10.2.1 GPS Surveying Standards

INTRODUCTION

Accuracies considerably better than 1 ppm are currently achievable with GPS if following special field and office procedures. Even relative accuracies of a few parts per million are achieved on a routine basis, which considerably exceed conventional standards for geodetic surveys using standard ground-based techniques. Nevertheless there is a need for the development of acceptable standards and specifications for GPS surveys.

What does "standards" mean? What does "specifications" mean?

The accuracy standards are essentially those developed for conventional surveys, augmented with several categories of higher accuracy. Specifications, or survey practices, on the other hand, relate to network design and planning, GPS instrumentation to be used, and field and office procedures that promote efficiency (and reliability), and facilitate the classification of surveys. They may take the form of strict requirements, recommendations or simply suggestions.

The U.S. National Geodetic Survey was the first national geodetic organisation to prepare standards and specifications (S& S) for GPS surveys (FGCC, 1988). These S& Swere be adopted in many other countries, including Australia, with minor modifications made to account for the different survey accuracy standards. (In Australia they are referred to as "standards & practices" -- see ICSM, 1994.) GPS field and technical experience has been drawn heavily in developing these specifications, and issues specific to the GPS technology have had to be addressed when framing the specifications:

• GPS provides three-dimensional information.
• Satellite geometry and ephemeris accuracy requirements need definition.
• Geoid slopes and accuracies need definition.
• Different measurement types and accuracies are available.
• Baseline, session, and network solutions are possible.
• GPS precisions are generally higher than the supposedly superior terrestrial control.
• GPS results are referred to a coordinate system that is geocentric, and is displaced in space relative to the origin of conventional geodetic datums.

However, must be emphasised that there are many aspects of the technology that are common to conventional geodetic techniques, including:

• The results are relative coordinates, as would be derived by conventional theodolite/EDM techniques.
• Networks can be built up in an analogous manner to geodetic traverses, and the quality control procedures are similar (no "hanging" traverses or "no check" lines!).
• The magnitude of the positional error between directly connected GPS stations is proportional to the distance separation of the stations.
• Redundancy can come from the number of occupations per stations, repeat baseline observations, measuring antenna height twice, etc.

A major difference between the U.S. S& S and the Australian S& S is that in the case of the former they are framed for GPS surveys alone, while the latter are general, and are intended to be valid for all survey technologies. Furthermore, the Australian S& S recognise the role of modern GPS techniques such as "rapid static" and "stop & go" (section 5.5.1), whereas the U.S. S& S predate these new techniques.

The accuracy standards can be easily modified, however, the recommended practices have had to be conservative in order that they need not have to be changed often. Although attempts have been made in some countries to differentiate the practices according to a hierarchy of: requirement, recommendation, or suggestion; this seems unnecessarily restrictive and does not reflect the changing nature of the technology (and with it, its capabilities). Hence it is preferable to simply state the specifications without making value-judgements, although wherever possible it should be indicated how they have been arrived at (by conjecture, experience or scientific fact?), whether they should be considered "optional", and whether they are likely to remain valid for the foreseeable future, etc. As a result, GPS specifications are likely to undergo continuous assessment and refinement.

Back to Chapter 10 Contents / Next Topic / Previous Topic

© Chris Rizos, SNAP-UNSW, 1999