Can RTK GPS Accuracy Really Be Trusted? What Error Factors and Comparative Testing Reveal

The figure "RTK GPS has an accuracy of 1-2 cm (0.4-0.8 in)" is often seen in catalogs and specifications, but many practitioners likely wonder whether that level of accuracy is always guaranteed in the field. In this article, we整理 the error factors that affect RTK GPS accuracy and explain what verification data reveal. By understanding the gap between catalog accuracy and field accuracy, you will be able to use RTK positioning more accurately.

Table of Contents

• The gap between RTK GPS catalog accuracy and field accuracy

• Error factor affecting accuracy 1: satellite geometry (PDOP value)

• Error factor affecting accuracy 2: multipath

• Error factor affecting accuracy 3: quality of correction data

• Error factor affecting accuracy 4: ionospheric and tropospheric variations

• Reliability of FIX solutions and verification methods

• What can be seen from actual comparative verification

• Creating on-site rules for accuracy assurance

• Proceeding with site work while verifying accuracy with LRTK

The Gap Between RTK GPS Catalog Accuracy and On-Site Accuracy

Many RTK GPS receiver datasheets state "horizontal accuracy ±1 cm (±0.4 in) + 1 ppm RMS" or "horizontal accuracy ±2 cm (±0.8 in)". This value is a statistical accuracy expressed as an RMS (root mean square) and represents the average error under good conditions.

However, in the field, this level of accuracy is not always achievable. Many factors affect actual measurement accuracy, including the measurement environment, satellite geometry, the quality of correction data, and equipment setup. It is important to understand that catalog accuracy refers to values under benchmark conditions (open ground, many satellites visible, and good reception of correction data).

Error Factor Affecting Accuracy 1: Satellite Geometry (PDOP Value)

PDOP (Position Dilution of Precision: position accuracy degradation rate) is an indicator that quantifies the effect of satellite geometry on positioning accuracy. The lower the PDOP value, the better the satellite geometry and the higher the positioning accuracy.

PDOP values of 1–2 indicate an ideal satellite geometry and the highest accuracy can be expected. PDOP values of 3–5 are within a practical range and provide accuracy adequate for most tasks. PDOP values of 6 and above lead to significantly reduced accuracy and should be avoided for critical surveying work.

In urban areas where high-rise buildings are densely packed or in deep valleys, the number and geometry of satellites in view can deteriorate, leading to higher PDOP values. Because a multi-GNSS receiver can use satellites other than GPS, the number of satellites received increases and this contributes to improving PDOP values.

Error factor affecting accuracy 2: Multipath

Multipath is a phenomenon in which satellite signals reflect off surrounding objects such as buildings, structures, and terrain, causing direct and reflected waves to arrive mixed at the receiver. Because the reflected waves travel a longer path than the direct waves, the apparent arrival time is delayed, which introduces errors into distance calculations.

Concrete walls, steel-frame structures, metal mounts, and water surfaces are common sources of multipath. When affected by multipath, a FIX solution can still have systematic errors of several centimeters to more than ten centimeters (several cm to over 3.9 in).

For on-site multipath mitigation, it is effective to install the antenna at a sufficient distance from obstacles・use a high-performance antenna that is resistant to multipath・and measure the same point at multiple times of day and average the results.

Error Factor Affecting Accuracy 3: Quality of Correction Data

When using network RTK, the quality of the service that delivers correction data directly affects positioning accuracy. If the distance between the reference station (electronic reference point) and the work site is large, the ionospheric and tropospheric conditions differ between them, so the corrections are not completely canceled and residual errors remain.

In general, as the baseline length (the distance between the reference station and the rover) increases, a distance-dependent error of about 1 ppm is added. For example, with a baseline length of 10 km (32808.4 ft) an additional error of 1 cm (0.4 in) occurs, and with 100 km (328084 ft) an additional error of 10 cm (3.9 in) occurs. In the network RTK VRS (Virtual Reference Station) method, this error is reduced by generating a virtual reference station near the work site, but in areas where the density of the electronic reference station network is low, VRS accuracy also decreases.

Error Factor 4 Affecting Accuracy: Ionospheric and Tropospheric Variations

GPS signals experience refraction and delay when passing through the atmosphere (the ionosphere and the troposphere). These effects are largely canceled out by standard differential processing, but during periods of high solar activity or the occurrence of geomagnetic storms the ionosphere becomes disturbed, which can leave larger residual errors than usual.

Solar activity varies on an approximately 11-year cycle, and ionospheric disturbances occur frequently during solar maximum. There have been reports domestically that the stability of RTK positioning decreased when large-scale geomagnetic storms occurred.

Because tropospheric delay depends on water vapor content, accuracy may be reduced during the rainy and typhoon seasons or in high-humidity environments.

Reliability and Verification Methods for FIX Solutions

Even when an RTK receiver indicates a "FIX" solution, the risk of a false FIX (a state in which an incorrect integer ambiguity has been fixed) is not zero. If a false FIX occurs, data may appear to be high-precision but actually contain errors on the order of several centimeters to several tens of centimeters.

As a practical method to detect and avoid false FIXes, cross-checking with known points (triangulation points, electronic reference points, etc.) is used. If known points exist near the work area, measure those points before and after measurements and confirm that the difference from the known coordinates is within the allowable error. This can reveal equipment anomalies, false FIXes, or setup mistakes.

Also, monitoring the receiver's DOP, the number of satellites received, and the delay time (age) of correction data, and establishing an operational rule to suspend data acquisition when abnormal values appear, will help ensure accuracy.

What can be learned from actual comparative testing

As a result of verifying RTK GPS accuracy in the field, the following trends are generally observed:

In open squares, farmland, and on paved roads, horizontal accuracy of approximately 1–2 cm (0.4–0.8 in) and vertical accuracy of approximately 2–4 cm (0.8–1.6 in), close to the catalog accuracy, can be obtained reliably.

In urban areas with dense clusters of buildings, it has been observed that obtaining a FIX solution can take time, that FIX is prone to being lost, and that even when obtained, horizontal errors of 3–5 cm (1.2–2.0 in) can occur.

In mountainous areas under tree cover and in valley bottoms, satellites are often blocked so FIX solutions cannot be obtained—or if obtained, are lost quickly—causing a significant drop in work efficiency.

Establishing On-site Rules to Ensure Accuracy

To consistently ensure RTK positioning accuracy in the field, it is important not only to rely on equipment performance but also to establish operational rules.

As pre-measurement checks, it is recommended to specify obtaining a FIX solution; PDOP of 5 or less; an age of correction data of 10 seconds or less; and confirmation of the minimum number of visible satellites (generally 6 or more). During measurements, periodically confirm that the FIX solution is maintained, and if FIX is lost, suspend measurements until it is reacquired. After measurements, verify accuracy by comparing with known points.

Documenting these rules as SOPs (Standard Operating Procedures) within the organization and establishing a system that enables all field staff to operate under the same standards will contribute to guaranteeing the organization's positioning quality.

Proceed with field work while checking accuracy with LRTK

The LRTK (iPhone-mounted GNSS high-precision positioning device) features an easy-to-use design that lets you take measurements while checking positioning status (FIX solution, FLOAT solution, number of satellites received) in real time from an iPhone app.

Network RTK support enables stable acquisition of FIX solutions over wide areas nationwide, supporting measurements under conditions that minimize sources of error. Because it can be used simply by attaching it to an existing iPhone, field staff who are not surveying specialists can obtain accurate positioning information while monitoring the positioning status.

To fully leverage the accuracy of RTK GPS, not only equipment performance but also an understanding of error sources and on-site operational management are essential. By correctly identifying error sources and implementing appropriate verification procedures, you can improve the reliability of RTK positioning in the field and stabilize survey quality.