The Surveyor General of New South Wales is a verifying authority under the National Measurement Act 1960 and responsible for ensuring that surveyors use verified measuring equipment. According to the Surveying and Spatial Information Regulation 2012, surveyors are required to verify their Electronic Distance Measurement (EDM) equipment in relation to an Australian standard of measurement of length at least once a year. For this purpose, Land and Property Information (LPI) provides and maintains several EDM baselines across the state. LPI is currently in the process of improving this infrastructure by upgrading existing baselines and building new baselines for the calibration of EDM instruments. This paper presents the current status of EDM baseline infrastructure in NSW and outlines the data processing performed by LPI in regards to EDM calibrations. The EDMCAL software currently employed by LPI is described and compared to a spreadsheet calculation generated by the University of New South Wales. Finally, LPI's new online EDM baseline booking system is introduced. This online system should now be used by surveyors to book access to all EDM baselines in NSW in order to allow efficient and effective use of existing and future baseline infrastructure.
Legal metrology covers all measurements carried out for any legal purpose, including measurements that are subject
to regulation by law or government decree. The National Measurement Act 1960 provides the legal basis for a
national system of units and standards of measurement of physical quantities
Practising surveyors in NSW are subject to the Surveying and Spatial Information Act 2002 (NSW Legislation,
2014a) and the Surveying and Spatial Information Regulation 2012
In this context, it is important to explain the difference between the terms verification and calibration. The verification of an EDM baseline is carried out periodically with precise EDM instrumentation carrying a current Regulation 13 certificate issued by NMI (the associated meteorological equipment is also calibrated against industry standards). This process determines the ‘true’ inter-pillar distances and establishes traceability because the EDM baseline becomes a subsidiary standard of the International Metre. The calibration of an EDM instrument on a verified baseline determines the corrections that need to be applied to the instrument in order to obtain the ‘true’ inter-pillar distances, thereby establishing traceability of its measurements to the national standard.
In order to assist the surveying profession in meeting these requirements, the Surveyor General has established several EDM baselines throughout New South Wales. On behalf of the Surveyor General, Land and Property Information (LPI) is currently in the process of improving this infrastructure by upgrading existing baselines and building new baselines for the calibration of EDM instruments.
This paper briefly presents the current status of EDM baseline infrastructure in NSW and outlines the data processing performed by LPI in regards to baseline verifications and EDM calibrations. The EDMCAL software currently employed by LPI is described and compared to a spreadsheet calculation generated by the University of New South Wales. Finally, a new online EDM baseline booking system is introduced to allow efficient and effective use of the baseline infrastructure in NSW.
The Surveyor General has established several EDM baselines consisting of between four and seven concrete pillars throughout NSW. Current best practice has established that EDM baselines should consist of at least five (and preferably six or seven) pillars to increase the number of distances observed, thereby allowing a more reliable determination of the instrument correction. As a result, LPI is in the process of rationalising and improving its EDM baseline infrastructure by upgrading existing baselines to include more pillars and building new 7-pillar baselines.
Figure 1 illustrates the location of the 16 EDM baselines presently maintained in NSW. The new 7-pillar Seaham baseline (constructed in December 2013) replaces the 4-pillar baseline at Newcastle, which will cease to be maintained by LPI in August 2014 (but will continue to be used for teaching purposes at the University of Newcastle). It is planned to upgrade the 4-pillar Armidale baseline to include seven pillars, and contracts have been signed to establish a new 7-pillar baseline at Coffs Harbour (replacing the 4-pillar baseline at Grafton). In addition, efforts are underway to establish new 7-pillar baselines at Wollongong (replacing the existing 4-pillar baseline) and the South Coast (replacing the 4-pillar baselines at Nowra and Bega).
All EDM baselines in NSW follow the Heerbrugg design (also known as the Schwendener design), which
features an almost equal distribution of the distances measured in all combinations over the baseline length as well
as over the unit length of the EDM and permits the detection of all distance-dependent errors, including cyclic
errors
LPI verifies these baselines on a 2-yearly basis and makes the current measurement reports available on the LPI
website
Naturally, the processing of EDM calibrations can be performed in different ways and with different tools. The
standard mathematical methods involved have been described in various textbooks
In NSW, different software called EDMCAL has been used for many years by LPI for this purpose. Alternatively, several surveyors utilise a spreadsheet calculation developed at the University of New South Wales (UNSW). This section describes the EDMCAL software, and section 4 compares EDMCAL to the UNSW spreadsheet in regards to processing outcomes.
The program EDMCAL determines the additive constant (also known as instrument/reflector constant) and scale
factor of EDM instruments using the parametric method of a rigorous least squares adjustment. This adjustment
includes:
• Data snooping after
The original program was written in the FORTRAN programming language by J.D. Love as part of an
undergraduate student project for the Bachelor of Surveying degree at UNSW under the supervision of Dr J.M.
Rüeger. The project report was submitted in April 1978. Since then, EDMCAL has undergone several modifications
at LPI. These include:
• Ensuring compatibility with modern operating systems.
• Improving system performance, data structure and output format.
• Providing the possibility to input the EDM’s modulation frequency, carrier wavelength and unit length as an
alternative to the first velocity correction parameters.
• Using the mean of forward and reverse distance observations as the distance measurement between pillars if
available.
• Output of individual corrections applied to the slope distances, i.e. instrument/reflector constant, atmospheric
correction, slope correction, height (or datum) correction and chord-to-arc correction.
• Generation of a baseline database containing verification data on all baselines in NSW.
• Additional optional output to confirm the results, i.e. scale factor computation by linear regression and output of
a HAVOC
The current version 5.1 of EDMCAL was created in December 2013. The mathematical and statistical procedures have been thoroughly tested over many years. During the most recent update, the mathematical algorithms used to determine the atmospheric corrections were improved to make them more readable and provide clear reference to their origin in the source code. Continuing EDMCAL usage and feedback from users may result in additional modifications to the statistical output in order to further improve the interpretation of results.
The input file is first checked for correctness of the number of pillars and number of observations. Several
corrections are applied to the observed slope distances in order to reduce these to horizontal distances at the height
of the lowest pillar: approximate additive constant (input by the user), atmospheric correction, slope correction,
height (or datum) correction, and chord-to-arc correction. The standard deviations of the measured distances are
then computed. This is followed by the formation and solution of the normal matrix derived from the observations
and the output of the results of calibration and data snooping according to
The reduced distances are used to form observation equations for a least squares adjustment
The additive constant is valid for a particular combination of instrument and reflector only, accounting for the distance measurement reference points of the EDM instrument and the reflector not being coincident with the vertical axes at either end of the distance. In this case, the approximate additive constant stated in the input file (e.g. sourced from a previous EDM calibration) is utilised. If unknown or not specified, a zero entry is used. Obviously, the final additive constant will be determined through a least squares adjustment later on.
The basic principle of EDM instruments is the indirect determination of the travel time of a wave of light from the instrument to the reflector and back. While the speed of light in a vacuum is well known, in practice measurements are (of course) not carried out in a vacuum. The EDM measurements must therefore be corrected for the ambient atmospheric conditions because the velocity of visible and infrared waves changes with temperature, pressure and relative humidity (for light waves, the humidity is usually ignored). The atmospheric correction (also known as first velocity correction) catm is calculated according to:
= VC1 − (27V3C.125p+ t) + 1(12.7237.1P5W+VtP) 10-6 d1 where
PWVP t p VC1 VC2 d1 = partial water vapour pressure (mb) = dry bulb temperature (ºC) = atmospheric pressure (mb) = reference refractive index (specific to EDM instrument) = instrument pressure factor (specific to EDM instrument) = observed slope (wave path arc) distance (m)
It should be noted that the second velocity correction (accounting for the fact that the light wave does not follow a circular path between the two pillars) is more important for microwaves than for light waves and ignored for EDM calibrations because it is insignificant over such distances. Also insignificant over such distances is the reduction from wave path arc distance to wave path chord distance (i.e. d1 to d2 in Figure 4).
The slope correction reduces the slope wave path chord to a horizontal distance at the mean elevation of the two pillars involved:
∆H2 ∆H4 ∆H6 cslope = − 2 d2 − 8 d23 − 16 d25 where ΔH = height difference between instrument and reflector (m) d2 = wave path chord distance (m)
The height (or datum) correction (also known as sea level correction) reduces the horizontal distance at the mean elevation of the two pillars to the chord distance referred to the datum plane through the lowest pillar elevation:
HM ∆H2 HM ∆H4 HM ∆H6
HM d2 + 2 d2 R + 8 d23 R + 16 d25 R cheight = − R (3) (4) where
HM = mean height of instrument and reflector above the lowest pillar (m) R = radius of curvature of the ellipsoid along the line (m), here assuming R = 6,370,100 m for
New South Wales
The chord-to-arc correction converts the chord distance at the lowest pillar elevation d3 to the datum arc distance d4 (this correction is generally zero over distances used for EDM calibrations):
The additive constant is then determined together with the distances between pillar 1 and the remaining pillars from the solution of the normal equations of a standard least squares adjustment. In order to determine the scale factor (and its ppm equivalent), a one-dimensional similarity transformation is carried out using the calibrated distances as coordinates.
An alternative tool for the calculation of EDM instrument calibrations was developed by Dr B.R. Harvey at the
University of New South Wales (UNSW) in form of an Excel spreadsheet
It is useful to investigate whether the UNSW spreadsheet and EDMCAL provide comparable results in regards to EDM calibrations. Several datasets observed by LPI legal metrology staff during EDM baseline verifications are used for this purpose. The comparison is mainly based on seven verification datasets for each of three EDM baselines: Wollongong (600 m, 4 pillars), Wagga Wagga (535 m, 5 pillars) and Dubbo (765 m, 6 pillars). In the absence of a history of observations on a particular baseline consisting of seven pillars, the comparison also incorporates four datasets recently collected on the 7-pillar baselines at Kingscliff (K, 721 m), Eglinton (E, 849 m) and Lethbridge Park (L, 984 m). The 25 datasets used in this study are summarised in Table 1. In order to provide maximum redundancy, all possible inter-pillar distances were observed. Most datasets were collected using a Leica TCA2003 total station, while a Leica TS30 total station was used for all datasets from 2011. Both instruments are similar in regards to the precision stated by the manufacturer (1.0 mm + 1 ppm and 0.6 mm + 1 ppm, respectively).
In this context, it should be noted that the UNSW spreadsheet requires the user to input the relative humidity for each measured line in the calculation of the first velocity correction, while this quantity is ignored in EDMCAL. However, LPI staff routinely observe and record these values in the field during baseline verifications, allowing the comparison to be undertaken without the need for estimated or externally sourced values. It should also be noted that the UNSW spreadsheet requires meteorological data to be corrected before it is entered, while EDMCAL is able to accept the raw observations and apply thermometer and barometer instrument corrections during processing.
The results of the comparison between the additive constants and scale factors calculated using the UNSW spreadsheet (no iteration performed) and EDMCAL are illustrated in Figures 5 and 6, respectively. It should be noted that the first Dubbo dataset (June 2000) was identified as an outlier and removed from the analysis. The results indicate good agreement between the two calculation tools. The additive constants generally agree within 0.3 mm for the 4-pillar (Wollongong) and 5-pillar (Wagga Wagga) baselines, while larger differences of up to 0.8 mm are obtained for the 6-pillar baseline at Dubbo. Much better agreement of 0.1 mm or better is achieved at the 7-pillar 0.6 0.4 ) m0.2 m ( A 0 C n i ce-0.2 n e r ffe-0.4 i D -0.6 -0.8 2.0 1.5 ) m (pp1.0 F S in0.5 e c n ree 0 iff D -0.5 -1.0 baselines, although it should be noted that three of these datasets represent the first verification after construction of the baseline and therefore use their own results as ‘known’ distances in the processing. The scale factors agree within about 0.5 ppm for the 4-pillar, 5-pillar and 7-pillar baselines and about 1.5 ppm for the 6-pillar baseline.
It is also useful to investigate the differences in the distances relative to the first pillar after the calculated additive constant and scale factor have been applied. As already mentioned, the June 2000 dataset at Dubbo was identified as an outlier and therefore removed from the analysis. Descriptive statistics on the comparison between the output of the UNSW spreadsheet and EDMCAL are summarised in Table 2. A high level of agreement between the two calculation tools is evident, and the stated differences can be assumed negligible for most EDM calibrations in practice. 5 pillars -0.4 0.7 1.1 0.03 0.20 6 pillars -0.7 0.6 1.3 -0.07 0.33
-0.6 0.4 1.0 -0.04 0.19
These results show that the UNSW spreadsheet provides results comparable to EDMCAL processing for general practical purposes. However, it is important to note that recently verified distances must be used as ‘known’ interpillar distances in order to obtain reliable results. Using the latest verification results (i.e. 2012 or 2013) as ‘known’ distances for the processing of all datasets investigated in this study provided considerable differences not only in the resulting scale factors due to slight pillar movement but also negatively affected the relative comparison between the two calculation tools.
In the past, EDM baselines in NSW were not subject to a booking requirement. Surveyors were generally able to visit a baseline at any time, provided no baseline specific access requirements were in place (e.g. prior approval and/or keys from baseline host required). Particularly at popular baselines, this can result in several surveyors attempting to use the baseline at the same time, thus negatively affecting the surveyors’ productivity and time management (particularly if considerable travel time to the baseline is involved).
In order to avoid these disadvantages, LPI has developed the EDM Baseline Booking System. This free online
booking system is now available via the LPI website
A screenshot of the booking system’s main page is shown in Figure 7. The process consists of the following three simple steps: 1. Select a booking date. 2. Select an EDM baseline. 3. Select an available booking time.
Once the booking is finalised, a confirmation will be sent by email, also outlining the general and baseline specific conditions of use that had to be accepted during the booking process. The user is required to carry a printout of this booking confirmation with them at all times when on the baseline site. This will provide proof of approved access to the baseline for the specified time period.
The EDM Baseline Booking System was launched on 29 October 2013 and should now be used by surveyors to book access to all EDM baselines in NSW in order to allow efficient and effective use of existing and future baseline infrastructure. The booking system will also assist LPI in monitoring the frequency of use of each baseline, thereby allowing more informed decision making in regards to the state’s EDM baseline infrastructure in the future. This paper has briefly described the status of EDM baseline infrastructure in NSW, which is currently being rationalised and improved by upgrading existing baselines to include more pillars and building new 7-pillar baselines. The EDMCAL software used by LPI for the processing of EDM baseline verifications and EDM calibrations was outlined in detail. Based on 25 datasets collected on baselines consisting of between four and seven pillars, it was shown that the EDM calibration spreadsheet developed by the University of New South Wales provides additive constants and scale factors comparable to the EDMCAL output.
Further research could investigate using slope distances in a 3D solution instead of distances reduced to the horizontal at the height of the lowest pillar. It would also be useful to examine the use of verified distances and their uncertainties in a Bayesian solution rather than holding these inter-pillar distances fixed. This is expected to yield better statistics but may not cause any significant change in the parameters.
It is important to note that calibrating an EDM instrument in prism mode does not calibrate the reflectorless EDM laser. These two modes generally have different additive constants and scale factors within any one instrument, i.e. testing in reflectorless mode must be performed separately. It should also be noted that if surveyors measure any distances that are longer than the longest line on their EDM calibration baseline, they should consider the reliability of the extrapolation of their calibration parameters.
Finally, LPI’s new EDM baseline booking system was introduced. This free online system should now be used by surveyors to book access to all EDM baselines in NSW in order to allow efficient and effective use of existing and future baseline infrastructure. By allowing LPI to monitor the frequency of use of each baseline, the booking system will also assist LPI in making more informed decisions regarding the state’s EDM baseline infrastructure.
LPI’s ICT Technical Services in general and Sam Osborne in particular are gratefully acknowledged for their assistance in the design of the online booking system. (Commonwealth), available at LPI
(2014a). EDM baseline certificates, available http://www.lpi.nsw.gov.au/surveying/surveying_services/edm_baseline_certificates (accessed Feb 2014). LPI (2014b). EDM baseline booking system, available at http://lpi.nsw.gov.au/edmbooking (accessed Feb 2014). NSW Legislation (2014a). Surveying and Spatial Information Act 2002, http://www.legislation.nsw.gov.au/viewtop/inforce/act+83+2002+cd+0+N (accessed Feb 2014). available NSW Legislation (2014b). Surveying and Spatial Information Regulation 2012, available http://www.legislation.nsw.gov.au/viewtop/inforce/subordleg+436+2012+cd+0+N (accessed Feb 2014). at at at