Transformation to Meta North Pole using rotation matrices

see also: How to set up a meta pole system by rotations

1 0 0 
R1 ()  0 cos  sin  

 
0  sin  cos  
cos  0  sin  
R 2 ( )   0 1 0 
 
 sin  0 cos  
 cos  sin  0 
R 3 ()    sin  cos  0 
 
 0 0 1 

Positive rotation:
Counterclockwise
(ccw) as seen from
the arrowhead
towards the origin

 Transformation using rotation matrices

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 63

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Transformation to Meta North Pole using rotation matrices
Remark: As a matter of course, formulae from spherical trigonometry cannot be
applied in order to make oblique ellipsoidal mappings. Instead, rotation matrices
a best suited to perform the corresponding transformations. Given ellipsoidal
latitude B and longitude L, Cartesian coordinates X,Y,Z (referring to the standard
equatorial system E with standard north pole) are easily computed via
2 A
X  N cos Bcos L , Y  N cos Bsin L , Z  N(1  E )sin B , N 
1  E 2 sin 2 B
Then, in the next step, X,Y,Z are transformed to the (metapole) E*-system using
* T
 X Y Z   R 3 (W0 )R 2 (90  F 0 )R 3 ( L 0 )  X Y Z
* * T

3 n 33 3 n

Finally, ellipsoidal meta-longitude AM and meta-latitude BM (referring to oblique

equatorial system E*, with Meta North Pole M) are obtained by
Y* Z*
A M  arctan * , BM  arctan
X (1  E 2 ) X*2  Y*2
 Transformation to Meta North Pole: Transverse case

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 64

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Transformation to Meta North Pole: Transverse case
If F0 = 0, W0 = /2 und L0 = 3/2 + L0 is set, the variation of L0 generates an
entire family of transverse mappings. E.g. in case of a mapping onto a transverse
cylinder the well-known Gauß-
Krüger- or UTM-strip system is
constructed. The angle W0 defines
the orientation of the coordinate
system.

L, F, L0=0, L0=3/2, F0=0, W0=/2


tanF
tanA =
cosL
sinB = -cosFsinL

 Transformation from Meta North Pole

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 65

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Transformation from Meta North Pole M(L0,F0,W0)

Second task: Given coordinates of Meta North Pole M(L0,F0,W0) and meta
coordinates A,B of a point P, compute its (standard) coordinates L,F referring to
(traditional) North Pole N.

sin(A  W0 )
L  L 0  arctan
sin F 0 cos(A  W0 )  cos F 0 tan B
F  arcsin[sin F 0 sin B  cos F 0 cos(A  W0 )cos B]

Important remark: The change of the aspect does not change the (special)
properties; only the appearance of parameter and coast lines change, as well as the
representation of L1=L2, e.g.. The latter now also include generalized meta
coordinates A,B instead of L,F. Recommended procedure: Express everything in
A,B and substitute the transformation L,F  A,B only at the very end.

 Example: Orthogonal projection in the transverse aspect

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 66

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Example 1: Orthogonal projection in the transverse aspect

Orthogonal projection in the normal aspect

(M=N: L0=0°,F0=90°,W0=0°)
General mapping equations and distortions
  L , r  R cos F 
x  R cos F cos L  X , y  R cos F sin L  Y
L1  1 , L 2  sin F

Orthogonal projection in the oblique aspect

(NM: L0,F0,W0 arbitrary)
General mapping equations and distortions
  A , r  R cos B 
x  R cos Bcos A , y  R cos Bsin A
L1  1 , L 2  sin B
y

x  Example 1: Orthogonal projection in the transverse aspect

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 67

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Example 1: Orthogonal projection in the transverse aspect
Orthogonal projection in the transverse aspect (MN: L0=180°, F0=0°, W0=90° 
metapole=intersection of equator and date line)
B  arcsin( cos F cos L ) , A  arctan(sin L cot F )  W 0
sin B   cos F cos L  cos B  1  sin 2 B
1 cot A
tan(A  W 0 )   cot A  sin L cot F  sin A  , cos A 
1  cot 2 A 1  cot 2 A

Mapping
equations and
distortions
x  R cos F sin L  Y
y  R sin F  Z
L1  1
y L 2  cos F cos L
x  Example 2: Mercator projection in the oblique aspect

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 68

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Example 2: Mercator projection in the oblique aspect
In example 2 of chapter on cylindrical projections,
mapping equations and distortions of the Mercator
projection in the normal aspect are given as:
 F 1
x  RL , y  R ln tan    , L1  L 2 
4 2  cos F
For the oblique case with given metapole M(L0,F0)
and meta prime meridian orientation W0, standard
coordinates L,F are simply replaced by A,B.
 B 1
x  RA , y  R ln tan    , L1  L 2 
4 2 cos B
Preferably, numerical calculations are done using
cos F cos L   X 
*
X  X 
L
     cos F sin L    Y*   R (W )R (90  F )R ( L )  Y  

F   Y  R
  Z     3 0 2 0 3 0  
See also
 sin F   Z 
*
   Z  oblique_mercator_projection
A    x 
* *
arctan(Y / X )  A 
     R 
 B  arctan(Z* / X*2  Y*2 )   y  ln tan   / 4  B / 2  
 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 69

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Few other map projections: General vertical perspective
Azimuthal map projection, e.g. "General vertical perspective projection in the
North polar aspect" ("Google™ Earth projection"):

Given: L, F, H (height of perspective center)

L
R cos F r HR cos F
  r
H  R  R sin F H H  R  R sin F
H cos F

H
 1  sin F
R
H cos F

c  sin F
H
R with c :  1
F min  arcsin
RH
 F  90 R
R x  r cos  , y  r sin 
rmax   H tan F min y
2R
1 x
H  Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 70

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Pseudo-azimuthal, equal area

}
Azimuthal mappings: =L , r=f(F)
Parallels  Circles, Meridians  Straight lines)
 x=r cos  , y=r sin 
Pseudo-azimuthal mappings: =c3L , r=c4f(L,F)
Parallels and Meridians  Complicated curves)

1  c 24 sin 2 F sin c3L c4 sin F

x  c1R 2 , y  c2R 2
1  1  c 42 sin 2 F sin c3L 1  1  c 24 sin 2 F sin c3L

Hammer Projection:
c1=2, c2=1, c3=0.5, c4=1
(parameter line images: algebraic
curves of degree 4)
y

x
 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 71

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Pseudo-azimuthal projections

Hammer-Wagner Projection:
c1=2.66723, c2=1.24104,
c3=1/3, c4=0.90631

Eckert-Greifendorff Projection:
c1=4, c2=1, c3=1/4, c4=1

x  Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 72

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Pseudo cylindrical
Cylindrical mappings: x=RL , y=f(F)
Pseudo-cylindrical mappings: x=RLcosFg(F) , y=Rf(F)
g(F)=1  equidistant parallels, f ' = g -1  equal area (equivalent) mapping

Elliptic pseudo-cylindrical map projection: "Mollweide projection", equal area

Special Kepler equation
2t  sin 2t   sin F
F [] t
0 0
10 0,137.24
20 0, 275.48
30 0, 415.85
40 0,559.74
50 0,709.10
y
60 0,866.98
x 70 1,039.00
80 1, 238.77
2 2
x RL cos t , y  R 2 sin t , 2t  sin 2t   sin F 90 / 2
  Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 73

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Pseudo cylindrical
Parabolic pseudo-cylindrical map
projection: "Craster projection",
equal area projection
3  2F 
x RL  2cos  1
  3 
y
F
y  3R sin
3 x

Rectilinear pseudo-cylindrical map

projection: "Eckert II projection",
equal area projection
2RL
x 4  3sin F
6
 2 
 
y
y R 2  4  3sin F  signF
 3
x

 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 74

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Pseudo cylindrical
Interrupted Goode Homolosine Projection
y

| F |  F 0  4044'11".8 | F |  F 0  4044'11".8
 i 8 
x   L 0  ( L  L i
0 )cos t  x   L i0  (L  L i0 )cos F 
 y  R     y  R  F 
     
 2 sin t  c signF 
2t  sin 2t   sin F, c  0.052 803 527 454 2  Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 75

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Pseudo cylindrical
Interrupted McBryde S3 Projection

| F |  F 0  5551' | F |  F 0  5551'
 i c2  x
x   L 0  ( L  L i
0 )(1  2cos t)     L i0  (L  L i0 )cos F 
 y  R  3
  y  R  F 
  c 2 t  c1signF    
 
2t  4sin t  (4  )sin F, c1  0.069 065 027 204 341 7, c 2  6 / (4  )  Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 76

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Pseudo conical
Conical mappings: =nL, n:=sinF0, r=f(D)=f(/2-F)
Pseudo-conical mappings: =LgDcosD, r=f(D
} 
x=r cos 
y=r sin 

"Bonne projection (1752)", equal area

cos F
 L , r  R(F 0  F  cot F 0 )
F 0  F  cot F 0

y
y F0=60°
x
x F0=20°
 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 77

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Pseudo conical

Waldseemüller map (1507): German postage stamp, special edition 2007

 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 78

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Pseudo conical

"Stab-Werner projection"
equal area

cos F
 L
/ 2  F
r  R( / 2  F )
x  r cos 
y  r sin 

y
x
 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 79

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections
"Berghaus-Star" "August-Epicycloidal" "Oblique Mercator"

"Cassini" "Craig"

 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 80

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Double projection
"Armadillo projection" (E.J. Raisz 1943): Sphere  Torus  Plane
L
x  R(1  cos F)sin
2
 1  sin F 0  cos F 0 L
y  R  sin F cos F 0  (1  cos F )sin F 0 cos 
 2 2
cos(L / 2)
FSouth max   arctan F=F0
tan F 0

Curve Fsouth max x

 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 81

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections
"Littrow" "Collignon"

"Eisenlohr"

"Wiechel"

 Few other map projections (oblique aspect)

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 82

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other oblique map projections (L0=9°,F0=49°,W0=45°)
"Plate-Carrée" "Mollweide"

"Larrivée" "Foucaut"

 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 83

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Other projections by re-numbering
Original map (Equal area
transverse azimuthal mapping) Choose a sector Re-number the sector

Reconstruct
original scale Affine transformation with added meridians

 Few other map projections

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 84

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Other projections by re-numbering

Re-numbered equal area transverse

azimuthal mapping

 Map optimization

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 85

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Optimal cylinder projections
Optimal projections of the sphere 2R 1 , e.g. cylindrical projection in the normal
aspect, conformal with 2 equidistant parallels (secant cylinder)
1st step: Define mapping region, L  [LW, LE], F  [FS, FN]
2nd step: Define local distortion measure. e.g. the Airy criterion  2A (departure from
isometry, squared, averaged)
3rd step: Compute global distortion
measure IA (overall distortion energy)

L FN
1 E
IA     A2 dF
F LW FS
LE F N

F    dF
L W FS

 (L E  L W )[sin(F N )  sin(F S )]
L FN
1 E

(L E  L W )[sin(F N )  sin(F S )] LW 

 IA   2A cos FdFdL
FS
 Map optimization

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 86

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Optimal cylinder projections
Showcase "conformal projection" (LW = -LE= -180, FN=:Fmax, FS= - Fmax)
Distortions: L1=L2=cosF0/cosF
 F max 2 F max 2
1  cos F 0  1  cos F 0 
IA 
4 sin F max  F  cos F   1 cos F dF dL 
sin F max 
0
 cos F

 1  cos FdF

max

1  2   F max  
  cos F 0 ln tan 4 2   2F max cos F 0  sin F max 
sin F max    

Determine optimal F0 in order to minimize IA:

dI A ˆ ˆ ln tan    F max  !
ˆ cos F
(F 0 )  2sin F ˆ
dF 0
0 0 4 2   2sin F 0  0
 
ˆ  F max
 cos F 0
 F 
ln tan   max 
4 2 
 Map optimization

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 87

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Optimal cylinder projections

 Map optimization

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 88

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Optimal conical projection

Standard parallels: B1=48°40', B2=52°40'

 IA = 3.110-7
Optimal parallels: B1=49°6', B2=52°54'
 IA = 2.710-7

 Map optimization

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 89

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Optimal cylinder projections
2
Optimal projections, e.g. transverse Mercator projection of the sphere R 1

Problem: Determine an (optimal) scale factor r of the central meridian L=L0 of a

region (LW,LE)(FS,FN) so as to minimize Airy distortion energy, IA.
Mapping equations (Scale factor r=1, A…meta longitude, B…meta latitude):
 B 1
x  RA , y  R ln tan    , L1  L 2 
4 2 cos B
tan F
tan A  , sin B   cos F sin L
cos L
Mapping equations (Scale factor r=?):
 B
x  rRA , y  rR ln tan   
4 2
r
L1  L 2 
cos B
 Map optimization

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 90

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 A few other map projections: Optimal cylinder projections
1
Target function using  2A  [(L1  1) 2  (L 2  1) 2 ]  (L1  1) 2 :
2
LE F N
1 2
IA     L1 (A, B)  1 cos FdFdL , F  (L E  L W )[sin(F N )  sin(FS )]
F L W FS
r r
Transition A,B  L,F: L1 (A, B)  L1 (L, F)  
2
1  sin B 1  cos 2 F sin 2 L
LE F N 2
1  r 
F LW F  1  cos2 F sin 2 L  cos FdFdL  Min
 IA   1 !
r
S

L FN L FN
d IA 2rˆ E cos FdFdL 2 E cos FdFdL !

F LW F 1  cos2 F sin 2 L  F L 

(rˆ )  ...  0
dr S W FS
2
1  cos F sin L 2

LE F N
cos FdFdL
 
L W FS 1  cos 2 F sin 2 L
 rˆ  LE F N L E  L W  6, FS  F N  80 : rˆ  0, 999 69
cos FdFdL

LW
F 1  cos2 F sin 2 L
S  Ellipsoid-of-revolution

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 91

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Earth model "Ellipsoid-of-revolution 2
"
A,E 2

Geometry: Major semi axis A, squared first eccentricity E2

Parameter lines: U1U=L (ellipsoidal longitude), U2V=B (ellipsoidal latitude)

dS2  N 2 cos 2 BdL2 

 M 2 dB2
 G11dL2  G 22 dB2
A
N
1  E 2 sin 2 B
A(1  E 2 )
M
(1  E 2 sin 2 B)3/ 2

N, M... radii of curvature in the prime vertical, in the meridian

 Mapping the Ellipsoid-of-revolution

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 92

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Mapping the Ellipsoid-of-revolution 2
A,E 2

Cylindrical conformal mappings, polar and transverse aspect

Polar aspect used for countries
with distinctive East-West
extension with central meridian
L0. Never used for global
maps, adopted mainly for legal
maps. North-South map
extension limited to medium
latitudes due to extreme
distortions towards the poles.
Poles are at infinite distance
from the equator.
y  A(L  L0 )  A
 E
 Attention: x and y sometimes
 1  E sin B  2   B  interchanged in geodetic
x  A ln   tan    
 1  E sin B   4 2  literature !
   Mapping the Ellipsoid-of-revolution

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 93

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Mapping the Ellipsoid-of-revolution 2
A,E 2

x, y :
Complicated
functions of
 L  L 0 , b  B  B0
and A, E 2 !!!
Attention: x and y
sometimes
interchanged in
geodetic literature !

Transverse aspect used for countries with distinctive South-North extension with
central meridian L0. Rarely used for global maps, adopted mainly for legal maps.
East-West map extension limited to a few degrees due to increasing distortions
with distance from central meridian L0. If necessary, several maps/mappings with
central meridians L01, L02,… are introduced.
 Mapping the Ellipsoid-of-revolution

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 94

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Mapping the Ellipsoid-of-revolution 2
A,E 2

Strategy for deriving conformal coordinates x, y

Replace well-known conformality postulate L1=L2  [tr(C G-1)]2 = 4 det(C G-1)
by the Cauchy-Riemann differential equations
u v u v
 ,  ,
U V V U
which hold for u1u, u2v, U1U, U2V being both orthogonal and isometric
parameters, accompanied by the Laplace differential equations, which guarantee
that u and v will be continuous functions of U and V.
2u 2u 2v 2v
Du(U, V) : 2
 2
 0 , Dv(U, V) : 2
 2
0
U V U V
Parameters, e.g. ,b, are called isometric if their arc element ds can
be represented in the form of ds2=l2(,b)(d2+db2). l2(,b) is a
position-dependent factor common to both d and db. The grid of
lines =const (b-lines) and b=const (-lines) then consists of
infinitesimal squares; it is an isometric grid. The metric matrix is a
multiple of the identity matrix: g(, b)  l 2 (, b) I 2
 Mapping the Ellipsoid-of-revolution

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 95

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Mapping the Ellipsoid-of-revolution 2
A,E 2

Obviously, planar (Cartesian) coordinates of the chart u1u=x, u2v=y are ortho-
gonal and isometric parameters because g(x, y)  I 2 , i.e. l 2 (x, y)  1. However,
ellipsoidal coordinates U1U=L, U2V=B are orthogonal but no isometric
parameters: G(L,B)  diag(N 2 cos 2 B,M 2 )  L 2 (L,B) I 2.
2
How can we construct an isometric grid on A,E 2
?

dS2  N2 cos2 BdL2  M2dB2

2 2 2 M2 2
 N cos B(dL  2 2
dB )
N cos B
 N cos B(dL  dQ2
2 2 2
)
M
dQ  dB 
N cos B
B  E

M()  1  E sin B  2  B
Q  d  ln   tan(  ) 
0
N() cos   1  E sin B  4 2 
 
Q... ellipsoidal isometric latitude
(ellipsoidal Mercator latitude)
 Isometric coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 96

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Isometric coordinates

L and Q are ellipsoidal isometric or

ellipsoidal Mercator coordinates !

 Conformal coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 97

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates
How to derive conformal coordinates x, y ?
Ansatz: Set up two homogeneous, bivariate polynomials of degree n, which map
a point P with given isometric longitude L, isometric latitude Q on 2A,E2 onto x,y
in the chart. These coordinates shall refer to the origin of a (local) coordinate
system fixed at a certain point P0(L0,Q0) with given isometric longitude L0,
isometric latitude Q0.
x (L  L 0 ,Q  Q 0 ) : x( , q)  (n,0) x q n  (n  1,1) x q n 1  (n  2, 2) x q n  2 2
 ...  (0, n) x n

n
 
i  0,1,...
(i, j) x q i j

2n+2 unknown coefficients

j 0,1,...
i  j n

y(L  L 0 ,Q  Q 0 ) : y( , q)  (n,0) y q n  (n  1,1) y q n 1  (n  2, 2) y q n  2 2

 ...  (0, n) y n

n
 
i  0,1,...
(i, j) y q i j

j 0 ,1,...
i  j n

Unknown coefficients are functions of ellipsoid geometry and latitude Q0.

 Conformal coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 98

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates
x (L  L 0 ,Q  Q 0 ) : x( , q)  (n,0) x q n  (n  1,1) x q n 1  (n  2, 2) x q n  2 2
 ...  (0, n) x n

y(L  L 0 ,Q  Q 0 ) : y( , q )  (n,0) y q n  (n  1,1) y q n 1  (n  2, 2) y q n  2 2

 ...  (0, n) y n

 Conformal coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 99

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates
Implement Laplace differential equations Dx(L,Q)=0, Dy(L,Q)=0

Showcase Dy(L,Q)=0 only !

1st and 2nd derivative wrt :
y  (n  1,1) y q n 1  2(n  2, 2) y q n 2  3(n  3,3) y q n 3 2
 4(n  4, 4) y q n 4 3
 ...  n(0, n) y n 1

y  2(n  2, 2) y q n 2  6(n  3,3) y q n 3  12(n  4, 4) y q n 4 2

 ...  n(n  1)(0, n) y n 2

1st and 2nd derivative wrt q:

y q  n(n,0) y q n 1  (n  1)(n  1,1) y q n 2  (n  2)(n  2, 2) y q n 3 2
 ...  (1, n  1) y n 1

y qq  n(n  1)(n,0) y q n 2  (n  1)(n  2)(n  1,1) y q n 3  ...  2(2, n  2) y n 2

Check : Dy( ,q)  y  y qq  0

 2(n  2, 2) y  n(n  1)(n,0) y 
 (n  k, k) y k(k  1)
 ...  (n  k  2, k  2) y   , k  2,..., n
(n  k  2)(n  k  1)
 n(n  1)(0, n ) y  2(2, n  2) y 
Conclusion 1: As a result of the Laplace differential equation Dy(L,Q)=0 unknown
coefficients (i,j)y are interrelated. Some of them can be eliminated in
favor of the others thus reducing their total number. The same is true for
unknown coefficients (i,j)x.  Conformal coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 100

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates
Implement Cauchy-Riemann differential equations x/L = y/Q, x/Q = - y/L
1st derivative of x wrt and y wrt q:
x  (n  1,1) x q n 1  2(n  2, 2) x q n 2  3(n  3,3) x q n 3 2
 4(n  4, 4) x q n 4 3
 ...  n(0, n) x n 1

y q  n(n,0) y q n 1  (n  1)(n  1,1) y q n 2  (n  2)(n  2, 2) y q n 3 2

 ...  (1, n  1) y n 1

Check: x  y q
 (n  1,1) x  n(n,0) y , 2(n  2, 2) x  (n  1)(n  1,1) y , ...
1st derivative of x wrt q and y wrt :
x q  n(n,0) x q n 1  (n  1)(n  1,1) x q n 2  (n  2)(n  2, 2) x q n 3 2
 ...  (1, n  1) x n 1

y  (n  1,1) y q n 1  2(n  2, 2) y q n 2  3(n  3,3) y q n 3 2

 4(n  4, 4) y q n 4 3
 ...  n(0, n) y n 1

Check: x q   y
 n(n,0) x  (n  1,1) y , (n  1)(n  1,1) x  2(n  2, 2) y ...

Conclusion 2: As a result of the Cauchy-Riemann differential equations, coefficients

(i,j)x and (i,j)y are mutually interrelated. Some coefficients (i,j)x can be
replaced by coefficients (i,j)y thus further reducing the total number of
unknown coefficients.
 Conformal coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 101

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates
Final result: y( ,q)   0  1q  b1   2 (q 2  2
)  b2 (2q )  ...
x( ,q)  b0  b1q  1  b2 (q 2  2
)   2 (2q )  ...
i,bi... i=0,... remaining unknown coefficients which must be determined
from additional conditions/postulates.
Postulate/case 1:
Equator (B = Q = Q0 = 0) or
a part of it shall be mapped
equidistantly !
 y(l,q=0) = Al
x(l,q=0) = 0
 b1A, all other coeffi-
cients vanish !

 y(l,q) = Al
x(l,q) = Aq
 Conformal coordinates – Case 1

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 102

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates – Case 1
Normal conformal mapping of the ellipsoid-of-revolution onto a cylinder/plane

y( , q)  A
  E
  E

  1  E sin Final
B  2 mapping
  Bequations
   1  E sin B0  2   B0  
x( , q)  Aq  A ln   tan     ln   tan    

  1  E sin BSee
 Lab3, 2 
 4 Exercise
 3   1  E sin B0  4 2  
 

 E

B0  0
 1  E sin B  2   B 
 A ln   tan    
 1  E sin B   4 2 
 

because q:= Q - Q0, with Q and Q0 , respectively, the isometric

latitudes computed from B and B0 , respectively, according to
B( 0 )
M(B* )
 "ellipsoidal Mercator latitude"
*
Q(0)  dB
0
N(B* )cos B*
 Conformal coordinates – Case 2

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 103

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates – Case 2
Postulate/case 2: Given meridian L=L0 ("reference meridian") or a part of it,
DG, shall be mapped equidistantly !

Mapping equations: y( ,q)   0  1q  b1   2 (q 2  2

)  b2 (2q )  ...
x( ,q)  b0  b1q  1  b2 (q 2  2
)   2 (2q )  ...
DG equidistant !  x(l=0,q) = DG(Q,Q0)  b01q+b
+ 1bq2+
q2b+2qb23+
q3b+3q...3 += ...
DG(Q,Q 0)
= DG(Q,Q 0)
If q=0, then x=0 !  b0 = 0 "no offset"
Origin at y=0 !  y(l=0,q) = 0  i = 0  i = 0,...
 Conformal coordinates – Case 2

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 104

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates – Case 2
! 1st step: Provide bi
b1q + b2 + b3 + ... = DG(Q,Q0)
q2 q3
2nd step: Provide qi
Taylor series expansion for the length G of a meridional arc starting at isometric
latitude Q0 (local origin P0 !) ending at isometric latitude Q:
dG 1 d 2G 2 1 d 3G
G(Q)  G(Q0 )  (Q  Q 0 )  (Q  Q 0 )  (Q  Q 0 ) 3  ...
dQ Q0 2! dQ 2 Q0
3! dQ3 Q0

 DG(Q,Q0 ) :
dG 1 d 2G 21 d 3G
G(Q)  G(Q0 )  (Q  Q 0 )  (Q  Q 0 )  (Q  Q 0 ) 3 ...
dQ Q0 2! dQ 2 Q0
3! dQ3 Q0

 b1 q  b2 q2  b3 q3 ...
1 diG
 bi 
i! dQi Q0

"Change of meridional arc length with isometric latitude"

 Conformal coordinates – Case 2

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 105

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates – Case 2
1 diG
1st step: Determine coefficients bi  in terms of B0(Q0)
i! dQi Q
0

Knowledge so far: (1) Length G(B) of a meridional arc up to ellipsoidal latitude B,

B B
G(B)   dSLconst   M(B')dB' and its change,
0 0
dG/dB=dSL=const/dB=M, respectively.
(2) change dQ/dB of isometric latitude Q through change of
ellipsoidal latitude B: dQ/dB=M/NcosB
1
dG dG dB dG  dQ  N(B0 )cos B0
b1       M(B0 )  N 0 cos B0
dQ Q0 dB B0 dQ Q0 dB B0  dB B0  M(B0 )
1
2
1d G 1  d dG  1  d dG dB  1 db1 dB 1 db1  dQ 
b2         
2 dQ 2 Q0
2  dQ dQ  Q 2  dB dQ dQ  Q 2 dB0 dQ Q0 2 dB0  dB B0 
0 0

1
  N 0 cos 2 B0 tan B0
etc. 2
 Conformal coordinates – Case 2

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 106

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates – Case 2
2nd step: Determine powers qi of q:= Q-Q0 in terms of b:=B-B0
thus replacing isometric by ellipsoidal latitudes

dQ 1 d 2Q
Q  Q0  (B  B0 )  2
(B  B0 ) 2  ...
dB B0 2! dB B
0


1 d mQ
 Q0   m
(B  B 0 ) m

m 1 m! dB B
0


1 d mQ
 Q0   m
bm
m 1 m! dB B
0


1 d mQ
 q : Q  Q 0   m
bm
m 1 m! dB B
0

i
  1 d mQ  dQ M
 q  
i
m
m
b  i  1,... and 
 m 1 m! dB B0  dB N cos B

 Conformal polynomials

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 107

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal polynomials

Mapping equation for conformal coordinate "Easting" (l:=LP-L0, b:=BP-B0)

i  0 i
2 2
y( ,q)   0  1q  b1   2 (q  )  b2 (2q )  ... 
 b1  2b2q  ...
Back trans-
formation
 (01) q b
+ (11) b
+ (21) b 2  (03) 3

 (31) b3  (13) b 3

 (41) b 4  (23)b 2 3
 (05) 5

 (33)b3 3
 (15) b 5

no index y anymore ?!  (25)b 2 5

...

Coefficients (ij) have to be evaluated at latitude B0 of the local origin !

See: http://www.gis.uni-stuttgart.de/lehre/campus-docs/geo2gk.pdf
 Conformal polynomials

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 108

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal polynomials

Mapping equation for conformal coordinate "Northing" (l:=LP-L0, b:=BP-B0)

b0  0, i  0 i
2 2
x( ,q)  b0  b1q  1  b2 (q  )   2 (2q )  ... 
 b1q  b2 (q 2  2
)  ... Back trans-
formation
 (10) b q b
+ (20) b 2  (02) 2

 (30) b3  (12)b 2

 (40) b 4  (22)b 2 2
 (04) 4

 (32)b3 2
 (14)b 4

no index x  (42)b 4 2
 (24)b 2 4
 (06) 6
anymore ?!
6
 (16)b  ...

Coefficients (ij) have to be evaluated at latitude B0 of the local origin !

See: http://www.gis.uni-stuttgart.de/lehre/campus-docs/geo2gk.pdf
 Conformal coordinates: Series coefficients

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 109

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates: Series coefficients
3m 0 m m
(10)  m 0 N (1  2  4  6 ), (20)  N t (2  2 4 ), (02)  0 N cos 2 B t, (32)  0 N cos 2 B (4  4 t 2 )
2 2 12
m m m
(30) = 0 N[2 (1  t 2 ) +4 (  2  7 t 2 )], (12) = 0 N cos 2 B (1  t 2  2 t 2  4 t 2 ), (40)   0 N 2 t
2 2 2
m m m
(22)  0 N cos 2 B t [4  3 2 (1  t 2 )], (04)  0 N cos 4 B t (5  t 2  9 2 ), (41)  0 N cos B
4 24 24
m m m
(14)  0 N cos 4 B[5  18 t 2  t 4  2 (9  40 t 2  t 4 )], (42)  0 N cos 2 B t, (24)  0 N cos 4 B t ( 7  5 t 2 )
24 3 6
m m
(06)  0 N cos6 B t (61  58 t 2  t 4 ), (16)  0 N cos 6 B (61  479 t 2  179 t 4  t 6 ), (01)  m 0 N cos B
720 720
m coefficients (ij) have to be
(11) =m 0 N cos B t (1  2  4 ), (21) = 0 N cos B[1  2 (1  3 t 2 )  4 (1  6 t 2 )] evaluated at latitude B0 of the
2 local origin !
m m m
(03)  0 N cos3 B (1  t 2  2 ), (31)  0 N cos B t [1  2 ( 10  3 t 2 )], (15)  0 N cos5 B t ( 61  58 t 2  t 4 )
6 6 120
m m
(13)  0 N cos3 B t [5  t 2  2 (4  t 2 )], (23)  0 N cos 3B[ 5  13 t 2  2 ( 4  8 t 2  3 t 4 )]
6 12
m0 3 2 2 E2
(33)  N cos B t (41  13 t ), t=tan(B),   2
cos 2 B and m 0 : scale factor of the reference meridian
36 1 E

see also: http://www.gis.uni-stuttgart.de/lehre/campus-docs/geo2gk.pdf

 Conformal coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 110

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates
given: P(LP,BP) [,L0]
choose: P0(L0,B0)
calculate: l=LP-L0, b=BP-B0
determine: x(l,b), y(l,b) and X(l,B)=X0+x(l,b)
Remark 1: X0/m0 is the meridional arc length
G of the local origin P0
Remark 2: choose P0 so that series for x(l,b)
and y(l,b) converge, e.g.
Strip width DL+
max  , b  1
0.5° overlap
depending on the expansion
degree of conformal polynomials
Remark 3: For B0=BP bivariate polynomials

! degenerate into univariate

polynomials (b  0) !

 Latitude  Meridional arc length

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 111

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Arc element dS  Meridional arc length G

B0 B0 B
A(1  E 2 ) 0
A 1  E '2
G(B0 )   M(B)dB   2 2 3/ 2
dB   2 2 3/ 2
dB
0 0
(1  E sin B) 0
(1  E ' cos B)
 elliptic integral of the second kind
 Series expansion of the integrand in terms of E 2 or E'2 and transition
from powers of cosB/sinB to sines/cosines of multiples of B
 Latitude  Meridional arc length

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 112

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Latitude B [rad]  Meridional arc length G

G(B)=
A[e0B+e2sin(2B)+e4sin(4B)+e6sin(6B)+e8sin(8B)+e10sin(10B)]

1 3 5 6 175 8 441 10 1 13 45 6 2577 8 9417 10

e0  1  E 2  E 4  E  E  E e0  1  E '2  E '4  E'  E'  E'
4 64 256 16384 65536 4 64 256 16384 65536

3 3 45 6 105 8 2205 10 3 9 237 6 819 8 23325 10

e2   E2  E4  E  E  E e2   E '2  E '4  E'  E'  E'
8 32 1024 4096 131072 8 32 1024 4096 131072

15 4 45 6 525 8 1575 10 15 4 75 1245 8 4905 10

e4  E  E  E  E e4  E'  E '6  E'  E'
256 1024 16384 65536 256 1024 16384 65536

35 6 175 8 3675 10 35 245 19985 10

e6   E  E  E e6   E '6  E '8  E'
3072 12288 262144 3072 12288 786432

315 2205 10 315 2835

e8  E8  E e8  E '8  E '10
131072 524288 131072 524288

693 693
e10   E10 e10   E '10
1310720 1310720

 Conformal coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 113

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates
Remark 4: since l is limited and distortions on the boundary of the mapped region
should be as small as possible, a system of neighboring reference
meridians is constructed.

 Conformal coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 114

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates: UTM Strip System
http://www.gis.uni-stuttgart.de/lehre/campus-docs/UTM_Strip_System.mp4

 Conformal coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 115

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates
Remark 5: Two meridian strips next to each other are overlapped artificially by
0,5°. A point P in the overlapping region is coordinated with respect
to both reference meridians L0i und L0i+1.

 Gauß-Krüger coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 116

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Gauß-Krüger coordinates: Northing and False Easting
The region west and east of the reference meridians has – in Germany and disre-
garding the overlap of 0,5° – a width of 1,5° (3°-systems). Other countries may
use different strip widths. Widely used are 6°-systems, e.g. in military applications
or in new reference systems, e.g. the European reference system ETRS89  UTM-
coordinates.
Structure of the coordinates:
In order to keep reference to the meridian L0 and in
order to avoid negative coordinates for points west
of it, the y-coordinate is augmented using the
following rules and converted to False Easting R :
1) Add the value (meridian reference number)
106  L0 / 3 [m]
2) Add a constant of 500 000 m = 500 km

 False Easting R= y(l,b) +106L0/3°+500 000 [m]

 Northing H = X0 + x(l,b) specifies the distance
from the equator.
Meridian strip system in Germany  Gauß-Krüger coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 117

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Gauß-Krüger coordinates: Northing and False Easting

Example 1:
Point 1 has ellipsoidal coordinates L=6.8°,
B=51.2° with respect to the Bessel
ellipsoid-of-revolution (ABessel,E2Bessel) 
y1GK = 55 909.151 m, X1GK = 5 674 057.263 m
Northing H = 5 674 057.263 m
False Easting R = 2 555 909.151 m

(Point 1 is located 55 909,151 m east of its

reference meridian with reference number
2, i.e. L0=6° and 5 674 057.263 m north of
the equator).

Meridian strip system in Germany

 Gauß-Krüger coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 118

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Gauß-Krüger coordinates: Northing and False Easting
Example 2:
Point 6 has ellipsoidal coordinates L=13.75°,
B=53.8° with respect to the Bessel ellipsoid-
of-revolution (ABessel,E2Bessel) 
y1GK = 115 287.428 m, X1GK = 5 964 460.428 m
Northing H1 = 5 964 460.428 m
False Easting R1 = 4 615 287.428 m
(Point 6 is located 115 287.428 m east of its
reference meridian with reference number 4, i.e.
L0=12° and 5 964 460.428 m north of the equator).
y2GK = -82 350.056 m, X2GK = 5 963 764.424 m
Northing H2 = 5 963 764.424 m
False Easting R2 = 5 417 649.944 m
(Point 6 is located 82 350.056 m west of its
reference meridian with reference number 5, i.e.
L0=15° and 5 963 764.424 m north of the equator).
Meridian strip system in Germany
 UTM coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 119

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 UTM coordinates: False Northing and False Easting

Gauß-Krüger coordinates as well as

Universal Transverse Mercator- (UTM-)
coordinates are conformal coordinates.
Both have identical theoretical
background. However, UTM-
coordinates are equipped with an
additional scale factor m0<1 with the
consequence that the strip width can be
extended (to usually 6°) while at the
same time keeping the distortions within
a reasonable limit. Thus the reference
meridian will not be mapped
equidistantly but two meridians west and
east of it ! Geometric interpretation:
Transverse elliptic secant cylinder.

 UTM coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 120

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 UTM coordinates: False Northing and False Easting
Scale factor m0 is computed from an optimization problem: Minimize Airy
distortion energy
B L B L
1 1 North East North East

 AdF  F B  L  [L(L, B)  1] N M cos BdL dB, F  B  L  N M cos BdL dB

2 2
I A :
F meridian South West South West
strip

over a region (LWest-LEast, Bsouth-BNorth). Scale factor m0 depends on the extension

of the region and on the geometry of the ellipsoid-of-revolution chosen. The
conventional international factor is m00,999 6, but also values such as m00,999
587, m0 0,999 585 or m00,999 579 can be found.
Structure of the coordinates:
Reference to the reference meridian is installed using Zone number (L0+3°)/6°+30
which completes the pair Northing/Easting. Reference meridians are
L0 = -177° = 177° western longitude (w.L.), L0  177°, ..., L0   3°, L0  3°, ...,
L0 = 177° = 177° eastern longitude (e.L.). Germany covers zones 32 and 33.
Eastings are modified by an offset of 500 000 m ( False Eastings), Northings for
southern latitudes (x < 0 !) by an offset of 10 000 000 m ( False Northings).
Strips are overlapped by 0,5°.
 UTM coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 121

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 UTM coordinates: False Northing and False Easting
Example: Point 4 has coordinates L=11.75°, B=54.3°
with respect to the GRS80-ellipsoid (AGRS80, E2GRS80)
 y1UTM = 178 940.792 m, X1UTM= 6 020 389.079 m.
False Northing = 6 020 389.079 m
False Easting = 678 940.792 m
Zone = 32
(Point 4 is located 178 940.792 m east of its
reference meridian of zone 32, i.e. L0=9°, and
6 020 389.079 m north of the equator).
y2UTM = -211 465.204 m, X2UTM = 6 021 773.311 m
False Northing = 6 021 773.311 m 32 33
False Easting = 288 534.796 m
Zone = 33
(Point 4 is located 211 465.204 m west of its
reference meridian of zone 33, i.e. L0=15°, and
6 021 773.311 m north of the equator). UTM-Meridian strip system in Germany

 UTM coordinates

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 122

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 UTM coordinates: False Northing and False Easting

Zone(s) and UTM-grid

Ellipsoidal coordinates with

respect to a reference ellipsoid
with a geometry specified in the
map legend

Easting and Northing with

respect to the reference
meridian of zone 32

Easting and Northing with

respect to the reference
meridian of zone 33

Topographische Karte © Bayer. Vermessungsverwaltung; 7/15

 UTM-grid

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 123

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 UTM-grid

 Conformal coordinates: Meridian convergence

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 124

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates: Meridian convergence

Which lines are straight ?

Which lines are orthogonal ?
Which lines are parallel ?
Which line(s) point north ?

How can north be found in P ?

How can azimuth  of the line
PQ be found ?

Grundlage: Abbildungen –  Landesamt für Geoinformation und Landentwicklung Baden-Württemberg (www.lgl-bw.de) 06.07.2015, Az.: 2851.3-A/896

 Conformal coordinates: Meridian convergence and distortion

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 125

© F. Krumm, Institute of Geodesy, Stuttgart University 2018
 Conformal coordinates: Meridian convergence
Meridian convergence c (Azimuth , grid bearing T:  =T+c)

x x / 
tan c  
y y / 

Grundlage: Abbildungen –  Landesamt für Geoinforma-tion und Landentwicklung Baden-Württemberg (www.lgl-bw.de) 06.07.2015, Az.: 2851.3-A/896

 Conformal coordinates: Meridian convergence and distortion

Map Projections and Geodetic Coordinate Systems Rev. 2.7d 126

© F. Krumm, Institute of Geodesy, Stuttgart University 2018