RTK Surveying Under Trees: How to Handle Canopy and Multipath Errors

Table of Contents

• Why is RTK surveying under trees difficult?

• How the canopy affects GNSS signals

• What is multipath error?

• Measures to achieve high-accuracy positioning under trees

• Utilizing multi-GNSS and the latest technologies

• Simple RTK surveying achievable with LRTK

• FAQ

Why is RTK surveying under trees difficult?

Even though GNSS surveying has become widespread, many people have experienced the frustration of “the position won’t stabilize…” when working in forests. Conducting RTK (Real-Time Kinematic) surveying in densely treed areas faces much greater challenges than surveying under open sky. RTK positioning can theoretically determine positions with centimeter-level accuracy, but to realize that performance it must receive strong, stable signals from multiple GNSS satellites. Under trees, branches and leaves cover the sky and block or attenuate satellite signals, so the number of receivable satellites decreases and signals tend to be weaker. As a result, the centimeter-level accuracy achievable on flat open ground may not be attainable, and in some cases positioning can become unstable or fail entirely.

Moreover, the problem of multipath error becomes prominent in forest environments. Signals reflected from trunks or the ground arrive at the receiver via multiple paths, causing errors in positioning data. These combined factors make RTK surveying in forests very challenging. However, if you correctly understand the issues and take countermeasures, there is room to improve accuracy even under difficult conditions. Below we explain in detail how the canopy and multipath errors affect measurements and introduce countermeasures and ways to leverage the latest technologies to mitigate those errors.

How the canopy affects GNSS signals

Tree branches and leaves (the canopy) block GNSS satellite radio waves and have a large impact on positioning accuracy. The moisture contained in leaves and branches particularly absorbs and attenuates radio waves, so under a dense canopy the signal strength (SNR) from satellites drops significantly. A reduction in the number of satellites received and a drop in signal strength are fatal to RTK positioning. Generally, RTK has difficulty obtaining a fixed solution (Fix) unless at least five satellites are being stably received; if the count drops to four or fewer, there is a risk that a position cannot be computed. Under trees, not only is the sky view restricted and the number of visible satellites reduced, but even received signals are weakened and noisier, so the solution may remain a float solution (Float) or the accuracy may degrade.

Satellite geometry also tends to be biased under a canopy. In environments where the sky is only partially open, the visible satellites may be concentrated in certain directions, worsening the geometric balance for position computation (DOP values). As a result, large errors can occur preferentially in particular directions such as east–west or north–south. For example, Japan has several satellites including quasi-zenith satellites overhead, but in a forest you may only catch some of them through narrow gaps above, making ideal positioning difficult.

To mitigate canopy effects, it is effective, where possible, to choose locations or times when the sky is more visible. For example, surveying in winter when trees have fewer leaves, or scheduling work for times of day when satellite geometry is favorable (when many satellites are at high elevation angles), can somewhat improve reception conditions. Also, avoid surveying in rain or fog if possible, because wet leaves increase radio attenuation; if you must work in such conditions, take extra time and wait for moments when reception stabilizes.

In practice, experiments using an RTK rover under trees showed that when six or more satellites were secured continuously the measured accuracy was nearly as specified (within a few centimeters (a few inches)), but when the count dropped to five or fewer the errors increased noticeably. The fix rate—the proportion of time a Fix solution is maintained—can remain relatively high (about 93%) in comparatively favorable forest conditions, but drops significantly in more obstructed environments. How many satellites you can continuously track is therefore key to maintaining accuracy under trees.

What is multipath error?

Multipath error occurs when signals from satellites are reflected or scattered by the ground or surrounding objects (buildings, rocks, tree trunks, etc.) and arrive at the receiver via multiple paths, causing positioning errors. GNSS receivers normally measure distance using the direct signal from each satellite, but if reflected waves are mixed in, the receiver computes an incorrect distance. For example, if a satellite signal is reflected off the ground or a tree trunk and takes a longer path to reach the receiver, there will be a time delay relative to the direct signal. That time delay corresponds to an extra distance in ranging calculations and can cause position offsets of a few centimeters (a few inches) to, in some cases, tens of centimeters (several inches to over a foot).

RTK uses carrier-phase measurements for higher accuracy than code-based positioning, but strong multipath can still prevent integer ambiguity resolution, making it difficult to obtain a fixed solution and, in the worst case, causing an incorrect fixed solution (false Fix). A false Fix can produce a coordinate that is severely erroneous, which is very dangerous. Fortunately, modern receivers include advanced multipath mitigation algorithms and the impact is reduced compared to the past, but multipath remains a non-negligible error source in forest environments. (Although not as extreme as in urban canyons, reflections from forest terrain and trunks still affect RTK accuracy.)

Surveyors can reduce multipath error by practical measures. First, consider antenna placement. Mounting the antenna in as open and high a position as possible reduces reception of ground-reflected waves. For example, extending the pole to raise the antenna height or fitting a metal ground plane under the antenna helps block unwanted low-elevation reflections. Also, if there are reflective objects (metal fences, vehicles, large rocks) near the survey point, move as far away from them as practical to reduce reflections.

As an observation technique, remaining stationary for a set period while measuring is effective. Multipath errors vary over time, so taking data over tens of seconds to minutes and averaging can cancel out errors and improve accuracy. Even in real-time RTK, waiting a few seconds until the Fix solution stabilizes before recording the point helps avoid outliers and increases reliability.

Measures to achieve high-accuracy positioning under trees

Considering the canopy obstruction and multipath issues described above, combine the following countermeasures when conducting RTK surveys in forests to maintain accuracy:

• High-performance antennas: Use surveying GNSS antennas with multipath mitigation. Choke-ring antennas or antennas with larger ground planes suppress reflections and improve signal-to-noise ratio (SNR). The rover’s antenna selection needs to balance portability, but choose as high-performance an antenna as is practical.

• Ensure antenna height: Place the antenna as high as possible. Extending the pole slightly can improve sky visibility and reduce reception of low-angle reflected waves. However, be cautious: taller poles can be less stable, so keep the pole vertical and work safely.

• Plan satellite geometry: Predict visible satellite count and geometry (PDOP) before surveying and choose times with favorable satellite conditions. Dedicated apps or web services can provide satellite prediction information so you can determine, for example, that a region has more satellites in the afternoon. In Japan, times when the quasi-zenith satellites (QZSS) are overhead are advantageous, as they provide strong signals from above even under a canopy.

• Adjust positioning mode: Consider changing measurement modes as needed. If real-time RTK proves difficult, record raw data and process later (PPK: Post-Processed Kinematic) to obtain a solution. Post-processing over longer data sessions can cancel out errors and potentially improve accuracy.

• Combine other methods: Don’t rely solely on GNSS; combine with other surveying methods such as total stations or terrestrial laser scanners when necessary. For example, establish GNSS control points at open locations and then use a total station to measure details within the forest—this hybrid approach can ensure accuracy.

Applying these measures makes it possible to obtain as stable a position as possible under trees. However, in extremely obstructed forests GNSS positioning may not be feasible despite precautions. In such cases, do not force measurements: consider moving the point, or revising the survey plan.

Utilizing multi-GNSS and the latest technologies

Recent advances in GNSS positioning have produced new solutions that help in harsh environments like forests. One is using multi-GNSS-capable equipment. Positioning with GPS alone limits the number of visible satellites, but multi-GNSS uses multiple systems (GPS, GLONASS, Galileo, BeiDou, QZSS, etc.) simultaneously. This greatly increases the total number of satellites visible in the sky, making it more likely to acquire sufficient satellites in forests or valleys. There are increasing reports where a location that could not obtain a solution with GPS alone was able to maintain a stable Fix after switching to multi-GNSS.

Receivers that support multiple frequencies are also advantageous. Devices that receive not only L1 but also L2 and L5 bands (and in Japan QZSS’s L6/CLAS signals) provide better ionospheric delay correction and shorten initialization time. Observing on multiple frequencies allows detection and mitigation of interference and multipath by comparing across bands, contributing to more stable positioning accuracy.

Also note the latest augmentation services available in Japan. Network RTK services (VRS, etc.) based on the Geospatial Information Authority of Japan’s continuous GNSS observation network allow high-accuracy positioning without installing your own base station. These services create a virtual reference near the rover and send correction data, so accuracy does not degrade significantly while moving across wide areas. Where communications are available, centimeter-level accuracy (half-inch accuracy) is achievable nationwide. The GSI’s GNSS Continuous Operating Reference Stations (about 1,300 electronic reference points) make it possible to obtain public coordinates (the World Geodetic System) instantly nationwide. Mobile carriers and surveying service providers also offer similar network RTK correction services, which are becoming the mainstream for high-accuracy positioning. In areas without communications, you can still utilize the quasi-zenith satellite “Michibiki”’s centimeter-level augmentation service (CLAS) to get RTK-like correction data without communications. Taking advantage of these latest services increases the likelihood of maintaining positioning accuracy under forest conditions.

Simple RTK surveying achievable with LRTK

Combining the technologies described above, all-in-one systems that make RTK surveying easy for anyone have recently appeared. A representative example is LRTK. LRTK is a system combining a smartphone and a compact high-precision GNSS receiver, designed so that even users without specialized knowledge can achieve centimeter-level positioning on site. Whereas traditional RTK required expensive, complex dedicated equipment, with LRTK you simply attach a dedicated device to a smartphone, launch the app, and start. LRTK supports network RTK (VRS), eliminating the need to set up a cumbersome base station.

LRTK also supports multi-GNSS and multi-frequency reception, designed to capture as many satellites as possible even in forests to secure accuracy. If cellular communication is unavailable in remote mountains, LRTK Pro series devices can directly receive Japan’s CLAS augmentation signals from satellites and continue positioning, providing a built-in backup when communications infrastructure is absent. This enables consistent high-accuracy positioning even in environments where surveying previously had to be abandoned.

Practical usability in the field is emphasized: LRTK devices are battery-powered, portable, and rugged enough for field use. You can check satellite reception and positioning results in real time on the smartphone screen, walk around with the survey pole, and record points single-handedly. Even those having difficulty with RTK surveys under trees can expect improved efficiency and accuracy by using LRTK. There are increasing examples of improved single-person surveying and infrastructure inspection workflows after adopting LRTK. Consider adopting the latest technologies to lower the hurdles of forest surveying.

Finally, here are frequently asked questions and their answers regarding RTK surveying under trees.

FAQ

• Q: Can RTK surveying be done in bad weather such as rain or fog? A: Yes, but accuracy may be affected. Rain or dense fog causes satellite signals to be attenuated or scattered by moisture, which can reduce positioning accuracy. When trees are wet, canopy-induced radio attenuation becomes even greater. Therefore, avoid surveying in rainfall if possible; if work is unavoidable, take extra care. For example, wait for precipitation to weaken before observing, or observe for longer than usual and average the data to reduce errors.

• Q: Is it really possible to achieve centimeter accuracy with RTK in densely treed areas? A: It is possible if conditions are favorable. If you can capture a sufficient number of satellites through canopy gaps and secure strong signals, centimeter-level accuracy (cm-level accuracy (half-inch accuracy)) can sometimes be achieved even in forests. There are reports of positioning with errors on the order of 1-2 cm (0.4-0.8 in) in somewhat sparse woodlands when satellite conditions were optimal. However, fixed solutions may not be obtainable depending on forest density and satellite geometry, so do not be overconfident. Use multi-GNSS-capable equipment and plan surveys carefully.

• Q: Can you tell on site whether multipath is affecting measurements? A: You can infer it from several signs. For example, if a stationary rover’s RTK solution swings by more than a few centimeters (a few inches), or if the SNR is high but a Fix won’t come, multipath may be the cause. If only satellites in a particular direction show large residuals or phase noise, that is also suspicious. As a countermeasure, try slightly moving the antenna position or inspect for potential reflectors; if the situation improves, multipath was likely the cause.

• Q: What alternative methods exist if RTK surveying in a forest fails? A: Several options exist. First, if network RTK is available use VRS to reduce baseline effects. If that still fails, try post-processing with PPK or move to a temporarily open area, obtain coordinates there, and use those as known points. If GNSS is impossible, use a total station to traverse. For example, you can establish RTK control at the forest edge and then use a total station to measure into the forest. Combining multiple methods flexibly to obtain the target point’s coordinates is often the practical solution.

• Q: Is LRTK really usable in forests? A: Yes; it performs comparably to other GNSS equipment. LRTK supports multi-GNSS and multiple frequencies and has specifications matching many high-performance GNSS receivers. In forest use, LRTK’s advantage is in the larger number of tracked satellites and rapid initialization. However, no equipment can work in completely opaque, extreme jungle conditions—LRTK is not omnipotent. Still, because of its portability and good support for correction services, LRTK is certainly a useful tool to have for forest surveys.