The Johnston Geodetic Station Fifty Years On
Compiled by Paul Wise 2014, updated 2017
Introduction
Prior to 1965, a rocky outcrop overlooking Mount Cavenagh Station homestead in the Northern Territory had no particular significance and was only shown as a small brown blob on the map. Even the earlier Nat Map geodetic trigonometrical survey had bypassed this point preferring the higher peak of Mt Cavenagh as the site for a survey station.
Figures 1 and 2 : Site of the Johnston Geodetic Station. Left photograph courtesy August Jenny 2014, right image courtesy Google Earth.
By the middle of 1966, however, this outcrop was marked by a specially constructed rock cairn in which was set a bronze plaque indicating that this point was the Johnston Geodetic Station, also sometimes called the Johnston Origin.
Fifty years on, this article explains why an unassuming lump of rock became the first calculated origin and centre of Australia.
Background
Until the early 1960s the origin for most of the Australian Geodetic Survey was the coordinates of the Sydney Observatory. However, because other agencies were also involved in completing this national task other local origins were also used. A single origin had to be found that was applicable to the whole of the Australian region.
To determine the mathematical surface or spheroid that best fitted the whole Australian landmass or geoid, a central origin near the centre of the continent was required. This point, where the spheroid and geoid intersected, would then locate the centre of the Australian spheroid. With the elements of the spheroid determined all points of the geodetic survey could then be referenced to a consistent datum providing a homogeneous network of horizontal positions.
The major loops of the geodetic survey were completed by the end of 1965. In addition, Tasmania and Papua New Guinea also had accurate survey connections to mainland Australia. These loops and connections were all the result of survey traverses of Tellurometer distance and azimuth (direction). However, as triangulation significantly contributed to the geodetic survey, the triangulated sections were converted to traverses. This was achieved by measuring a continuous string of selected triangulation sides through the triangulation network with Tellurometer electronic distance measuring equipment.
Figure 3 : Shows how a section of a triangulation network can be converted to a traverse by measuring a continuous string of selected triangulation sides with Tellurometer electronic distance measuring equipment.
An additional program of observations was also undertaken. At selected trigonometrical stations on the geodetic survey, highly accurate astronomical observations were recorded to determine latitude, longitude and azimuth. Such observations used precision theodolites and sophisticated apparatus for time signal reception and time recording. At these so called Laplace stations the difference between the astronomic and geodetic coordinates was thus obtained. When these values were used in the Laplace equation the correction to the observed astronomic azimuth was found and used in the geodetic adjustment to correct the observed survey traverse azimuths. Annexure A describes the Laplace correction in greater detail.
In practical terms the difference between the astronomic and geodetic coordinates is caused by random variations in the Earth's gravity from place to place. These variations mean that the plumb line at a point does not always point towards the centre of the Earth. Lambert in his 1964 paper The Role of Laplace Observations in Geodetic Survey, discussed the use of the Laplace observation in the Australian context.
Further, as detailed by Bomford in his 1964 paper Small Corrections to Astronomic Observations, all Laplace astronomic observations had to be reduced to a common standard. Specifically, this standard involved the following three elements. Firstly, the celestial reference frame for the observed stars had to be as defined in the Fourth Fundamental Katalog (FK4). Secondly, movement of the Pole was as set by the Bureau International de l'Heure Mean Pole of Epoch. Finally, longitudes determined before 1962 were corrected to the Greenwich meridian of 1962.0 which was when new conventional longitudes were adopted to bring the zero of longitude back on to the former meridian of Greenwich as closely as possible. Although these corrections seemed very small in themselves, their cumulative effect could negatively impact the outcome of a national adjustment if the corrections were not applied beforehand.
The Central Origin for an Australian Spheroid
To make the Laplace computations described above, there had to be an interim cohesive set of coordinates for the geodetic survey. As no specific Australian spheroid then existed these coordinates had to be based on a suitable interim spheroid.
For a short period in 1962, the necessary geodetic computations were performed on the NASA (National Aeronautical and Space Administration) spheroid, with the central origin being the coordinates of trigonometrical station Maurice. The Maurice origin emerged from the astronomic/geodetic comparison of the coordinates at 54 Laplace stations along the vicinity of the 32º parallel between Sydney and Perth. Trigonometrical station Maurice was on Maurice Hill some 16 kilometres southwest of Orroroo, South Australia.
Figure 4 : Map showing the Laplace stations (red dots), mainland loops and network of the geodetic survey and places mentioned in the text in green (indicative only).
As a result of these computations, new central origin and spheroidal parameter values were determined and from April 1963 to April 1965 computations were made on the 165 spheroid with the 165 central origin. The 165 central origin was based on the astronomic/geodetic comparison of the coordinates at 155 Laplace stations spread over the whole of Australia with the exception of Cape York and Tasmania.
The coordinates of Maurice as determined by astronomical observation or the spheroid and origin were :
165 spheroid, 165 central origin : 
S 32° 51' 13.979", E 138° 30' 34.062" 
165 spheroid, Maurice central origin : 
S 32° 51' 13.000", E 138° 30' 34.000" 
Clarke 1858 spheroid, Sydney origin : 
S 32° 51' 11.482", E 138° 30' 42.290" 
Observed Astronomic : 
S 32° 51' 11.341", E 138° 30' 25.110" 
In 1965 the International Astronomical Union adopted the parameters of a spheroid for astronomical use. The National Mapping Council was satisfied that these parameters were also an appropriate fit for Australia and adopted these values for the Australian National Spheroid (ANS). Although this spheroid was used as the best fit to the Australian regional geoid, its centre did not coincide with the centre of mass of the earth i.e. it was nongeocentric, by some 200 metres.
The major parameters of the various abovementioned spheroids are :
Spheroid name 
Semimajor (equatorial) axis (a) 
Flattening 1/f 
NASA 
6,378,148 metres 
298.3 
165 
6,378,165 metres 
298.3 
Clarke 1858 
6,378,293.645 metres 
294.26 
ANS 
6,378,160 metres 
298.25 
Frederick Marshall Johnston, former Commonwealth Surveyor General and the first Director of National Mapping passed away in 1963. The following year the National Mapping Council (that he did much to create) resolved:...that a special geodetic station be established and suitably monumented in the centre of Australia as the origin of the National Geodetic Survey and that this station be named Johnston in memory of Frederick Marshall Johnston former Commonwealth Surveyor General and the first Director of National Mapping.
In May 1965, a complete recomputation of the geodetic survey of Australia commenced, emanating from trigonometrical station Grundy, whose coordinates on both the 165 spheroid and the Australian National spheroid were 25° 54' 11.078"S and 134° 32' 46.457"E. Trigonometrical station Grundy was on Mount Grundy about 35 kilometres south of Finke in the Northern Territory and 135 kilometres almost due east of the later determined Johnston Geodetic Station. It is now interesting to note that Mount Grundy is less than 50 kilometres south of the Lambert Gravitational Centre that was determined in 1988 and named after Bruce Philip Lambert, the longest serving Director of National Mapping.
By December 1965, further Laplace observations had also been undertaken to then include Cape York and Tasmania, bringing the total number of Australian Laplace stations to 533. Of these Laplace stations, 275 stations were judiciously selected (please refer to Figure 5 below) and an astronomic/geodetic comparison of their coordinates revealed that:…random undulations in the geoid [earth’s surface] make it impossible to locate a centre for the spheroid with a standard error of less than 0.5 seconds, about 15 metres, even with a very large number of stations. This finding saw the central origin calculated using these 275 Laplace stations being adopted as the best available at the time. Thus, as resolved by the National Mapping Council the coordinates of this central origin would determine the location of the Johnston Geodetic Station, or Johnston Origin. Its centrality later gave this point notoriety as the first calculated centre of Australia!
Figure 5 : Map showing the distribution of the 275 Laplace stations within the first order geodetic network (after Bomford, 1967).
The Johnston Geodetic Station was established and constructed on a granite outcrop about one kilometre from Mount Cavenagh homestead. Officers of the Division of National Mapping undertook the task during October and November 1965. First order survey connections were made to the nearest, intervisible, existing trigonometrical stations on Mt Cavenagh and Mt Cecil. The precise coordinates of the origin station once calculated were then used in the final adjustment.
The 1966 national adjustment of the Australian Geodetic Survey comprised some 2,506 stations, 53,000 kilometres of Tellurometer traverse, the geodetic triangulation nets, and 533 Laplace stations. This adjustment was carried out by National Mapping for the National Mapping Council after intensive verification of the input data. Nat Map had access to the Control Data Corporation 3600 computer of the Commonwealth Scientific and Industrial Research Organization (CSIRO) in Canberra. The computer allowed such a volume of raw survey data to be converted into a continental wide series of very accurately related coordinates in latitude and longitude. This adjustment used the ANS with its central origin at the Johnston Geodetic Station, having the three dimensional coordinates of :
Latitude: 
25° 56' 54.5515" S 
Longitude: 
133° 12' 30.0771" E 
Elevation: 
571.2 metres above the Spheroid 
This Johnston Origin, together with the adopted spheroid then provided the basis for the Australian Geodetic Datum (AGD66), and this datum was adopted by the National Mapping Council on 21 April 1966, and proclaimed in the Commonwealth Gazette No. 84 on 6 October 1966.
The Australian Geoid
The term geoid is often used to describe the equipotential surface of the Earth; the surface of the Earth's gravity field which best corresponds with mean sea level. In the context of a figure of the Earth, the spheroid is a mathematical shape that best fits the geoid overall. Owing to variations in the composition and density of the Earth’s crust, differences in mass will cause the geoid to dip below or rise above the surface of the spheroid. In Australia where heights are related to the Australian Height Datum (AHD) the AHD is about 0.5 metres above the geoid in northern Australia and roughly 0.5 metres below the geoid in southern Australia.
Today heights obtained from Global Navigation Satellite System (GNSS) receivers are known as ellipsoidal heights and are referenced to a simplified mathematical representation of the Earth known as the ellipsoid (for practical purposes an ellipsoid and a spheroid are the same). Ellipsoidal heights differ from geoid/AHD heights by between 30 and +70 metres across Australia. This is known as the geoidellipsoid separation, or N value. To convert ellipsoidal heights to geoid/AHD heights, a geoid model can be used. AUSGeoid09 is Australia's newest geoid model for converting ellipsoidal heights to AHD heights and is accurate to 0.03m across most of Australia. AUSGeoid09 supersedes early geoid models such as AUSGeoid98 and AUSGeoid93.
Mather (1971) reported that the first attempt at an astrogeodetic solution for the geoid in Australia was by Bomford for the Woomera region. As described in Bomford (1963) the real impetus for a new datum came from the requirements of the Woomera rocket range. Their requirement was to be able to accurately track rockets in flight. In 1962 and 1963 a geoid survey of the Woomera area confirmed that the existing geodetic datum being the Clarke 1858 spheroid Sydney origin did not fit the geoid. The spheroidal surface was found to be tilted by about 8" in an eastwest direction relative to the geoid. The Woomera survey then permitted computations on the Clarke 1858 spheroid Sydney origin, to be adjusted to the required accuracy. However, it was Bomford's conclusion that it will therefore be wise to discontinue the use of the Clarke 1858 spheroid Sydney origin, as soon as the adjustment of the Geodetic Survey of Australia is completed in 1965.
Figure 6 : Members of Natmap Astro Section mid1960s; Dr Peter Bardulis (seated, Head of Section), Klaus Leppert (left) and Tony Bomford (right), unfortunately the names of the ladies are unknown.
An astrogeodetic geoid for Australia was produced by the United States Army Map Service in late 1966 and referred to the Australian Geodetic Datum (Fischer & Slutsky, 1967). This 1966 geoid was based on astrogeodetic observations recorded at approximately 600 Laplace stations and is known as the 1967 astrogeodetic solution. A further solution was made by Nat Map 1971. The 1971 solution used the astrogeodetic observations from approximately 1150 Laplace stations (Fryer, 1971).
The Johnston Geodetic Station
The Johnston Geodetic Station consisted of a fourteen foot pole with vanes, the same as used by Nat Map for the massive beacons in the New Guinea highlands, within a large stone cairn. However, it was decided that the cairn would have a conical shape instead of the usual Nat Map pudding shape, and for appearances be covered by a veneer of thin flat stone. This stone had to be transported by vehicle from some distance away. The people constructing the cairn had had previous experience with erecting Nat Map cairns and there was some disquiet at the time at this seemingly nonstandard approach.
Two officers of the Northern Territory Lands and Survey department assisted with the monumentation. Twentyone bags of cement, together with sufficient sand to make a threetoone mixture, plus the necessary water was carried up the hill in containers strapped to Yukon packs. In addition, the rock for the veneer surface and much of the rock for filling the centre of the cairn also had to be carried up the hill. The high prevailing temperature saw all members of the building team suffer the effects of heat exhaustion. All team members were grateful to the people at the nearby Mt Cavanagh homestead, for not only showing great interest in the work, but for also making shower facilities available.
1965 Nat Map Field Survey Party Members 

First order survey connection at the Johnston Geodetic Station : 

O J (Bob) Bobroff 
Surveyor Class 2 
Bob Goldsworthy 
Technical Assistant Grade 2 


First order survey connection at Cavenagh and Cecil : 

John Harris 
Surveyor Class 1 
David Yates 
Field Assistant 


Monumentation : 

Ed Burke 
Technical Assistant Grade 2 
John Ely 
Field Assistant (occupied Cavenagh with Ed Burke) 
Bill Sutherland 
Field Assistant 
Eddy Ainscow 
Field Assistant 
H A (Bill) Johnson 
Surveyor Class 3, Supervising Surveyor, Geodetic Survey Branch 
Two officers of the Northern Territory Department of Lands and Survey both names unknown. 
Figure 7 : John Ely watching Bill Sutherland and Eddy Ainscow installing the vanes. The tent at right of picture is to protect the theodolite from sun and wind during first order angle observations to Cavenagh and Cecil (courtesy Ed Burke 1965).
It was fitting that HA (Bill) Johnson (19071990) was a member of the monumentation team as this was the last cairn that Nat Map constructed. Almost 12 years earlier Johnson had been a member of the field party that built Nat Map’s first cairn at Moorkaie outside Broken Hill, New South Wales. During construction of the Johnston Geodetic Station Cinematographer John Carter of the Commonwealth Film Unit filmed the field survey part of the documentary Mapping Australia.
Figures 8 and 9 : (LR) Cinematographer John Carter with MRA2 Tellurometer in centre background and filming camera right and Johnston Geodetic Station in 1966 prior to installation of the bronze plaque (courtesy Denyse Goldsworthy 1965 and Peter Langhorne 1966).
In May 1966, a Nat Map field party comprising R A Ford (19141994), Senior Technical Officer Grade 1 and P H Langhorne, Surveyor Class 1 and others while enroute from Melbourne to their survey area at Helen Springs installed the bronze plaque in the cairn.
Closing remarks
On 1 January 2000, Australia adopted an Earth centred or geocentric datum, namely the Geocentric Datum of Australia 1994 (GDA94). The GDA is a part of a global coordinate reference framework and is directly compatible with the Global Navigation Satellite Systems (GNSS). The change to GDA recognised how much our world now relies on global satellite positioning. So, fifty years on, the Australian Geodetic Datum has already been obsolete for 15 years. Its physical representation in the landscape, however, the Johnston Geodetic Station, still stands as a reminder of the first time that Australia had a calculated origin or centre and homogeneous network of horizontal positions.
It is therefore somewhat unexpected today to find that the Johnston Geodetic Station now overlooks part of the technology that replaced it. To the west of the granite outcrop on which the Johnston Geodetic Station was constructed, is an AuScope site. This is one of over one hundred such sites. Each site includes a GNSS receiver, antenna and meteorological sensors. Sites were selected such that the antenna monuments could be anchored to bedrock and support the most precise positioning applications. Remote sites are solar powered and are capable of streaming data in realtime. The meteorological sensors are to aid atmospheric studies.
The AuScope GNSS receivers are able to track existing and planned GPS constellations. The data gathered by the receiver’s aid in any refinement of the National Geospatial Reference System for monitoring crustal movement and continental deformation. The future is to ultimately remove the need for national systems having instead a single, global, homogenous and stable reference system.
Figure 10 : The Johnston Geodetic Station in 2012 (courtesy Laurie McLean).
Figure 11 : The Mt Cavenagh (identifier MTCV) AuScope site on the plain below the Johnston Geodetic Station in 2012. Far left is the GNSS antenna atop its monument and to the right from front to back are the data uplink transmitter, cabinets for the receiver, meteorological sensors and ancillary electronics and the solar panel assembly (courtesy Laurie McLean).
Figures 12 and 13 : (LR) The Johnston Geodetic Station courtesy August Jenny, 2014, and plaque
References
AuScope (2014), accessed December 2014 at : http://www.auscope.org.au/site/
Bomford, Anthony G (1963), The Woomera Geoid Surveys, 196263, Technical Report 3, Division of National Mapping.
Bomford, Anthony G (1964), Small Corrections to Astronomic Observations, The Australian Surveyor, Vol. 20, Issue 3, pp. 199211.
Bomford, Anthony G (1967), The Geodetic Adjustment of Australia, 19631966, Survey Review, Vol. XIX, No. 144, pp. 5271.
Bomford, Anthony G, Cook, David P and McCoy Frank J (1970), Astronomic Observations in the Division of National Mapping 19661970, Technical Report 10.
Bomford, Guy (1971), Geodesy, Oxford University Press, London, Ed. 3, 1971.
Fischer, Irene (2005), Geodesy? What's That? My Personal Involvement in the AgeOld Quest for the Size and Shape of the Earth, iUniverse Inc., USA.
Fischer, Irene & Slutsky, Mary (1967), A preliminary geoid chart of Australia, Australian Surveyor, Vol.21, No.8, pp.327332.
Fischer, Irene, Slutsky, Mary, Shirley, Francis Raymond and Wyatt, Phillip Y (1968), New pieces in the picture puzzle of an astrogeodetic geoid map of the world, Bulletin Geodesique, Vol.88, No.1, pp.199221.
Ford, Reginald A (1979), The Division of National Mapping’s part in the Geodetic Survey of Australia, The Australian Surveyor, Vol. 29, No. 6, pp. 375427; Vol. 29, No. 7, pp. 465536; Vol. 29, No. 8, pp. 581638.
Fricke, W., & Kopff, A. (1963), Fourth Fundamental Katalog (FK4), Astron. RechenInst. Heidelberg.
Fryer, JG (1971), The Geoid in Australia  1971, Technical Report 13, Division of National Mapping.
Leppert, Klaus (1973), Geodesy in Australia, 195672, Proceedings 16th Australian Survey Congress, Canberra, pp.A1A6.
Lambert, Bruce P (1963), The Australian Geodetic Survey – An interim report on the utilisation of electronic distance measuring equipment, In Report of Proceedings, Conference of Commonwealth Survey Officers, Cambridge, 1963, Paper 8, pp. 7275.
Lambert, Bruce P (1964), The Role of Laplace Observations in Geodetic Survey, The Australian Surveyor, Vol.20, Issue 2, pp.8196.
Lines, John D (1992), Australia on Paper – The Story of Australian Mapping, Fortune Publications, Box Hill.
Mather, Ronald S (1972), The 1971 geoid for Australia and its significance in global geodesy, Journal of the Geological Society of Australia, Vol.19, No.1, pp.2129.
National Mapping Council of Australia, (1972), The Australian Map Grid – Technical Manual, Special Publication 7.
National Mapping Council of Australia, (1986), The Australian Map Grid – Technical Manual, Special Publication 10.
Annexure A
Laplace Correction
The Laplace correction, after PierreSimon Laplace (1749–1827), is used to relate a geodetic azimuth to an astronomical azimuth. Such a correction is made at a point called a Laplace Station, where the deflection of the vertical components is known. In a broader sense, the Laplace equations are used to connect the real physical world with a mathematical representation.
Figure A1 : Deflection of the Vertical
The mathematical ellipsoid/spheroid normal is perpendicular to the tangent to the curve of spheroid. In the physical world, the direction of the plumbline is the result of physical forces acting on an object. Refer Figure A1. Although the difference is not significant for many applications, at any given point in the terrain the direction of the spheroid normal and the direction of the plumbline differ and must be considered in geodetic computations. The direction of the normal at any point in the terrain is well defined and computable. Given the nonuniform distribution of mass within the Earth, however, the precise direction of the plumb line at the same point is less predictable. The difference between the spheroid normal and the vertical plumb line is called deflection of the vertical.
The spheroid surface is not necessarily parallel with the level surface of the terrain or geoid. The angular amount by which the two surfaces are not parallel is given by the deflection of the vertical and is expressed in terms of a north/south or in meridian component and an east/west or in prime vertical component.
Astronomical observations of directions to stars are gravity based and yield astronomical positions. Survey (azimuth and distance from a known point to an unknown point) derived positions are spheroidal based and provide geodetic positions. The two systems are related by the Laplace equations.
Deflection in meridian (ξ or east/west direction) :
ξ = Astronomic latitude  Geodetic latitude (1)
Deflection in prime vertical (η or north/south direction) :
η = (Astronomic longitude  Geodetic longitude) cos (Geodetic latitude) (2)
Azimuth
η = (Astronomic azimuth  Geodetic azimuth) cot (Geodetic latitude) (3)
Equations (2) and (3) show that deflection in prime vertical (η) can be deduced from the difference of either the astronomic and geodetic longitudes or of the astronomic and geodetic azimuths. As the two values for η must be the same then combining equations (2) and (3) gives :
Geodetic azimuth = Astronomic azimuth –
[(Astronomic longitude  Geodetic longitude) sin (Geodetic latitude)] (4)
Equation (4) is most important as it enables the geodetic azimuth at any Laplace station to be determined from a combination of astronomical azimuth and longitude observations. Laplace stations therefore permit an independent check to be made on a survey’s derived azimuth especially on long geodetic traverses such as occurred in Australia. In the 1966 Australian adjustment, of the 2,506 geodetic stations 533 were also Laplace stations.
In Australia, however, the Laplace observations were also used to establish the central origin for the Australian spheroid following the 1965 adoption of the parameters for the Australian National Spheroid (ANS). The central origin would ultimately have geodetic coordinates such that the resulting means of the deflections of the vertical (ξ and η from above) at a selected 275 Laplace stations were close to zero.
The starting position was the May 1965 adjustment which used the coordinates of trigonometrical station Grundy as the central origin. Grundy was chosen as its coordinates, as determined on the 165 spheroid, remained unchanged when they were recalculated on the ANS. This coincidence was a strong indication that the coordinates of Grundy were close to those of the central origin; within 1¼° as it turned out.
With the geodetic coordinates of the 275 Laplace stations now all originating from Grundy, simplistically an azimuth and distance could be calculated to each Laplace station from Grundy, the starting central origin. At each Laplace station the deflections of the vertical could also be determined as explained above. By inspection of the deflections a new set of coordinates for the central origin could now be determined.
Using these new coordinates for the central origin and the previously computed and now fixed azimuths and distances to the Laplace stations, new geodetic coordinates for each of the Laplace stations could be determined. As the astronomic coordinates for the Laplace stations would remain static, new deflections of the vertical at each Laplace station could then be determined. By iteration, an optimum set of coordinates for the central origin could be found such that the consequential geodetic coordinates for the Laplace stations resulted in the means of the deflections of the vertical being close to zero.
In 1965, the accepted coordinates for the central origin, based on the 275 Laplace stations, resulted in a mean deflection of +0.15" in meridian and 0.36" in prime vertical. Later testing revealed that this result could not be improved even if a much larger number of Laplace stations were used. The central origin so determined became known as the Johnston Origin and was monumented as the Johnston Geodetic Station.