Contents of this page may be automatically translated, we take no responsibility for the accuracy of the translation. Feel free to contact our customer support centre if you have any questions.
GNSS measurements are affected by various sources of error. If the impact of the sources of error is large, it also means that the uncertainty in the position of your GNSS receiver increases. In some cases, you may not be able to get a position at all. Here we describe the main sources of error that can affect the uncertainty of your estimated GNSS position.
Uncertainty in the timing of satellite and receiver clocks
Clock error means that the satellite does not transmit the signal, or that the receiver does not receive the signal exactly at the time indicated. However, satellite clock errors are much smaller in absolute terms than receiver clock errors, as satellites use atomic clocks with very high precision and receivers use much simpler clocks. The clock errors can be reduced or eliminated by relative measurements.
Uncertainty in the orbits and positions of satellites
Orbit errors mean that the GNSS satellite is not exactly in the position predicted in the transmitted orbital data. This source of error can be reduced by relative measurement, or by calculating the position afterwards when you have access to final orbits of good quality.
Influence of the ionosphere on satellite signals
The influence of the ionosphere on satellite signals is generally the largest source of error in GNSS surveying. The ionosphere is an area in the upper part of the atmosphere that contains charged particles due to solar radiation. As the GNSS signal travels through the ionosphere, it is affected in a way that is directly proportional to the frequency, which allowing you to reduce the uncertainty of calculated positions using multi-frequency or relative measurement.
Influence of the troposphere on satellite signals
Under normal Swedish conditions, the effect of the troposphere on the satellite signals is less than the influence of the ionosphere. But the influence of the troposphere is more difficult to reduce because it is not frequency dependent. The troposphere is the lowest part of the atmosphere, where weather occurs, and extends up to 7–17 kilometres above the earth's surface depending on latitude. In the troposphere, the GNSS signal is mainly affected by water vapour, which varies greatly in time and space.
Multipath error due to satellite signals not going straight from satellite to receiver
Sometimes GNSS signals do not travel the shortest distance between satellite and receiver but are reflected via other objects on the way. This phenomenon is called multipath error, and the impact on calculated positions is highly dependent on local conditions around the GNSS receiver. When surveying GNSS in an environment with tall buildings, trees or other objects that obscure the line of sight to the satellites, multipath errors are more common than when surveying in open landscapes.
You can partially reduce the impact of multipath errors by measuring over a longer period. Instrument suppliers are also constantly developing algorithms to identify and reduce the impact of multipath errors in the receiver, so for accurate positioning, it is important that you use good quality GNSS equipment with up-to-date software.
Other sources of error that may affect the calculated GNSS position
There are also other sources of error that must be taken into account in GNSS surveying, such as signal interference, hardware delay and relativistic effects. In most cases, these are handled automatically in the receiver.
In addition, the handling of equipment and the execution of the measurement affect the measurement uncertainty that you can expect in GNSS surveying. Advice (in Swedish) on measurement methodology and instrument management can be found in the HMK documents on GNSS-based detail surveying and control surveying.
Questions and answers
With the low-cost receivers found in, for example, smartphones and navigation equipment for cars, the measurement uncertainty at absolute positioning is in the range of 5-20 meters, depending on how much the satellite signals are disturbed at the time of measurement.
Construction surveying and other professional GNSS surveys often require lower measurement uncertainty than that. Then you need a better GNSS receiver in combination with supporting data from other GNSS receivers or services to reduce as much as possible the impact of error sources on calculated positions. Under good measurement conditions, it is possible to calculate a GNSS position with a few centimetres of standard uncertainty.
The fact that the uncertainty in GNSS surveying is greater in height than in plane is partly due to the geometry of the satellites in relation to the receiver. The satellites have a good spread in a north-south and east-west direction but are only above the receiver. A good distribution of satellites in the height component would require satellites below us, but for obvious reasons there are none.
GNSS surveying does not actually determine height above sea level, but height above a reference surface called the ellipsoid. These different heights differ from each other by 20–40 metres in Sweden. To convert from height above the ellipsoid to height above sea level, a so-called geoid model is needed, which also has an uncertainty that needs to be added to the uncertainty in the GNSS surveying itself.
The fact that GNSS signals are disturbed by atmosphere and reflections in the immediate environment also has a greater impact on height than the coordinates in plane.