometres found at high elevations while levelling lines connected benchmarks a few kilometres apart along roads and railways.  As a result, our national horizontal and vertical networks were connected at a limited number of junction points.

The vertical points were observed by spirit leveling over a period of almost 100 years. They are linked to mean sea-level at a few tide gauges along Canada’s Atlantic, Pacific and Arctic coasts. In 1928, a first national scale adjustment of all levelling lines was completed and the Canadian Geodetic Vertical Datum of 1928 (CGVD28) was established. This initial network was extended and densified until the end of the century by adding new levelling lines and re-adjusting them to update the fabric in a piece-wise fashion. Currently, the CGVD28 primary network consists of about 80,000 benchmarks covering less than half of Canada’s landmass.  To date, no practical or cost-effective means have been found to extend leveling lines northward.

Recently, the release of CGVD2013 and its reference geoid, the Canadian Gravimetric Geoid of 2013, has become the means to access the height system across Canada.  By applying  geoid heights to GNSS ellipsoidal elevations,  mean sea-level elevations consistent with the direction of water flow are now readily available across Canada.  With national coverage and centimetre-level precision, CGVD2013 offers a uniform vertical datum for all Canadians.  For users within CGVD28 network coverage, the transition to the new datum may not appear to provide significant short-term benefits over local areas.  It is expected that CGVD2013 will prove beneficial in the years to come as high-resolution datasets observed over large areas and extended corridors using airborne and satellite sensors become more readily available.

While CGVD28 elevations were affected by metre-level systematic errors from coast to coast, the CGVD2013 geoid-based realization is unbiased across Canada and offers superior performance over most spatial scales.  From a maintenance perspective, it is more sustainable in the long term. More information about the modernization of the height reference system can be found at the NRCan height reference system modernization page.

The availability of signals from navigation satellites has greatly reduced the need for inter-visibility between control stations and made possible the measurement of coordinate differences over longer distances.   It has also shown that better relative precisions could be achieved at almost all spatial scales. With the wide footprint of GNSS signals, users can now access distant geodetic control points from their project area, greatly simplifying field logistics and reducing associated costs. One of the most important advantages of the satellite navigation systems is that users readily obtain 3-dimensional coordinates.  With richer content, GNSS delivers both the horizontal and vertical components of position and can therefore satisfy a broader range of applications.

In Canada, the geodetic control points updated with 3D coordinates form the Canadian Spatial Reference System realization of NAD83(CSRS).  This CSRS designation reflects the 3D nature of the control points and the significantly higher level of precision.  The CSRS can therefore support both geomatics and geoscience users who exploit the centimetre to millimetre positioning accuracies of GNSS technologies and require long-term reliability.  Control points in the CSRS realization also offer elevations with respect to the ellipsoid at an accuracy level compatible with that of modern geoid models.  In that respect, the CSRS realization of NAD83 was an essential requirement for the modernization of Canada’s height system, as it offered the foundational geometric structure on which geoid heights could be anchored.

Accuracy refers to the “absolute” uncertainty of a point, meaning with respect to the coordinate system. For example, in a global terrestrial system, the accuracy of a control point reflects the uncertainty of its position with respect to the reference frame origin, i.e. the Earth’s geocenter. On the other hand, relative accuracy is the uncertainty of a point with respect to others in its surroundings. Alternatively, precision describes the ability to reproduce or repeat measurements with small variance between them.

Control points established using terrestrial methods were not accurate although most were precise at local to regional scales.  Traditionally, control points have been classified according to their relative precision using different “orders”. The order of a control point would reflect the size of the error ellipse within which the coordinate should be found, relative to other surrounding points, at a 95% confidence level.  Different classification schemes have been used by different jurisdictions across Canada to describe the precision of the control points they publish.   In general, three or four orders were defined to regroup control points with relative precisions varying from about 1 to 30 parts per million (1 ppm=1cm/10km). With GPS surveys now routinely delivering relative precision of 1 ppm or better, lower order control points established by conventional methods are now of limited use, although they still contribute to verifying the quality of lower resolution imagery.

Over the past 30 years, the use of space-based technologies has improved the accuracy and stability of terrestrial reference frames by three orders of magnitude, from a few metres to a few millimeters.  This is enabling long-term monitoring of small displacements over large areas, such as global earth dynamics and sea-level change. Realistic precision estimates given with respect to the reference frame itself also facilitate the computation of relative precision over different regions.  This new reality is resulting in agencies adopting a single-point mode of control point classification based on nominal network spacing (level) and accuracy (class).  This new classification scheme better reflects the quality of 3D control points.