10.2.2 GPS Surveying Standards

SURVEY ACCURACY STANDARDS

GPS accuracy standards are generally country specific.

Typically accuracy standards for horizontal coordinates are based on the ratio of the relative positional error of a pair of control stations to the horizontal separation of the points, at the 95% confidence level. As this ratio increases, the classification of the survey decreases. In the case of GPS this measure of positional accuracy may be extended to all three of the components (as in the U.S. S& S, while only the horizontal components are considered in the Australia S& S).

It is possible to measure relative position using GPS carrier phase data routinely to the 0.5-1cm + 1-2ppm level (section 2.4.4). With careful observation procedures, and the appropriate data processing, precisions of the order of 0.01ppm have been achieved for scientific GPS surveys. This is 1000 times better than the existing "1st order" accuracy standard. Hence the need for new survey categories. This leads to the problem of how to measure the quality of a GPS survey, and the secondary concern of classification.

Tables 1 and 2 below give the Australian and U.S. survey accuracy standards for GPS (and, in the case of Australia, for other types of horizontal surveys as well), expressed in terms of maximum allowable base error and line-length error for relative position.

Table 1. Australian horizontal ( 2-D ) survey accuracy classifications.

CLASS* Minimum geometric accuracy standard (95% confidence level) a -- Base error (mm) b -- Line-length error (ppm)
3A 0.2 2 2A 0.6 8 A 1.5 18 B 3 35 C 6 75 D 10 125 E 20 250

*CLASS is a function of field and reduction procedures.
ORDER is a function of both CLASS and fit to existing control.

Table 2. U.S. three-dimensional GPS survey accuracy classifications.

ORDER-CLASS Minimum geometric accuracy standard (95% confidence level) a -- Base error (mm) b -- Line-length error (ppm)
AA 3 0.01 A 5 0.1 B 8 1 1 10 10 2-I 20 20 2-II 30 50 3-1 50 100

In both the Australian and U.S. standards, since the previous survey standards were not adequate for classifying high precision GPS surveys, a two-tier system of classification has been adopted:

• A general classification for standard surveys, incorporating the present classifications for geodetic control, engineering, mapping and cadastral surveys. The results of general classification surveys are based on a constrained adjustment of GPS stations tied to the local network control. In the case of Australia, the long standing classifications "A", "B", "C", "D", "E" have been preserved.

   

• A special classification for high precision geodetic surveys such as, for example, crustal motion determination, land subsidence monitoring and precise engineering surveys. Special classification surveys are adjusted independently of the local network control and are therefore based only on the internal three-dimensional consistency of the GPS network. In the case of the U.S. new categories "AA", "A", "B"; and for Australia, new categories of "3A", "2A", have been established.

To cater for these classifications, a new terminology was necessary to distinguish between the survey results:

that were a function only of the field survey methods, reduction techniques and the results of a minimally constrained network adjustment, and those

that were a function of the above AND the conformity of the new survey results with the existing network coordinate set after the transformation process required to convert results from the original datum (such as WGS84 in the case of GPS) to the local geodetic datum.

The U.S. GPS standards refer to the "geometric" classification based on the internal consistency of the survey, and the "NGRS" (National Geodetic Reference System) classification based on the fit to existing geodetic control. In a similar fashion, in Australia the CLASS of any survey is defined by the quality of its internal consistency, whether the survey was intended for general or special classification (for example, photogrammetric control purposes, or crustal motion monitoring), and the ORDER is a function of the class and the fit to the existing horizontal and/or vertical control.

How does the survey classification system function in Australia? The "Recommended Standards and Practices for Control Surveys" (ICSM, 1994) suggests survey techniques designed to achieve a specific CLASS of survey. If the survey methods or reduction techniques used are not commensurate with the desired class of survey, or if the network adjustment fails to achieve the desired class, the stations in the survey should be assigned the highest CLASS common to all three aspects. Therefore, the allocation of CLASS to a survey on the basis of the results of a minimally constrained Least Squares network adjustment (after the usual statistical testing for outliers, etc.) is strictly defined according to the test (ICSM, 1994):

" whether the semi-major axis of each relative, one-sigma, standard error ellipse (for 2-D) or ellipsoid (for 3-D), is less than or equal to the length of the maximum allowable semi-major axis (r) using the following formula:

r = c.( d + 0.2 )

(10.2-1)

where

c	is an empirically derived factor represented by historically accepted precision for a particular standard of survey, and
d	is the distance to any station in kilometres (minimum of 1km). "

The values of c assigned to various classes of survey are shown in Table 3. (The values of base error and line-length error shown in Table 1 are derived from eqn (10.2-1) and the values of c given in Table 3 below.)

Table 3. Australian classification of horizontal control surveys.

CLASS	c	Typical applications

	(1-)	(95%)

3A	1	2	Special high precision surveys
2A	3	8	High precision geodetic surveys
A	7.5	18	National and state geodetic surveys
B	15	35	Densification of geodetic survey
C	30	75	Survey coordination projects
D	50	15	Lower CLASS projects
E	100	250	Lower CLASS projects

Experience has shown that the overall pattern of error propagation is not strictly proportional to distance (as implied by the formula above), but is a combination of instrumental and centring errors. A graph of the length of the maximum allowable semi-major axis against distance between any two stations is given in Figure below, showing clearly the different error propagation characteristics for lines above and below one kilometre.

Class related values of "c" for 2-D surveys.
(ICSM, 1994)

Back to Chapter 10 Contents / Next Topic / Previous Topic

© Chris Rizos, SNAP-UNSW, 1999