Presented by: Richard Knippers

This website provides information on concepts of spatial referencing. A brief introduction is given,
followed by more in-depth notes on coordinate systems, reference surfaces, map projections and
coordinate transformations. It includes publications, links to external resources and a question form.
The website is used in the ITC educational programmes as an on-line help for students.

International Institute for Geo-Information Science and Earth Observation (ITC),
Enschede
l ast modi f i ed: August , 2009

1.0 Introduction
In the process of map making ellipsoidal or spherical surfaces are used to represent the surface of the
Earth. These curved reference surfaces are then projected on a map formed into a cylinder, cone or flat
plane (figure below). Since a map is a small-scale representation of the Earth's surface it is necessary
to apply some kind of scale reduction.

The process of representing the Earth on a flat map.
1.1 Reference surfaces
Two main reference surfaces (or Earth figures) are used to approximate the shape of the Earth. One is
called the ellipsoid, the other is the Geoid. The Geoid is the equipotential surface at mean sea level
and is used for measuring heights represented on maps. The starting point for measuring these heights
are mean sea level points established at coastal places. These points represent an approximation to
the Geoid. There are several realizations of local mean sea levels in the world. These are called local
vertical datums or height datums.
The ellipsoid (also called spheroid) provides a relatively simple mathematical figure of the Earth. It is
used to measure locations, the latitude (|) and longitude (), of points of interest. These locations on
the ellipsoid are then projected onto a mapping plane. There are many different ellipsoids defined in the
world, some well-known are the WGS84, GRS80, International 1924 (also known as Hayford),
Krasovsky, Bessel, or the Clarke 1880 ellipsoid.

A cross section of an ellipsoid, used to represent the Earth surface, defined by its semi-major axis a and semi-
minor axis b. For maps at small scales we can use the mathematically simpler sphere.
To measure locations accurately, the selected ellipsoid should fit the area of interest. Therefore
a horizontal datum (also called geodetic datum) is established, which is an ellipsoid but positioned
and oriented in such a way that it best fits to the area or country of interest. There are a few hundred of
these local horizontal datums defined in the world. Recent years have seen that globalisation is leading
to the definition of global (or geocentic) datums, such as the ITRF or WGS84.
1.2 Map projections
To produce a map the curved reference surface of the Earth, approximated by an ellipsoid or a sphere,
is transformed to the flat plane of the map by means of a map projection. In other words, each point on
the reference surface of the Earth with geographic coordinates (|,) may be transformed to set
of Cartesian coordinates (x,y) or map coordinates representing positions on the map plane.

Example of a map projection where the reference surface with geographic coordinates (|,) is projected onto the
2D mapping plane with 2D Cartesian coordinates (x, y).
Hundreds of map projections are developed in order to accurately represent a particular map or to best
suit a particular type of map. Examples of map projections are Transverse Mercator (also known as
Gauss-Krger), equidistant cylindrical and conic projection, Lambert's azimuthal, conic and cylindrical
projection, stereographic projection, and various others. Map projections are typically classified
according to the geometric surface from which they are derived: cylinder, cone or plane. The three
classes of map projections are respectively cylindrical, conical and azimuthal.

Three classes of map projections.

Furthermore map projections are typically classified according to the distortion properties of a map. The
three distortion properties of map projections are respectively:equal-area (or
equivalent), equidistant or conformal . Equal-area projections correctly represent area sizes,
equidistant map projections correctly represent distances (in certain directions), while conformal map
projections correctly represent angles and shapes (of small areas).
1.3 Map coordinate systems
A map coordinate system can be created by choosing a projection and then tailoring its parameters to
fit any region on the Earth. An example is the coordinate system used in the Netherlands. It is
called Rijksdriehoekstelsel (RD). This 2D Cartesian system is based on the azimuthal stereographic
projection centred in the middle of the country and the Bessel ellipsoid is used as reference surface.
The horizontal datum, with underlying Bessel ellipsoid, is called Amersfoort datum. The origin of the
coordinate system has been shifted (false origin) from the projection centre (Amersfoort) towards the
South-West to avoid negative coordinates inside the country.

The coordinate system of the Netherlands is derived from an oblique azimuthal stereographic projection.
Standard coordinate systems have been developed to simplify the process of choosing a system. The
most important standard map coordinate system used is the Universal Transverse Mercator (UTM).
Recent years have seen that globalisation is leading to the establishment of global 3D coordinate
systems. These spatial reference systems can be realized thanks to advances in satellite-based
positioning. The most important standard 3D system for the GIS community is theInternational
Terrestrial Reference System (ITRS).

2.0 Introduction
Different kind of coordinates are used to position objects in a two- or three-dimensional space. Spatial
coordinates (also known as global coordinates) are used to locate objects either on the Earths surface
in a 3D space, or on the Earths reference surface (ellipsoid or sphere) in a 2D space. Specific examples
are the geographic coordinates in a 2D or 3D space and the geocentric coordinates, also known as
3D Cartesian coordinates. Planar coordinates on the other hand are used to locate objects on the flat
surface of the map in a 2D space. Examples are the 2D Cartesian coordinates and the 2D polar
coordinates.
2.1 2D geographic coordinates (|,)
The most widely used global coordinate system consists of lines of geographic latitude (phi
or | or ) and longitude (lambda or ). Lines of equal latitude are calledparallels. They form circles on
the surface of the ellipsoid. Lines of equal longitude are called meridians and they form ellipses
(meridian ellipses) on the ellipsoid. Both lines form the graticule when projected onto a map plane. Note
that the concept of geographic coordinates can also be applied to a sphere as the reference surface.

The latitude (|) and longitude () angles represent the 2D geographic coordinate system.
The latitude (|) of a point P (figure section 2.2) is the angle between the ellipsoidal normal
through P' and the equatorial plane. Latitude is zero on the equator (| =0), and increases towards the
two poles to maximum values of | =+90 (90N) at the North Pole and | =- 90 (90S) at the South
Pole.
The longitude () is the angle between the meridian ellipse which passes through Greenwich and the
meridian ellipse containing the point in question. It is measured in the equatorial plane from the meridian
of Greenwich (=0) either eastwards through =+180 (180E) or westwards through =-180
(180W).
Latitude and longitude represent the geographic coordinates (|,) of a point P' (figure section 2.2)
with respect to the selected reference surface. They are also calledgeodetic coordinates or ellipsoidal
coordinates when an ellipsoid is used to approximate the shape of the Earth. Geographic coordinates
are always given in angular units. An example, the coordinates for the City hall in Enschede are:
| = 52 13' 26.2" N, = 6 53' 32.1" E.
These latitude and longitude coordinates are related to the Amersfoort datum. Note that the use of a
different reference surface will result in a different latitude and longitude (section 5.2 on datum
transformations).
There are several formats for the angular units of geographic coordinates. The Degrees: Minutes:
Seconds (4930'00"N, 12330'00"W) is the most common format, another the Decimal Degrees
(49.5000, -123.5000), generally with 4-6 decimal numbers.
Geographic coordinates are often used to store and manage, and interchange spatial data. The data
are projected onto a local map coordinate system for editing, analysis and mapping. As example, the
internal coordinate system of Google Earth are geographic coordinates (latitude/longitude) on the
World Geodetic System of 1984 (WGS84) datum. When the data are displayed on the monitor they are
projected using the equidistant cylindrical (or simple cylindrical) map projection.
Next to the geodetic (or geographic) latitude(|) there are two other type of latitudes. These are the
astronomic latitude and the geocentric latitude. The astronomic latitude (u) (figure below) is the angle
between the equatorial plane and the normal to the Geoid (i.e. a plumb line). It differs from the geodetic
(or geographic) latitude only slightly, due to the slight deviations of the Geoid from the reference
ellipsoid. The astronomic latitudeis the latitude which results directly from observations of the stars,
uncorrected for vertical deflection (section 3.5), and applies only to positions on the Earth's surface.
Astronomic observations are used to establish local horizontal (or geodetic) datums. The geocentric
latitude (|') is the angle between the equatorial plane and a line from the center of the ellipsoid (used
to represent the Earth). This value usually differs from the geodetic latitude, unless the Earth is
represented as a perfect sphere. Both geocentric and geodetic latitudes refer to the reference ellipsoid
and not the Earth.

Three different latitudes: the geodetic (or geographic) latitude (| ), the astronomic latitude (u) and the geocentric
latitude (|' ).
2.2 3D geographic coordinates (|,, h)
3D geographic coordinates (|, ,h) are obtained by introducing the ellipsoidal height h to the system.
The ellipsoidal height (h) of a point is the vertical distance of the point in question above the ellipsoid.
It is measured in distance units along the ellipsoidal normal from the point to the ellipsoid surface. 3D
geographic coordinates can be used to define a position on the surface of the Earth (point P in figure
below).

The latitude (| ) and longitude () angles and the ellipsoidal height (h) represent the 3D gegraphic coordinate
system.
2.3 Geocentric coordinates (X,Y,Z)
An alternative method of defining a 3D position on the surface of the Earth is by means of geocentric
coordinates (x,y,z), also known as 3D Cartesian coordinates. The system has its origin at the mass-
centre of the Earth with the X- and Y-axes in the plane of the equator. The X-axis passes through the
meridian of Greenwich, and the Z-axis coincides with the Earth's axis of rotation. The three axes are
mutually orthogonal and form a right-handed system. Geocentric coordinates can be used to define a
position on the surface of the Earth (point P in figure below).

An illustration of the geocentric coordinate system

It should be noted that the rotational axis of the Earth changes its position over time (referred to as polar
motion). To compensate for this, the mean position of the pole in the year 1903 (based on observations
between 1900 and 1905) has been used to define the so-called 'Conventional International Origin'
(CIO).
2.4 2D Cartesian coordinates (X,Y)
A flat map has only two dimensions: width (left to right) and length (bottom to top). Transforming the
three dimensional Earth into a two-dimensional map is subject of map projections and coordinate
transformations (section 4 and 5). Here, like in several other cartographic applications, two-dimensional
Cartesian coordinates (x, y),also known as planar rectangular coordinates, are used to describe the
location of any point in a map plane, unambigiously.
The 2D Cartesian coordinate system is a system of intersecting perpendicular lines, which contains
two principal axes, called the X- and Y-axis. The horizontal axis is usually referred to as the X-axis and
the vertical the Y-axis (note that the X-axis is also sometimes called Easting and the Y-axis
the Northing). The intersection of the X- and Y-axis forms the origin. The plane is marked at intervals
by equally spaced coordinate lines, called the map grid. Giving two numerical coordinates x and yfor
point P, one can now precisely and objectively specify any location P on the map.

An illustration of the 2D Cartesian coordinate system.
Normally, the coordinates x =0 and y =0 are given to the origin. However, sometimes large positive
values are added to the origin coordinates. This is to avoid negative values for the x and y coordinates
in case the origin of the coordinate system is located inside the area of interest. The point which then
has the coordinates x =0 andy =0 is called the false origin.
An example is the map coordinate system used in the Netherlands. It is
called Rijksdriehoekstelsel (RD). The system is based on an azimuthal stereographic projection
(section 4 on map projections) and the Bessel ellipsoid is used as reference surface. The origin of the
coordinate system has been shifted (false origin) from the projection centre (Amersfoort) towards the
south-west to avoid negative coordinates inside the country (figure below).

The coordinate system of the Netherlands represented by the map grid and graticule. The origin of the coordinate
system has been shifted (false origin) from the projection centre (Amersfoort) towards the South-West.

The grid on a map represents lines having constant 2D Cartesian coordinates (x,y). It is almost always
a rectangular system and is used on large and medium scale maps to enable detailed calculations and
positioning. The map grid is usually not shown on small scale maps (about one to a million or smaller).
Scale distortions that result from transforming the Earths curved surface to the map plane are so great
on small-scale maps that detailed calculations and positioning are difficult.
The graticule on a map represents the projected position of the geographic coordinates (|,) at
constant intervals, or in other words the projected position of selected meridians and parallels. The
shape of the graticule depends largely on the characteristics of the map projection used and the scale
of the map (figure below).

The world mapped in the Transverse Mercator projection with a 15 degrees graticule.
The map sheet limit or neat line (the line enclosing the mapped area) can either be formed by the
outline of the grid or by the outline of the graticule. The grid as outline of the mapped area has the
advantage of being rectangular, hence the map face of each map sheet will be exactly the same size,
allowing a completely uniform arrangement of the marginal information. This type of sheet index system
is used for topographic mapping of the Netherlands. The graticule as outline of the mapped area might
give a curved outline and the size of the mapped area, covered by a particular geographic coordinate
unit (e.g. 15' of latitude by 15' of longitude), will vary according to the location on Earth and the projection
system. However, it shows immediately the extent of the area in the geographic system.

Two different map grids are shown on the 1:25,000 topographic maps of the Netherlands. The grid lines and the
coordinate values of the Dutch RD system are portrayed in black. The coordinate values of the UTM grid are
portrayed in blue. The geographic coordinates are shown at every 5 minutes and at the corners of the map.
Grid and graticule spacings on a map vary depending on the scale of the map. The topographic maps
of the Netherlands at scale 1:25,000 show grid lines at every kilometer and the graticule lines (or ticks)
are given at every 5 minutes (figure above).
2.5 2D polar coordinates (o,d)
Another possibility of defining a point in a plane is by polar coordinates (o,d). This is the
distance d from the origin to the point concerned and the angle o between a fixed (or zero) direction
and the direction to the point. The angle o is called azimuth or bearing and is measured in a clockwise
direction. It is given in angular units while the distance d is expressed in length units.

An illustration of the 2D Polar coordinate system.
Bearings are always related to a fixed direction (initial bearing) or a datum line. In principle, this
reference line can be chosen freely. However, in practice three different directions are widely used: True
North, Grid North and Magnetic North (FAQ's nr. 30). The corresponding bearings are called: true (or
geodetic) bearing, grid bearing and magnetic (or compass) bearing.
Polar coordinates(o,d) are often used in land surveying. For some types of surveying instruments it is
advantageous to make use of this coordinate system. The development of precise remote distance
measurement techniques has led to the virtually universal preference for the polar coordinate method
in detailed surveys.
The transformation of polar coordinates (o,d) into Cartesian coordinates (x,y) is done when field
measurements, angular and distance measurements are transformed into map coordinates. The
equation for this transformation is:

The inverse equation is:

A more realistic case makes use of a translation and a rotation to transform one system to the other.
2.6 Main references
R.A. Knippers. Geometric Aspects of Mapping. Non-published educational notes. ITC, Enschede,
1999.
A. Mehlbreuer. Geometric Fundamentals of Mapping. Non-published educational notes. ITC,
Enschede.
3.0 Introduction
The surface of the Earth is anything but uniform. The oceans, can be treated as reasonably uniform,
but the surface or topography of the land masses exhibits large vertical variations between mountains
and valleys. These variations make it impossible to approximate the shape of the Earth with any
reasonably simple mathematical model. Consequently, two main reference surfaces have been
established to approximate the shape of the Earth. One reference surface is called the Geoid, the other
reference surface is the ellipsoid. These are illustrated in the figure below.

The Earth's surface, and two reference surfaces used to approximate it: the Geoid, and a reference ellipsoid. The
deviation between the Geoid and a reference ellipsoid is called geoid separation (N).
3.1 The Geoid and the vertical datum
We can simplify matters by imagining that the entire Earths surface is covered by water. If we ignore
tidal and current effects on this global ocean, the resultant water surface is affected only by gravity.
This has an effect on the shape of this surface because the direction of gravity - more commonly known
as plumb line - is dependent on the mass distribution inside the Earth. Due to irregularities or mass
anomalies in this distribution the 'global ocean' results in an undulated surface. This surface is called
the Geoid. The plumb line through any surface point is always perpendicular to it.

The Geoid, exaggerated to illustrate the complexity of its surface.
Where a mass deficiency exists, the Geoid will dip below the mean ellipsoid. Conversely, where a mass
surplus exists, the Geoid will rise above the mean ellipsoid. These influences cause the Geoid to deviate
from a mean ellipsoidal shape by up to +/- 100 meters. The deviation between the Geoid and an ellipsoid
is called thegeoid separation (N) or geoid undulation. The biggest presently known undulations are
the minimum in the Indian Ocean with N =-100 meters and the maximum in the northern part of the
Atlantic Ocean with N =+70 meters (figure below).

Deviations (undulations) between the Geoid and the WGS84 ellipsoid.
Tool for the calculation of a geoid undulation (compared to WGS84) at a point whose latitude and
longitude is specified.
The Geoid is used to describe heights. In order to establish the Geoid as reference for heights, the
oceans water level is registered at coastal places over several years using tide gauges (mareographs).
Averaging the registrations largely eliminates variations of the sea level with time. The resulting water
level represents an approximation to the Geoid and is called the mean sea level.
Every nation or groups of nations have established those mean sea level points, which are normally
located close to the area of concern. For the Netherlands and Germany, the local mean sea level is
realized through the Amsterdam tide-gauge (zero height). We can determine the height of a point in the
Netherlands or Germany with respect to the Amsterdam tide gauge using a technique known as
geodetic levelling (figure (b) below). The result of this process will be the height above local mean sea
level. The height determined with respect to a tide-gauge station is known as the orthometric
height (height H above the Geoid).
Obviously, there are several realizations of local mean sea levels (also called local vertical datums)
in the world. They are parallel to the Geoid but offset by up to a couple of meters. This offset is due to
local phenomena such as ocean currents, tides, coastal winds, water temperature and salinity at the
location of the tide-gauge.
Care must be taken when using heights from another local vertical datum. This might be the case in the
border area of adjacent nations. An example, the tide gauge (zero height) of the Netherlands differs -
2.34 metres from the tide gauge (zero height) of the neighbouring country Belgium (figure below). Even
within a country, heights may differ depending on to which tide gauge, mean sea level point, they are
related. An example, the mean sea level from the Atlantic to the Pacific coast of the USA increases by
0.6 to 0.7m.

Fragment of a topographic map showing the border area of Belgium and the Netherlands. The heights in both
countries refer to different tide gauges (zero heights). As a result, height contours (represented by brown
lines) are abruptly ending at the border.
The local vertical datum (or height datum) is implemented through a levelling network (figure (a) below).
A levelling network consists of benchmarks, whose height above mean sea level has been determined
through geodetic levelling. The implementation of the datum enables easy user access. The surveyors
do not need to start from scratch (i.e. from the Amsterdam tide-gauge) every time they need to
determine the height of a new point. They can use the benchmark of the levelling network that is closest
to the point of interest (figure (b) below).

A levelling network implements a local vertical datum: (a) network of levelling lines starting from the Amsterdam
tide-gauge, showing some of the benchmarks; (b) how the orthometric height (H) is determined for a point,
working from the nearest benchmark.

The use of satellite-based positioning equipment (e.g. GPS) to determine heights with respect to a
reference ellipsoid (e.g. WGS84) is becoming more in use. These heights are known as the ellipsoidal
heights (height h above the ellipsoid). Ellipsoidal heights have to be adjusted before they can be
compared to orthometric (mean sea level) heights. Geoid undulations (N) are used to adjust the
ellipsoidal heights (H =h - N).

Height h above the reference ellipsoid and height H above the Geoid for two points on the Earth surface. The
ellipsoidal height is measured orthogonal to the ellipsoid. The orthometric height is measured orthogonal to the
Geoid.
As a result of satellite gravity missions, it is currently possible to determine the orthometric height (height
H above the Geoid) with centimetre level accuracy. It is foreseeable that a global vertical datum may
become ubiquitous in the next 10-15 years. If all published maps are also using this global vertical
datum by that time, heights will become globally comparable, effectively making local vertical datums
redundant for GIS users.
3.2 The ellipsoid
Above, we have defined a physical surface, the Geoid, as a reference surface for heights. We also need
a reference surface for the description of the horizontal coordinates (i.e. geographic coordinates) of
points of interest. Since we will later project these horizontal coordinates onto a mapping plane, the
reference surface for horizontal coordinates requires a mathematical definition and description. The
most convenient geometric reference is the oblate ellipsoid (figure below). It provides a relatively
simple figure which fits the Geoid to a first order approximation, though for small scale mapping
purposes a sphere may be used. An ellipsoid is formed when an ellipse is rotated about its minor axis.
This ellipse which defines an ellipsoid or spheroid is called a meridian ellipse (notice that ellipsoid and
spheroid are used here as equivalent and interchangeable words).

An oblate ellipse, used to represent the Earth surface, defined by its the semi-major axis a and semi-minor axis b.

The shape of an ellipsoid may be defined in a number of ways, but in geodetic practice the definition is
usually by its semi-major axis a and flattening. Flattening f is dependent on both the semi-major
axis a and the semi-minor axis b.

The ellipsoid may also be defined by its semi-major axis a and eccentricity e, which is given by:

Given one axis and any one of the other three parameters, the other two can be derived. Typical values
of the parameters for an ellipsoid are:

3.3 The sphere
As can be seen from the dimensions of the Earth ellipsoid, the semi-major axis a and the semi-minor
axis b differ only by a bit more than 21 kilometres (figure below). A better impression on the Earth's
dimensions may be achieved if we refer to a more "human scale". Considering a sphere of
approximately 6 metre in diameter then the ellipsoid is derived by compressing the sphere at each pole
by 1 cm only. This compression is rather small compared to the dimension of the semi-major axis a.

The sphere compared to the ellipsoid.
The consequence is that instead of using the ellipsoid, the sphere might be sufficient for certain mapping
tasks. In practice, maps at scale 1:5,000,000 or smaller can use the mathematically simpler sphere
without the risk of large distortions. At larger scales, the more complicated mathematics of ellipsoids
are needed to prevent these distortions in the map.

The mathematically simpler sphere may be used as reference surface for maps at small-scale.
3.4 Local and global ellipsoids
Many different ellipsoids have been defined in the world. Local ellipsoids have been established to fit
the Geoid (mean sea level) well over an area of local interest, which in the past was never larger than
a continent. This meant that the differences between the Geoid and the reference ellipsoid could
effectively be ignored, allowing accurate maps to be drawn in the vicinity of the datum (figure below).

The Geoid, a globally best fitting ellipsoid for it, and a regionally best fitting ellipsoid for it, for a chosen region.
With increasing demands for global surveying, work is underway to develop global reference
ellipsoids. In contrast to local ellipsoids, which apply only to a specific country or localized area of the
Earths surface, global ellipsoids approximate the Geoid as a mean Earth ellipsoid. The International
Union for Geodesy and Geophysics (IUGG) plays a central role in establishing these reference figures.
In 1924, the general assembly of the IUG in Madrid introduced the ellipsoid determined by Hayford in
1909 as the International ellipsoid. However, according to present knowledge, the values for this Earth
model give an insufficient approximation.
At the general assembly 1967 of the IUGG in Luzern, the 1924 reference system was replaced by
the Geodetic Reference System 1967 (GRS 1967). It represents a good approximation (as of 1967)
to the mean Earth figure.
For some time, the Geodetic Reference System 1967 was used in the planning of new geodetic surveys.
For example, the Australian Datum 1966 and the South American datum 1969 are based upon this
ellipsoid. However, at its general assembly 1979 in Canberra the IUGG recognized that the Geodetic
Reference System 1967 no longer represented the size and shape of the Earth to an adequate
accuracy. Consequently, it was replaced by the Geodetic Reference System 1980(GRS80) ellipsoid.

Three global ellipsoids defined by a semi-major axis a, semi-minor axis b, and flattening f. The GRS80 and
WGS84 can be considered identical for all practical purposes.
3.5 The local horizontal datum
Ellipsoids have varying position and orientations. An ellipsoid is positioned and oriented with respect to
the local mean sea level (or Geoid) by adopting a latitude (|)and longitude () and ellipsoidal height
(h) of a so-called fundamental point and an azimuth to an additional point. We say that this defines
a local horizontal datum.Notice that the term horizontal datum and geodetic datum are being treated
as equivalent and interchangeable words.

Examples of local horizontal datums with their underlying ellipsoid and difference in position (datum shift) with
respect to WGS84.
Several hundred local horizontal datums exist in the world. The reason is obvious: Different local
ellipsoids with varying position and orientation had to be adopted to best fit the local mean sea level in
different countries or regions. An example is the Potsdam Datum, the local horizontal datum used in
Germany. The fundamental point is in Rauenberg and the underlying ellipsoid is the Bessel ellipsoid
(a =6,377,397,156 m, b =6,356,079.175 m). We can determine the latitude and longitude (|,) of any
other point in Germany with respect to this local horizontal datum using geodetic positioning techniques,
such as triangulation and trilateration. The result of this process will be the geographic (or horizontal)
coordinates (|,) of the new point in the Potsdam Datum. Other examples of local horizontal datums
(or geodetic datums) are the North American Datum 1927 is used for the North American countries, the
Tokyo Datum for J apan, the European Datum for the European countries, and the Amersfoort datum
for the Netherlands.
The local horizontal datum is realized through a so-called triangulation network (or survey network).
Such a network consists of monumented points forming a network of triangular mesh element (figure
below). The angles in each triangle are measured in addition to at least one side of a triangle; the
fundamental point is also a point in the triangulation network. The angle measurements and the adopted
coordinates of the fundamental point are then used to derive geographic (or geodetic)
coordinates (|,) for all monumented points of the triangulation network.
Within this framework, users do not need to start from scratch (i.e. from the fundamental point) in order
to determine the geographic coordinates of a new point. They can use the monument of the triangulation
network that is closest to the new point. The extension and re-measurement of the network is nowadays
done through satellite measurements.

The old primary triangulation network in the Netherlands made up of 77 points (mostly church towers). The
extension and re-measurement of the network is nowadays done through satellite measurements (click for town
names).
Computations of a triangulation network for a local horizontal datum were usually made with optical
survey instruments such as a theodolite or total station. These instruments are levelled by means of
spirit bubbles. The bubbles follow the influence of the Earth's gravity which means that the computations
have to be corrected with respect to the ellipsoid. Without these corrections the computations may be
distorted by some centimeters or even decimeters because of the local difference between the direction
of the plumb line (the normal to the Geoid) and the vertical direction on the ellipsoid (the normal to the
ellipsoid). This difference in direction is known as the deflection of the vertical (figure below). Gravity
measurements are used to determine these vertical deflections. Satellite-based measurements (e.g.
GPS) are not involved with these vertical deflections, because the GPS receiver directly relates the
measurements to the reference ellipsoid.

The angle between normal to the Geoid and the normal to the ellipsoid is known as the deflection of the vertical
(the angle is exaggerated to illustrate it). Vertical deflections are caused by Geoid undulations and amount to 10"
(flat areas) or 20-50" (steep mountain slopes).
3.6 The global horizontal datum
Local horizontal datums have been established to fit the Geoid well over the area of local interest, which
in the past was never larger than a continent. With increasing demands for global surveying activities
are underway to establish global reference surfaces. The motivation is to make geodetic results
mutually comparable and to provide coherent results also to other disciplines like astronomy and
geophysics.
The most important global (or geocentric) spatial reference system for the GIS community is
the International Terrestrial Reference System (ITRS). It is a three-dimensional coordinate system
with a well-defined origin (the centre of mass of the Earth) and three orthogonal coordinate axes (X,Y,Z).
The Z-axis points towards a mean Earth north pole. The X-axis is oriented towards a mean Greenwich
meridian and is orthogonal to the Z-axis. The Y-axis completes the righthanded reference coordinate
system (figure (a) below).

(a) The International Terrestrial Reference System (ITRS), and; (b) the International Terrestrial Reference Frame
(ITRF) visualized as a distributed set of ground control stations (represented by red points).
The ITRS is realized through the International Terrestrial Reference Frame (ITRF), a distributed set
of ground control stations that measure their position continuously using GPS (figure (b) above).
Constant re-measuring is needed because of the involvement of new control stations and ongoing
geophysical processes (mainly tectonic plate motion) that deform the Earths crust at measurable
global, regional and local scales. These deformations cause positional differences in time, and have
resulted in more than one realization of the ITRS. Examples are the ITRF96 or the ITRF2000 datums.
The ITRF96 datum was established at the 1st of J anuary, 1997. This means that the measurements
use data up to 1996 to fix the geocentric coordinates (X,Y and Z in metres) and velocities (positional
change in X,Y and Z in metres per year) at the different stations. The velocities are used to propagate
the measurements to other epochs (times). The trend is to use the ITRF everywhere in the world for
reasons of global compatibility. The World Geodetic System of 1984 (WGS84) datum has been
refined on several occasions and is now aligned with the ITRF to within a few centimetres worldwide.
The Global Positioning System (GPS) uses the WGS84 as its reference system.
Global horizontal datums, such as the ITRF2000 or WGS84, are also called geocentric
datums because they are geocentrically positioned with respect to the centre of mass of the Earth.
They became available roughly after the 1960's, with advances in extra-terrestrial positioning
techniques. These techniques include Satellite Laser Ranging (SLR), Lunar Laser Ranging (LLR),
Global Positioning System (GPS), and Very Long Baseline Interferometry (VLBI), among others. Since
the size and shape of satellite orbits is directly related to the centre of mass of the Earth, observations
of natural or artificial satellites can be used to pinpoint the centre of mass of the Earth, and hence the
origin of the ITRS. This technique can also be used for the realization of the global ellipsoids and datums
at the accuracy level required for large-scale mapping.
To implement the ITRF in a region, a densification of control stations is needed to ensure that there are
enough coordinated reference points available in the region. These control stations are equipped with
permanently operating satellite positioning equipment (i.e. GPS receivers and auxiliary equipment) and
communication links. Examples for (networks consisting of) such permanent tracking stations are the
AGRS in the Netherlands and the SAPOS in Germany.
We can easily transform ITRF coordinates (X,Y and Z in metres) into geographic coordinates (|,,h)
with respect to the GRS80 ellipsoid without the loss of accuracy. However, the ellipsoidal height h,
obtained through this straightforward transformation, is has no physical meaning and contrary to our
intuitive human perception of height. Therefore we use the height H, above the Geoid (figure below). It
is foreseeable that global 3D spatial referencing, in terms of (|,,H), could become ubiquitous in the
next 1015 years. If all published maps are also globally referenced by that time, the underlying spatial
referencing concepts will become transparent and hence redundant for GIS users.


Height h above the geocentric ellipsoid, and height H above the Geoid. The first is measured orthogonal to the
ellipsoid, the second orthogonal to the Geoid.
Hundreds of existing local horizontal and vertical datums are still relevant because they form the basis
of map products all over the world. For the next few years, we will be required to deal with both local
and global datums until the former are eventually phased out. During the transition period, we will
require tools to transform coordinates from local horizontal datums to a global horizontal datum and
vice versa (section 5 on coordinate transformations). The organizations that usually develop
transformation tools and make them available to the user community are provincial or National Mapping
Organizations (NMOs) and cadastral authorities.
3.7 Main references
de By, R.A. (editor), Georgiadou, P.Y., Knippers, R.A., Kraak, M.J ., Sun, Y., Weir, M.J .C. and van
Westen, C.J . Principles of geographic information systems(Chapter 4.2 on spatial referencing), 2nd
edition, ITC Educational Textbook, ITC, Enschede, 2001.
A. Mehlbreuer. Geometric Fundamentals of Mapping. Non-published educational notes, ITC,
Enschede.
R.A. Knippers. Geometric Aspects of Mapping. Non-published educational notes, ITC, Enschede,
1999.
P. Stefanovic. Georeferencing and Coordinate Transformations. Non-published educational notes.
ITC, Enschede, 1996.


4.0 Introduction
Maps are one of the worlds oldest types of document. For quite some time it was thought that our
planet was flat, and during those days, a map simply was a miniature representation of a part of the
world. Now that we know that the Earths surface is curved in a specific way, we know that a map is in
fact a flattened representation of some part of the planet. The field of map projections concerns itself
with the ways of translating the curved surface of the Earth into a flat map.
4.1 What is a map projection?
A map projection is a mathematically described technique of how to represent the Earths curved
surface on a flat map. To represent parts of the surface of the Earth on a flat paper map or on a computer
screen, the curved horizontal reference surface must be mapped onto the 2D mapping plane. The
reference surface for large-scale mapping is usually an oblate ellipsoid, and for small-scale mapping, a
sphere. Mapping onto a 2D mapping plane means transforming each point on the reference surface
with geographic coordinates (|,) to a set of Cartesian coordinates (x,y) representing positions on the
map plane (figure below).

Example of a map projection where the reference surface with geographic coordinates (|,) is projected onto the
2D mapping plane with 2D Cartesian coordinates (x, y).
The actual mapping cannot usually be visualized as a true geometric projection, directly onto the
mapping plane as illustrated in the figure above. This is mostly achieved through mapping equations.
A forward mapping equation transforms the geographic coordinates (|,) of a point on the curved
reference surface to a set of planar Cartesian coordinates (x,y), representing the position of the same
point on the map plane:
(x, y) =f (|,)
The corresponding inverse mapping equation transforms mathematically the planar Cartesian
coordinates (x,y) of a point on the map plane to a set of geographic coordinates(|,) on the curved
reference surface:
(|, ) =f (x, y)
Following are two examples of mapping equations for the sphere (equations for the ellipsoid are
generally more complex).
I. The first example are the mapping equations used for the Mercator projection:
The forward mapping equation is:

The inverse mapping equation is:

where R is the radius of the spherical reference surface; | and are given in radians; o is the central
meridian of the projection; e = 2.7182818, the base of the natural logarithms, not the eccentricity.
Suppose a point, located at 60
o
N and 130
o
W, is projected on a map that uses the Mercator projection
(where the reference surface is a sphere with a radius of 6371000 m. and the central meridian (o) is
0
o
, equal to the Greenwich meridian). Using the forward mapping equation of the Mercator projection,
the values found for the Cartesian coordinates are for x =-14,455,340m and for y =8,390,339m.
II. The second example is the mapping equations used for the north polar stereographic
projection:
The forward mapping equation is:

The inverse mapping equation is:

where R is the radius of the spherical reference surface;o is the longitude of the projection centre.
Suppose the same point, located at 60
o
N and 130
o
W, is projected on a map that uses the stereographic
projection (where the reference surface is a sphere with a radius of 6371000 m., the centre of the
projection is located on the North pole and the longitude of the centre (o) is 0
o
). Using the forward
mapping equation of the north polar stereographic projection, the values found for the Cartesian
coordinates are for x =-2615435m and for y =2194610m.
Map projection equations have a significant role in projection change (section 5 on coordinate
transformations). Interested readers can find an extensive list of mapping equations with numerical
examples in 'Map Projections - A Working manual' by J . P. Snyder. A number of equations are also
given at World of Mathematics and in the OGP Guidance note 7: Coordinate Conversions and
Transformations including Formulas.
4.2 Classification of map projections
Map projections can be described in terms of their:
i. class (cylindrical, conical or azimuthal),
ii. point of secancy (tangent or secant),
iii. aspect (normal, transverse or oblique), and
iv. distortion property (equivalent, equidistant or conformal).
i. The three classes of map projections are cylindrical, conical and azimuthal. The Earth's reference
surface projected on a map wrapped around the globe as a cylinder produces a cylindrical map
projection. Projected on a map formed into a cone gives a conical map projection. When projected
directly onto the mapping plane it produces an azimuthal (or zenithal or planar) map projection. The
figure below shows the surfaces involved in these three classes of projections.

The three classes of map projections: cylindrical, conical and azimuthal. The projection planes are respectively a
cylinder, cone and plane.
ii. The planar, conical, and cylindrical surfaces in the figure above are all tangent surfaces; they touch
the horizontal reference surface in one point (plane) or along a closed line (cone and cylinder) only.
Another class of projections is obtained if the surfaces are chosen to be secant to (to intersect with)
the horizontal reference surface; illustrations are in the figure below. Then, the reference surface is
intersected along one closed line (plane) or two closed lines (cone and cylinder). Secant map surfaces
are used to reduce or average scale errors because the line(s) of intersection are not distorted on the
map (section 4.3 scale distortions on a map).

Three secant projection classes.

A method to calculate the lines of intersection in a normal conical or cylindrical projection (i.e. standard
parallels) could be by determining the range in latitude in degrees north to south and dividing this range
by six. The one-sixth rule places the first standard parallel at one-sixth the range above the southern
boundary and the second standard parallel minus one-sixth the range below the northern limit (figure
below). There are other possible approaches.

A conical projection with a secant projection plane. The lines of intersection (standard parallels) are selected at
one-sixth below and above the limit of the mapping area.
iii. Projections can also be described in terms of the direction of the projection plane's orientation
(whether cylinder, plane or cone) with respect to the globe. This is called the aspect of a map projection.
The three possible apects are normal, transverse and oblique. In a normal projection, the main
orientation of the projection surface is parallel to the Earth's axis (as in the figures above for the cylinder
and the cone). A transverse projection has its main orientation perpendicular to the Earth's axis. Oblique
projections are all other, non-parallel and non-perpendicular, cases. The figure below provides two
examples.

A transverse and an oblique map projection.
The terms polar and equatorial are also used. In a polar azimuthal projection the projection surface is
tangent or secant at the pole. In an equatorial azimuthal or equatorial cylindrical projection, the
projection surface is tangent or secant at the equator.
iv. So far, we have not specified how the Earth's reference surface is projected onto the plane, cone or
cylinder. How this is done determines which kind of distortion properties the map will have compared to
the original curved reference surface. The distortion properties of map are typically classified according
to what is not distorted on the map:
- In a conformal (orthomorphic) map projection the angles between lines in the map are
indentical to the angles between the original lines on the curved reference surface. This
means that angles (with short sides) and shapes (of small areas) are shown correctly
on the map.
- In an equal-area (equivalent) map projection the areas in the map are identical to the
areas on the curved reference surface (taking into account the map scale), which
means that areas are represented correctly on the map.
- In an equidistant map projection the length of particular lines in the map are the same
as the length of the original lines on the curved reference surface (taking into account
the map scale).
A particular map projection can have any one of these three properties. No map projection can be both
conformal and equal-area. A projection can only be equidistant (true to scale) at certain places or in
certain directions.
Another descriptor of a map projection might be the name of the inventor (or first publisher) of the
projection, such as Mercator, Lambert, Robinson, Cassini etc., but these names are not very helpful
because sometimes one person developed several projections, or several people have developed
similar projections. For example J .H.Lambert described half a dozen projections. Any of these might be
called 'Lambert's projection', but each need additional description to be recognized.

Based on these discussions, a particular map projection can be classified. An example would be the
classification conformal conic projection with two standard parallels having the meaning that the
projection is a conformal map projection, that the intermediate surface is a cone, and that the cone
intersects the ellipsoid (or sphere) along two parallels; i.e. the cone is secant and the cones symmetry
axis is parallel to the rotation axis. This would amount to the projection of the figure above(conical
projection with a secant projection plane). Other examples are:
- Polar stereographic azimuthal projection with secant projection plane;
- Lambert conformal conic projection with two standard parallels;
- Lambert cylindrical equal-area projection with equidistant equator;
- Transverse Mercator projection with secant projection plane.
4.3 Scale distortions on a map
A map projection without distortions would correctly represent shapes, angles, areas, distances and
directions, everywhere on the map. Unfortunately, any map projection is associated with scale
distortions. There is simply no way to flatten out a piece of ellipsoidal or spherical surface without
stretching some parts of the surface more than others (figure below). The amount and which kind of
distortions a map will have depends largely - next to size of the area being mapped - on the type of the
map projection that has been selected.

Scale distortions after flattening a piece of the ellipsoidal or spherical reference surface.

Since there is no map projection that maintains correct scale all over the map, it may be important to
know the extent to which the scale varies on a map. On a world map, the scale variations are evident
where landmasses are wrongly sized or out of shape and the meridians and parallels do not intersect
at right angles or are not spaced uniformly. These maps may have a scale reduction diagram to
indicate the map scale at different locations, helping the map-reader to become aware of the distortions.
On maps at larger scales, maps of countries or even city maps, the distortions are not evident to the
eye. However, the map user should be aware of the distortions if he or she computes distances, areas
or angles on the basis of measurements taken from these maps.
Scale distortions can be measured and shown on a map by ellipses of distortion. The ellipse of
distortion, also known as Tissot's Indicatrix, shows the shape of an infinite small circle with a fixed
scale on the Earth as it appears when plotted on the map. Every circle is plotted as circle or an ellipse
or, in extreme cases, as a straight line. The size and shape of the ellipse shows how much the scale is
changed and in what direction.
The indicatrices on the map in the figure below have a varying degrees of flattening, but the areas of
the indicatrices everywhere on the map are the same, which means that areas are represented correctly
on the map. The distortion property of the map projection is therefore equal-area (or equivalent).

The ellipses of distortion plotted on the cylindrical equal-area projection. The flattening of the ellipses towards the
polar regions shows that shape distortions increase towards the polar regions.
When the indicatrices are circles everywhere on the map, the angles and consequently shapes (of small
areas) are shown correctly on the map. The distortion property of the map projection is therefore
conformal (e.g. the Mercator projection).
Scale distortions on a map can also be shown by means of a scale factor (ratio of the scale at a given
point to the true scale). Scale distortions exist at locations where the scale factor is smaller or larger
than 1. E.g. a scale factor at a given point on the map is equal to 0.99960 signifies that 1000 metres on
the reference surface of the Earth will actually measure 999.6 metres on the map. This is a contraction
of 40 centimetre per kilometre.
The nominal map scale (given map scale) divided by the scale factor will give the actual scale. E.g. a
scale factor of 0.99960 at a given point on a map with a nominal scale of 1:10,000 (one to ten thousand)
will give a scale of 1:10,004 (10,000 divided by 0.99960) at the given point. This is a smaller scale than
the nominal map scale. A scale factor of 2 at a given point on a map with a nominal scale of 1:10M (one
to ten million) will give a scale of 1:5M (10 million divided by 2) at the given point. This is a larger scale
than the nominal map scale.
Scale distortions for both, tangent and secant map surfaces, are illustrated in the figures below.
Distortions increase as the distance from the central point (tangent plane) or closed line(s) of
intersection increases.

Scale distortions on a tangent map surface. The central point is not distorted on the map.

Scale distortions on a secant map surface. Line(s) of intersection are not distorted on the map.
On a secant map projection - the application of a scale factor of less than 1.0000 to the central point or
the central meridian has the effect of making the projection secant - the overall distortions are less than
on one that uses a tangent map surface. Most countries have derived there map coordinate system
from a projection with a secant map surface for this reason.
4.4 Choosing a map projection
Every map must begin, either consciously or unconsciously, with the choice of a map projection and its
parameters. The cartographer's task is to ensure that the right type of projection is used for any
particular map. A well chosen map projection takes care that scale distortions remain within certain
limits and that map properties match to the purpose of the map.
Generally, normal cylindrical projections are typically used to map the world in its entirety (in particular
areas near the equator are shown well). Conical projections are often used to map the different
continents (the mid-latitudes regions are shown well), while the polar azimuthal projections may be used
to map the polar areas. Transverse and oblique aspects of many projections can be used for most parts
of the world, though they are usually more difficult to construct.
In theory, the selection of a map projection for a particular area can be made on the basis of:
i. the shape of the area,
ii. the location (and orientation) of the area, and
iii. the purpose of the map.

i.) Ideally, the general shape of the mapping area should match with the distortion pattern of a
specific projection. If an area is approximately circular it is possible to create a map that
minimizes distortion for that area on the basis of an azimuthal projection. The cylindrical
projection is best for a rectangular area and a conic projection for a triangular area (figure
below).

The choice of the map projection class (cylindrical, conical or azimuthal) depends largely on the general shape of
the mapping area.

ii.) The choice of the aspect of a map projection depends largely on the location (and orientation) of
the geographic area to be mapped. Optimal is when the projection centre coincides with centre
of the area, or when the projection plane is located along the main axis of the area to be mapped
(figure below).

An oblique Mercator projection is used for mapping the Alaska State (zone 5001). The cylindrical projection plane
is located along the main axis of the area to be mapped (source: P. H. Dana).

iii.) Once the class and aspect of the map projection have been selected, the distortion property of
the map projection has to be chosen. The most appropriate type of distortion property for a map
depends largely on the purpose for which it will be used.
Map projections with a conformal distortion property represent angles and local shapes correctly,
but as the region becomes larger, they show considerable area distortions. An example is the Mercator
projection. Although Greenland is only one-eighth the size of South America, Greenland appears to be
larger (figure below). Maps used for the measurement of angles (e.g. aeronautical charts, topographic
maps) often make use of a conformal map projection.

The Mercator projection is a cylindrical map projection with a conformal property. The area distortions are
significant towards the polar regions. An example, Greenland appears to be larger but is only one-eighth the size
of South America.
Map projections with a equal-area distortion property on the other hand, represent areas correctly,
but as the region becomes larger, it shows considerable distortions of angles and consequently shapes
(figure below). Maps which are to be used for measuring areas (e.g. distribution maps) often make use
of an equal-area map projection.

The cylindrical equal-area projection after Lambert is a cylindrical map projection with an equal-area property.
The shape distortions are significant towards the polar regions.
The equidistant distortion property is achievable only to a limited degree. That is, true distances can
be shown only from one or two points to any other point on the map or in certain directions. If a map is
true to scale along the meridians (i.e. no distortion in North-South direction) we say that the map is
equidistant along the meridians (e.g. the equidistant cylindrical projection in the figure below). If a map
is true to scale along all parallels we say the map is equidistant along the parallels (i.e. no distortion in
East-West direction). Maps which require correct distances measured from the centre of the map to any
point (e.g. air-route, radio or seismic maps) or maps which require reasonable area and angle
distortions (several thematic maps) often make use of an equidistant map projection.

The equidistant cylindrical projection (also called Plate Carre projection) is a cylindrical map projection with an
equidistant property. The map is equidistant (true to scale) along the meridians (in North-South direction). Both
shape and area are reasonably well preserved with the exception of the polar regions.
In summary, the ideal map projection for any country would either be an azimuthal, cylindrical, or
conic projection, depending on the shape of the area, with a secant projection plane located along the
main axis of the country or the area of interest. The selected distortion property depends largely on the
purpose of the map.
Some map projections have rather special properties. The Mercator projection was originally designed
to display accurate compass bearings for sea travel. Any straight line drawn on this projection
represents an actual compass bearing. These true direction lines are rhumb
lines (or loxodromes). Thus, the route of constant direction between two locations is a always a straight
line. For navigation, this is the easiest route to follow, but not necessary the shortes route (figure below).

The rhumb lines (lines of constant direction) are shown as straight lines on the Mercator projection. The shortest
distance between two points - the great circle path - is shown as a curved line.
The gnomonic projection is a useful projection for defining routes of navigation for sea and air travel,
because great circles - the shortest routes between points on a sphere - are shown as straight lines.
Thus, the shortest route between any two locations is always a straight line. No other projection has
this special property. In combination with the Mercator map where all lines of constant direction are
shown as straight lines it assist navigators and aviators to determine appropriate courses. Changes in
direction for following the shortest route can be determined by plotting the shortest route (great circle
or orthodrome) from the Gnomonic map onto the Mercator map (figure above).

All great circles - the shortest routes between points on a sphere - are shown as straight lines on the gnomonic
projection.
In the 15th, 16th and 17th centuries, during the time of great transoceanic voyaging, there was a need
for conformal navigation charts. Mercator's projection - conformal cylindrical - met a real need, and is
still in use today when a simple, straight course is needed for navigation. Because conformal projections
show angles correctly, they are suitable for sea, air, and meteorological charts. This is useful for
displaying the flow of oceanic or atmospheric currents, for instance.
For topographic and large-scale maps, conformality and equidistance are important properties. The
equidistant property, possible only in a limited sense, however, can be improved by using secant
projection planes. The Universal Transverse Mercator (UTM) projection is a conformal cylindrical
projection using a secant cylinder so it meets conformality and reasonable equidistance. Other
projections currently used for topographic and large-scale maps are the Transverse Mercator (the
countries of Argentina, Colombia, Australia, Ghana, S-Africa, Egypt use it) and the Lambert Conformal
Conic (in use for France, Spain, Morocco, Algeria). Also in use are thestereographic (the Netherlands)
and even non-conformal projections such as Cassini or the Polyconic.
Suitable equal-area projections for thematic and distribution maps include those developed
by Lambert, whether azimuthal, cylindrical, or conical. These do, however, have rather noticeable
shape distortions. A better projection is the Albers equal-area conic projection with two standard
parallels, which is nearly conformal. In the normal aspect, they are excellent for mid-latitude distribution
maps and do not contain the noticeable distortions of the Lambert projections.
An equidistant map, in which the scale is correct along a certain direction, is seldom desired. However,
an equidistant map is a useful compromise between the conformal and equal-area maps. Shape and
area distortions are often reasonably well preserved. An example is the equidistant
cylindrical projection (also called Plate Carre projection), where the meridians are true to scale map
(i.e. no distortion in North-South direction).
The projection which best fits a given country is always the minimum-error projection of the selected
class. The use of minimum-error projections is however exceptional. Their mathematical theory is
difficult and the equidistant projections of the same class will provide a very similar map.
4.5 Map projections in common use
A variety of map projections have been developed, each with its own specfic qualities. Only a limited
amount are frequently used. Here are some well-known projections described and illustrated. They are
grouped into cylindrical, conical and azimuthal projections.
4.5.1 Cylindrical projections
Probably one of the best known cylindrical projection is Mercator's cylindrical projection. The
transverse case and occasionally the oblique case of the Mercator projection are used in several
countries for topographic mapping purposes. The Transverse Mercator and Univeral Transverse
Mercator (UTM) projection are the best known examples. Two other well-known normal cylindrical
projections are the equidistant cylindrical (or Plate Carre) projection and Lambert's cylindrical
equal-area projection. Normal cylindrical projections are typically used to map the world in its entirety
(in particular areas near the equator are shown well).
1.Mercator projection:
The Mercator projection is a normal cylindrical projection. The property of the projection is conformal.
Parallels and meridians are straight lines intersecting at right angles, a requirement for conformality.
Meridians are equally spaced. The parallel spacing increases with distance from the equator (figure
below).

Mercator projection is a cylindrical map projection with a conformal property. The loxodromes in black are straight
lines. The great circle lines (orthodromes) in blue are curved.

The projection was originally designed to display accurate compass bearings for sea travel. Any straight
line drawn on this projection represents a constant compass bearing or a true direction line (loxodrome
or rhumb line). Sailing the shortest distance course along the great circle means that the direction
changes every moment. These changes in course direction can be deternined by plotting the great
circle onto the Mercator projection (figure above).
The Mercator projection is sometimes inappropriately used in atlases for maps of the world, and for
wall-maps as area distortions are significant towards the polar regions. The ellipses of distortion appear
as circles (indicating conformality) but increase in size away from the equator (indicating area
distortion). This exaggeration of area as latitude increases makes Greenland appear to be as large as
South America when, in fact, it is only one eight of the size.
The oblique Mercator projection is sometimes used to align the cylindrical projection plane with a region
that is oblique and follows neither a north-south nor an east-west axis. For example, this projection is
used for mapping the Malaysian peninsula and the Alaska State, zone 5001 (figure section 4.4).

2.Tranverse Mercator projection:
The Transverse Mercator projection is a transverse cylindrical conformal projection. The projection is
also known as the Gauss-Krger or Gauss conformal. Angles and shapes (of small areas) are shown
correctly, as a result of conformality. The figure below shows a part of the world mapped on the
Transverse Mercator projection.

A part of the world mapped on a transverse cylinder in the Transverse Mercator projection.
Versions of the Transverse Mercator (TM) projection are used in many countries as the local map
coordinate system on which the topographic mapping is based. Ghana uses TM projection with the
central meridian located at 1W of Greenwich. The projection is also used for aeronautical charts and
recommended to the European Commission for conformal pan-European mapping at scales larger than
1:500,000.
3.Universal Tranverse Mercator (UTM) projection:
The Universal Transverse Mercator (UTM) projection uses a transverse cylinder, secant to the
reference surface (figure below). It is recommended for topographic mapping by the United Nations
Cartography Committee in 1952. The UTM divides the world into 60 narrow longitudinal zones of 6
degrees, numbered from 1 to 60. The narrow zones of 6 degrees (and the secant map surface) make
the distortions so small that they can be ignored when constructing a map for a scale of 1:10,000 or
smaller.

The projection plane of the UTM projection is a secant cylinder in a transverse position.
The UTM projection is designed to cover the world, excluding the Arctic and Antarctic regions. The
areas not included in the UTM system, regions north of 84N and south of 80S, are mapped with
the Universal Polar Stereographic (UPS) projection. The figure below shows the UTM zone
numbering system. Shaded in the figure is UTM grid zone 3N which covers the area 168 - 162W (zone
number 3), and 0 - 8N (letter N of the latitudinal belt).

The UTM zone numbering system (click to enlarge).
Each zone has it's own central meridian. E.g. zone 11 extends from 120W to 114W, therefore the
central meridian has a longitude value of 117W (figure below).

Part of the world mapped in UTM Zone 11. The central meridian is located at 117 degrees west of Greenwich.
The zone extends from 120W to 114W.

A scale factor of 0.99960 is given to the central meridian of a UTM zone. To avoid negative coordinates
for positons located west of the central meridian, the central meridian has been given a (false) Easting
value of 500,000m. The equator has been given a Northing value of 0m for positions north of the
equator, and a (false) Northing value of 10,000,000m is allocated to the equator for positions south of
the equator.
If a map series covers more than one UTM zone it is inconvenient to have the Eastings changing
suddenly at a zone junction. For this reason a 40 kilometre overlap into an adjacent zone is allowed
(figure below). Mapping beyond this area will result in distortions at the edges of a UTM zone which
may not be acceptable for the larger map scales.

2 adjacent UTM-zones of 6 degrees longitude with a 40km overlap into the adjacent zone.
4.Equidistant cylindrical projection:
The equidistant cylindrical projection, also called simple cylindrical, or Plate Carre, has a true
scale along all meridians (i.e. no distortion in north-south direction). The projection is also known as
the latitude/longitude projection because the latitude and longitude are directly mapped into y and x
respectively. Meridians are spaced at the same distances as the parallels, forming a grid of equal
rectangles. Both shape and area are reasonably well preserved with the exception of the polar regions.
Used for simple portrayals of the world or regions with minimal geographic data such as index
maps.Google Earth uses the equidistant cylindrical (or simple cylindrical) projection for the display of
its imagery base. The transverse version of this projection is known as the Cassini projection.

The equidistant cylindrical projection (also called Plate Carre projection). The map is equidistant (true to scale)
along the meridians (in north-south direction). Both shape and area are reasonably well preserved with the
exception of the polar regions.


5.Lambert's cylindrical equal-area projection:
The Lambert's cylindrical equal-area projection represents areas correctly, but it does have rather
noticeable shape distortions towards the poles. Meridians are equally spaced and 0.32 times the length
of the equator. Parallels are unequally spaced and farthest apart near the equator (opposite to the
Mercator projection). Because of the distortions it is of little use for world maps. The Mollweide
projection, a pseudo-cylindrical equal-area projection, would be a better choice for this purpose. The
Gall-Peters projection is a version of Lambert's projection, but with standard parallels at 45 degrees
North and South, and also the Behrmann projection, but with standard parallels at 30 degrees North
and South.

Lambert's cylindrical equal-area projection. The shape distortions are significant towards the polar regions.
6.Pseudo-cylindrical projections:
Pseudo-cylindrical projections are projections in which the parallels are represented by parallel
straight lines, and the meridians by curves. The central meridian is the only meridian that is straight.
Examples are Mollweide, Sinusoidal (Sanson-Flamsteed), Goode Homolosine, McBryde-Thomas
series, Eckert's series (I -VI), Winkel (I, II), Denoyer and Robinson. They are typically used to map the
world in its entirety. Most pseudo-cylindrical projections are equal-area (certainly not conformal because
the parallels and meridians do not always cross at right angles).
The Mollweide projection (figure below) is a classic equal-area projection, keeping parallels as straight
lines while still preserving areas. All meridians but the central one are elliptical arcs. Suitable for
thematic or distribution mapping of the entire world.

Mollweide's pseudo-cylindrical projection with an equal-area property.
The Robinson projection (figure below) is neither conformal nor equal-area and no point is free of
distortion, but the distortions are very low within about 45 of the center and along the Equator and
therefore recommended and frequently used for thematic world maps. The projection provides a more
realistic view of the world than rectangular maps such as the Mercator.

Robinson's pseudo-cylindrical projection. Shapes and areas are reasonable well preserved.
7.Interrupted projections:
Interrupted projections show the globe in one sheet with interrupted forms of graticules. They are
typically used to map the different continents or oceans of the world in one map. Examples are Goode's
projection (an interrupted Mollweide projection) or the interrupted Sanson-Flamsteed projection.

Interrupted Mollweide projection for continental masses (after J.P. Goode).
4.5.2 Conic projections
Four well-known normal conical projections are the Lambert conformal conic projection, the simple
conic projection, the Albers equal-area projection and thePolyconic projection. They give useful maps
of mid-latitudes for countries which have no great extent in latitude.

1.Lambert conformal conic projection:
The Lambert conformal conic projection is confomal. The parallels and meridians intersect at right
angles (as in any conformal projection). Areas are, of course, inaccurate in conformal projections. Like
with other conformal projections, Lambert's conical is also widely used for topographic maps. It is
adapted in France and recommended to the European Commission for conformal pan-European
mapping at scales smaller or equal to 1:500,000.

Lambert Conformal Conic projection (standard parallels 10 and 30 degrees North).
2.Simple conic projection:
The simple conic projection (figure below) is a normal conical projection with one standard parallel. All
circular parallels are spaced evenly along the meridians, which creates a true scale along all meridians
(i.e. no distortion in north-south direction). The map is therefore equidistant along the meridians.Both
shape and area are reasonably well preserved. Whereas small countries are possibly shown on this
projection, larger areas, such as Russia or Europe are better portrayed on the conic projection with two
standard parallels.

Simple conic (or equidistant conic) projection (standard parallel 15 degrees North). The meridians are true to
scale.
3.Albers equal-area projection:
The Albers equal-area projection uses two standard parallels. It represents areas correctly and has
reasonable shape distortions in the region between the standard parallels as compared with the
noticeable distortions of the Lambert's equal-area conic projection with one standard parallel. This
projection is best suited for regions predominantly east-west in extent and located in the middle
latitudes. Used for small regions or countries but not for continents. It is adapted for maps of the United
States, for thematic maps and for world atlases.

Albers equal-area conic projection (standard parallels 10 and 30 degrees North).
5.Polyconic projection:
The polyconic projection is neither conformal nor equal-area. The projection is a derivation from the
simple conic projection, but with every parallel true to scale (similar to the Bonne's equal-area
projection). The polyconic projection is projected onto cones tangent to each parallel, so the meridians
are curved, not straight (figure below). The scale is true along the central meridian and along each
parallel. The distortion increase rapidly away from the central meridian. This disadvantage makes the
projection unsuitable for large areas on a single sheet. It is adaptable for topographic maps, and is
earlier used for the International Map of the World, a map series at 1:1,000,000 scale published by a
number of countries to common internationally agreed specifications, and also for large-scale mapping
of the United States until the 1950's and coastal charts by the U.S. Coast and Geodetic Survey.

Polyconic projection, with true scale along each parallel.
6.Pseudo-conical projections:
Pseudo-conical projections are projections in which the meridians are represented by curves, and the
parallels are equally spaced concentric circular arcs (unlike the pseudo-cylindrical projections in which
the parallels are represented by straight lines). The central meridian is the only meridian that is straight.
Examples are the Bonne and Werner projection.
Bonne's projection (figure below) is a pseudo-conical equal-area projection, with every parallel true to
scale (similar to the polyconic projection). The projection was once popular for large-scale topographic
maps and to map the different continents. The Werner projection is a variant of Bonne's projection with
the standard parallel at the North or South pole.

Bonne's projection (standard parallel 60 degrees North), with true scale along each parallel.
4.5.3 Azimuthal projections
Azimuthal (or zenithal or planar) projections are made upon a plane tangent (or secant) to the reference
surface. All azimuthal projections possess the property of maintaining correct azimuths, or true
directions from the centre of the map. In the polar cases, the meridians all radiate out from the pole at
their correct angular distance apart. A subdivision may be made into perspective and non-perspective
azimuthal projections. In the perspective projections, the actual mapping can be visualized as a true
geometric projection, directly onto the mapping plane; illustrations are in the figure below. For
the gnomonic projection, the perspective point (like a source of light rays), is the centre of the Earth.
For the stereographic this point is the opposite pole to the point of tangency, and for
the orthographic the perspective point is an infinite point in space on the opposite side of the Earth.
Two well known non-perspective azimuthal projections are the azimuthal equidistantprojection (also
called Postel projection) and the Lambert azimuthal equal-area projection.

Three perspective azimuthal projections: Gnomonic, stereographic and orthographic (source: ESRI).
1.Stereographic projection:
The azimuthal stereographic projection is a conformal projection. Since the projection is conformal,
parallels and meridians intersect at right angles. In the polar aspect the meridians are equally spaced
straight lines, the parallels are unequally spaced circles centered at the pole (figure below). Spacing
gradually increases away from the pole. The scale is constant along any circle having its centre at the
projection centre, but increases moderately with distance from the centre. The ellipses of distortion
remain circles (indicating conformality). Areas increase with distance from the projection center. The
polar stereographic projection is used in combination with the UTM coordinate system as Universal
Polar Stereographic (UPS) for mapping regions north of 84N and south of 80S. Recommended for
conformal mapping of regions approximately circular in shape; the Netherlands uses a modified
version of the stereographic projection (Dutch double stereographic) known as RijksDriehoekstelsel
(RD).

Polar azimuthal stereographic projection is a planar projection with a conformal property.

2.Orthographic projection:
The orthographic projection is a perspective projection that views the globe from an infinite distance.
Distortion in size and area near the projection limit appears more realistic than almost any other
projection. In the polar aspect, meridians are straight lines radiating from the center, and the lines of
latitude are projected as concentric circles that become closer toward the edge of the globe. Only one
hemisphere can be shown.
Google Earth shows the Earth as it looks from an elevated platform such as an airplane or orbiting
satellite. The projection used to achieve this effect is called the general perspective. This is similar to
the orthographic projection, except that the point of perspective is a finite (near earth) distance rather
than an infinite (deep space) distance.

Polar azimuthal orthographic projection.
3.Gnomonic projection:
The gnomonic projection (also known as central azimuthal projection) is neither conformal nor equal-
area. The scale increases rapidly with the distance from the center. Area, shape, distance and direction
distortions are extreme, but all great circles (or orthodromes) - the shortest distances between two
places on a sphere - are shown as straight lines. It's wise to orient the centre of the map at the point of
interest (figure below), since scale distortions increase rapidly away from the centre of the gnomonic
map projection and the direction of the shortest route (azimuth) is shown by a straight line from the
centre of the map only (this applies to all azimuthal maps). The projection is useful for defining routes
of navigation for sea and air travel, because the shortest route between any two locations is a always
a straight line (figure below). It should however not be used for regular geographic maps or for distance
measurements. An example, the measured distance on the gnomonic projection between Amsterdam
airport and New York airport is 8415 kilometres, but in reality it is around 5884 kilometres (second figure
below).

Air-routes plotted in red on a the gnomonic projection. The centre of the projection is oriented at Amsterdam
airport. The shortest air-routes - great circle routes - are shown by a straight line and the directions of the shortest
air-routes are true from the centre of the projection. However, the distance distortions are extreme.
4.Azimuthal equidistant projection:
The non-perspective azimuthal equidistant projection (also known as Postel projection) is an
equidistant projection. Distances measured from the centre of the map to any point are correct; the
bearing of any point from the center is also correct (this applies to all azimuthal maps). Distortion of
other properties increases away from the center point, but are not very large compared to the distortions
of the gnomonic projection. The projection is frequently used to show air-route distances (figure below).

Air-routes plotted in red on a the azimuthal equidistant projection. The centre of the projection is oriented at
Amsterdam airport to show correct distances (and directions) from this point to the other airports. The shortest
route - great circle - is shown by a straight line from the centre of the map only (in blue a curved great circle line).
5.Lambert azimuthal equal-area projection:
The non-perspective Lambert azimuthal equal-area projection preserves areas while simultaneously
maintaining a true direction from the center. The general pattern of distortion is radial. Scale decreases
with distance from the center. It is best suited for maps of continents or regions that are equally extended
in all directions from the centre, such as Asia and the Pacific ocean. Recommended to the European
Commission for statistical analysis and display. A variant of this projection is the Hammer-Aitoff
projection.

Polar azimuthal equal-area projection (after Lambert).
4.6 Map projections overview
In summary, a short list of map projections grouped by class:
Projection
Description
Cylindrical

Central cylindrical
Map projection is perspective but not conformal nor equal area. Projected perspectively from the center of the Earth
onto a cylinder tangent to the equator. Only used for teaching purposes.
Equidistant cylindrical
Also known as simple cylindrical or Plate Carre. The projection is equidistant in the direction of the meridians.
Parallels and meridians (half as long as the parallels) are equally spaced straight lines forming square blocks. This
projection maps longitude and latitude directly into x and y, hence is sometimes called the latitude-longitude
projection. In Google Earth used for display of imagery. The transverse version is known as the Cassini projection.


Equirectangular
Also known as Plate rectangle, a variant of Plate Carre. Used for raster maps which store information of the whole
world: each pixel represents a rectangular block of latitude-longitude coordinates.
Gall-Peters
Similar to Lambert's cylindrical equal-area projection, but with standard parallels at 45 degrees North and South.
Lambert cylindrical equal-area
It is of little use for world maps because of the distortions. Mainly used for educational purposes.
Miller cylindrical
Modified Mercator projection proposed by O.M. Miller. Compromise between Mercator and other cylindrical
projections. Shape, area and scale distortion increases moderately away from the equator. Used in numerous world
maps.
Mollweide
Pseudo-cylindrical projection. Map is equal area. Occasionally used in thematic world maps.
Mercator
Conformal map projection. Designed for navigational use; standard for marine charts. Recommended use for
conformal mapping of regions predominantly bordering the equator. Often inappropriately used as a world map.
Trasverse Mercator
Also called Gauss Conformal, or Gauss Krger. Transverse form (transverse cylinder) of the Mercator projection.
Used for topographic maps at scales from 1: 20,000 to 1: 250,000. Recommended for conformal mapping of regions
that are predominantly north-south in extent.
Universal Transverse Mercator (UTM)
A version of the Transverse Mercator, but one with a secant map surface. It divides the world into 60 narrow
longitudinal zones of 6 degrees. Widely used standard for topographic maps and military maps.
Azimuthal
Azimuthal equidistant
Distances measured from the centre of the map to any point are correct and the bearing of any point from the
center is correct (this applies to all azimuthal maps). Commonly used in the polar aspect for maps of polar regions
and the Northern and Southern hemispheres. The oblique aspect is frequently used for world or air-route maps
centered on important cities and occasionally for maps of continents.
Gnomonic
Map is perspective and neither conformal nor equal area. Area, shape, distance and direction distortions are
extreme. It is used to show great circle paths as straight lines and thus to assist navigators and aviators.
Hammer-Aitoff
A variant of Lambert azimuthal equal-area. Used for thematic maps of the whole world.
Lambert azimuthal equal-area
Used for maps of continents and hemispheres. Also suited for regions extending equally in all directions from a
center point, such as Asia and the Pacific Ocean. Recommended to the European Commission for statistical
analysis and display.
Orthographic
Known by Egyptians and Greeks 2000 years ago. Map is perspective and neither conformal nor equal area. Only
one hemisphere can be shown. The Earth appears as it would on a photograph from space.
Stereographic
Apparently invented by Hipparchus (2nd century bc). Used in combination with UTM projection as Universal Polar
Stereographic (UPS) for mapping poles and in navigation charts for latitudes above 80. Recommended for
conformal mapping of regions that are approximately circular in shape; a modified version of the stereographic
projection is used in the Netherlands for large-scale and topographic maps.
Conical
Albers equal area conic
It is equal to Lambert's equal area conic, but has two standard parallels (secant cone). Excellent for mid-latitude
distribution maps. The projection does not contain the noticeable distortions of the Lambert projections. Frequently
used for maps of the United States, for thematic maps and for world atlases.
Lambert conformal conic
Lambert conformal conic, also called conical orthomorphic (Lambert, 1972). Extensively used for large-scale
mapping of regions predominantly east-west in extent. Further widely used for topographic maps.
Polyconic
or American polyconic (Hassler, 1820). Map is neither conformal nor equal area, but each parallel is true to scale.
The sole projection used for large scale mapping of the United States by the USGS until the 1950's.
Simple conic
Also known as equidistant conic. Meridians are true to scale (i.e. no distortion in north-south direction). The most
common projection in atlases for small countries.
Other projections
Sinusoidal
Used since 16th century. Also called Sanson-Flamsteed or Mercator equal area projection. Pseudo-cylindrical
projection. Map is equal area. Used in atlas maps of South America and Africa. Occasionally used for world maps.
Modifications are called sinusoidal interrupted and sinusoidal 3x interrupted.
Van der Grinten
Shows the entire Earth within one circle. All areas, shapes and angles are greatly distorted.
Winkel Tripel
Used in several atlases. A triple compromise of reduced shape, area and distance distortion. Selected by the
National Geographic Society (NGS) for its new reference world map, in place of the Robinson projection.
4.7 Main references
R.A. Knippers. Coordinate systems and Map projections.. Non-published educational notes,
ITC, Enschede, 1998.
R.A. Knippers. Geometric Aspects of Mapping. Non-published educational notes, ITC, Enschede,
1999.
P. Stefanovic. Georeferencing and Coordinate Transformations. Non-published educational notes.
ITC, Enschede, 1996.


5.0 Introduction
Map and GIS users are mostly confronted in their work with transformations from one two-dimensional
coordinate system to another. This includes the transformation of polar coordinates delivered by the
surveyor into Cartesian map coordinates (section 2.5) or the transformation from one 2D Cartesian (x,y)
system of a specific map projection into another 2D Cartesian (x,y) system of a defined map projection.

Integration of spatial data into one common coordinate system.
Datum transformations are also important, usually for mapping purposes at large and medium scale.
An example, map and GIS users are often collecting spatial data in the field using satellite navigation
technology and need to represent this data on published maps on a local horizontal datum.
5.1 Changing map projection
Forward and inverse mapping equations (see section 4 on map projections) are generally used to
transform data from one map projection to another. The inverse equation of the source projection is
used first to transform source projection coordinates (x,y) to geographic coordinates (|,). Next,
the forward equation of the target projection is used to transform the geographic coordinates (|,) to
target projection coordinates (x,y). The first equation takes us from a projection A into geographic
coordinates. The second takes us from geographic coordinates (|,)to another map projection B. This
is illustrated in the figure below.

The principle of changing from one into another projection using the mapping equations.
The coordinate system (map projection) of the input data must be known to use the mapping equations
for a projection change. If the coordinate system of the input data is not known it may be possible to
use a 2D Cartesian transformation. 2D ground control points (GCPs) or common points are then
required to determine the relationship between the unknown and the known coordinate system. The
transformation may be conformal, affine, polynomial, or of another type, depending on the geometric
differences between the two map projections. This is illustrated in the figure below. Refer to section 5.4
on 2D Cartesian transformations for more details.

The principle of changing from one unknown projection into a known projection using a 2D Cartesian
transformation. A number of 2D control points are required to determine the relation between both systems.
Two-dimensional Cartesian transformations have a different accuracy compared to the transformations
based on projection equations. The latter take into account the Earth curvature. This is especially
important in the case of large areas and small scale. However, if the control points are coplanar and
the extent of the area is not too large, the 2D Cartesian transformation could yield a better model of
coordinate relations than the presumed set of projection equations would do.
5.2 Datum transformations
A change of map projection may also include a change of the horizontal datum (also called geodetic
datum). This is the case when the source projection is based upon a different horizontal datum than the
target projection. If the difference in horizontal datums is ignored, there will be no perfect match between
adjacent maps of neighbouring countries or between overlaid maps originating from different
projections. It may result in up to several hundred metres difference in the resulting coordinates.
Therefore, spatial data with different underlying horizontal datums may need a so-called datum
transformation. Datum transformations are transformations from a 3D coordinate system (i.e.
horizontal datum) into another 3D coordinate system.

Datum shift between two geodetic datums. Apart from different ellipsoids, the centres or the rotation axes of the
ellipsoids do not coincide.
Suppose we wish to transform spatial data from the UTM projection to the Dutch RD system, and that
the data in the UTM system are related to the European Datum 1950 (ED50), while the Dutch RD
system underlies the Amersfoort datum. In this example the change of map projection must be
combined with a datum transformation step for a perfect match. This is illustrated in the figure below.

The principle of changing from one into another projection combined with a datum transformation from datum A
to datum B.
The inverse mapping equation of projection A is used first to take us from the map coordinates (x,y) of
projection A to the geographic coordinates (|,) in datum A. Next, the datum transformation takes us
from the geographic coordinates (|,) in datum A to the geographic coordinates (|,) in datum B.
Finally, the forward equation of projection B is used to take us from the geographic coordinates (f,l) in
datum B to the map coordinates (x,y') of projection B. A height coordinate (h or H) may be added to the
geographic coordinates.
The transformation parameters to take us from one datum system to another datum system are
estimated on the basis of a set of selected points whose coordinates are known in both datum systems.
If the coordinates of these points are not correct - often the case for points measured on a local datum
system - the estimated parameters may be inaccurate. As a result the datum transformation will be
inaccurate. This is often the case when we transform coordinates from a local horizontal datum to a
global geocentric datum. The coordinates in the local horizontal datum may be distorted by several tens
of metres because of the inherent inaccuracies of the measurements used in the triangulation network.
These inherent inaccuracies are also responsible for another complication: the transformation
parameters are not unique. Their estimate will depend on which particular common points are chosen,
and they also will depend on the amount of transformation parameters estimated.
Mathematically a datum transformation can be realized by relating the geographic coordinates (f,l,h) of
both datum systems directly, or indirectly by relating the geocentric coordinates (x,y,z) of the datums.
The first method is discussed in section 5.2.2, the latter in section 5.2.1.
5.2.1 Datum transformations via geocentric coordinates
Datum transformations via the geocentric coordinates (x,y,z) are 3D similarity transformations.
Essentially, these are transformations between two orthogonal 3D Cartesian spatial reference frames
together with some elementary tools from adjustment theory. This is illustrated in the figure below.

The principle of changing from one datum into another datum via the geocentric coordinates.
The three most applied methods for a datum transformation via the 3-dimensional geocentric
coordinates are:
i. the geocentric translation,
ii. the Helmert 7-parameter transformations (position vector or coordinate frame), and
iii. the Molodensky-Badekas 10-parameter transformation.
Refer to the OGP Guidance note 7: Coordinate Conversions and Transformations including Formulas
(page 105-110) for the formulas and a detailed explanation of the formulas.
i. The geocentric translation relates two datum systems through three translations. The method
applies a shift between the centres of the two geocentric coordinate systems. This shift is defined by
the parameters AX,AY and AZ (or Xo,Yo and Zo).
An example, the ITRF (X,Y,Z) coordinates of a point in the state of Baden-Wrttemberg (Germany) are:
X (ITRF) = 4,156,939.96m
Y (ITRF) = 671,428.74m
Z (ITRF) = 4,774,958.21m
The three translation parameters for transforming the point from ITRF to the Potsdam datum (the local
datum of Germany with underlying Bessel ellipsoid) are given as follow:AX= -635.00m, AY= -
27.00m and AZ= -450.00m. This set of parameters provided by the National Imagery and Mapping
Agency (NIMA) has been calculated using common points distributed throughout Germany and based
on the ITRF. Applying these geocentric translations to the given point results in the following Potsdam
(X,Y,Z) coordinates:
X (Potsdam) = 4,156,939.96 - 635.00 = 4,156,304.96m
Y (Potsdam) = 671,428.74 - 27.00 = 671,401.74m
Z (Potsdam) = 4,774,958.21 - 450.00 = 4,774,508.21m
ii. The Helmert 7-parameter transformation relates two datum systems through a rotation, an origin
shift and a scale factor. The transformation is expressed with seven parameters: three rotation angles
(o,|,), three origin shifts (AX,AY and AZ) and one scale factor (s). An example, the ITRF (X,Y,Z)
coordinates of the given point in the state of Baden-Wrttemberg are transformed to the Potsdam
datum. The 7 parameters for transforming the point from ITRF to the Potsdam datum are given as
follow:o=+1.04sec,|=+0.35sec,=-3.08sec, AX=-581.99m, AY=-105.01m,AZ=-414.00m and s=1-(8.3 *
10
-6
)=0.9999917. This set of parameters provided by the federal mapping organization of Germany
was calculated using common points distributed throughout Germany. Applying these parameters to
the given point results in the following Potsdam (X,Y,Z) coordinates:
X (Potsdam) = 4,156,305.34m
Y (Potsdam) = 671,404.31m
Z (Potsdam) = 4,774,508.25m
The two sets of transformed coordinates agree at a level of a few meters. The difference in X is 0.38m,
in Y around 2.57m, and in Z around 0.04m. These differences can be explained because of the different
set of transformation parameters. In a different country, the agreement could be at the level of
centimetres, or tens of meters and this depends primarily on the quality of implementation of the local
horizontal datum.
The Helmert 7-parameter transformation is considered to be reversible, i.e. the same parameter values
can be used to execute the reverse transformation. Thus, the reverse parameters can be applied,
whereby o=-1.04sec, |=-0.35sec,=+3.08sec, AX=+581.99m, AY=+105.01m, AZ=+414.00m and s=1 +
(8.3 * 10
-6
)=1.0000083, to transform the given point from Potsdam coordinates to ITRF coordinates.
The Helmert 7-parameter transformation can be either a position vector transformation or
a coordinate frame transformation. Both transformations are based on the same definition of
translation and scale parameters, but a different definition of the rotation parameters. The coordinate
frame transformation assumes that the rotations are applied to the coordinate reference frame, while
the position vector transformation (also called Bursa-Wolf transformation) assumes that the rotations
are applied to the point's vector. The only difference between the two methods is that the rotation has
changed sign. The position vector transformation is commonly used, but is not universally accepted. It
is therefore essential to communicate the related coordinate transformation method when exchanging
datum transformation parameters.
iii. The Molodensky-Badekas 10-parameter transformation relates two datum systems through a
rotation, an origin shift and a scale factor. This is the same as for the Helmert transformation methods,
but instead of deriving the rotations about the origin of the geocentric coordinate system, they are
derived at a location within the points used in the determination of the parameters. Three additional
parameters, the coordinates of the rotation point (Xp,Yp,Zp), are then required. The transformation is
therefore expressed with 10 parameters: three rotation angles (o,|,), three origin shifts (AX,AY and AZ),
one scale factor (s) and the coordinates of the rotation point (Xp,Yp,Zp) given in the source geocentric
coordinate system. Compared to the Helmert transformation, the Molodensky-badekas provides usually
a better approximation, but the transformation is not reversible.
5.2.2 Datum transformations via geographic coordinates
Datum transformations via the geographic coordinates directly relate the ellipsoidal latitude (|) and
longitude (), and possibly also the ellipsoidal height (h), of both datum systems. This is illustrated in
the figure below.

The principle of changing from one datum into another datum via the 3D geographic coordinates.
The applied methods for a datum transformation via the 3-dimensional geographic coordinates are:
i. the geographic offsets,
ii. the Molodensky and Abridged Molodensky transformation, and
iii. the multiple regression transformation.
i. The simplest method uses geographic offsets. It relates both datum systems with only two
parameters, the difference in the geographic latitude (A|) and the difference in the geographic
longitude (A). The ellipsoidal height (h) is mostly not included. The method is only used for
purposes where low accuracy can be tolerated. The equation is:
|datum B=|datum A+A|
datum B=datum A +A
ii. A second method of directly transforming latitude, longitude, and height is the
standard Molodensky and Abridged Molodensky transformation (it should not be confused with
the Molodensky-Badedas transformation). The standard equations directly relate ellipsoidal
latitude and longitude coordinates and ellipsoidal height of two datums by deriving the geographic
coordinate offsets. The abridged form is found by dropping any terms that are second order in
small parameters, and the ellipsoidal height (h) is ignored. Refer to the OGP Guidance note
7 (page 112) for the abridged versions of the formulas and a detailed explanation. The simplified
equation is:
A| = f (Ax,Ay,Az,Aa,Af)
A = f (Ax,Ay,Az,Aa,Af)
Ah = f (Ax, Ay,Az, Aa, Af)
where Ax,Ay,Az are the geocentric translation parameters (centre offset between the two datums), Aa is
the difference in the semi-major axes of the target and source ellipsoids, Af is the difference in the
flattening of the two ellipsoids.
The Molodensky equations are commonly used to relate ellipsoidal latitude and longitude coordinates
and ellipsoidal height of a local geodetic datum to those of theWGS84 datum. An example is illustrated
in the figure below.

The principle of changing from a local datum (Bahamas) into the WGS84 datum via the 3D geographic
coordinates using Molodensky. Given are the centre offset parameters (Ax, Ay, Az) , the difference in the semi-
major axes (Aa) and the difference in the flattening (Af).
The Molodensky equations are easy to use to transform GPS coordinates in WGS84 to a local map
coordinate system (map projection) with it's own ellipsoid and datum. Simply the ellipsoid parameters
and the datum shift values (centre offset between the two datums) are required to relate the two datums.
The list defines a datum as: datum name = ellipsoid name, shift X, shift Y, shift Z, name geographic
area or as: datum name = ellipsoid name (in this case a subsection [datum name] provides the shift).
The datum shift values refer to WGS84. Tools for the transformation of coordinates from one 3D
geographic coordinate system into another using the Molodensky methods:
- Molodensky (program that finds |,, and h for a point in the WGS84 datum system. The input
are the major-axis (a), inverse flattening (f) and the |,, and h of the point in a local datum
system). MolodenskyGeneralized (program that finds |, l, and h for apoint in a target datum
system. The input are the major-axis (a), inverse flattening (f) and the |, , and h of the point in
a source datum system). The program uses the abridged version of the molodensky
transformation.
- MolodenskyInverse (program that finds AX, AY and AZ, the centre offset between two datums.
The input are the major-axis (a), inverse flattening (f) and 3 points (|, , and h) in both the
WGS84 and the local system).
The problem with the Molodensky transformation is the limited amount of parameters (Ax, Ay, Az, Aa, Af)
that it uses. This is quite satisfactory for small areas, but for larger areas such as large countries or
continents (e.g. US with NAD27, Europe with ED50) significant errors occur (tens of meters). It requires
another method to provide good datum shifts for these original types of large area datums.
iii. A third method uses multiple regression equations. A series of best-fit equations provide the
local shifts in latitude and longitude as a function of position. It is commonly used to relate
ellipsoidal latitude and longitude coordinates of continental size datums to those of the WGS84
datum and involve polynomial expressions in the two ellipsoidal coordinates which go up to
degree 9 for the time being. The coefficients (transformation parameters) are determined on the
basis of coordinate differences of a set of selected points whose coordinates are known in both
datum systems. The main advantage of this method over Molodensky (often implemented in
geo-software) is that a better fit over continental size land areas can be achieved. The simplified
equation is:
A| = f (|,, a1, a2, a3, .......)
A = f (|,,b1, b2, b3, ...... )
Ah = f (|,, c1, c2, c3, .......)
5.3 Conversions from geographic to geocentric coordinates and visa versa
The three geocentric transformations decribed in section 5.2.1 are usually combined with conversions
between the geocentric coordinates (x,y,z) and the ellipsoidal latitude (|) and longitude () coordinates
and height (h) in both datum systems. This is illustrated in the figure below.

The principle of a datum transformation via the geocentric coordinates. The datum transformation is combined
with conversions between the 3D geographic coordinates and geocentric coordinates in both datum systems.
5.3.1 Geographic to geocentric conversion
The conversion from the latitude and longitude coordinates into the geocentric coordinates is rather
straightforward and turns ellipsoidal latitude (|) and longitude (), and possibly also the ellipsoidal
height (h), into X,Y and Z, using 3 direct equations. If the ellipsoidal semi-major axis is a, semi-minor
axis b, and inverse flattening 1/f, then
X = (u+ h) cos | cos
Y = (u+ h) cos | sin
Z = [(1- e
2
) u + h] sin |
where u is the prime vertical radius of curvature at latitude | and is equal to u = a /(1 e
2
sin
2
\)
0.5
,
e is the eccentricity of the ellipsoid where e
2
= (a
2
b
2
) / a
2
= 2f f
2
.
An example, the ITRF(|,,h) coordinates of the given point in the state of Baden-Wrttemberg are:
| (ITRF) = 48 46' 59.6564" N
(ITRF) = 9 10' 30.6113" E
h (ITRF) = 330.397m
Applying these formulas to the given point in the state of Baden-Wrttemberg (Germany) results in the
following ITRF (X,Y,Z) coordinates:
X (ITRF) = 4,156,939.96m
Y (ITRF) = 671,428.74m
Z (ITRF) = 4,774,958.21m
To apply the required conversion we need the ellipsoidal height. Datum transformation programs
implemented in GIS software often simplify this transformation: i.e. ellipsoidal heights (h) are taken
equal to 0. It is obvious that such a simplification will affect the accuracy of the datum transformation.
Instead of setting the ellipsoidal heights (h) to zero, the mean sea level heights (H) may be used as
input.
5.3.2 Geocentric to geographic conversion
The inverse equations for the reverse conversion are more complicated and require either an iterative
calculation of the latitude and ellipsoidal height, or it makes use of approximating equations like those
of Bowring. These last have millimetre precision for 'Earth-bound' points, i.e. points that are at most 10
km away from the ellipsoidal surface (any point on the Earth surface). Refer to the OGP Guidance note
7 (page 76) for the formulas and a detailed explanation of the formulas. An example, the Potsdam
(X,Y,Z) coordinates of the given point in the state of Baden-Wrttemberg are:
X (Potsdam) = 4,156,305.34m
Y (Potsdam) = 671,404.31m
Z (Potsdam) = 4,774,508.25m
These Potsdam (X,Y,Z) coordinates were computed from the ITRF (X,Y,Z) coordinates by applying a
Helmert datum transformation (section 5.2.1). Applying the inverse equations to the given point results
in the following Potsdam (|,,h) coordinates:
| (Potsdam) =48 47' 3.2752" N
(Potsdam) = 9 10' 34.3870" E
h (Potsdam) = 278.825m
5.4 2D Cartesian coordinate transformations
2D Cartesian coordinate transformations can be used to transform 2D Cartesian coordinates (x,y) from
one 2D Cartesian coordinate system to another 2D Cartesian coordinate system. The three primary
transformation methods are:
i. the conformal transformation,
ii. the affine transformation, and
iii. the polynomial transformation.
i. A conformal transformation is a linear (or first-order) transformation and relates two 2D
Cartesian coordinate systems through a rotation, a uniform scale change, followed by
a translation. The rotation is defined by one rotation angle (o), and the scale change by one
scale factor (s). The translation is defined by two origin shift parameters (xo,yo). The equation is:
X' = s X cos(o) - s Y sin(o) + xo
Y' = s X sin(o) + s Y cos(o) + yo
The simplified equation is:
X' = aX - bY + xo
Y' = bX + aY + yo
where a=s cos(o) and b=s sin(o). The transformation parameters (or coefficients) are a,b,xo,yo.
ii. An affine transformation is a linear (or first-order) transformation and relates two 2D Cartesian
coordinate systems through a rotation, a scale change in x- andy-direction, followed by
a translation. The transformation function is expressed with 6 parameters:one rotation angle
(o), two scale factors, a scale factor in the x-direction (sx) and a scale factor in the y-direction
(sy), and two origin shifts (xo,yo). The equation is:
X' = sx X cos(o) - sy Y sin(o) + xo
Y' = sx X sin(o) + sy Y cos(o) + yo
The simplified equation is:
X' = aX - bY + xo
Y' = cX + dY + yo
where the transformation parameters (or coefficients) are a,b,c,d,xo,yo.
The diffference between a conformal and an affine transformation is illustrated in the figure below. Both
are linear transformations which means that the lines of the grid remain straight after the transformation.

a) The uniform scale change of the conformal transformation retains the shape of the original rectangular grid. b)
The different scale in x and y-direction of the affine transformation changes the shape of the original rectangular
grid, but the lines of the grid remain straight.
iii. A polynomial transformation is a non-linear transformation and relates two 2D Cartesian
coordinate systems through a translation, a rotation and a variable scale change. The
transformation function can have an infinite number of terms. The equation is:
X' = xo + a1X+ a2Y+ a3XY + a4X
2
+ a5Y
2
+ a6X
2
Y+ a7XY
2
+ a8X
3
+........
Y' = yo+ b1X+ b2Y+ b3XY + b4X
2
+ b5Y
2
+ b6X
2
Y+ b7XY
2
+ b8X
3
+........
Polynomial transformations are sometimes used to georeference uncorrected satellite imagery or aerial
photographs or to match vector data layers that don't fit exactly by stretching or rubber sheeting them
over the most accurate data layer. The figure below shows a grid with no uniform scale distortions. It
may occur in an aerial photograph, caused by the tilting of the camera and the terrain relief (topography).
An approximate correction may be derived through a high-order polynomial transformation. The
displacements caused by relief differences can be corrected using a Digital Elevation Model (DEM).

A grid with no uniform scale distortions. An approximate correction may be derived through a high order
polynomial transformation.
2D Cartesian coordinate transformations are generally used to assign map coordinates (x,y) to an
uncorrected image or scanned map. The type of transformation (usually an affine transformation)
depends on the geometric errors in the data set. This is illustrated in the below figure (a). After
georeferencing, the image can be aligned (rectified) so that the pixels are exactly positioned within the
map coordinate system (figure (b)). For each image pixel in the new coordinate system, a new pixel
value has to be determined by means of an interpolation from surrounding pixels in the old image. This
is called image resampling.

a) Coordinates are assigned to a raster image by means of a 2D transformation using a set of control points. The
image is georeferenced; b) The georeferenced image is rectified to match it with the map coordinate system.
The parameters (or coefficients or unknowns) of a conformal, affine or polynomial transformation are
usually computed with ground control points (GCPs) or common points (also called tie points) such
as corners of houses or road intersections, as long as they have known coordinates in both systems.
The number of control points required to determine the 4 parameters (a,b,xo,yo) of a conformal
transformation must be at least 2. An affine transformation requires at least 3 control points to determine
the 6 parameters (a,b,c,d,xo,yo), and 6 control points are required to determine the 12 parameters (xo,a1-
a5,yo,b1-b5) of a simple second-order polynomial transformation..
If there are more control points (at least one) than actually required to determine the parameters, we
can compute the Root Mean Squares Error (RMSE) using the least squares adjustment. The RMSE
is a statistical measure of accuracy for a coordinate measurement, similar to standard deviation,
indicating the spread of the measured values around the true value. It is a way to quantify the difference
between the computed positions and the true positions (residuals) and it represents the average
deviation of all the measured positions from their true positions. It is used as an independent check of
the quality of the transformation. If the RMSE is not acceptable, then the transformation can be repeated
using a different selection or more control points, or a different transformation method can be used.
5.5 Summary
In the previous sections we discussed several types of coordinate transformations: projection change,
3D datum transformations and 2D Cartesian transformations. An overview of these transformations is
given in the figure below.
The illustration shows how the inverse and forward mapping (or projection) equations are used to
transform coordinates from one map coordinate system (projection A) to another (projection B). The
projection change may include a datum transformation from one datum (datum A) to another datum
(datum B) in case the source projection (projection A) is based upon a different horizontal datum than
the target projection (projection B). The datum transformation may take place via a 3D geocentric
transformation or directly via a 3D geographic transformation. Alternatively, 2D Cartesian
transformations may be used to transform coordinates from one map coordinate system to another (e.g.
in case the projection of the input map coordinates is unknown).

Overview of several types of coordinate transformations.
5.6 Main references
Knippers, R.A and Hendrikse J. Coordinate transformations. Kartografisch Tijdschrift, KernKatern
2000-3, 2001.
de By, R.A. (editor), Georgiadou, P.Y., Knippers, R.A., Kraak, M.J ., Sun, Y., Weir, M.J .C. and van
Westen, C.J . Principles of geographic information systems(Chapter 4.2 on spatial referencing), 2nd
edition, ITC Educational Textbook, ITC, Enschede, 2001.

8.0 Introduction
This section intends to answer your questions concerning Geometric Aspects of Mapping. Answers are
given on a number of commonly asked questions related to coordinate systems, reference surfaces,
map projections, and coordinate transformations.
8.1 Ask a question
Questions concerning Geometric Aspects of Mapping can be sent to us by filling in the Question Form:
8.2 FAQ on coordinate systems
1. What is a coordinate system? A method to locate the position of objects in two or three
dimensions into correct relationship with respect to each other.
2. What kind of coordinate systems are used in mapping? Spatial coordinate systems (also known
as global coordinate systems) are used to locate data either on the Earths surface in a 3D space,
or on the Earths reference surface (ellipsoid or sphere) in a 2D space. Examples are the geographic
coordinates in a 2D or 3D space and the geocentric coordinates, also known as 3D Cartesian
coordinates. Planar coordinate systems on the other hand are used to locate data on the flat surface
of the map in a 2D space. Examples are the 2D Cartesian coordinates and the 2D polar
coordinates. Refer to section 2 for details.
3. What is a graticule? The graticule represents the projected position of the geographic
coordinates (|,) on a map at constant intervals, or in other words the projected position of selected
meridians and parallels. The shape of the graticule depends largely on the characteristics of the
map projection used and the scale of the map.
4. What is a grid? The map grid represents lines having constant 2D Cartesian coordinates (x,y). It
is almost always a rectangular system and is used on large and medium scale maps to enable
detailed calculations and positioning. The map grid is usually not shown on small scale maps (about
one to a million or smaller). Scale distortions that result from transforming the Earths curved surface
to the map plane are so great on small-scale maps that detailed calculations and positioning are
difficult.
8.3 FAQ on reference surfaces
5. What kind of reference surfaces are used in mapping? In mapping different surfaces or Earth
figures are used. These include a geometric or mathematical reference surface, the ellipsoid or the
sphere, for measuring locations, and an equipotential surface called the Geoid or vertical datum for
measuring heights. See section 3 for details.
6. Can we approximate the Geoid with a mathematical model? The Geoid is extremely undulated
due to the large vertical variations between mountains and valleys and mass anomalies. This makes
it impossible to approximate the shape of the Geoid with any reasonably simple mathematical
model. You also have to realize that the Geoid is a physical surface. We can only measure it, but
not describe it mathematically. Another difficulty is that there are several realizations of the Geoid
due to local phenomena such as ocean currents, tides, coastal winds, water temperature and
salinity at the location of the tide-gauge.
7. What is a vertical datum? The vertical datum, an approximation of the Geoid, is defined as natural
reference surface for land surveying. A vertical datum fits the mean sea level surface throughout
the area of interest and provides the surface to which height ground control measurements are
referred.
8. Why are there several vertical datums defined? Historically, the vertical datum (or height datum)
is fixed locally not globally by establishing a local mean sea level. Several of these local vertical
datums exist in order to minimize distortions due to ocean current, water/sea tides, winds, water
temperature, and salinity at the place of tidal gauge. These local measurements of the mean sea
level height (zero height) are an approximation of the Geoid. Nowadays there are possibilities to
measure the Geoid with gravity satellite missions. They might lead in the future to the establishment
of one global Geoid (height datum) with centimeter level accuracies.
9. Why do we need an ellipsoid as reference surface in mapping? The physical surface of the
Earth is a complex shape. In order to represent it on a plane it is necessary to move from the
physical surface to a mathematical one, close to the former.
10. When can we use the sphere as the Earth figure? The surface of the Earth may be taken
mathematically as a sphere instead of ellipsoid for maps at smaller scales. In practice, maps at
scale 1:5,000,000 or smaller can use the mathematically simpler sphere without the risk of large
distortions. At larger scales, the more complicated mathematics of ellipsoids are needed to prevent
these distortions in the map. A sphere can be derived from the certain ellipsoid corresponding either
to the semi-major or semi-minor axis, or average of both axes or can have equal volume or equal
surface than the ellipsoid.
11. What is a horizontal (or geodetic) datum? A horizontal (or geodetic) datum is defined by the size
and shape of an ellipsoid as well as several known positions on the physical surface at which
latitude and longitude measured on that ellipsoid are known to fix the position of the ellipsoid. It is
used for geodetic control and as reference surface for the planimetric (horizontal) measurements
on the Earth surface.
12. What is the difference between an ellipsoid and a horizontal (or geodetic) datum? A horizontal
datum is the realization of an ellipsoid with a certain position and orientation by means of survey
methods. E.g. a local geodetic datum is determined by the dimensions of the ellipsoid, adopted
coordinates of one fundamental point and the orientation of the ellipsoid (the azimuth from the
fundamental point to another point). A local horizontal datum is determined by the dimensions of
the ellipsoid (a,b), adopted coordinates (geographic coordinates) of one fundamental point (a point
of the national triangulation network), and the orientation of the ellipsoid (the azimuth from the
fundamental point to another point).
13. Why are there several ellipsoids and datums defined? An ellipsoid and a horizontal datum serve
as geometric models of the Earth surface. Different local ellipsoids with varying position and
orientation had to be adopted to best fit the Earth's surface over an area of local interest. This is
important to minimize distortions on maps. A large number of reference ellipsoids and many more
local datums may be encountered in world mapping. Some well-known ellipsoids are the
International, Krasovsky, Bessel, Clark 1880, GRS80 and the WGS84 ellipsoid. Local ellipsoids
serve as reference only for a local area of the Earth's surface. Global ellipsoids (e.g. WGS84) serve
as mean reference for the entire Earth surface.
14. What is the difference between a local and global horizontal datum? A local datum is
determined by the dimensions of the ellipsoid (a,b), adopted coordinates (geographic coordinates)
of one fundamental point (a point of the national triangulation network), and the orientation of the
ellipsoid (the azimuth from the fundamental point to another point). Positions in terms of the locally
adopted datum are propagated through the country, with varying degrees of accuracy, by survey
networks. In the past local datums were generally based on one or more astronomically determined
positions and the best model of the Earth for the local area. Global (or geocentric) datums (e.g.
ITRS, WGS84, ETRS89) are determined by means of satellite-based positioning techniques. By
observing satellites one can pinpoint the centre of mass of the Earth, and hence the origin of the
geocentric datum. Permanently operating satellite positioning equipment (e.g. GPS receivers) allow
the implementation of a global datum (e.g. ITRF2000).
15. What is a Terrestrial Reference System? A Terrestrial Reference System (TRS) is a spatial
reference system co-rotating with the Earth in its diurnal motion in space. In such a system,
positions of points anchored on the Earth solid surface have coordinates which undergo only small
variations with time, due to geophysical effects (tectonic or tidal deformations).
16. What is a Terrestrial Reference Frame? A Terrestrial Reference Frame (TRF) is a set of physical
points with precisely determined coordinates in a specific coordinate system (cartesian, geographic,
mapping...) attached to a Terrestrial Reference System. Such a TRF is said to be a realization of
the TRS.
17. What is the International Terrestrial Reference System? The International Terrestrial Reference
System (ITRS) is a global datum for the entire world defined by a 3D geocentric coordinate system.
Its origin is located in the centre of mass of the Earth. The X-axis is oriented towards the Greenwich
meridian, and is orthogonal to the Z-axis and the Y-axis. The ITRS is realized/established through
the ITRF, a distributed set of ground control stations that measure their position continuously (using
GPS). Constant re-measuring is needed because of the continuously involvement of new control
stations and the tectonic plate motion that causes positional differences in time. This results in a
more than one time realization of the ITRS. Examples are the ITRF96 or the ITRF2000.
18. What is WGS84? WGS84 is one of the World Geodetic Systems which provides the basic
reference frame and geometric figure for the Earth. WGS84 provides a positional relation of various
local geodetic systems to an Earth-centered, Earth-fixed coordinate system, through reports of the
DMA of U.S. Defense (D.O.D). GPS measurements use the WGS84 as reference surface for their
measurements.
19. What is Mean Sea level? Refer to extern link: <http://www.pol.ac.uk/psmsl/puscience>.
20. How do we determine the geodetic latitude (|) and longitude() of a point? Satellite based
measurements (e.g. GPS) are nowadays used to precisely determine the geodetic coordinates of
a point. Before the introduction of GPS the latitude and longitude of a point was measured using
geodetic positioning techniques such as triangulation and trilateration, and astronomic
observations.
8.4 FAQ on map projections
21. What is a map projection? A map projection is a mathematically described technique of how to
represent the Earths curved surface on a flat map. To represent parts of the surface of the Earth
on a flat paper map or on a computer screen, the curved horizontal reference surface must be
mapped onto the 2D mapping plane. The reference surface for large-scale mapping is usually an
oblate ellipsoid, and for small-scale mapping, a sphere. Mapping onto a 2D mapping plane means
transforming each point on the reference surface with geographic coordinates (|,) to a set of
Cartesian coordinates (x,y) representing positions on the map plane.
(x, y) map projection =f (|,)
The corresponding inverse mapping equation transforms mathematically the planar Cartesian
coordinates (x,y) of a point on the map plane to a set of geographic coordinates (|,) on the curved
reference surface. See section 4 for details.
22. What are map projection parameters? Map projection parameters are part of the projection
equations. Common parameters are: R=radius of the sphere;a=equatorial radius or semi-major
axes of the ellipsoid of reference; b=polar radius or semi-minor axes of the ellipsoid of
reference; e=eccentricity of the ellipsoid;f=flattening of the ellipsoid; ho=scale factor at central
meridian; h=relative scale factor along a meridian of longitude; ko=scale factor at standard
parallel(s); k=relative scale factor along a parallel of latitude; o=central meridian or longitude of
origin; o=latitude of origin; Xo=false Easting; Yo=false Northing. Projection parameters have a
significant role in defining a map coordinate system. An extended list of parameters is given in 'Map
Projections - A Working manual', p.viii-ix by J .Snyder.
23. Why do we need a map projection? If you are mapping a significant portion of the Earth's surface
it is impossible to project it on a flat piece of paper without scale distortions. Map projections take
care that the scale distortions remain within certain limits and the distortion pattern of a map
projection determines the property of the projection. Each projection has its own characteristics.
For example a map projection may have the property that all angles are correctly represented
(conformal projection property). A map projection is not of major importance for city or street maps,
which cover a relatively small surface of the Earth.
24. How do we classify map projections? Map projections can be classified in terms of their class
(cylindrical, conical, azimuthal), their property (equivalent, equidistant, conformal), their aspect
(normal/polar, transverse, oblique), and its tangent or secant map surface. An example would be
the classification conformal conic projection with two standard parallels having the meaning that
the projection is a conformal map projection, that the intermediate surface is a cone, and that the
cone intersects the ellipsoid (or sphere) along two parallels; i.e. the cone is secant and the cones
symmetry axis is parallel to the rotation axis. Other examples are:
- Polar stereographic azimuthal projection with secant projection plane
- Lambert conformal conic projection with two standard parallels
- Lambert cylindrical equal-area projection with equidistant equator
- Transverse Mercator projection with secant projection plane.
25. Why are some map projections using mapping zones? There are map projections that divide
the mapping area into zones in order to keep scale distortions within acceptable limits. A good
example is the Universal Transverse Mercator (UTM) projection. Each zoning systems has its own
map coordinate system (x,y) with a specific origin. Coordinate transformations are required to match
areas that are located in different zones.
26. How to determine the zon(s) of the UTM for a certain country? You need to know the east-west
extent of the country and compare it to the UTM zone numbering system. The zone numbering
system of the UTM projection specifies 60 zones of 6 degrees longitude each. The zones are
numbered from 1 to 60 starting with zone number 1 at -180o (West of Greenwich) until -174o (West
of Greenwich) and so on. An example, the Netherlands is located between the longitudes 3 and 7
degrees East of Greenwich, so it covers 2 UTM zones, zone number 31 and zone number 32. Zone
number 31 covers the area from 0o (Greenwich) until 6o (East of Greenwich), and zone 32 the area
between 6-12o E.
27. How do we match adjacent maps? In order to fit two or more separate maps exactly along their
edges, a number of parameters must be maintained: 1. the maps must be constructed with the
same projection and projection parameters; 2. they must be at the same scale; and 3. they should
be based on the same reference datum.
28. How to select a suitable map projection? The choice of a map projection class largely depends
on the size and shape of the geographical area to be mapped; cylindrical projections are often used
for large rectangular areas (and to map the world); conic projections for medium size triangular
areas (and to map the different continents); azimuthal projections small-size circular areas (and to
map the poles). The choice of a map projection property has to be made on the basis of the purpose
of map; conformal projections are often used for sea, air and meteorological charts, topographic
and large scale maps; equidistant projections are often used for air-route, radio or seismic maps
but also for topographic and large scale maps; equal-area projections are often used for distribution
maps and also for historical, population, geological and soil maps (more information in section 4).
29. What type of information concerning the map coordinate system should be shown on a
map? Most important items are the scale (graphic and/or numerical), the map projection (incl.
ellipsoid and datum), the grid and/or the graticule, and in the border area the grid and graticule
values. Other items might be necessary to put a map (e.g. topographic map) in its full use: items
such as the vertical (height) datum, unit of elevation (meters/feet), contour interval, graphic scales
in kilometers/statute miles/nautical miles, projection details (parameters such as the false
easting/northing, central meridian, etc), geographic coordinates of sheetcorners, instructions on the
use of the grid, information on true, grid and magnetic north, etc. Examples of coordinate
information on a topographic map:

Notes concerning the map coordinate system in the legend of a German topographic map (scale 1:25 000).

Notes concerning the map coordinate system in the legend of maps distributed by the NIMA.
30. What are True North, Magnetic North, and Grid North? The True North is the direction of the
meridian to the North Pole at any point on the map. The Grid North is the northern direction of the
north-south grid lines on a map. The Magnetic North is the direction of the Magnetic North Pole as
shown on a compass free from error or disturbance. The corresponding bearings are called: true
bearing or geodetic bearing, grid bearing and magnetic or compass bearing.

Three different angles are in use: The Magnetic Declination is the angle between Magnetic North
and True North at any point (in the figure 7 degrees and 36 minutes at the sheet centre of a map).
The Grid Convergence is the angle between Grid North and True North. The Grid Magnetic Angle is
the angle between Grid North and Magnetic North (in the figure 4 degrees and 58 minutes at the
sheet centre of a map). This is the angle required for conversion of grid bearings to magnetic
bearings or vice versa. The Annual Magnetic Change (8 minutes East in the figure above) is the
amount by which the magnetic declination changes annually because of the change in position of
the magnetic north pole
8.5 FAQ on coordinate transformations
31. What is a coordinate transformation? A coordinate transformation is a conversion of coordinates
from one to another coordinate system. Transformations can be between plane coordinate systems,
between geographic and plane coordinate systems, between geographic coordinates and
geocentric coordinate systems, etc. See section 5 for details.
32. What is a map projection change? The transformation of coordinates from a plane system based
on one projection type into a plane system based on another type. An example, a projection change
can transform x, y coordinates from the UTM coordinate system into the Lambert Conformal Conical
projection system. Forward and inverse mapping equations (see section 4 on map projections) are
normally used to transform data from one map projection to another.
33. How to convert a data set from one UTM zone to another UTM zone? Make use of a projection
change (see section 5.1). Project the data onto the reference surface (geodetic datum) with the
inverse equations of the UTM projection and thereafter project the data onto the target (adjacent)
UTM zone with the forward UTM projection equations. The conversion into another UTM zone may
also be applied via a 2D Cartesian transformation, but it requires a set of control points (common
points) to determine the relationship between the two UTM zones, and the transformation will give
a different (perhaps unacceptable) accuracy.
34. How do you assign coordinates to a data set with unknown coordinates? You can use control
points, such as the corners of houses, or road intersections, with known coordinates taken from a
map or another data source to determine the relationship between the unknown and a known
coordinate system. The transformation may be conformal, affine or polynomial depending on the
systematic errors in the data set.
35. How do you georeference a map (e.g. for digitizing), if the only gi ven coordinate system are
geographic coordinates? Select four graticule corner points and convert these points from
geographic coordinates into map coordinates (x,y) using the forward equations of the map
projection. Directly using the geographic coordinates for the georeferencing may cause geometric
and alignment errors.
36. Why do we need datum transformations? Numerous maps are projected onto various ellipsoids
or reference datums. Very often, it is needed to transform one datum to another to avoid alignment
problems. See section 5.2 for details.