PE&RS July 2016 Public - page 474

474
July 2016
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
For instance, the 1874 Samarkand Baseline was 1,605
sazhens or 3,424 meters.
The early baselines were measured with wooden rods,
however no information is available on what kind of wood
was used. In the United States during the same time
period, Magnolia wood was used for leveling rods because
it was believed to have the lowest coefficient of expansion
(from variations in temperature AND humidity) of all wood
species. Note that the Americans (Coast & Geodetic Survey)
still boiled their wood in paraffin just to make sure. (There’s
an interesting geodetic story on boiled wood I’ll have to tell
someday about Palmdale, CA.)
In order to make a topographic survey at 1:4,000 scale
of the City of Tashkent, a traverse net was established
in 1885 by Pomerantsev. The revised coordinates of the
Tashkent Observatory SW pillar used for the city survey
are: Φ
o
= 41
o
19’ 30.42” North, Λ
o
= 38
o
58’ 00.99” East of
Pulkovo, (or 69
o
17’ 39.54˝ East of Greenwich). The vertical
datum of Tashkent was determined by means of barometric
observations at the Meteorological Station of the Military
Topographic Department located in the residence of Colonel
Zhemchuznikov. The transfer of this vertical datum point
value to other horizontal datum origins with trigonometric
leveling techniques produced geodetic problems of mind-
boggling proportions!
The physical connection between the Tashkent Datum
of 1875 and the Indian Datum of 1916 in the Pamir region
was made by an exploratory triangulation party of the
International Geodetic Union. The expedition was led
by Professor Finsterwalder in the early part of the 20th
century. The 3-parameter shift from Tashkent to Indian is:
dX = -223.632 meters, dY = –281.310 meters, dZ = +304.059
meters. (The fit of two points agree to better than a meter.)
The Tashkent Datum of 1875 origin (at the Meridian Circle)
is: Φ
o
= 41
o
19’ 30.42” North, Λ
o
= 69
o
17’ 39.54˝ East of
Greenwich. The defining azimuth at the point of origin to the
North Stone Pillar (probably the mire) of the Observatory is:
α
o
= 00
o
52’ 08.25”, and the ellipsoid of reference is the Bessel
1841. The Indian Datum of 1916 origin at Kalianpur Hill
Station is: Φ
o
= 24
o
07’ 11.26” North, Λ
o
= 77
o
39’ 17.57˝ East
of Greenwich. The defining azimuth at the point of origin
to station Surantal is: α
o
= 190
o
27’ 05.10”. The ellipsoid of
reference is the Everest 1830, where a = 6,377,276.345
meters, and
1
/
f
= 300.8017.
The Kazalinsk Datum of 1891 origin at the finial Cross on the
Town Church is: Φ
o
= 45
o
45’ 46.450” North, Λ
o
= 62
o
06’ 01.66˝
East of Greenwich). The defining azimuth at the point of
origin to station Sulutan is: α
o
= 20
o
34’ 07.34”. The ellipsoid
of reference again is Bessel 1841. The 3-parameter shift from
Kazalinsk 1891 to Tashkent 1895 is: dX = +530 meters, dY =
– 160 meters, dZ = – 104 meters. (The fit of four points agrees
to about 20 meters in each geocentric component.)
TheOshDatumof 1901 origin is definedat Point II,Northwest
Base as: Φ
o
= 40
o
37’ 16.670” North, Λ
o
= 72
o
56’ 11.175” East
of Greenwich). The defining azimuth at the point of origin to
Point I, Southeast Base is: α
o
= 152
o
54’ 01.86”. The ellipsoid
of reference again is Bessel 1841. The 3- parameter shift from
Osh 1901 to Tashkent 1895 is: dX = – 146.633 meters, dY
= +472.553 meters, dZ = +508.352 meters. (The fit of seven
points agrees to 4.19 meters in each geocentric component.)
In order to better model the shift and reduce the fit errors,
I decided to try a Bursa-Wolfe 7-parameter shift. The results
yielded an average fit error in Latitude of 0.48 meters, the
error in Longitude was 1.09 meters, and the error in Height
was 0.09 meters. The parameters that give this fit are: dX =
+26.82 meters, dY = –183.11 meters, dZ = – 186.25 meters,
scale = 39.25 x 106, R
z
= – 07.79”, R
y
= +2.48”, R
x
= +21.49”.
As a computational “check point,” Point II, Northwest Base
on the Tashkent Datum of 1875 has the geodetic coordinates:
φ = 40
o
36’ 55.740” N, λ = 72
o
56’ 22.880” E.
Uzbekistan has only one Grid system, and that is based
on the Russia Belts, a series of Gauss-Kruger Transverse
Mercator projections that use the exact same rules and
parameters as UTM with two exceptions. The Krassovsky
1940 ellipsoid is used everywhere in the former USSR where
a = 6,378,245.0 meters, and 1/ f = 298.3. The scale factor at
origin is 1.0 rather than the 0.9996 scale factor used for UTM.
The single unifying datum used in all of the former USSR
is the “Coordinate System 1942,” a consistent marvel that is
the largest classical datum in the world! This datum is often
improperly referred to after its origin, Pulkovo Observatory,
where: Φ
o
= 59
o
46’ 18.55” North, Λ
o
= 30
o
19’ 42.09” East
of Greenwich. The defining azimuth at the point of origin
to Signal A is: α
o
= 317
o
02’ 50.62”. The latest 3-parameter
shift values published by the National Imagery and Mapping
Agency for the vicinity listed as Kazakhstan is from System
42 to WGS84, such that: dX = +15 meters, dY = – 130 meters,
dZ = – 84. The stated accuracy is 25 meters in each component,
which is the maximum error useable for 1:50,000 mapping,
and that was based on only two points.
U
pdate
A recent publication by Mirmakhmudov, E., Prenov Sh.,
Magdiev, Kh., and Fazzilova, D. on (the)
Intermediate
reference frame for Uzbekistan topographic maps,
United Nations/Russian Federation Workshop on the
Applications of Global Navigation Satellite Systems,
Krasnoyarsk, 18-22.05.2015 states that Fazilova found in
2002 the transformation parameters
from
CS-42
to
WGS84
are:
D
X = +23 m,
D
Y = – 125 m,
D
Z = – 87 m. However, no
accuracy estimate is offered.
The contents of this column reflect the views of the author, who is
responsible for the facts and accuracy of the data presented herein. The
contents do not necessarily reflect the official views or policies of the
American Society for Photogrammetry and Remote Sensing and/or the
Louisiana State University Center for GeoInformatics (C
4
G).
This column was previously published in
PE&RS
.
447...,464,465,466,467,468,469,470,471,472,473 475,476,477,478,479,480,481,482,483,484,...582
Powered by FlippingBook