lOMoARcPSD|43072212

Map projections - NOTES

Geomatic and geospatial information system (GEGIS) (Jomo Kenyatta University of

Agriculture and Technology)

Scan to open on Studocu

Studocu is not sponsored or endorsed by any college or university

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

Geographical data projections

Map projections

A map projection is a geometrical transformation of the earth's spherical or ellipsoidal surface onto
a flat map surface, it is an attempt to portray the surface of the earth or a portion of the earth on a
flat surface.
Some distortions of conformity, distance, direction, scale, and area always result from this process.
Some projections minimize distortions in some of these properties at the expense of maximizing
errors in others. Some projection are attempts to only moderately distort all of these properties.

GLOBES VERSUS FLAT MAPS

Of all maps, globes give us the most realistic picture of the earth as a whole. Basic geometric
properties such as distance, direction, shape, and area are preserved because the globe is the same
scale everywhere.

Globes have a number of disadvantages, however.

i. They don't let you view all parts of the earth's surface simultaneously the most you can see
is a hemisphere
ii. Globes limit the kind of detail –too much generalization.
iii. Globes also are bulky and don't lend themselves to convenient handling and storage.
iv. The instruments and techniques that are suited for measuring distance, direction, and area
on spherical surfaces are relatively difficult to use.
v. Computations on a sphere are far more complex than those on a plane surface.

Types of Map Projection

Although the term 'projection' is used to describe the various transformations that enable us to
View the spherical surface of the earth on a planar map, only a few actually are true 'projections'
in the optical sense of the word. Indeed, there are two broad classes of "projections": true
projections, and those based on mathematical transformations.
True projections
These are formed by the projection of the earth's graticule of meridians and parallels on to some

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

sort of screen or projection surface. Thus it is useful to imagine the earth as a transparent globe,
perhaps made of clear glass, on which the meridians and parallels are painted as black lines. The
position of the light source may be inside or outside the globe and of course the screen or projection
surface can vary in shape as well.
Projection surfaces are of two general types:
(a) Plane surface
(b) Developable surfaces.
Plane surface is the simplest projection surface and provides the most direct link between
the global geography and the map. A plane can only touch the surface of a sphere at a single point
of tangency and for this one unique point there is no distortion of the spherical grid on the map.
For all other points, however, map distortion occurs and increases with distance from the point of
tangency. Because directions from the point of tangency are shown correctly on a plane surface,
these projections are also known as azimuthal projections. It is the type of projection surface
usually used to map the polar regions of the planet.
Developable surfaces are projection surfaces which need to be 'unfolded' in order to produce a
plane surface; there are two principal types: a cone and cylinder . The cone is developed by
unrolling from a cut from the base to the apex of the cone; the resulting shape is a plane segment
of a circle. The cylinder is developed by unrolling from a cut along the length of the cylinder; the
resulting shape is a rectangle. The cone and cylinder form a continuum of geometry in the sense
that a cylinder is a zero-angle cone.
Light-source locations for producing true projections commonly are fixed at
(a) The centre of the globe, called the gnomonic position;
(b) The antipode point of tangency of the projection surface, called the stereographic position; and
(c) At a point at infinite distance, called the orthographic position.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

Orientation and tangency of projection surfaces also has an important effect on the character
of the resulting map. It may be useful to think of the various possibilities here as being either

(a) Normal,

(b) Transverse;

(c) oblique

Normal projection (a) Normal orientation and tangency: Refers to those commonly used in
mapping the globe on a particular kind of projection surface. For example, a plane projection
surface tangent to the north or south geographic pole is a standard for mapping the Arctic and
Antarctic, respectively (Figure7.3e). This choice of projection for the polar positions is not only
recommended by its simplicity but it has the advantage of covering the land and ocean areas in
question (within 30o of the polar tangency) with relatively little distortion. Cylindrical projections
normally are employed so that they are tangent to Earth's equator. For this reason, only along the
equator of a cylindrical projection are there no projection distortions.

Thus, cylindrical projections in their normal orientation and tangency are ideally suited to the
mapping of equatorial regions, especially within ±10o strip about the equator, where the distortion

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

of scale is less than +2% (distance between two points on the map parallel is 2% longer than the
corresponding positions on the globe).

The normal projection orientation of a cone is such that its axis and that of the globe's rotation is
coincident. It is tangent to a parallel, higher in latitude as the cone angle increases. Thus the cone
in its normal orientation and tangency is ideally suited to the mapping of middle latitudes. Cones,
cylinders and planes are part of a geometric continuum, reflected in the fact that cones become
planes and cylinders as the cone angle approaches its limits at 180o and 0o, respectively.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

(b) Transverse orientation of the projection surfaces, as this class name suggests, involves a 90o
rotation from the normal orientation. Thus, planes are tangent to the equator, cylinders are tangent
to a meridian of longitude and cones have axes passing through the equator.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

(c) Oblique orientation refers to any intermediate orientation and tangency of the projection
surface between the limiting classes of normal and transverse transformations

Mathematical projections

Not all 'projections' are true optical projections. Many are generated mathematically and the
character of these transformations often are difficult to visualize. The spherical graticule of the
globe is transformed in various ways to satisfy certain criteria.

General properties of projections in relation to the globe

On a sphere, features of the earth's surface - their shape, area, and distance and directions between
them - are shown correctly. The manner and degree to which these properties are distorted can be
evaluated by reference to the spherical graticule forming the parallels of latitude and meridians of
longitude

Properties of spherical grid system

(a) Parallels of latitude measure an angle from Earth's centre in the polar plane, north and south
with respect to the equator (0o); thus the geographic north and south poles are at respective latitudes
90oN and 90o S and the total latitudinal angular sweep from pole to pole is 180o.
(b) Parallels of latitude are indeed parallel, each with all others.
(c) Parallels of latitude are equally spaced; the polar circumference is approximately 40 008 km
so that 1o of latitude corresponds with a spacing of 20 004km 180o or 111.13 km on Earth's surface.
(d) Parallels of latitude are of unequal length. The longest parallel is the equator (40 076 km) and
others decline in length north and south.
(e) Meridians of longitude measure an angle from Earth's centre in the equatorial plane, 180o east
and west with respect to the prime meridian (0o) through Greenwich, England; thus the total
meridianal angular sweep is a full circle of 360o.
(f) All meridians pass through both poles and are equal in length (40008 km).
(g) Meridians are not parallel and meridian spacing varies from 0 km where meridians meet at the
poles to a maximum at the equator. 1o of equatorial longitude corresponds to a spacing of 111.32
km.
(h) Parallels of latitude and meridians of longitude everywhere cross at right angles.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

An ideal map projection retains all these graticule characteristics through the translation to the
map. But the ideal map projection can never be achieved and if a projection is designed to
guarantee a particular characteristic as true, others on it necessarily will be distorted. As a result
map projections will possess certain specific qualities of the globe but never all of them. In
particular, map projections may be evaluated in terms of how well they preserve several global
spatial dimensions: shape, area, distance, and direction.

Conformality is the term used to describe the property of correct shape retention on a map;
such maps are said to be orthomorphic. The importance of conformality is that map features can
be recognized by their distinctive shapes. If shapes are correct then directions must also be correct.
It is necessary that, on all conformal maps, lines of latitude and longitude must cross at right angles
and that the scale must be the same in all directions at any given point, just as is the case on the
globe.
Equivalence refers to the retention of correct relative global areas on a map. Although it is possible
to achieve equivalence in global map, such a projection always creates severe distortions in shape
of the lands involved. To retain equivalence, any scale changes that occur in one direction must be
offset by appropriate changes in the normal direction. Such projections obviously are important
for maps depicting accurate relative areas.
Distance relationships on a map can be correctly depicted only if the length of a straight line
between two points on the map projection represents the great circle distance between the same
two points on the globe. It is only possible to depict correct distances on a map from one, or at
most, two points. The azimuthal equidistant projection achieves this for all lines from the polar
position (but for no others).
Direction is correctly retained on a projection when a straight line drawn between two points on
the map shows the correct azimuth of the line; the azimuth is defined by the angle formed at the
starting point of the straight line. In other words, a projection depicting true direction must show
the great circle routes between points as straight lines. This is one of the properties of a gnomonic
projection. Note that true direction or azimuth does not mean true bearing, an important distinction
we will discuss later in the context of describing Mercator's projection.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

THE MAP PROJECTION PROCESS

The concept of map projections is somewhat more involved than is implied in the previous
discussion. Not one but a series of geometrical transformations is required. The irregular
topography of the earth's surface is first defined relative to a much simpler three-dimensional
Surface, and then the three-dimensional surface is projected onto a plane.
1. The first step is to define the earth's irregular surface topography as elevations above or sea
depths below a more regular surface known as the geoid. This is the surface that would
result if the average level of the world's oceans (mean sea level) were extended under the
continents. It serves as the datum, or starting reference surface, for elevation data on our
maps.
2. The second step is to project the slightly undulating geoid onto the more regular oblate ellipsoid
surface. This new surface serves as the basis for the geodetic control points determined by surveyors
and as the datum for the geodetic latitude and longitude coordinates found on maps . An additional
step taken in making a small-scale flat map or globe is to mathematically transform geodetic
coordinates into geocentric coordinates on a sphere, usually equal in surface area to the ellipsoid.
3. The third step involves projecting the ellipsoidal or spherical surface onto a plane through
the use of map projection equations that transform geographic or spherical coordinates into
planar (x,y) map coordinates. The greatest distortion of the earth's surface geometry occurs
in this step.
MAP PROJECTION PROPERTIES
i. Scale Because of the stretching and shrinking that occurs in the process of transforming the
spherical or ellipsoidal earth surface to a plane, the stated map scale is true only at selected
points or along particular lines called points and lines of tangency
Actual scale—this is the scale that you measure at any point on the map; it will differ from one
location to another. Variation in actual scale is a consequence of the geometrical distortion that
results from flattening the earth.
Principal scale of the map. This is the scale of the generating globe, a globe reduced to
the scale of the desired flat map.

Scale factor (SF), which is defined as the following:

SF = Actual Scale/Principal Scale

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

ii. Completeness
Completeness refers to the ability of map projections to show the entire earth. You'll find the most
obvious distortion of the globe on "world maps" that don't actually show the whole world. Such
incomplete maps occur when the equations used for a map projection can't be applied to the entire
range of latitude and longitude. The Mercator world map is a classic example. Here the y-
coordinate for the north pole is infinity,' so the map usually extends to only the 80th parallel north
and south. Omitting these high latitudes may be acceptable for maps showing political boundaries,
cities, roads, and other cultural features, but maps showing physical phenomena such as average
temperatures, ocean currents, or landforms normally are made on globally complete.

iii. Correspondence relations

You might expect that each point on the earth would be transformed to a corresponding point on
the map projection. Such a point-to-point correspondence would let you shift attention with equal
facility from a feature on the earth to the same feature on the map, and vice versa. Unfortunately,
this desirable property can't be maintained for all points on many world map projections. One or
more points on the earth may be transformed into straight lines or circles on the boundary of the
map projection, most often at the north and south poles. You may notice that the SF in the east
west direction must be infinitely large at the poles on this projection, yet it is 1.0 in the north south
direction.

iv. Continuity
To represent an entire spherical surface on a plane, the continuous spherical surface must be
interrupted at some point or along some line. These breaks in continuity form the map border on a
world projection. Where the mapmaker places the discontinuity is a matter of choice. On some
maps, for example, opposite edges of the map are the same meridian. Since this means features
next to each other on the ground are found at opposite sides of the map, this is a blatant violation
of proximity relations and a source of confusion for map users. Similarly, a map may show the
north and south poles with lines as long as the equator. This means features adjacent to each other,

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

but on opposite sides of the meridian used to define the edge of the map, will be far apart along
the top or bottom edge of the map, while on the earth's surface they are at almost exactly the same
location. Maps of individual continents (except Antarctica) or nations almost always show these
areas without breaks in continuity. To do otherwise would needlessly complicate reading,
analyzing, and interpreting the map.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

Map Projection
A one-to-one relationship between points on the surface of the generating globe
and on the two-dimensional map is established for most of the points. This one-
to-one relationship is never satisfied over all points of the earth since the earth is
a continuous surface and the map is bounded. These points where a one-to-one
correspondence is not retained are called singular points. Examples are when the
poles are represented by lines on the map, some areas may not be shown at all, or
where there are interruptions in the map.
The basic equations to project coordinates on a sphere (φ λ) onto a map (x,y) are:

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

The process of representing a part of the surface of the earth as a picture on the
plane is represented in figure 1 below.

There are many different type of map projections, but ALL introduce some
distortions.
The principal scale can be retained along certain lines or at certain points on the
map. These are then known as lines or points of zero distortion. If one visualises
the projection surface as a piece of paper which touches the generating globe at a
certain point or along certain lines, then these lines or points where the two
surfaces touch will be distortionfree.
Map Projections: Fundamental Development Surfaces and Lines/Points of Zero
Distortion

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

Point of Zero
Distortion

Possible Lines of Zero

Distortion

Plane or Gnomic Projection

L ine of Zero
Distortion

Cylindrical Projection

L ine of Z ero
Distortion

Conica l Projec tion

GAUSSIAN DISTORTION THEORY OF MAP PROJECTIONS

Objective is to determine scale and angular distortion on our map projection.
One of the most important criteria/principle to consider when
choosing a suitable map projection is to minimize the relevant

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

distortions (e.g. scale, angle or area) according to the intended use of

the map.
In order to do this one needs to consider how these deformations are distributed
over the map area. The mathematical tools of map projections provide us with the
necessary tools to study the deformation characteristics of a projection in a general
sense. The fundamentals of the theory were first laid down by Gauss. This was later
extended by the French mathematician, Tissot. Ultimately what we need is to
examine a small unit circle on the generating globe and examine the
shape of a sample of such circles at various points projected onto the
plane. These circles tend to be projected in the shape of an ellipse. We
are going to use Gauss’s theory as a means to achieve this.

Key elements of our ellipse of distortion are:

i. Gaussian fundamental quantities;
ii. scale distortion on meridians and parallels, maximum and minimum
scale at a point;
iii. directions of lines on a globe as projected on a plane,
iv. the azimuth and directions of maximum and minimum scale (to find
the orientation of our ellipse of distortion) – the principal directions.

We require a simple set of equations and algorithms that we can use to design and
analyse the different qualities of particular map projections. In this section we
formulate the Gaussian fundamental quantities. We use these to establish the scale
distortion along the meridians and parallels, the magnitude, azimuth and plane
direction of maximum and minimum scale distortion at a point and the angular
distortion at a point on the globe.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

Gauss’s theory applies to the projection of any curved surface on another curved
surface. The following diagram shows the line element AC=dS on the surface of the
generating globe.

If , at A, at a given latitude Φ and longitude λ, then the latitude and longitude at C

is given by Φ + d Φ and λ + λ l . The intersection of the meridians and parallels
which pass through A and C form a quadrilateral. The line dS forms an angle θm
with the meridian, the azimuth of dS, and an angle θp with the parallel. As parallels
and meridians intersect at right angles (are orthogonal) the sum of θm and θp is
90°. The linear difference along the parallel through point A is called dSp , and the
linear difference in the meridian through A is called dSm.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

First Gaussian Fundamental Quantities

These are the fundamental elements required to determine the magnitudes of
distortion parameters. As we shall see, these are used throughout our discussion
on distortions in map projections and it is necessary to be able to derive these.
On the Globe…
The length of the meridian arc is dSm =RdΦ
The length of a parallel circle arc is dSp = RcosΦ dλ
The first assumption is that the quadrilateral is infinitely small. This allows the use
of an approximation by Pythagoras’ theorem:

If this quadrilateral on the sphere is mapped onto a plane it will be distorted.

Generally the direction and length of the lines as well as the direction
between them will be distorted. The points A and C on the sphere are mapped
to A’ and C’ on the plane respectively. The lengths of the parallel and meridian
elements are denoted by small symbols: ds. The angle of intersection of the
quadrilateral sides at A is given by θ’ (which is the distortion of the sum of θ m and
θ p is 90°).

Because an assumption that the quadrilateral is infinitely small has been made, the
mapped sides and diagonals can also be considered to be straight lines.
Figure 3: Quadrilateral Projected onto the Plane

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

dx = length along the x axis from A’ to C’

dy = length along the y axis from A’ to C’
ds = length from A’ to C’
Pythagoras ‘rule applies again: ds2 = dx2 + dy2

Now x = f(φ,λ) and y = g(φ,λ) by definition. We want to express x and y in terms of

φ and λ to relate objects on the globe to the shape of the same object on the plane.
On the projected plane....more detailed view
need ds in terms of lat. and long to get scale along meridian, parallel and along ds.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

Differentiating, we get:

(The objective is to write our line increment ds in terms of latitude and longitude
increments, dφ and dλ)

Substituting:

E, F, and G are
known as the First Gaussian Fundamental Quantities. They are used to
study and evaluate the various distortions that take place in a map projection.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

SCALE DISTORTION

By what ratio are lines on the globe enlarged or contracted on the plane? To determine
this we need to determine the scale distortion along the meridians and parallels first, then
we can use these quantities to determine the scale µ along any arc on the globe. We then
use this quantity to determine the magnitudes and plane directions of the maximum and
minimum scale distortions at a particular point. This provides us with the distortion
characteristics of our map projection.

Scale Distortion along the Meridian ( µφ )

Referring to the previous figures, an infinitely small quadrilateral projected from the
generating globe to the plane.

A′B′
µφ =
AB

( A′B′)2 = (B′P′)2 + ( A′P′)2

(B′P′) = ∂x dφ AB
∂φ

( A′P′) = ∂y dφ AB
∂φ

2 2
( A′B′)2 =  ∂x dφ AB   ∂y
+ 

dφ AB 
 ∂φ   ∂φ 
(orthogonal components)
( A′B′)2 = Edφ AB 2

A′B′ = E dφ AB

and

AB = dS m

therefore

E dφ AB
µφ =
dS m

E dφ AB E
µφ = =
Rdφ R

Page 8/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

Scale Distortion along the Parallel Circle ( µλ )

A′D′
µλ = On the sphere: AD = dSp (see previous figures)
AD

( A′D′)2 = (S ′D′)2 + ( A′S ′)2

(S ′D′) =  ∂y dλ AD 
 ∂λ 

( A′S ′) =  ∂x dλ AD 
 ∂λ 

2 2
( A′D′)2 =  ∂x dλ AD   ∂y
+

dλ AD 
 ∂λ   ∂λ 

( A′D′)2 = Gdλ AD 2

A′D′ = G dλ AD

and

AD = dS p

therefore

G dλ AD G dλ AD G
µλ = = =
dS p R cos φdλ AD R cos φ

Page 9/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

Scale Distortion along an arbitrary arc through point A ( µ )

A′C ′ ds
µ= =
AC dS

dS = (R2dφ 2 + R 2 cos2 φdλ2 )

ds = (Edφ 2 + 2Fdφdλ + Gdλ2 )
Now, if R=1 for a unit spheroid (generating globe has a normalized unit radius):

µ=
ds
=
(Edφ 2 + 2Fdφdλ + Gdλ2 )
dS (dφ 2 + cos2 φdλ2 )
If, at a certain point on the map, µ = 1, then the true scale S in that point equals the
nominal scale ( µ 0 ). If µ ≠ 1 then the projection has a distortion in the line element and
the true scale is given by: S = µ 0 . µ

This principle is very important to understand - the scale indicated on a map is the
nominal scale, and is only applicable at those points, or along those lines where µ = 1 .
Generally, scale is not preserved over the entire map projection, but varies from point to
point and is often different in all directions at each point.

These three particular scale factors ( µφ , µλ , µ ) relate an infinitesimal linear distance on

the map projection with the corresponding linear distance on the globe.

Page 10/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

MAXIMUM AND MINIMUM PARTICULAR SCALES

Objective: Our first objective is to find the azimuths of maximum and minimum scale
distortion at a point and the plane directions of these distortions on the projected mapping
plane. We can use these to determine the distortion characteristics of a particular set of
map projection equations by plotting ellipses of distortion at a select number of points on
the globe. The plane directions give us the orientation of these ellipses and the maximum
and minimum scale factors give us the magnitude of the scale distortion.

Tissot’s theorem is stated as follows:

Whatever the system of projection there are at every point on one of the surfaces
orthogonal directions which on the projection plane the vectors which correspond to them
also intersect one another at right angles. If angles are not preserved in a particular
projection there are only two such directions.

Our first objective is to find the azimuths of the maximum and minimum scales at a
particular point, and then compute the plane directions of these maximum and minimum
scales on the plane. This gives us the means to plot our ellipses of distortion at a selected
sample of points on the mapping plane.

Since the scale distortion varies in different directions, it is interesting to determine in

which directions it reaches its maximum and minimum at a particular point on the map.
These directions are the axes of our ellipse of distortion. We now need to derive formulae
to determine these.

The azimuth angle, A of a line AC (which is equal to angle θ m on our initial sketch of the
quadrilateral on the generating globe in figure 2) is given by:

Page 11/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

R cos φdλ cos φdλ

tan A = =
Rdφ dφ

cos φdλ
dφ =
tan A

tan Adφ
dλ =
cos φ

Substituting these into the equation for scale distortion:

ds (Edφ 2 + 2Fdφdλ + Gdλ2 )

µ=
dS
=
(R 2dφ 2 + R2 cos2 φdλ2 )
However, this last equation is not useful as we do not know the values of dφ and dλ .
We need to eliminate these from our equation. To do this, we express scale in terms of
azimuth and the Gaussian fundamental quantities and differentiate scale with respect to
azimuth, A, to obtain our azimuth of maximum and minimum distortion.

Let’s ignore the square root for the meantime. Taking each term separately and
dφ 2
computing (note that as a trick we multiply the bottom term by repeatedly in all
dφ 2
cases);

First Term:

Edφ 2 E  dφ 2 

(R2dφ 2 + R 2 cos2 φdλ2 ) =
(
R 2  dφ 2 1 + cos 2 φdλ2 / dφ 2 ) 


E  1  E 
=  1  E  1 
=
( ) ( )
=  
R 2  1 + cos 2 φdλ2 / dφ 2  R 2  1 + tan 2 A  R 2  sec2 A 
  
E
= cos 2 A
2
R

Note Sec 2 A = 1 + tan 2 A is a standard trigonometrical relationship. To recap, cos φdλ is

the length increment of our infinitesimally small quadrilateral along the parallel on the
cos φdλ
sphere (the radius R cancels out!). And so tan A = . We will use these relationships
dφ
again in manipulating the second term:

Page 12/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

Second Term:

2 Fdφdλ 2 F  dφdλ 

( 2 2 2
R dφ + R cos φdλ 2 2 )
2 2
=
2
R  dφ + cos φdλ  2


2 F  dφdφ tan A / cos φ  2F  tan A 
=  
=
2 2 ( 2 2 2 2 )
R  dφ 1 + cos φdλ / dφ  R  cos φ 1 + cos φdλ / dφ 
   2 2 2 ( )
2 F  tan A  2 F  tan A  2F
=   =   = sin A cos A
R 2 cos φ  1 + tan 2 A  R 2 cos φ  sec2 A  R 2 cos φ

Third Term:

Gdλ2 G  dλ2 

( 2 2 2
R dφ + R cos φdλ 2 2 ) =
2 2 2
R  dφ + cos φdλ  2

G   2
tan Adφ 2  G  tan 2 A 
=  
=
( )
R 2  cos 2 φdφ 2 1 + cos 2 φdλ2 / dφ 2  R 2 cos 2 φ  1 + tan 2 A ( ) 

 tan 2 A 
=
G  = G
sin 2 A
2 2  2 
R cos φ  sec A  R cos 2 φ 2

Thus, these terms can now be recombined to find an expression for µ in terms of
azimuth A:

E 2F G
µ2 = cos2 A + sin A cos A + sin 2 A
R2 R 2 cos φ R 2 cos2 φ

2F
To simplify matters, let p = ; and remembering the expressions for scale along
2
R cosφ
the meridian, µφ , and scale along the parallel, µλ , the above expression can be rewritten
as:

µ 2 = µφ2 cos 2 A + p sin A cos A + µλ2 sin 2 A

We need to find the maximum and minimum scales to plot our ellipse of distortion. Now,
to find the extremes of scale at a point, the following expression for the extremum must
hold true:

d µ2( )=0 (we assume direction of max/min of µ2 is same as for µ)

dA

Page 13/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

So, differentiating with respect to azimuth A:

First Term:

(
d µφ2 cos2 A ) = −2µ 2 sin A cos A
φ
dA

Second Term: (Note1)

d ( p sin A cos A)
dA
(
= p cos 2 A − p sin 2 A = p cos 2 A − sin 2 A = p cos 2 A )
Third Term: (note2)

(
d µλ2 sin 2 A )
= 2 µλ2 sin A cos A
dA

Combining these terms and equating to 0:

0 = −2µφ2 sin A cos A + 2 µλ2 sin A cos A + p cos 2 A

( )
0 = −2 sin A cos A µφ2 − µλ2 + p cos 2 A

( )
0 = − sin 2 A µφ2 − µλ2 + p cos 2 A

p
tan 2 AM =
(µφ2 − µλ2 )
This indicates that there are two orthogonal azimuths A M and A M − 90 o where
maximum and minimum scale distortion occur. These directions are called the principal
directions (or more correctly, the principal azimuths as they are directions on the globe).

Plug the values of A M and A M − 90 o into µ 2 = µφ2 cos 2 A + p sin A cos A + µλ2 sin 2 A
to get maximum and minimum scales.

Let maximum distortion = a Let minimum distortion = b

1 d dv du
Rule of differentiation: (uv) = u +v
dx dx dx
2 d n du
Rule of differentiation: (u ) = nu n −1
dx dx

Page 14/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

DIRECTIONS ON THE PROJECTED MAP AT A POINT P ON THE GLOBE

Objectives: We need the plane direction on the map of an infinitesimal segment on the
globe. We can then determine the principal directions of our ellipse of distortion on the
plane. Furthermore we are interested in the angle between the meridians and the parallels
on the projected plane. This gives us an idea of angular distortion.

A straight line on the globe will often be projected as a curve on the map. The direction at
a point P on the globe is given by the azimuth or by the angles θ m ;θ p in figure 1. On the
plane projection the direction of the curves (e.g. meridians and parallels) at the point, P’,
are given by the directions αm and αp of the tangents at P’.

First objective: If we have an angle θ=90o between the meridian and parallel on the
globe, what is the distortion in this angle at a point on the plane given by the plane angle
θ’? (Note to sketch below3.)

Me
r id
ia n
Me
rid
Y i an

θ’

P’

l
lle
ra
Pa

Parallel

n
tio
ir ec
ed
an
Pl

αp αm
X

Angle of Intersection θ’ between parallels and meridians on mapping plane

3
Plane coordinate systems: Remember that in most countries directions α are measured
from the x-axis toward the y axis in both left and right handed coordinate systems. In a
right handed systems as the one above, directions are measured from the x axis
increasing anti-clockwise. In our normal geodetic (LH system) they increase clockwise).

Page 15/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

y
x xd
d
Q’ C'
yd
x
d
B'
P'
y
d
dy
ds
y

α
d x
d
αm D' R’
' yd yd

αp x
A'
S'
xd

dx
Directions α On The Plane of a Small Increment ds
(projected from dS on the Sphere)

In the figure above for any line increment A’C’

 ∂y ∂y 
 dφ + dλ 
dy  ∂φ ∂λ 
tan α = =
dx  ∂x ∂x 
 dφ + dλ 
 ∂φ ∂λ 

Substituting the equations for dφ and dλ expressed in terms of azimuth determined

earlier into the equation for tanα :

 ∂y cos φdλ ∂y tan Adφ 

 + 
∂φ ∂λ cos φ 
tan α = 
tan A
 ∂x cos φdλ ∂x tan Adφ 
 + 
 ∂φ tan A ∂λ cos φ 

Remember azimuth of A’C’ projected onto the plane is the angle B’A’C’.

dλ
Taking out a common term :
sin A

Page 16/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

 ∂y 2 
 cos A cos φ + ∂y sin Adφ 
 ∂φ ∂λ cos A cos φdλ 
tan α =  
 ∂x 2 
 cos A cos φ + ∂x sin Adφ 
 ∂φ ∂λ cos A cos φdλ 
 

but :

sin 2 Adφ sin A sin Adφ tan A sin Adφ dλ dφ

= = = sin A = sin A
cos A cos φdλ cos A cos φdλ cos φdλ dφ dλ

 ∂y ∂y 
 cos A cos φ + sin A 
∂φ ∂λ
tan α =  
 ∂x ∂x 
 cos A cos φ + sin A 
 ∂φ ∂λ 

Remember from our earlier discussion of the Gaussian fundamental Quantities that we
∂y ∂y ∂x ∂x
can determine expressions for , , , and calculate values for these terms for
∂φ ∂λ ∂φ ∂λ
particular values of φ and ?.

Angular Distortion: Angle between Meridian and Parallels

We have the direction on the plane of any line segment of azimuth A on the globe. We
now want the angle between the meridian and the parallel expressed in terms of the
Gaussian fundamental quantities and the scale distortion along the meridians and the
parallels.

Now, the new directions of the meridian and parallel can be found by setting A=0° and
A=90° respectively:

∂y / ∂φ
tan α m =
∂x / ∂φ

∂y / ∂λ
tan α p =
∂x / ∂λ

We want the magnitude of θ’ without having to calculate the directions of the meridian
and parallel projected onto the plane. So, the angle of intersection between the projected
meridians and parallels θ’can be calculated from:

Page 17/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

(
tan θ ′ = tan α m − α p )
tan α m − tan α p
tan θ ′ =
1 + tan α m tan α p

∂y / ∂φ ∂y / ∂λ
−
∂x / ∂φ ∂x / ∂λ
tan θ ′ =
∂y / ∂φ ∂y / ∂λ
1+
∂x / ∂φ ∂x / ∂λ

But now from the First Gaussian Fundamental Quantities:

 ∂x  2  ∂y  2   ∂x 2
 ∂y  
2
   
EG =   +     +  
 ∂φ   ∂φ    ∂λ   ∂λ  
 

2 2 2 2 2 2 2 2
 ∂x   ∂x   ∂x   ∂y   ∂y   ∂x   ∂y   ∂y 
EG =     +     +     +    
 ∂φ   ∂λ   ∂φ   ∂λ   ∂φ   ∂λ   ∂φ   ∂λ 

 ∂x ∂x   ∂y ∂y   ∂x ∂x   ∂y ∂y  
F 2 =   +     +   
 ∂φ ∂λ   ∂φ ∂λ   ∂φ ∂λ   ∂φ ∂λ  

2 2 2 2
 ∂x   ∂x  ∂x ∂x ∂y ∂y  ∂y   ∂y 
F 2 =     + 2 +   
 ∂φ   ∂λ  ∂φ ∂λ ∂φ ∂λ  ∂φ   ∂λ 

2 2 2 2
 ∂x   ∂y   ∂y   ∂x  ∂x ∂x ∂y ∂y
EG − F 2 =     +     − 2
∂
   
φ ∂λ ∂
   
φ ∂λ ∂φ ∂λ ∂φ ∂λ

 ∂x ∂y ∂x ∂y 
EG − F 2 =  − 
 ∂λ ∂φ ∂φ ∂λ 

So now:

EG − F 2
tan θ ′ =
F

EG − F 2
It can similarly be proved that sin θ ′ =
EG

Solving for cosθ ′ :

Page 18/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

sin 2 θ ′ EG − F 2
=
cos2 θ ′ F2
F 2 sin 2 θ ′
cos2 θ ′ =
EG − F 2

cos θ ′ =
2 (
F 2 EG − F 2 )
(
EG − F 2 EG )
F2
cos2 θ ′ =
EG

F F
cosθ ′ = =
EG µφ µλ cos φ

Thus the angle made at A, at a latitude of φ , by the graticule intersection4 of the meridian
and the parallel can be expressed in terms of a First Gaussian Fundamental Quantities and
the scale along the meridian and along the parallel.

Conformal Projections

If F is set to 0 then θ ′ = 90 o , and it is evident that the projected meridians and parallels
also intersect at right angles (although they may still be projected as curves). Further, if F
= 0 then, from the equation for directions of the maximum and minimum scale
distortions:

2F
p R 2 cos φ
tan 2 AM = =0=
µφ2 − µλ2 E2 G
− cos 2 φ
R2 R2

2 AM = arctan(0) = 0o / 180o / 360o

AM = 0o / 90o / 180o.....

Thus the principal directions are along the meridians and parallels. Scale distortion in a
conformal projection is the same in all directions i.e. It is independent of azimuth. In this
E G
case F = 0 and 2 = 2 . In this case µφ = µλ = µ . Length distortion is the same
R R cos2 φ
in all directions and therefore there is no angular distortion (hence the name conformal).

4
By definition a graticule is the position of the intersection of a meridian and a parallel
plotted on a map.

Page 19/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

Equal-Area Projections

In an equal area (equivalent) projection, elementary areas are preserved. An elementary

surface on the generating globe is given by:

dS m dS p = R 2 cos φdφdλ

The same area of the parallelogram5 on the map is given by: dsm ds p sin θ ′ .

Area distortion factor σ

dsm ds p sin θ ′
The surface distortion is thus given by6: σ= = µφ µλ sin θ ′
dS m dS p
which, as we shall see later, can also be expressed as: σ = ab

EG − F 2
And we know that sin θ ′ = ;
EG

and remembering the formulae for µφ and µλ :7

EG − F 2
Area scale σ= .
R 2 cos φ

This parameter σ is defined in the same units as the particular scales; therefore it is
known as the area scale.

Setting this equal to 1 for an equal area projection: EG − F 2 = R 2 cosφ .

5
6
see previous footnote for parallelogram area expression.
E G
7
µφ = and µ λ =
R R cos φ

Page 20/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

TISSOT’S INDICATRIX

Tissot’s theory of distortions states that

A circle on the datum surface with a centre P and a radius ds may be assumed
to be a plane figure within its infinitely small area. This area will remain
infinitely small and plane on the projection surface. Generally the circle will be
portrayed as a ellipse.

This ellipse is called Tissot’s Indicatrix as it indicates the characteristics of a projection in

the direct environment of a point.

The axes of Tissot’s Indicatrix correspond to the two principal directions and the
maximum and minimum particular scales, a and b, at any point, occur in these directions.

Proof That The Projected Circle Is An Ellipse

Notes: In figure, the X axis is directed east-west; Y axis is directed north-south.

Remember that capital letters denote elements on the generating globe, and small letters
elements on the projection.

Y
y

dS ds
X ' x

Sphere Plane

dY = dS .sin θ = G . dλ 
on. the. generating. globe
dX = dS .cosθ = E . dφ 
dy = ds.sin θ ′ = g . dλ 
on. the. projection
dx = ds.cosθ ′ = e . dφ 
1
dφ = dS .cosθ
E
1
dλ = dS .sin θ
G
1
∴ dy = g . dS .sin θ
G
1
and : dx = e . dS .cosθ
E

Page 21/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

dx 2 dy 2
= dS 2 .cos2 θ ;.. = dS 2 .sin 2 θ
E /e G/g
dx 2
+
dy 2
E /e G/g
(
= sin 2 θ + cos2 θ dS 2)
dx 2 dy 2
+ = dS 2
E /e G/g

If dS=1 then the elementary circle on the globe has a radius of 1

(remember that capital letters denote elements on the generating globe, and small letters
elements on the projection.)

dx 2 dy 2
+ =1 This is an equation of an ellipse.
E /e G/g

Analysis of Deformation Characteristics using Tissot’s Indicatrix

If we call the semi-major and semi-minor axes of the ellipse a, and b, then these are the
directions of maximum and minimum distortion i.e. the principal directions. a and b are
also thus called the principal scale factors.

x2 y2
+ =1
b a

For convenience we will consider the plane x and y axes to be in the principal directions.

Length Distortion

µ x = ds cosθ ′ on the plane

µ X = dS cosθ = 1 on globe
(There is no distortion on the globe)

ds
Remember from previous sections: µ = , so:
dS
 ds  µ cosθ ′
a=  =
 dS  x cos θ
 
ds µ sin θ ′
b=  =
 dS  y sin θ

Or:

Page 22/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

a cos θ = µ cosθ ′
b sin θ = µ sin θ ′
a 2 cos2 θ = µ 2 cos 2 θ ′
b 2 sin 2 θ = µ 2 sin 2 θ ′

(
a 2 cos2 θ + b 2 sin 2 θ = µ 2 cos 2 θ ′ + sin 2 θ ′ )
µ 2 = a 2 cos 2 θ + b 2 sin 2 θ

This formula expresses the length distortion in any direction as a function of the original
direction θ , and the principal scale factors, a and b.

The angle θ indicates the direction of the parallel with respect to the x axis. The direction
of the meridian with respect to the x axis is thus

θ + 90o = θ + π / 2 = β

The scale distortions along the parallels and meridians (note: not necessarily equal to the
maximum and minimum distortions along a and b) are thus:

µλ2 = a 2 sin 2 β + b 2 cos 2 β

µφ2 = a 2 sin 2 α + b 2 cos 2 α = a 2 cos 2 β + b 2 sin 2 β

µλ2 + µφ2 = a 2 + b 2

This is known as the First Theorem of Appolonius:

The sum of the squares of the two conjugate diameters of an ellipse is constant.

Angular Distortion 2 Ω

Without derivation:
a −b
2Ω = 2 arcsin , where 2 Ω is the maximum angular distortion. The maximum
a +b
angular deformation occurs in each of the four quadrants.

Page 23/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

x
Unit circle on generating globe
superimposed on indicatrix, showing
maximum change in direction Ω.

Figure: Maximum Angular Distortion

If 2 Ω = 0 then no angular distortion occurs and the projection is called conformal. The
property of a conformal projection is that a = b and Tissot’s Indicatrix is a circle with
equal scale distortion in all directions. This is consistent with the previously derived
π
conditions for conformality, namely that µφ = µλ , and θ ′ = . The area is not preserved
2
and the projected circle increases in size as one moves away from the line of zero
distortion.

Areal Distortion ( σ ): Second Theorem of Appolonius.

This is found by dividing the projected area by the area of the circle on the globe (radius
=1):
πab
σ= = ab
πR 2

When looking at equal area projections earlier, it was found that:

σ = µφ µλ sin θ ′ , thus µφ µλ sin θ ′ = ab .

This is called the Second Theorem of Appolonius. When ab = 1 then the projection is
equal-area or equivalent.

Note: conformality and equivalence are exclusive: ab = 1 and a = b cannot occur at the
same time.

Page 24/34 ENGO 421 - Map Projections

Downloaded by Francis Mbae ([email protected])
 lOMoARcPSD|43072212

48 Chapter 3 MAP PROJECTIONS

COMMONLY USED MAP

PROJECTIONS

As stated above, one way that map projections are

commonly grouped into families is based on the projec-
tion surface—a plane, cone, or cylinder. We'll use these
surfaces to structure our discussion of commonly used
map projections.

Planar projections

Orthographic
The orthographic projection is how the earth would
appear if viewedfrom a distant planet (figure 3.13). Since
the light source is at an infinite distance from the gener-
ating globe, all rays are parallel. This projection appears
to have been first used by astronomers in ancient Egypt, Figure 3.13 The orthographic projection best shows the
but it came into widespread use during World War II spherical shape of the earth.
with the advent of the global perspective provided by the
air age. It is even more popular in today's space age, often
used to show land-cover and topography data obtained
from remote sensing devices. The generating globe and
half-globe illustrations in this book are orthographic
projections, as is the map on the front cover of the book.
The main drawback of the orthographic projection is that
only a single hemisphere can be projected. Showing the
entire earth requires two hemispherical maps. Northern
and southern hemisphere maps are commonly made, but
you may also see western and eastern hemisphere maps.

Stereographic
Projecting a light source from the antipodal point on
the generating globe to the point of tangency creates the
stereographic projection (figure 3.14). This is a confor-
mal projection, so shape is preserved in small areas. The
Greek scholar Hipparchus is credited with inventing
this projection in the second century BC. It is now most
commonly used in its polar aspect and secant case for Figure 3.14 Polar stereographic projection of the northern
maps of polar areas. It is the projection surface used for hemisphere. Since this is a conformal projection, tiny circles
the Universal Polar Stereographic grid system for polar on the generating globe are projected as circles of the same
areas, as we will see in the next chapter. A disadvantage size at the point of tangency to four times as large at the
of the stereographic: conformal projection is that it is gen- equator.
erally restricted to one hemisphere. If it is not restricted
to eine hemisphere, then the distortion near the edges
increases to such a degree that the geographic features
in these areas are basically unrecognizable. In past cen-
turies, it was used for atlas maps of the western or eastern
hemisphere.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

COMMONLY USED MAP PROJECTIONS 49

Gnomonic Azimuthal equidistant

Projecting with a light source at the center of the The azimuthal equidistant projection in its polar aspect
generating globe to a tangent plane produces the gno- has the distinctive appearance of a dart board—equally
monic projection. One of the earliest map projections, spaced parallels and straight-line meridians radiating
the gnomonic projection was first used by the Greek outward from the pole (figure 3.16). This arrangement of
scholar Males of Miletus in the sixth century BC for parallels and meridians results in all straight lines drawn
showing different constellations on star charts, which from the point of tangency being great circle routes.
are used to plot planetary positions throughout the year. Equally spaced parallels mean that great circle distances
The position of constellations in the sky over the year are correct along these straight lines. The ancient Egyp-
was used as a calendar, telling farmers when to plant and tians apparently first used this projection for star charts,
harvest crops, and when floods would occur. Horoscopes but during the air age it also became popular for use
and astrology also began with the ancient Greeks over by pilots planning long-distance air routes. In the days
two thousand years ago. Many believed that the position before electronic navigation, the flight planning room
of the sun and the planets had an effect on a person's life in major airports had a wall map of the world that used
and that future events in their lives could be predicted an oblique aspect azimuthal equidistant projection cen-
based upon the location of celestial bodies in the sky. tered on the airport. You will also find them in the public
The gnomonic projection is the only projection with areas of some airports. All straight lines drawn from the
the useful property that all great circles on the globe are airport are correctly scaled great circle routes. This is one
shown as straight lines on the map (figure 3.15). Since of the few planar projections that can show the entire
a great circle route is the shortest distance between two surface of the earth.
points on the earth's surface, the gnomonic pro jection is
especially valuable as an aid to navigation (see great circle
directions in chapter 12). The gnomonic projection is also
used for plotting the global dispersal of seismic and radio
waves. Its major disadvantages are increasing distortion
of shape and area outward from the center point and the
inability to project a complete hemisphere.

Figure 3.16 Polar aspect azimuthal equidistant world

map projection tangent at the pole. Great circle distances
are correct along straight lines outward from the point
of tangency at the north pole since the north-south SF
is always 1. The east-west SF increases to a maximum of
infinity at the south pole, where there is a point-to-line
correspondence.
Figure 3.15 Polar gnomonic projection of the northern
hemisphere from 15°N to the pole. All straight lines on the
projection surface are great circle routes. Note the severe
shape distortion of this projection compared to the polar
stereographic projection shown in figure 3.13.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

50 Chapter 3 MAP PROJECTIONS

Lambert azimuthal equal area You may see world maps showing elevation data or
In 1772 the mathematician and cartographer Johann satellite imagery on this projection. This choice of pro-
Heinrich Lambert published equations for the tan- jection is not based on any geometrical advantages, but
gent case planar Lambert azimuthal equal area pro- rather on the simplicity of creating flat world maps when
jection, which, along with other projections he devised, they had to be done by hand.
carries his name. This planar equal area pro jection is
usually restricted to a hemisphere, with polar and equa-
torial aspects used most often in commercial atlases
(figure 3.17). More recently, this projection has been
used for statistical maps of continents and countries that
are basically circular in overall extent, such as Australia,
North America, and Africa. You will also see the oceans
shown on maps that use the equatorial or oblique aspects
of this projection. The Lambert azimuthal equal area pro-
jection is particularly well suited for maps of the Pacific
Ocean, which is almost hemispheric in extent.

Cylindrical projections

Equirectangular
The equirectangular projection is also called the
equidistant cylindrical or geographic projection. This
simple map projection, nearly 2,000 years old, is attrib-
uted to Marinus of Tyre, who is thought to have con-
structed the projection about in 100 AD. Parallels and
meridians are mapped as a grid of equally spaced horizon-
tal and vertical lines twice as wide as high (figure 3.18). Figure 3.17 The Lambert azimuthal equal area projection is
The equal spacing of parallels means that the projection is often used for maps of continents that have approximately
equidistant in the north–south direction with a constant equal east-west and north-south extents. This map of
SF of 1.0. In the east–west direction the SF increases North America in the box is part of an oblique aspect of the
steadily from a value of 1.0 at the equator to infinity at projection centered at 45°N, 100°W.
each pole, which is projected as a straight line.

— SF
-2 0
-. , 111110 4W 41110

1 ..,.
: 0 411
,
Q
.., •
..
u) 0 .• •

1 5
IlIln /Ilk Ii
m
VW VW
------EP----
r

20
..... ...
,

Figure 3.18 Equirectangular world map projection.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

COMMONLY USED MAP PROJECTIONS 51

Mercator course on the map and simply maintain the associated

Constructed by Gerhardus Mercator in 1569, the bearing during passage to arrive at the plotted location.
Mercator projection is a tangent case cylindrical con- You can see how navigators would prefer a map on which
formal projection. As with all conformal projections, compass bearings would appear as straight rhumb lines
shape is preserved in small areas. This projection offers a (see chapter 13 for more on rhumb line plotting). The
classic example of how a single projection can be used both Mercator projection has been used ever since for nautical
poorly and well. Looking at the projection (figure 3.19), charts, such as small-scale piloting charts of the oceans.
we can imagine Mercator started the construction of his The large-scale nautical charts used for coastal navigation
projection with a horizontal line to represent the equator can be thought of as small rectangles cut out of a world
and then added equally spaced vertical lines to represent map that uses the Mercator projection.
the meridians. Mercator knew that meridians on the Of course, using these maps for navigation is not really
globe converge toward the poles, so the meridians he had as simple as this. Recall that the gnomonic projection is
drawn as parallel vertical lines must become progressively the only projection on which all great circles on the gen-
more widely spaced toward the poles than they would be erating globe are shown as straight lines on the flat map.
on the generating globe. He progressively increased the Lines drawn on a Mercator projection show constant
spacing of parallels away from the equator so that the compass bearing, but they are not the same as the great
increase matched the increased spacing between the circle route, which is the shortest distance between two
meridians. As a result of this extreme distortion toward points on a globe. Therefore, the Mercator projection is
the poles, he cut his projection off at 80°N and S. This often used in conjunction with the gnomonic projection
not only produced a conformal map projection, but also to plot navigational routes. The gnomonic projection is
the only projection on which all lines of constant com- used first to determine the great circle route between two
pass direction, called rhumb lines, are straight lines on points, and then the route is projected to transform the
the map. great circle to a curve on a Mercator projection. Finally,
Navigators who used a magnetic compass immediately the curve is translated into a series of shorter straight
saw the advantage of plotting courses on the Mercator line segments representing portions of the route, each
projection, since any straight line they drew would be a with constant compass direction (see figure 12.28 in
line of constant bearing. This meant they could plot a chapter 12 for an illustration of this process).

Figure 3.19 Mercator

projection shown with
rhumb lines between
selected major world
cities.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

52 Chapter 3 MAP PROJECTIONS

The use of the Mercator projection in navigation is an The projection generated intense debate because
example of a projection used for its best purpose. A poor of Peters's assertion that this was the only "nonracist"
use of the Mercator projection is for wall maps of the world map. Peters claimed that his map showed devel-
world. We saw earlier that this projection cannot cover oping countries more fairly than the Mercator projec-
the entire earth, and is often cut off . at 80°N and S. Cut- tion, which distorts and dramatically enlarges the size of
ting off part of the world does create a rectangular projec- Eurasian and North American countries. His assertion
tion surface with a height-to-width ratio that fits walls was a bit of a straw dog, since the Mercator projection was
very well. The problem, of course, is the extreme scale designed and admirably suited for navigation and never
enlargement and consequent area distortion at higher lat- intended for comparing country sizes.
itudes. The area exaggeration of North America, Europe, The first English version of the Gall-Peters projection
and Russia gives many people an erroneous impression of map was published in 1983, and it continues to have
the size of the land masses. passionate fans as well as staunch critics.
Although the relative areas of land masses are
Gall-Peters maintained, their shapes are distorted. According to
The Gall-Peters projection is a variation of the prominent cartographer Arthur Robinson, the Gall
cylindrical equal area projection. Its equations were Peters map is "somewhat reminiscent of wet, ragged long
published in 1885 by Scottish clergyman James Gall winter underwear hung out to dry on the Arctic Circle."
as a secant case of the cylindrical equal area projection Although several international organizations have
that lessens shape distortion in higher latitudes by plac- adopted the Gall-Peters projection, there are other equal
ing lines of tangency at 45°N and 45°S. Arno Peters, a area world projections, such as the Mollweide projection
German historian and journalist, devised a map based (figure 3.24), that distort the shapes or land masses far
on Gall's projection in 1967 and presented it in 1973 as less. Maps based on the Gall-Peters iprojection continue
a "new invention" superior to the Mercator world projec- to be published and are readily available, though few
tion (figure 3.20). major map publishers use the projection today.

Figure 3.20 Gall-Peters projection world map.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

COMMONLY USED MAP PROJECTIONS 53

Transverse Mercator
We saw earlier that Lambert constructed his azimuthal
equal area projection in 1772. That same year, he also
constructed the transverse Mercator projection, along
with the Lambert azimuthal equal area projection and
another described later in this chapter. In Europe this
projection is called the Gauss-Kruger, in honor of the
mathematicians Carl Gauss and Johann Kruger who
later worked out formulas describing its geometric dis-
tortion and equations for making it on the ellipsoid.
Lambert's idea for the transverse Mercator projection
was to rotate the Mercator projection by 90 0 so that the
line of tangency became a pair of meridians—that is, any
selected meridian and its antipodal meridian (figure 3.12,
bottom right). The resulting projection is conformal, as
is the Mercator projection, but rhumb lines no longer are
straight lines. Along the central meridian of the projec-
tion (the vertical meridian that defines the y-axis of the
projection), the SF is 1.0, and the scale increases perpen-
dicularly away from the central meridian. Thus, narrow
north–south strips of the earth are projected with no
local shape distortion and little distortion of area.
You're likely to see the transverse Mercator projec-
tion used to map north–south strips of the earth called
gores (figure 3.21), which are used in the construction
of globes. Because printing the earth's surface directly
onto a round surface is very difficult, instead, a map of
the earth is printed in fiat elongated sections and then
attached to a spherical object. The narrow, 6°-wide
zones of the universal transverse Mercator grid system
(described in chapter 4) are' based on a secant case trans-
verse Mercator projection. North–south trending zones
of the U.S. state plane coordinate system (also explained
in chapter 4) are also based on secant cases of the projec-
tion. Most 1:24,000-scale U.S. Geological Survey (US GS)
topographic maps are projected on these state plane
coordinate system zones.

Figure 3.21 Gores of the globe on transverse Mercator

projections 30° wide at the equator centered at 90°W and
120°W. To make a world globe, the highlighted portion of
each map would be cut out and pasted onto the globe and
other gores would be used to cover the remaining area.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

54 Chapter 3 MAP PROJECTIONS

Conic map projections Albers equal area conic

Mathematically devised in 1805 by the German
Lambert conformal conic mathematician Heinrich C. Albers, the Albers equal
The Lambert conformal conic projection is another of area conic projection was first used in 1817 for a map of
the widely used map projections constructed by Lambert Europe. This is probably the projection used most often
in 1772. It is a secant case normal aspect conic projection for statistical maps of the conterminous United States
with its two standard parallels placed so as to minimize and other mid-latitude east–west trending regions. For
the map's overall scale distortion. The standard parallels example, you will see this projection used for U.S. sta-
for maps of the conterminous United States are placed at tistical maps created by the Census Bureau and other
33°N and 45°N to keep scale distortion at the map's edges federal agencies. The secant case version that has been
to less than 3 percent (figure 3.22). Just as the transverse used for nearly 100 years for the conterminous United
Mercator projection is used as the basis for the state plane States has standard parallels placed at 29.5°N and 45.5°N
coordinate system zones in north–south trending states, (figure 3.23). This placement reduces the scale distortion
the Lambert conformal conic projection is used as the to less than 1 percent at the 37th parallel in the middle of
basis for system zones in east–west trending states in the map, and to 1.25 percent at the northern and south-
the United States like Oregon and Wisconsin (see chap- ern edges of the country. USGS products, such as the
ter 4 for more on the state plane coordinate system using national tectonic and geologic maps also use the Albers
examples from these two states). As noted earlier, these projection, as do recent reference maps and satellite image
projections are in turn used for the 1:24,000-scale USGS mosaics of the country at a scale of around 1:3,000,000.
topographic maps within the state. The reason for the widespread use of the Albers
One major use of the Lambert conformal conic projection is simple—people looking at maps using
projection is for aeronautical charts. All U.S. 1:500,000- this projection can assume that the areas of states and
scale sectional charts can be thought of as smaller rectan- counties on the map are true to their areas on the earth.
gles of the 'national map described in the above paragraph.
Recall that navigators (like aviators) prefer navigational Pseudocylindrical and other projections
charts that use conformal projections, which preserve We have seen that map projections can be classed into
shapes and directions locally. Equally important is the families based on the nature of the surface used to con-
fact that straight lines drawn on the 1:500,000-scale struct the projection. This classification results in planar,
charts are almost great circle routes on the earth's surface conic, and cylindrical families. But these three fami-
(figure 3.22). lies constitute only a small portion of the vast array of

Figure 3.22 Lambert

conformal conic
projection used for a
map of the United States.
Straight lines are very
close to great circle
routes, particularly on
north-south paths.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

COMMONLY USED MAP PROJECTIONS 55

Figure 3.23 Albers equal area

conic projection of the United
States.

projections that have been constructed by cartographers. the generating globe. Instead, they reduce in distance as
We also noted that map projections could be classified the poles are approached. The elliptical shape of this pro-
based on their geometric distortion properties. Using jection makes it look more "earth-like," and the overall
this classification, we have shown that some preserve distortion in shape is less than on other equal area world
areas or shapes, while others have no special property projections such as the Gall-Peters.
aside from holding overall scale distortion to a minimum You'll find the Mollweide projection used for maps
or presenting a pleasing visual image. For many purposes, that show a wide range of global phenomena, from popu-
a projection that "looks right" is more important than a lation to land cover to major diseases. Cartographers have
projection that rigidly provides area, distance, shape, or devised other orientations of the projection by adjust-
direction fidelity. ing the central meridian to better show the oceans or to
Let's look at a last set of projections that can be center attention on a particular continent.
described using one or both of the classifications above.
These include pseudocylindrical and other projections
that are of special interest. Pseudocylindrical map pro-
jections can be conceived of as juxtaposing a number of
partial cylindrical maps. They are similar to cylindri-
cal projections in that parallels are horizontal lines and
meridians are equally spaced. The difference is that all
meridians except the vertical-line central meridian are
curved instead of straight.

Mollweide
You've probably seen world maps in the shape of an ellipse Figure 3.24 Mollweide projection used for a world map.
# twice as wide as it is high. Most likely you were looking at
the Mollweide projection, constructed in 1805 by the
German mathematician Carl B. Mollweide. This ellipti-
cal equal area projection most commonly uses the equa-
tor as the standard parallel and the prime meridian as the
central meridian of the projection (figure 3.24). Parallels
are horizontal lines, but they are not equally spaced as on

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

56 Chapter 3 MAP PROJECTIONS

Sinusoidal distortion within any lobe (section) of the projection.

According to some sources, Jean Cossin of Dieppe, Shape distortion at the edges of the map can be lessened
France, appears to be the originator of the sinusoidal considerably by interrupting the composite projection
projection, which he used to create a world map in into separate lobes that are pieced together along a central
1570. Others suggest that Mercator devised it, since it line, usually the equator. With interruption, the better
was included in later editions of his atlases. This easily parts of the projection are repeated within each lobe.
constructed equal area projection (figure 3.25) was The Goode interrupted homolosine projection
used by Nicholas Sanson (ca. 1650) of France for atlas (figure 3.26, bottom), created in 1923 by Goode from the
maps of the world and separate continents, and by John uninterrupted homolosine projection, is an interrupted
Flamsteed (1729) of England for star maps. Hence you pseudocylindrical equal area composite map projection
may also see it called the Sanson-Flamsteed projection. used for world maps as an alternative to portraying the
In addition to correctly portraying the relative areas earth on the Mercator world map projection (figure 3.19).
of continents and countries, the sinusoidal world projec- The projection is a composite of twelve segments that
tion has an SF of 1.0 along the central meridian, and the form six interrupted lobes. The six lobes at the top and
east—west SF also is 1.0 anywhere on the map. This means bottom are Mollweide projections from 40°N or 40°S
that the projection is equidistant in the east—west direc- to the pole, eaeh with a different central meridian. The
tion and in the north—south direction along the central six interior regions from the equator to 40°N or 40°S
meridian, but only in these directions. Note the severe are sinusoidal projections, each with a different central
shape distortion at the edges of the map (figure 3.25). meridian. If you look carefully along the edges of the
lobes, you can see a subtle discontinuity at the 40th par-
Homolosine allels. The two northern sections are usually shown with
You may have seen world map projections made by some land areas repeated in both regions to show the
compositing different projections along certain parallels Greenland land mass without interruption.
or meridians. The uninterrupted homolosine projec- You will find the Goode homolosine projection used
tion (figure 3.26, top), constructed in 1923 by American for maps in commercial world atlases to show a variety of
geography professor J. Paul Goode, is a composite of global information. It is a popular projection for showing
two pseudocylindrical projections. The sinusoidal pro- physical information about the entire earth, such as land
jection is used for the area from 40°N to 40°S latitude, elevations and ocean depths, land-cover and vegetation
while Mollweide projections cover the area from 40°N types, and satellite image composites. When viewing
to 40°S to the respective poles. Since the component these maps, remember that the continuity of the earth's
projections are equal area, the homolosine projection is surface has been lost completely by interrupting the pro-
as well. Notice in figures 3.25 and 3.26 that the shapes jection. This loss of continuity may not be apparent if the
of the continents look less distorted on the homolosine graticule is left off the projection, as may be the case with
projection than on the sinusoidal projection, particularly some global distribution maps found in world atlases.
in polar areas.
An interrupted projection is one in which the Robinson
generating globe is segmented in order to minimize the In 1963, the American academic cartographer Arthur H.
Robinson constructed a pseudocvlindrical projection
that is neither equal area nor conformal but makes the
continents "look right." A "right-appearing" projection
is called orthophanic. For the Robinson projection
(figure 3.27), Robinson visually adjusted the horizontal
line parallels and curving meridians until they appeared
suitable for a world map projection to be used in atlases
and for wall maps. To do this, Robinson represented the
poles as horizontal lines a little over half the length of
the equator. You may have seen this projection used for
world maps created by the National Geographic Society.
It has also been used for wall maps of the world that show
the shape and area of continents with far less distortion
Figure 3.25 Sinusoidal equal area world map projection. than wall maps that use the Mercator projection.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

COMMONLY USED MAP PROJECTIONS 57

Figure 3.26 Uninterrupted Uninterrupted Homosoline Projection

homolosine and Goode
interrupted homolosine equal
area world projections.

Goode Homosoline Projection

_--Lobes

IV16,11weide
40°N ON

40°6, 1 —40°S
MollWeid6
I ,
Lobes - Lobes

Figure 3.27 Robinson

pseudocylindrical world map
projection.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

58 Chapter 3 MAP PROJECTIONS

Aitoff The Winkel tripel projection was not used widely until
Other map projections can be constructed by mathe- 1998 when the National Geographic Society announced
matically modifying widely used projections developed that it was adopting the Winkel tripe! projection as its
centuries earlier. In 1889, the Russian cartographer standard for maps of the entire world. As a result, use of
David Aitoff published a modification of the equato- the Winkel tripel projection has increased dramatically
rial aspect azimuthal equidistant projection that today over the last few years.
carries his name (figure 3.28). For the Aitoff projec-
tion, Aitoff simply doubled the horizontal scale of the
azimuthal equidistant projection, creating an elliptical
projection with the same two-to-one width-to-height
ratio as the Mollweide projection (figure 3.24). Unlike
the Mollweide projection, parallels are not straight
horizontal lines, and the map is neither equal area nor
equidistant. The Aitoff projection is an interesting com-
promise between shape and area distortion, suggesting
the earth's shape with less polar shearing than on maps
that use the Mollweide projection.

Figure 3.29 Winkel tripel projection used to create a

world map.

PROJECTIONS ON THE SPHERE

AND ELLIPSOID

Figure 3.28 Aitoff projection used to create a world map. Mapmakers have a general rule that small-scale maps
can be projected from a sphere, but large-scale maps
always must be projected from an ellipsoidal surface
Winkel tripel such as the WGS 84 ellipsoid. We saw in chapter 1 that
New map projections may also be constructed as small-scale world or continental maps such as globes and
mathematical combinations of two existing projections. world atlas sheets normally use coordinates based on
Perhaps the best known example is the Winkel tripel an authalic or other auxiliary sphere. This was because
projection constructed in 1921 b y the German cartog- prior to using digital computers to make these types of
rapher Oswald Winkel. The term tripel is not someone's maps numerically, it was much easier to construct them
name, but rather a German word meaning a combina- from spherical geocentric coordinates. Equally impor-
tion of three elements. Winkel used the term to. empha- tant, the differences in the plotted positions of spherical
size that he had constructed a compromise projection and corresponding geodetic coordinates are negligible
that was neither equal area, conformal, nor equidistant, on small-scale maps.
but rather minimized all three forms of geometric dis- Large-scale maps must be projected from an
tortion. He accomplished this by averaging the x- and ellipsoidal surface because, as we saw in chapter 1, the
yloordinates computed for the equirectangular and spacing of parallels decreases slightly but significantly
Aitoff world projections at the same map scale. The result- from the pole to the equator. We noted that on the
ing projection is similar to the Robinson projection, but WGS 84 ellipsoid the distance between two points one
if you look closely you will see that parallels are not the degree apart in latitude (between 00 and 1°) at the equa-
straight horizontal lines characteristic of pseudocylindri- tor is 68.703 miles (110.567 kilometers), shorter than the
cal projections. Rather, they are slightly curving, nonpar- 69.407 mile (111.699 kilometer) distance between two
allel lines (figure 3.29). points at 89°N and 90°N.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

PROJECTIONS ON THE SPHERE AND ELLIPSOID 59

Let's see the differences in length when we project

UTM Projections of One Degree
one degree of geodetic latitude along a meridian at the of Geodetic Latitude along a Meridian
pole, at the 45th parallel, and at the equator using a
transverse Mercator projection at a scale of 1:1,000,000 90° 45° 1°
(figure 3.30). At this scale, the length of a degree of lati-
tude at the pole is projected as slightly over I millime-
ter longer than a degree of latitude at the equator. This
difference on these two projections may seem minimal,
but on the ground it represents a distance of nearly a
kilometer! At larger map scales the difference in length
becomes more noticeable—for instance, slightly over
1 centimeter on polar and equatorial maps at a scale of
1:100,000. Topographic and other maps at this scale and
larger are projected from an ellipsoid so that accurate dis-
tance and area measurements can be made on them. The
same holds true for large-scale equidistant projections
such as the polar aspect azimuthal equidistant projection
(figure 3.16), since the spacing of parallels on the projec-
tion must be slightly lessened from the pole outward to
reflect their actual spacing on the earth.
We saw in chapter 1 and in table C.2 in appendix C
that the only place on the earth where the 69.05 mile
(111.12 kilometer) per degree spacing of parallels on the
—E
authalic or rectifying sphere is essentially the same as on -I-
if.
CD
CD
CD
the WGS 84 ellipsoid is in the mid-latitudes close to the = CD-

45th parallel. You will find that the transverse Mercator 89° 44° 777

or another map projection based on the sphere or the

WGS 84 ellipsoid will have the same spacing of parallels Figure 3.30 Transverse Mercator map projections of one
in locations that straddle the 45th parallel. degree of geodetic latitude on the WGS 84 ellipsoid along
Every map projection has its own virtues and a meridian at the pole, midway between pole and equator,
limitations. You can evaluate a projection only in light of and at the equator.
the purpose for which a map is to be used. You shouldn't
expect that the best projection for one situation will be
the most appropriate for another. Fortunately, the map the individual sheets match in a north–south direction,
projection problem effectively vanishes if the cartogra- they don't fit together in an east-west direction. Yet even
pher has done a good job of considering projection prop- this difficulty won't be a serious handicap unless you try
erties and if you are careful to take projection distortion to create a large map mosaic.
into consideration in the course of map use. The age of computers has changed how we think about
At the same time, however, there is the very real map projections. You no longer need to "make do" with
possibility that the mapmaker through ignorance or lack inappropriate projections. Computer-generated projec-
of attention can choose an unsuitable projection or pro- tions are available for almost any use. Furthermore, it
jection parameters. Therefore is it of utmost importance is now practical for you to sit down at a computer and
to map users to understand the projection concepts dis- construct your own projections. Most important, per-
cussed in this chapter to be assured that the maps we use haps, you can manipulate projections on the computer
• have been made with careful consideration of all the deci- in search of the ideal projection base for a given applica-
sions required to make an appropriate map projection. tion. All these benefits can only be realized, of course, if
Since most map use takes place at the local level, where you know enough about projections to take advantage of
earth curvature isn't a big problem, global map projec- these opportunities that computers provide.
tions aren't a great concern for many users. With regions Being map-savvy also helps you evaluate projections
as small as those covered by topographic map quadran- published in magazines and newspapers. World maps
gles, your main projection-related problem is that, while in the popular media often lack latitude-longitude grids.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

60 Chapter 3 MAP PROJECTIONS

This masks the extreme spatial distortion inherent in NOTES

such maps and gives the impression that the continents
are truly represented. Thus, the map user is well advised 1. The equation for the sphere isy = in ('tan (x/4 +
to reconstruct the latitude-longitude grid mentally as a where p is the latitude in radians (there are 2 ,7 radians in
first step in map reading. a 360 0 circle, so 1 radian is approximately 57.295 degrees).
Understanding the concepts above will go a long way The value of this equation at 90° (7r/2 radians) is infinity.
toward helping you make the right decisions, but there
are also books and Web sites dedicated to map projec-
tions. National mapping agencies such as the USGS and
swisstopo, the Swiss national mapping agency, also offer
useful map projection information on their Web sites
and in their publications.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

GEODETIC REPORT OF KENYA

1. INTRODUCTION

This paper details the existing geodetic network, the Datum used and how they were
derived, the problems experienced currently and the improvements necessary. The
recent activities in Global positioning through the effort of Survey of Kenya and Kenya
Institute of Surveying & Mapping (KISM) is considered as a means of starting future
extension to the GPS Network in Kenya.

1.1 Location

Kenya is in the Eastern region of Africa located between longitudes 34º and 40º East and
latitude 5º 30' North and 5º South. It is bounded by the indian ocean on the East, Somalia
Republic on the North East, Sudan and Ethiopia on the North, Uganda on the West and
Tanzania on the South.

Area

The area of Land only is 571416 square km while Hinter land water (lakes) is 11,230
square km the country has 3km exclusive territorial sea area of 2,717 square km. The
total Area with the above included is therefore 585,363 square km.

1.2 EARLY GEODETIC ACTIVITIES - A HISTORICAL BACKGROUND

Since 1892 several major triangulation networks have been observed and computed by
various organisations for specific purposes. Each organisation chose the co-ordinate
system that appeared suitable resulting in a number of different systems. For each
sizeable network, the readily available datum was used resulting in many datums for
both planimetric and height control.

At the beginning of the 20th century the Anglo German Boundary Commission (AGBC)
of 1893 was established to carry out delimitation of boundaries of colonies in Central
and East Africa. This involved the Belgians, Portuguese, British and Germans.

The interterritorial boundaries required geodetic survey points on which they would be
based. As is expected, no geodetic control points existed. The colonial powers hence
carried out triangulation work along agreed boundaries. The triangulation chains
established were not soundly observed and were shaped by the boundaries rather than by
geodetic requirements. The first such triangulation in East Africa was observed between
Kenya and Tanganyika between 1892 and 1893 - see diagram No. 1.

The triangulation had the longitude of Jombo adopted as the

origin. This had been obtained from Admiralty chart No. 1390. The latitude observed at
vilima viwili was adopted and carried down by triangulation to Jombo. The height
datum was the high water spring tide at vanga and the trigonometric heighting was done

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

by reciprocal heighting. This triangulation was also used to height kibo, the peak of Mt.
Kilimanjaro from vilima viwili.

The second triangulation chain was conducted through the Anglo German Boundary
Commission of 1902-1906. This was a triangulation from Zanzibar to Ruwenzori. The
Kenyan portion of the same work covered Taveta to Misoma - see diagram 2.

1.2.1 THE KENYA MAJOR TRIANGULATION

By 1906 only two triangulation chains of the Anglo-German Boundary commission

existed. The chains however ran along the Southern boundary of the country and the
edge of Lake Victoria. This was considered too far from the areas where surveys for
land registration were required. Kenya major triangulation was then started in 1906 to
provide control for title surveys as well as control for topographical mapping. Kenya
major used cassini co-ordinate system. The diagram No. 3 shows the extent of this
network. Kenya major derived its datum from AGBC of 1902 at Kisumu where a base
existed and its latitude at Athi River. The longitude was derived at Elemoborasha. After
this initial work, several other triangulation chains were conducted between 1906 and
1914. After 1914 the next major work was done between 1939 and 1941 as extensions
to the Kenya major triangulation - see diagram No. 4,5,6.

Kenya major triangulation had common points with the AGBC chains. The common
points enabled re-computation of AGBC 1902-1906 triangulation based on Kenya major
sections. The purpose was to generate co-ordinates consistent with Kenya major
triangulation data. This is indicated in Diagram, No.6.

Survey of Kenya has adopted Kenya major datums for Emali to Tsavo, Central Embu,
Meru, Riftvalley, the Uasin Gishu and Trans-Nzoia triangulations.

1.3 ESTABLISHMENT OF KENYA PRIMARY, SECONDARTY AND

LOWER ORDER TRIANGULATION

The Directorate of overseas surveys was the arm of the British Government through
which Britain provided survey and mapping services in overseas countries. Originally
the Survey Department was known as the Directorate of Colonial Surveys (DCS) but as
British colonies attained independence the name was changed to DOS. The DOS began
operations in Kenya at around 1950.

The DOS established the present Kenya primary, secondary and lower order
triangulation and also observed traverses. The DOS have also mapped this country from
1950 until their departure at the end of 1983. The D.O.S. used U.T.M. as the system of
co-ordinates and clarke 1880 as the reference ellipsoid. The geodetic control Diagram
covering points in East Africa and the SK81F show the extent of the present controls.

1.4 DATUM USED IN KENYA

The 30th meridian arc datum is a chain of geodetic triangulation running almost parallel
to the 30th meridian east from Port Elizabeth in South Africa to Cairo in Egypt - see
diagram 0.7.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

During 1950 the DOS computed part of this chain upto Uganda having held fixed some
points in Zimbabwe which had been computed from South Africa. By holding the arc
results fixed upto Uganda, the rest of East African triangulation chains were adjusted.
Their results were then being dependent on the arc results of 1950, referred to as
originating from' 1950 Arc Datum'. Any re-adjustment of the arc means that all the
triangulation chains depending on it would have to be re-adjusted. By 1960 this arc had
been revised and the East African triangulation chains were accordingly re-adjusted.
The new values were then designated as originating from "1960 Arc Datum".

The current Geodetic network in Kenya is based on the 1960 Arc Datum. The
coordinates is in U.T.M.

1.5 DOPPLER OBSERVATIONS

Survey of Kenya, the Defence Mapping Agency of the U.S.A. and the Directorate of
Military Surveys of the United Kingdom first carried out Doppler observations in Kenya
in 1972 and 1973. The aim of this exercise was to:-

(i) Evaluate the accuracy of the primary control in Kenya;

(ii) Provide geodetic control in remote and un-surveyed area in the country;

(iii) Strengthen the triangulation with precise position at Optimum Spacing;

(iv) Contribute to the development of a single well fitting datum for the
African continent.

The ITT 5500 Doppler equipment was used. Doppler positioning using precise
ephemeris fixed fifteen (15) stations distributed over the country. Further tests show that
surveys carried out with this equipment are accurate to 3 meters in each X, Y, Z co-
ordinates. The Doppler station monuments are still intact.

1.6 VERTICAL CONTROL

A geodetic leveling programme was proposed in 1947 but leveling was carried out
between 1950 and 1958. The datum for all heighting in Kenya is the mean sea level
referred to a tide gauge at Kilindini Harbour in Mombasa. This gauge has however been
submerged and a new one constructed and connected to some fundamental bench marks.
Upto date about 3570km has been precisely leveled, the last 70km having been carried
out by Kenya institute of Surveying and Mapping during 1992 and 1993.

2.0 REFERENCE ELLIPSOIDS

The two main reference ellipsoids used in kenya are (I) Clarke 1858 and (ii)
Clarke1880.

2.1 System of Co-ordinates

There are three co-ordinate systems that have been used in Kenya:-

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

The cassini-soldner co-ordinate system

The East African war system
The U.T.M. co-ordinate system.

2.1.1 Cassini-soldner co-ordinate system

The origins of this co-ordinate system are the intersections between the equator and
the odd meridians. The odd meridians served as the central meridian for each 2º belt
which extends one degree to east and west. To give an example, the Central
meridians can be 35ºE, 37ºE, 39ºE etc, with belt limits extending from 34ºE to 36ºE,
36ºE to 38ºE, 38ºE to 40ºE etc. The reference ellipsoid used is Clarke 1858, and the
unit of measurement is the British foot.

Before 1950, nearly all triangulation network was based on this system. The cadastral
surveys in Kenya were also based on this system of co-ordinates.

2.1.2 The East African War System of Co-ordinates

This system of co-ordinates was introduced as a military system for East Africa. The
extended triangulation to Kenya major triangulation north of and around Mt. Kenya, was
based on this system. The main object of East Africa war system was to unify the co-
ordinate system for the British commonwealth territories in the South, East and Central
Africa to avoid discontinuity in topographical mapping and grid references across
territorial boundaries.

The co-ordinates were based on traverse mercator projection of 5 wide belts. The belts
were designated C,D,E, etc and Kenya was covered by Belts H and J as follows:-

Belt H: - Central meridian 37º30’E

Western limits 35ºE
Eastern Limits 40ºE

Belt J: - Central Meridian 42º30’E

Western Limit 40ºE
Eastern Limit 45ºE

The projection had the following parameters:-

Origin: Equator and Central meridian

False co-ordinates: Eastings +400 000 meters

Northings +4,500,000 meters
Scale factor: 1999/2000 approximately
equals 0.9995

Spheroid: Clarke 1880.

The Co-ordinates in this system have been converted to the current U.T.M. system.

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

2.1.3 The U.T.M. Co-ordinate System

The U.T.M. system was introduced in Kenya in 1950 by the Directorate of overseas
surveys (D.O.S). This was when D.O.S. began providing control Survey work in Kenya.

The system used Clarke 1880 spheroid. The unit of measurement was the international
metre. The survey department has made effort to convert all points to this co-ordinate
reference system.

3.0 PROBLEMS EXPERIENCED WITH THE CURRENT GEODETIC

NETWORK

The main problems that have been identified with the current network include:-

(a) The existence of different co-ordinate systems this has caused the need for
regular co-ordinate conversion especially from Cassini to U.T.M. system of co-
ordinates.

(b) Pillars located in hills have been difficult to access and many have been
destroyed. The Network density is also low.

(c) Lack of suitable points to base the transformation especially from Cassini to
UTM.

(d) Lack of information or data of previous work. The records of survey work
carried out before 1950 are not readily available.

(e) Height data has been found not consistent having been derived from different
datums.

(f) Equipment used earlier had lower level of precision and the network is generally
weak.

(g) Hydrographic Charts are not fully developed due to lack of controls near the
indian ocean, and around the lakes.

(h) Points established by space techniques are relatively few.

(i) Re-establishment of destroyed pillars has not been carried out fully due to high
cost of classical triangulation method.

4.0 MINIMIZING CURRENT PROBLEMS OF GEODETIC NETWORK

4.1 Initial Activities to solve the Geodetic Network Problems using GPS

In August, 1996 Kenya Institute of Surveying and Mapping (KISM) under the mini-
project started to carry out GPS surveys on First order triangulation pillars in an
attempt to establish transformation parameters between the world Geodetic System 84

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

ellipsoid (WGS − 84) and Clarke 1880 reference ellipsoid used in East Africa. The
work was not completed due to lack of funding.

As an initial effort, between 1996 and 1997, twelve First Order Triangulation points
were occupied and observation made using GPS receivers. The data is available but
further work is needed in order to have several points well distributed all over the
country.

The United states of America’s NIMA (National Imaging and Mapping Agency) has
carried out some GPS work to determine the transformation parameters between
WGS-84 and Clarke 1880 (Arc ’60 Datum) using GPS stations and have determined
the transformation parameters as:-

Shift in X = - 157m
Shift in Y = - 2m
Shift in Z = - 299m

The error estimates are EX = 4m, EY = 3m, and EZ = 3m.

The Co-ordinates in WGS-84 have also been requested by the Airport Authorities
here in Kenya as it is the co-ordinate system used by all major Airlines. The
establishment of proper transformation parameters is crucial to optimizing the results
of any GPS work.

4.2 Processing Softwares

KISM is in the process of computing the transformation parameters using the

following softwares.

1. SKI − This is Leica software suitable for computations of GPS Surveys

using Leica’s GPS receivers. The software is suitable for small and
medium scale surveys. The software is however limited to short baseline
computations.

2. BERNESE − This is also a commercial software suitable for long lines.

3. GARMIT/GLOBK − This is academic software running on Linux operating

system. In all cases precise ephemeries is used other than the broadcast
ephemeries.

4.3 Existence of IGS Stations

In Kenya there are two continuous GPS observation stations connected to the IGS
(International Geodynamic GPS Services)stations. One is in Malindi while the is in
KISM.

The one in KISM was established by KISM in 1997. The data is collected using Leica’s
SR 399E sensor run by Multistation V 2.3 software. The antenna is placed high enough
such that it is able to obtain data without obstruction, 15 degrees above the horizon. The

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

data from these continuous recording stations are used for the computations of the
baselines and eventual co-ordination of all other points.

The data from the IGS stations is available through the internet.

4.3.1 Data Processing

For ordinary surveys, data is processed using SKI software which has the option of
processing RINEX (Receiver Independent data Ex change format) data. For geodetic
surveys the data require processing by GAMIT/GLOBK or BERNESE software, these
have means of dealing with finer errors caused by procession and radiation.

4.3.2 KISM IGS Co-ordinates

The co-ordinates of KISM IGS was determined from 6 months of data using precise
ephemeries and other IGS stations using the ITRF94 (International Terrestrial Reference
Frame of 1994) Co-ordinates System. This was later transformed to WGS-84 Co-
ordinates with assistance of the GSI − Japan (Geographical Survey institute of Japan)
using GAMIT software.

The WGS-84 geographical co-ordinates are given as:-

Hat 1 15 0.777775
Hang 36 51 24.03925
Ellip.Height 1630.229

4.3.3 Malindi IGS Station

Information and data about this station is available from the Internet.

The co-ordinates of Malindi IGS which was used to Compute the co-ordinates of KISM
IGS is given as:-

X = 4865366.8060
Y = 4110737.5960
Z = -331121.7800

The co-ordinates are referenced to ITRF94.

4.3.4 The GPS Equipment at KISM

The GPS equipment available are:-

4 No. - SR399 sensors (L1/L2 with C/A, P1, P2

capabilities) CR 399 control

2 No. - SR 260 Sensors (LI band only) CR 261 control; and

Downloaded by Francis Mbae ([email protected])

 lOMoARcPSD|43072212

2 No. SR 399E sensors for continuos observations (LI/L2 bands)with antenna. The
multistation 2.3 or apc is used as a control, running on windows 95/98.

4.4 Establishment of Geodetic controls with GPS

Once the transformation parameters between the local Datum and the WGS 84 is
established to an accurate level, and the geoid undulations over the country, then the rest
of the control points can be established using GPS which is faster, accurate and
convenient to work with. For accurate Geodetic control, three or more GPS receivers
would be required, such a combination would enable a network adjustment to be
possible.

5.0 WHAT NEEDS TO BE DONE TO IMPROVE ON GEODETIC

NETWORK

The following listed points may contribute to improvement on the current network:-

(1) The formation of Geoinformation system for Geodetic control. This can be used
to store the current Geodetic points and their conditions.

The inventory created can be used effectively for decision making on the type of
monumentation analysis of the data quality, the adjustment of the network,
reconstruction and modification of the network, measurement of all needed
gravity differences by modern gravimeters and the measurement of absolute
gravity by special instruments.

(2) Establishment of modern three dimensional network based on satellite

observations. This is easily achieved owing to remarkable accuracy,
independence of weather conditions, and high performances of Global
positioning system (GPS).

(3) The potential of GPS can be realized through projects such as:-

(1) Establishment of more GPS permanent stations at convenient locations.

(2) Establishment of national reference GPS networks
(3) Realization of urban network projects
(4) Provision of WGS 84 co-ordinates to all Airports.

(4) Direct connection of our fundamental networks with neighbouring networks is

necessary in order to provide Cartographic continuity and to participate in
international projects.

(5) The choice of receivers is an important factor and specifications need to be

studied in relation to the processing software and datatypes it can handle.

(6) More receivers should be purchased based on specifications that meet Geodetic
standards.
(7)

Downloaded by Francis Mbae ([email protected])