Elwyn McLachlan
Trimble Navigation Ltd
Forest canopy is one of the most limiting factors in using the global positioning system for positioning and mapping. However, forestry and natural resource applications are also the single largest users of GPS technology. Over the years much research has been put into technology improvements and data collection techniques to improve the ability to get good GPS positions under canopy. This paper will discuss the ongoing research and recommendations for using GPS in a forest environment.
Forest and natural resource applications can achieve significant gains through employing GPS data collection technologies. Where previously data was collected using pen and paper, estimating locations on paper maps, or using a compass and chain to determine positions, GPS systems have allowed data collection to be stream-lined, and result in more accurate and consistent data. However, as the use of GPS as a data collection tool has expanded rapidly, the limitations of the technology are better understood. The use of GPS to provide location information is now common-place, and cases are coming to court where proof of ownership or liability is tied to GPS-derived data. As such, it is increasingly important that GPS users employ the best possible data collection techniques to achieve the highest quality data. This paper outlines a number of data collection techniques, tools and tips for getting the best GPS data in a canopy situation.
GPS for GIS data collection was pioneered in the mid 1980s by Trimble Navigation Ltd, as a solution to the needs of the US Forest Service (USFS). The USFS needed to record approximate positions of sample sites and timber stands, and calculate acreages for planting and harvesting. Prior to the development of the GPS system, this work was done using a transit and chain—both time-consuming and inaccurate.
Today, the USFS is still the largest single of user of “resource” or “mapping-grade” GPS—that is, GPS receivers capable of accuracies of 0.5 to 5 meters (differential). In addition to mapping tree stands and acreages, GPS systems are deployed across the organization for research efforts, such as mapping and locating sample plots, taking water and soil samples. Conservation efforts, such as mapping animal habitat areas, and tracking environmental change over time, also make use of GPS data collection, and base map generation, including recording forest roads and trails, water features and watershed areas, and boundary lines, is done primarily using GPS tools.
Private forestry companies, and other natural resource organizations, including the federal Environmental Protection Agency and state-level Departments of Environmental Protection, have also adopted GPS systems as a essential tool in their work. These users, along with local government and utility organizations that collect GPS data to populate their GIS databases, must collect GPS data in all kinds of environments, including canopy.
Relying on GPS in their data collection efforts, all of these organizations are faced with the issue of recording quality GPS data in a canopy environment.
There are a number of issues affecting GPS data quality that are particularly problematic in canopy environment. These include:
Low SNR — The signals transmitted by GPS satellites are extremely low power, and as such, have very little penetrating ability. Materials that have a high water content, such as leaves from deciduous trees, cause the GPS signal to be attenuated to the point it becomes unusable. In the field this will be seen as low SNR (signal to noise ratio) for satellites.
High PDOP — Working in a tree canopy environment the user will typically have limited views of the sky, causing the GPS receiver to view only satellites that are high overhead. As satellites are clustered together, this results in poor PDOP (Position Dilution of Precision), which has a large detrimental affect on the quality of the positional data.
Multipath — When the GPS signal hits a physical barrier it is possible for it to be reflected, this is known as multipath. Similar to speaking in a room with an echo, where the listener can have difficulty discerning between the real voice and the echo, the GPS antenna has to determine which is the real GPS signal, and which the “echo”. This can result in errors of up to 10 meters and worse. Tree trunks and branches are a significant source of multipath, particularly when wet.
Constantly changing satellite constellations — In forested environments, especially when the user is moving, the view to the sky can change rapidly and frequently. This causes different satellites to be used in the position computation. Satellite constellation has a large affect on the quality of the data collected, as different satellite constellations cause different bias in the data. Constantly changing constellations result in data that is inconsistent and has poor relative accuracy.
Forest canopy is arguably the most challenging environment in which to collect GPS data, however there are a number of techniques and methods that can be applied to ensure the best possible data is collected.
Reject multipath — Good quality GPS data collection systems typically include some kind of multipath rejection technology. Trimble’s patented Everest technology detects multipath signals, and rejects it before it can be used to compute GPS positions. In various studies, GPS receivers that incorporate this technology show results up to 50% better than those without.
Use HDOP filters — If vertical accuracy is not important in your application, consider ignoring the vertical component of the PDOP value, and collect acceptable horizontally accurate data. PDOP is made up of a horizontal, vertical and temporal component. By ignoring the vertical component, more positions will pass the standard test, resulting in more positions being collected in the field without sacrificing horizontal accuracy.
Always post-process —Even if your application requires you to use real-time data collection in the field for navigation, it is recommended that you always differentially correct your data after you return to the office. Post-processed differential correction techniques, including such options as velocity filtering and smoothing, can significantly improve the quality of the data collected in the field.
Do mission planning — Before going out into the field, check your mission planning software to ensure you go out at the best time of the day. Generally the more satellites that are available, the less likely you are to encounter poor PDOP situations. With the current GPS constellation, depending on where in the world you work, you may be fortunate enough to find times of the day when 8 or 10 satellites are visible in the sky. In-field mission planning capabilities can save you time in the field by allowing you to determine whether you are likely to see improvement in the conditions, or whether you should accept the data recorded and move on.
Other data collection techniques — Most GPS data collection systems allow you to supplement the GPS data with other information to allow you to record data when you conditions get bad. Use offsets to record position data when you can’t get GPS right at the feature. For example, stand in the clear spot in the forest, and use a laser rangefinder to shoot the trees that you need to collect data on. The data collection system automatically calculates the distance and bearing from your GPS position to the feature you need to record, and logs the adjusted GPS position in the file.
The productivity vs precision trade off — In recent years awareness has developed of the inverse correlation between productivity and precision: as the need for precision increases, the expected productivity decreases significantly. When working in a difficult GPS environment, such as the forest, you may need to sacrifice good data in order to get more data. By relaxing the tolerances on SNR, PDOP and elevation, you are more likely to record data underneath canopy. However, when you do this, the result is lower quality data. Trimble’s SuperCorrect™ technology provides an ideal solution, allowing you to open settings and record every possible position in the field, but then allowing the re-setting of mask values in the office and filtering out data that does not meet quality requirements.
Purchase the right tool for the job — If absolute accuracy is of paramount importance in your application, then make sure you choose a product that can achieve this accuracy. For example, a conventional survey instrument might be a more appropriate tool if you cannot afford the risk of poor positional accuracy. Not all GPS is equal, different systems have quite different capabilities. Make sure you document your data quality requirements, and evaluate different systems against your requirements. When testing GPS accuracy, make sure you use a “truth” point that has been surveyed in correctly (ideally using a conventional survey instrument), and if you are going to be working in canopy, run your tests against a truth point in canopy.
Store all the data about your data — Regardless of what environment you collect your data in, be sure to keep metadata about your GPS data. This can be particularly important if data is used in legal proceedings, where issues of ownership and liability frequently arise. Metadata is your data’s defence. In the future as other people use the data you collect, it is important for them to be aware of how accurate it really is, so that data is used in applications appropriate to its accuracy. And as rapidly changing technology allows improved data accuracy in the future, you can use the metadata to easily determine which of the data in your database is of lower quality, and re-record as necessary.
GPS and canopy are not a great match. However, with improvements in technology, like multipath rejection, and a better awareness of the issues involved in recording data in canopy, it is possible to collect quality data. GPS data collection in canopy requires more careful and thorough planning. But despite these limitations GPS is an enabling technology, allowing foresters and natural resource users to record and store information that was not available to them previously.
For best results in canopy, remember these simple rules: