8 TRANSVERSE MERCATOR

PROJECTIONS

The transverse Mercator projection was developed by Lambert in 1772, analytically derived
by Gauss 50 years later, and then formulas more suitable for calculations were derived by Kruger in
1912. This is perhaps the reason that it is one of the most widely used conformal map projections.

8.1 Transverse Mercator

A transverse Mercator projection is the ordinary Mercator projection turned through a 900
angle so that it is related to a central meridian in the same way that the ordinary Mercator
projections is related to the equator. Its properties are:

1. Cylinder is tangent to the globe at a meridian

2. scale is true along the meridian, which is called the central meridian
3. origin of the map X coordinate.is at the central meridian
4. The origin of the map Y coordinate is the equator.
5. The projection is conformal like the regular Mercator projection.
GdE 321 Transverse Mercator Projections page 2

9.1.1 Direct Problem

The mapping equations for the direct problem are given by (9.1) to (9.4) where:
𝑿 ∆𝝀𝟑 𝐜𝐨𝐬 𝟑 𝝋 𝟐 𝟐
∆𝝀𝟓 𝐜𝐨𝐬 𝟓 𝝋 𝟓 − 𝟏𝟖𝒕𝟐 + 𝒕𝟒 + 𝟏𝟒𝜼𝟐 − 𝟓𝟖𝒕𝟐 𝜼𝟐
= ∆𝝀 𝐜𝐨𝐬 𝝋 + (𝟏 − 𝒕 + 𝜼 ) + ( )
𝑵 𝟔 𝟏𝟐𝟎 +𝟏𝟑𝜼𝟒 + 𝟒𝜼𝟔 − 𝟔𝟒𝜼𝟒 𝒕𝟐 − 𝟐𝟒𝜼𝟔 𝒕𝟐

∆𝜆7 cos 7 𝜑
+ (61 − 479𝑡 2 + 179𝑡 4 − 𝑡 6 ) (8.1)
5040
and
𝒀 𝑺𝝋 ∆𝝀𝟐 ∆𝝀𝟒
= + 𝐬𝐢𝐧 𝝋 𝐜𝐨𝐬 𝝋 + 𝐬𝐢𝐧 𝝋 𝐜𝐨𝐬 𝝋 (𝟓 − 𝒕𝟐 + 𝟗𝜼𝟐 + 𝟒𝜼𝟒 )
𝑵 𝑵 𝟐 𝟐𝟒

∆𝜆6 61 − 58𝑡 2 + 𝑡 4 + 270𝜂 2 − 330𝑡 2 𝜂2 + 445𝜂 4

+ sin 𝜑 cos5 𝜑 ( )
720 +324𝜂6 − 680𝜂 4 𝑡 2 + 88𝜂 8 − 600𝜂 6 𝑡 2 − 192𝜂 8 𝑡 2
∆𝜆8
+ sin 𝜑 cos7 𝜑 (1385 − 311𝑡 2 + 543𝑡 4 − 𝑡 6 ) (8.2)
40320

where:
𝑡 2 = tan2 𝜑
𝑒2
𝜂2 = cos2 𝜑
1 − 𝑒2
∆λ = ( λ - λCM ) = longitude difference in radians
N = radius of the prime vertical
Sϕ is the length of the meridian arc from the equator to latitude ϕ having
𝑎
𝑆𝜑 = [𝑎 𝜑 − 𝑎2 sin 2𝜑 + 𝑎4 sin 4𝜑] (8.3)
1+𝑛 𝑜
For a= 6,378,137 and f= 0.00335281068118 (GRS 80)
𝑒 2 =0.00669438
𝑛 = 0.00167922 ;
𝑎𝑜 = 1.00000070 ;
𝑎2 = 0.00251883 ;
𝑎4 = 0.00000264
where:
𝑓 𝑎
𝑛= = 0.00167922; = 6,367,444.66 𝑚
2−𝑓 1+𝑛
𝑛2 𝑛4
𝑎𝑜 = 1 + + = 1.00000070;
4 64
GdE 321 Transverse Mercator Projections page 3

3 𝑛3
𝑎2 = (𝑛 − ) = 0.00251883;
2 8
15 2 𝑛4
𝑎4 = (𝑛 − ) = 0.00000264
16 4

For a 1 meter precision equation 8.1 and 8.2 can be simplified to

∆𝜆3 cos 3 𝜑
𝑋 = 𝑁 [∆𝜆 cos 𝜑 + (1 − 𝑡 2 + 𝜂 2 ) ] (8.4)
6

and

∆𝜆2
𝑌 = 𝑆𝜑 + 𝑁 [ sin 𝜑 cos 𝜑] (8.5)
2

The convergence of the meridians in terms of geodetic coordinates

∆𝜆2 cos2 𝜑(1 + 3𝜂 2 + 2𝜂 4 ) ∆𝜆4 cos4 𝜑(2 − 𝑡 2 )

𝛾 = ∆𝜆 sin 𝜑 [1 + + ] (8.6)
3 15

where Δλ and γ are in radians

The scale factor in terms of geodetic coordinates

∆𝜆2 cos2 𝜑 2)
∆𝜆4 cos4 𝜑
𝜇 =1+ (1 +𝜂 + (5 − 4𝑡 2 ) (8.7)
2 24

8.1.2 Inverse Problem

Footpoint Latitude
The footpoint latitude (φ1) is a point on the central meridian obtained by drawing a parallel to the x
axis through the point P.
GdE 321 Transverse Mercator Projections page 4

C.
M. Parallel of
through y point
Footpoint in question
Parallel of point in
Latitude φ1 x
P question

Latitude
y

Equator

Figure 8.2 Footpoint

Latitude

Recalling that the length of the meridian arc is Sϕ (9.3) and that the central meridian is y then our
footpoint latitude will be given by the meridian arc length y along the central meridian. To solve for
the footpoint latitude, we use the Newton-Raphson method. The procedure as follows:
1. Obtain an initial approximate value for φ by letting φo=y/a
2. Evaluate an improved solution from
𝑆(𝜑) − 𝑦
𝜑1 = 𝜑𝑜 − (8.8)
𝑆 ′ (𝜑)
where:
𝑎
𝑆(𝜑) = (𝑎 𝜑 − 𝑎2 sin 2𝜑𝑜 + 𝑎4 sin 4𝜑𝑜 )
1+𝑛 𝑜 𝑜
𝑎
𝑆 ′ (𝜑) = (𝑎 − 2𝑎2 cos 2𝜑𝑜 + 4𝑎4 cos 𝜑𝑜 )
1+𝑛 𝑜

3. Designate the computed 𝜑1 𝑎𝑠 𝜑𝑜 and continue the iterative procedure until

|𝜑1 − 𝜑𝑜 | ≈ 0.00000000 𝑟𝑎𝑑𝑖𝑎𝑛𝑠
depending on required precision.
4. Final φ1 is the footpoint latitude

Given the x and y coordinates, we are to calculate the φ and λ. Mapping equations in terms of
grid coordinates
GdE 321 Transverse Mercator Projections page 5

𝑋2 𝑋4 5 + 3𝑡12 + 𝜂12
+ ( )
2𝑀1 𝑁1 24𝑀1 𝑁13 −4𝜂14 − 9𝜂12 𝑡12
61 − 90𝑡12 + 46𝜂12 + 45𝑡14
𝜑 = 𝜑1 − 𝑡1 6 (8.9)
𝑋 −252𝑡12 𝜂12 − 3𝜂14 + 100𝜂16
+
720𝑀1 𝑁15 −66𝑡12 𝜂14 − 90𝑡14 𝜂12 + 88𝜂18
4 4 2 6 2 8
[ (+225𝑡1 𝜂1 + 84𝑡1 𝜂1 − 192𝑡1 𝜂1 ) ]

𝑋 1 𝑋 3
− ( ) (1 + 2𝑡12 + 𝜂12 )
𝑁1 6 𝑁1
1 𝑋 5 5 + 6𝜂12 + 28𝑡12 − 3𝜂14 + 8𝑡12 𝜂12
𝜆 = sec 𝜑1 + ( ) ( ) (8.10)
120 𝑁1 +24𝑡14 − 4𝜂16 + 4𝑡12 𝜂16 + 24𝑡12 𝜂16
1 𝑥 7
− ( ) (61 + 662𝑡12 + 1320𝑡14 + 720𝑡16 )
[ 5040 𝑁1 ]

For lower precision to 1”estimates (8.9) and (8.10) can be truncated to:
𝑋2
𝜑 = 𝜑1 − 𝑡1 [ ] (8.11 𝑎)
2𝑀1 𝑁1
𝑥 1 𝑥 3
𝛥𝜆 = sec 𝜑1 [ − ( ) (1 + 2𝑡12 + 𝜂12 ) + ⋯ ] (8.11 𝑏)
𝑁1 6 𝑁1
where:
𝜑1 = 𝑓𝑜𝑜𝑡𝑝𝑜𝑖𝑛𝑡 𝑙𝑎𝑡𝑖𝑡𝑢𝑑𝑒
𝑎(1 − 𝑒 2 )
𝑀1 = 𝑟𝑎𝑑𝑖𝑢𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑚𝑒𝑟𝑖𝑑𝑖𝑎𝑛 𝑜𝑓 𝜑1 =
(1 − 𝑒 2 sin2 𝜑1 )3⁄2
𝑎
𝑁1 = 𝑟𝑎𝑑𝑖𝑢𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑝𝑟𝑖𝑚𝑒 𝑣𝑒𝑟𝑡𝑖𝑐𝑎𝑙 𝑜𝑓 𝜑1
√1 − 𝑒 2 sin2 𝜑1
𝑡1 = tan 𝜑1
𝑒2
𝜂12 = cos2 𝜑1
1 − 𝑒2

Meridian convergence in terms of grid coordinates

𝑋 1 𝑋 3 1 𝑋 5
tan 𝛾 = 𝑡1 [ − ( ) (1 + 𝜂12 − 2𝜂14 ) + ( ) (2 + 2𝜂12 + 9𝜂14 )] (8.12. 𝑎)
𝑁1 3 𝑁1 15 𝑁1

A simplified expression which is sufficient for most geodetic work

𝑋 1 𝑋 3
tan 𝛾 = 𝑡1 [ − ( ) ] (8.12. 𝑏)
𝑁1 3 𝑁1
GdE 321 Transverse Mercator Projections page 6

Scale Factor in terms of grid coordinates

1 X 2 1 X 4
μ = 1 + ( ) (1 + η12 ) + ( ) (1 + 6𝜂12 + 9𝜂14 + 4𝜂16 − 24𝑡12 𝜂14 − 24𝑡12 𝜂16 ) (8.13. 𝑎)
2 N1 24 N1
A simplified expression would be:
1 x 2
μ = 1 + ( ) (1 + η12 ) (8.13. b)
2 N1
9.2 Universal Transverse Mercator Projection (UTM)

For the state plane coordinate system, the Mercator projection cylinder is made to cut the surface of
the sphere along two standard lines parallel to the central meridian.

Central
Meridia N
n

Scale 1:1
Figure 10.4 Universal Transverse Mercator
Projection

The universal transverse Mercator (UTM) projection is based on the transverse Mercator projection.
Its specifications are:
1. Transverse Mercator projection is in zones that are 60 wide.
2. The reference ellipsoid is GRS 80 in North America
3. The origin of longitude is at the central meridian
4. The origin of latitude is at the equator.
5. The unit of measure is the meter.
6. False Northing of 0 m is used for northern hemisphere; 10,000,000.00 for southern
hemisphere
7. A false Easting of 500,000.00 is used for the central meridian of each zone.
8. The scale factor at the central meridian is 0.99960000
9. The zones are numbered with 1 for the zone between 180 W and 1740 W meridians and
increasing to 60 for the zone between meridians 1740 E and 180°E
GdE 321 Transverse Mercator Projections page 7

10. The latitude for the system varies from 80oN to 80oS.

Zone Zone Zone Zone Equator

1 2 31 60

1800 E
1800W 174 0 W 168 0 W 00

Grid Zones

The map above shows how the UTM system divides the earth into 60 longitude zones, each six degrees
wide. The numbering begins at Zone 1 at 180 degrees west and proceeds eastward. To find the grid zone for
any longitude:

• Treat west longitude as negative and east as positive.

• Add 180 degrees; this converts the longitude to a number between zero and 360 degrees.
• Divide by 6 and round up to the next higher number.
GdE 321 Transverse Mercator Projections page 8

9.3.1 Direct Problem

The mapping equations for the direct problem for the UTM in terms of the Transverse Mercator
equation is

𝑥 𝑥
{𝑦} = 𝜇𝑜 {𝑦} (8.14)
𝑈𝑇𝑀 𝑇𝑀

For 1 meter precision grid computations :

𝑿 ∆𝝀𝟑 𝐜𝐨𝐬 𝟑 𝝋
= 𝝁𝒐 [∆𝝀 𝐜𝐨𝐬 𝝋 + (𝟏 − 𝒕𝟐 + 𝜼𝟐 )] (𝟖. 𝟏𝟓𝒂)
𝑵 𝟔

𝒀 𝑺𝝋 ∆𝝀𝟐
= 𝝁𝒐 [ + 𝐬𝐢𝐧 𝝋 𝐜𝐨𝐬 𝝋] (𝟖. 𝟏𝟓 𝒃)
𝑵 𝑵 𝟐

The meridian convergence for the UTM is the same as for the TM

𝛾𝑈𝑇𝑀 = 𝛾𝑇𝑀 (8.16)

By choosing a scale factor of 0.9996 at the central meridian, the scale factor for TM projection is
then modified,

∆𝜆2 cos2 𝜑
𝜇 = 𝜇𝑜 [1 + (1 + 𝜂 2 ) + ⋯ … ] (8.17)
2
The distance from the central meridian to the meridian of true scale along the equator can be
calculated as :

1 𝜇 − 𝜇𝑜 2 0.0282899
∆𝜆 = √ ∙ 2
= (8.18)
cos 𝜑 𝜇𝑜 (1 + 𝜂 ) cos 𝜑 √1 + 𝜂 2

For φ =0, 𝜂 2 = 0.00669438

Δλ=0.02819568 radians or 1.6155° or a distance of approximately 180,000 meters.
GdE 321 Transverse Mercator Projections page 9

Δλ=6°

ko=0.9996
180 km. 180 km.

Isoscale curve

9.3.2 Inverse Problem

To determine the latitude and longitude of a point, first find the footpoint latitude φ1 using iterative
procedure, then find N1, M1, t1, and 𝜂12

𝜑1 𝑋 2 1
𝜑= − 𝑡1 [( ) ] (8.19 𝑎)
𝜇𝑜 𝜇𝑜 2𝑀1 𝑁1

𝑋 1 𝑋 3
𝜆 = sec 𝜑1 [ − ( ) (1 + 2𝑡12 + 𝜂12 ) + ⋯ … … .. ] (8.19 𝑏)
𝜇𝑜 𝑁1 6 𝜇𝑜 𝑁1
Scale Factor
1 X 2
μ = 𝜇𝑜 (1 + ( ) (1 + η12 ) + ⋯ … . ) (8.20)
2 N1

8.1 Transverse Mercator

A transverse Mercator projection is the ordinary Mercator projection turned through a 900
angle so that it is related to a central meridian in the same way that the ordinary Mercator
projections is related to the equator. Its properties are:

1. Cylinder is tangent to the globe at a meridian

2. scale is true along the meridian, which is called the central meridian
3. origin of the map X coordinate.is at the central meridian
4. The origin of the map Y coordinate is the equator.
GdE 321 Transverse Mercator Projections page 10

5. The projection is conformal like the regular Mercator projection.

9.4 – Philippine Transverse Mercator Projection (PTM)

Central
Meridian N

Scale 1:1
Philippine Transverse Mercator Projection

The Philippine Transverse Mercator projection has been adopted as the official projection
system for surveying and mapping in the Philippines in 1961. It has the following characteristics:

1. Transverse Mercator projection is in zones of 2o wide.

2. The reference ellipsoid is Clarke Spheroid of 1866
a= 6,378,206.4 1/f = 294.9786982
3. The point of origin is the intersection of the central meridian and the equator.
5. The unit of measure is the meter.
6. False Northing of 0 m from the equator
7. A false Easting of 500,000.00 is used for the central meridian of each zone.
8. The scale factor at the central meridian is 0.99995
GdE 321 Transverse Mercator Projections page 11

9. The zones are numbered I to V with the following longitude of CM and extent of zones

Zone No. Central Meridian Extent of zone

I 117o 116o 00 to 118o 30’
II 119o 117o 30’ to 120o30’
III 121o 119o 30’ to 122o 30’
IV 123o 121o 30’ to 124o 30’
V 125o 123o 30’ to 127o 00

9.4.1 Geographic to Grid Transformation

The parameters N, t, 𝜂 2 for the direct problem are given by (9.1) to (9.4).
Substituting ko=0.99995, mapping equations become

𝑋 ∆𝜆3 𝑐𝑜𝑠 3 𝜑 ∆𝜆5 𝑐𝑜𝑠 5 𝜑 5 − 18𝑡 2 + 𝑡 4 + 14𝜂 2 − 58𝑡 2 𝜂 2

= 𝑘𝑜 [∆𝜆 𝑐𝑜𝑠 𝜑 + (1 − 𝑡 2 + 𝜂 2 ) + ( ) + ⋯.]
𝑁 6 120 +13𝜂 4 + 4𝜂 6 − 64𝜂 4 𝑡 2 − 24𝜂 6 𝑡 2

𝑌 𝑆𝜑 ∆𝜆2 ∆𝜆4
= 𝑘𝑜 [ + 𝑠𝑖𝑛 𝜑 𝑐𝑜𝑠 𝜑 + 𝑠𝑖𝑛 𝜑 𝑐𝑜𝑠 𝜑 (5 − 𝑡 2 + 9𝜂 2 + 4𝜂 4 ) + ⋯ ]
𝑁 𝑁 2 24

Where:

N=prime vertical

𝑘𝑜 = 𝑠𝑐𝑎𝑙𝑒 𝑎𝑡 𝑡ℎ𝑒 𝑐𝑒𝑛𝑡𝑟𝑎𝑙 𝑚𝑒𝑟𝑖𝑑𝑖𝑎𝑛

Using Helmert formula to evaluate for the meridian arc Sφ

𝑎
𝑆(𝜑) = (𝑎 𝜑 − 𝑎2 sin 2𝜑𝑜 + 𝑎4 sin 4𝜑𝑜 )
1+𝑛 𝑜 𝑜

Where
𝑓
𝑛 = 2−𝑓 =0.00169792

𝑎
= 6,367,395.1; ;
1+𝑛
GdE 321 Transverse Mercator Projections page 12

𝑛2 𝑛4
𝑎𝑜 = 1 + + = 1.000000721
4 64
3 𝑛3
𝑎2 = (𝑛 − ) = 0.002546873
2 8
15 2 𝑛4
𝑎4 = (𝑛 − ) = 0.000002703
16 4

The scale factor at any point within the zone is designated as:
∆𝜆2 cos2 𝜑 ∆𝜆4 cos4 𝜑
𝜇 = 0.99995 [1 + (1 + 𝜂 2 ) + (5 − 4𝑡 2 )] (8.22)
2 24
9.4.2 Inverse problem

The geographic equations for the PTM are:

2
𝜑1 𝑋 1
𝜑= − 𝑡1 [( ) + ⋯ … .] (8.23 𝑎)
0.99995 0.99995 2𝑅1 𝑁1
3
𝑋 1 𝑋
𝜆 = sec 𝜑1 [ − ( ) (1 + 2𝑡12 + 𝜂12 ) + ⋯ … … .. ] (8.23 𝑏)
0.99995𝑁1 6 0.99995𝑁1
Scale Factor
1 X 2
μ = 0.99995 (1 + ( ) + ⋯ . ) (8.24)
2 N1
N1 can be approximated to
𝑅𝑚 = √𝑀𝑁
Eq.8.24 becomes
𝑥2
μ = 0.99995 (1 + + ⋯.) (8.24)
2𝑀𝑁

Azimuths
GdE 321 Transverse Mercator Projections page 13

2
α

Central Meridian
Projected
Projected geodesic
meridian γ T

1
x
Equator

There are three different azimuths to contend with

T= the grid azimuth of the projected geodesic

t = the grid azimuth of the chord
α = the geodetic azimuth of the projected geodesic
The following relationships are valid.

𝑇12 = 𝛼12 − 𝛾
𝑡12 = 𝑇12 − (𝑇 − 𝑡)12
𝑡12 = 𝛼12 − 𝛾 − (𝑇 − 𝑡)12
Where:
γ = convergence of meridian
T-t = arc to chord correction

Arc to chord correction (Bomford,1962)

(𝒚𝟐 − 𝒚𝟏 )(𝒙𝟐 + 𝟐𝒙𝟏 ) (𝟐𝒙𝟏 + 𝒙𝟐 )𝟐

𝑻−𝒕= [𝟏 − ]𝝆 (𝟖. 𝟐𝟓)
𝟔𝑴𝑵 𝟐𝟕𝑴𝑵

Where:

2
𝑅𝑚 = 𝑀𝑁 (8.26)
GdE 321 Transverse Mercator Projections page 14

1
Evaluated at 2 (𝜑1 + 𝜑2 )

Line Scale

A line scale is the average scale over the line. It allows us to obtain the difference between the lengths of the
geodesic and the chord distance.

𝑑 = 𝜇̅ 𝑆 (8.27)

where:
𝜇1 + 4𝜇𝑚 + 𝜇2
𝜇̅ = (8.28)
6

where:
𝑥12
𝜇1 = 𝜇𝑜 [1 + ]
2𝑀1 𝑁1
𝑥22
𝜇2 = 𝜇𝑜 [1 + ]
2𝑀2 𝑁2

𝑥 +𝑥 2
( 1 2 2) 𝑥12 + 2𝑥12 𝑥22 + 𝑥22
𝜇𝑚 = 𝜇𝑜 [1 + ] = 𝜇𝑜 [1 + ]
2𝑀𝑚 𝑁𝑚 8𝑀𝑚 𝑁𝑚

Combining the above equations and simplifying

𝑥12 + 𝑥1 𝑥2 + 𝑥22
𝜇̅ = 𝜇𝑜 [1 + ]
6𝑀𝑁

a more precise value is given by Bomford with slight modification

𝑥𝑢2 𝑥𝑢2
𝜇̅ = 𝜇𝑜 [1 + (1 + )] (8.28)
6𝑀𝑁 36𝑀𝑁

where:

𝑥𝑢2 = 𝑥12 + 𝑥1 𝑥2 + 𝑥22