UTM & MGRS Coordinate System History
In 1994 a friend of mine made a request to the Pentagon for information on the origin of the UTM and MGRS coordinate systems. My friend handed off the documents he received to me in 2015. I've scanned them into pdf files and had them retyped so they would be both more readable and searchable. I've also replaced a missing page and line that's not in the pdf files.
The documents received date from the 1940's and clearly show that the requirements of artillery gunners were the driving force in the design of map coordinate systems.
Response from John W. Hager Geodesist with the Defense Mapping Agency
A link to a pdf of the original document
DEFENCE MAPPING AGENCY
HYDROGRAPHIC/TOPOGRAPHIC CENTER
SCIENTIFIC DATA DEPARTMENT
GEOPOSITIONING DIVISION
GEOPOSITIONING BRANCH V
Details of the origin of the Military Grid Reference
System (MGRS) have become obscure with the passage of time.
The earliest information available in this office, the Study
and Discussion of Military Grids, Probably early 1947,
contains all the specifications of the Universal Transverse
Mercator (UTM) grid. This study does not mention the MGRS.
A memorandum from the Commanding Officer of the Army Map
Service to the Chief of Engineers, 6 December 1946, (Encl 2
to the above study) gives examples of grid references. It is
noted that the 100,000-meter grid square letters, an integral
part of the MGRS, are not here used in the grid references.
The letter from Brigadier Martin Hotine, Directorate of
Military Survey , (Encl 3 to the above study) states that the
specifications for a six-degree Transverse Mercator world-
wide system were communicated to the British in October 1945.
On 27 February 1948, the Joint Mapping Photography Committee
Ad-hoc Committee on Universal Military Grid Referencing
System proposed a grid referencing system containing all
the specifications of the MGRS as it is currently used.
The Joint Intelligence Committee (JIC) of the Joint
Chiefs of Staff issued JIC Papers 410/1 and 410/2, approved
14 November 1949, prescribing the use of the Universal
Transverse Mercator Grid, the Universal Polar Stereographic Grid
and the Military Grid Reference System by all branches of the
Armed Forced for joint operations use.
AGAO-S 061.3 (28 Dec. 49) CSGID-M, Issued 29 December
1949 by the Office of the Adjutant General, Department of the
Army, prescribed the use of those grids and grid reference
system for the Department of Army.
Army Map Service Technical Manual No.36, Grids and Grid
References, January 1950, elaborated the details of the
military grids and grid reference system. This was subsequently
replaced by Department of the Army Technical Manual 5-241-1 ,
same title. The latest elaboration of these details is
contained in Defense Mapping Agency Technical Manual 8358.1,
Datums, Ellipsoids, Grids, and Grid Referencing Systems.
JOHN W. HAGER
Geodesist
Brigadier Hotine's letter, London 1945
This is one of the early documents articulating the desire for a world-wide grid system.
A link to a pdf of the original document
Directorate of Military
Survey, War Office.
12th November, 1945.
To:-
Colonel A.G. Matthews,
Chief, Intelligence Division,
c/o ST & D (RE)
1801 K Street N.W.
Washington, D.C.
Dear Matty,
Many thanks for your letter of October 26th regarding proposed new world-
wide grid systems.
I think there can be no doubt that the polyconic is a bad military grid
because it is not orthomorphic and does not therefore give the required
degree of accuracy in rapid "plane" computations for range and bearing; or the
same facility as an orthomorphic projection in rapid small instrumental
surveys interpolated by "plane" methods.
The present haphazard plaster of grids which has grown up all over the
world (particularly in British areas of responsibility) is also a headache
even though these are orthomorphic. They frequently lead to junctions
in awkward places, although there is no reason to suppose that the junction bug-
bear will be overcome by any cast-iron 6-degree system; or indeed by any grid
system at all.
The main point I think is that before we change at all we want to make
quite certain that we are changing in the direction of the far future and
not merely to meet some transient consideration. It is bound to be many years
before we get on to a new system and we shall merely have had the vast labour
and confusion for nothing if by then we have changed our minds again.
We have been considering this matter at this end and have decided at any
rate to try out a "mesh" system based upon the graticule rather that the use
of a "plane" grid at all.
The advantage of a mesh system are as follows:
(a) It would obviate all grid junctions everywhere at any rate on
the smaller scale maps. There would, however, be come
discontinuity remaining on large scale maps across the boundary of
disparate survey systems (e.g.:- the frontier of two countries)
where fundamental geographic positions are not in sympathy.
(b) It would vastly facilitate inter-service cooperation. The Navy,
for instance, always work on some sort of graticule system of
reference if they can and are only induced to accept "plane"
grids for bombardment purposes under protest. The Air Force similarly
hate grid junctions which always fall awkwardly for such purposes as
fighter defense. Many anti-air defence systems cannot in any case
operate across a grid junction, e.g.:- the use of "fruit machines"
for vectoring defending aircraft.
The question has for instance arisen in particularly acute form in
relation to Coast Artillery which may otherwise be forced to switch at
very short notice between no less than three grids; one for landward
firing in support of ground operations; one for seaward firing in
conjunction with the Navy; and one at the shortest notice for employment
in an A.A. role.
The disadvantages of a mesh system may be summarized as :-
(a) Computations for range and bearing will not be simple,
although it is likely that for the most rapid purposes the
introduction of a scale factor in one direction will be sufficient:
the value of the scale factor being suitably broadcast by, for
instance, marginal information on maps.
(b) All trig. lists would have to be cast into the graticule. We
should no longer be able to use the results of foreign surveys
neat in their own native projection.
(c) The vast cost and probable confusion over a long period of time
involved in any change.
We are, however, trying out a mesh system in experimental areas in
conjunction with the Artillery. I do not know what the answer will be
but I certainly think that we must go into this much broader question
in detail before we make any alteration whatever.
If it is decided to stick to grid and to introduce a world-wide
system then I think the six-degree Transverse Mercator proposal is as
good as any. One advantage of it which has probably been brought to your
notice is that the Russians do this and have adopted the six-
degree belts of the International 1/1M map. The Germans were also
proposing to do so, We get a considerable portion of the globe already
covered for us on this system therefore (always assuming that we can get
any data out of the Russians, which is doubtful). Conversely it may be
considered an advantage to the Russians for us to facilitate their study
of our surveys and to utilise them. This aspect of question very
soon runs into deep water.
Although the introduction of a world system of grids such as the
Transverse Mercator proposal looks very tidy i doubt very much whether
it would work out quite as tidily. A meridian boundary, in the nature
of things must always ignore such factors as a grid junction falling
awkwardly in a possible battle area and also such factors as the
straight utilisation of National surveys, which of course are placed on
a National projection rather than a purely geographic one. We might
accordingly find that we had undertaken all the disadvantages of change
for very little if we were to adopt such a stereotyped proposal.
Another question that arises is the choice of a unit. We are
likely to standardize on the metre of the grid systems. The main reason
is that the British Army must accustom itself to training, and even
maybe fighting overseas where it will frequently get foreign maps dished
up with the least possible alteration in the shortest possible time.
In the majority of the cases foreign material and trig. data would be in
metres. There is moreover a growing world tendency (not as yet very
evident in America) to get on to such an International unit as the
metre for basic surveys, even though the common linear units of the
country may be different. For instance the new surveys of Great Britain
are coordinated in metres and all post-war O.S. maps will carry a metric
grid. It would of course be a great advantage if we could both do the
same but I do not know how you view the chances of getting the metre
adopted for such purposes in America. At first I should not have
thought the chances were very great.
If we go on to a "mesh" system the question of degrees or grades or
mils or possibly some other systems altogether will arise. It is necessary
to have some decimal sub-division of angle for this purpose but the
centesimal system works out too small and the mil works out too large
as applied to latitude and longitude on the earth's surface. The
sexagesimal minute is about right but is not decimally sub-divided, nor does
it spring decimally from a parent unit. The answer may be to adopt, as
the unit tens and decimals of sexagesimal minutes. We are adopting the
latter for our preliminary trials.
I will let you know this question progresses and shall be grateful
for any further developments your side. I think it is important
that we should keep in close touch with one another oven though we may
not finally be able to adopt the same system. In fact I feel a little
guilty about not having briefed you sooner.
M.HOTINE
Brig. D. Survey
*COPY*
STUDY AND DISCUSSION OF MILITARY GRIDS
by
Army Map Service
and
Military Intelligence Division, Office, Chief of Engineers, U.S. Army
A link to a pdf of the original document
STUDY AND DISCUSSION OF MILITARY GRIDS
by
Army Map Service
and
Military Intelligence Division, Office, Chief of Engineers, U.S. Army
1. Purpose of Study
The purpose of the following study is to determine the characteristics
required in a military grid and to select a system most nearly answering these
requirements. Marked disadvantages are inherent in most grid systems now used.
These disadvantages are complicated by the existence of many incompatible systems.
2. Existing Conditions
2.01. The polyconic military grid is prescribed by Section VII, AR 300-15,
for use on all military maps of the United States. This system is laid out in
zones 9 degrees wide in longitude with 1 degree of overlap between zones. It
is so inaccurate at long ranges in certain directions that it cannot be used
satisfactorily for the control of the fire of coast artillery weapons or heavy
field artillery.
2.02. Because of this inaccuracy, the coast artillery harbor defense grid
for area in the neighbourhood of the harbor defences in the continental United
States is also authorized in paragraph 28, Change No.4, AR 300-15. This harbor
defense grid system is a Lambert conic conformal designed particularly to
serve the guns of the harbor defense concerned. It is not only not connected
to the military grid system in the same area but is incompatible therewith.
2.03. Many other grid systems are in use not only in the Unites States
but also in the rest of the world. Twenty states of the Unites States have
adopted the state plane coordinate systems measured in feet and especially
designed to serve the particular state in question. Each such system is hardly
extensible beyond the borders of the state without the introduction of mater-
ial inaccuracies. The enclosed map shows the overall picture of the grid
systems used in allied military operations during the recent campaigns. (Incl. 1)
2.04. It is obvious that the presence of so many systems complicates map
preparation and impose material confusing handicaps on actual combat operations.
The presence of more than one grid system covering one area presents no particular
problem in peace time or on manoeuvres involving but one arm. When the
fog of war confuses men's minds, the presence of several coordinate systems in
one area for use of different arms in fraught with potent opportunity for
disasters resulting from uncoordinations attributed to mistakes in using the
military grid. Involved in this matter are branch pride and branch stubbornness,
each branch feeling justified in having a special grid system designed to the
particular capabilities and needs of that branch. An example of this occurred
in the United Kingdom whore the British coast artillery, Navy and Air Force
covered the coastal area with three incompatible, incommensurable grid systems.
Intolerable confusion which resulted from the use of these grids during
the numerous German air raids in the Battle of Britain makes it highly probable that
these conflicting systems would have led to at east a few local disasters
has there been an invasion of Great Britain.
3. Basic Requirement
3.01. Primary Purpose of a Grid. The primary purpose served by the
military grid on a map is to provide quick solutions to problems of distance and
azimuth for the firing of weapons. It provides a quick simple means for
referring to spot locations and for designating targets. It is an essential tool
in coordination of military operations.
3.02. The coordination of the efforts of the many arms used on land, sea,
and air, is a problem so complex as to make mandatory a single simple solution
for problems of target designation and determination of range and azimuth. This
requirement is believed to be so important in war that the use of a single system
of limited but adequate accuracy is held to be better than the simultaneous
use of two incompatible but otherwise more accurate systems.
3.03. The characteristics of the using arms and weapons which affect the
design of the system to the adopted involve relatively little research. As a
general rule, it has been assumed that permanently emplaced batteries will be
more accurate in their fire than batteries temporarily emplaced in the fields.
Therefore, the following table appears to provide sufficient criteria to
determine the desirable accuracies of the grid system adopted.
Probable Errors of Different Caliber
Permanently Emplaced Guns at Ranges Shown
------------------------------------------------------------------------
Minimum Probable
Probable Errors in Yards Relative Errors
--------------------------------------------------------------
Range in 6" Gun 8" Gun 16" Gun
Yards Range Defl. Range Defl. Range Defl. Range Deflection
10,000 22 2 68 3 18 3 1:555 1:5,000
15,000 35 4 70 5 28 4 1:535 1:3,750
20,000 52 6 73 8 40 6 1:500 1:3,333
25,000 68 8 77 13 52 7 1:480 1:3,571
30,000 83 19 63 9 1:476 1:3,333
35,000 73 10 1:479 1:3,500
40,000 80 10 1:500 1:4,000
45,000 *77 *7 *1:584 *1:6,428
------------------------------------------------------------------------
* These values are appear unusual
4. Desirable Characteristics
4.01. Primarily a grid system should be accurate enough for all weapons
and all military uses other than for very long distance missiles, should be
quickly applicable to any previously ungridded native, map, should yield readily
to simple computing methods and should provide simple numerical designators for
location of targets.
4.02. Plane System. The system of coordinates desired is one with which
all computations for the most accurate artillery firing can be simply yet
accurately performed and especially one in which the integrity of angles is
preserved. A mathematically exact graticule, such as that presented by the
meridians and parallels, requires the use of geodetic functions to solve the
spherical triangles involves, and entails a long, time-consuming complicated
computation. Moreover, due to the convergence of the meridians, the arc of the
parallel intercepted between any two meridians becomes shorter as the latitude
increases. Complicated geodetic formulas would be necessary in the computation
of any distance except one along a meridian. Complex fire control instruments
would be needed, named by personnel highly trained in a branch of advanced
mathematics. Neither the personnel, the instruments, nor the time are normally
available. As a consequence, the system adopted should be one in which plane
trigonometry can be employed in the solution of triangles. In such a plane
system for general application to large area, the simplest and quickest
computations can be secured through use of a grid network of equally spaced
parallel and mutually perpendicular lines.
4.03. Grid Accuracy. A high degree of accuracy is, of course, desirable.
However, grid accuracies which are greatly in excess of the accuracies for the
most precise weapon using the grid appear to be neither necessary nor practicable.
By reference to paragraph 3.03 above, it will be noted that the probable
errors of artillery weapons are much greater in range than are their probable
errors in deflection. The minimum probable error of permanently emplaced
guns rarely is less than 1/555 in range and 1/5,000 in deflection. Consequently,
a suitable military grid should be one designed to conform to these minimum
probable errors.
4.04. Adaptability to Various Projections. The grid system selected
should be adaptable for use of native maps without complicated recomputation
or redrafting of that map. There are many map projections used in the making
of large scale maps throughout the world. It is desirable to be able to
overprint the adopted grid system on any or all of these projections without
the introduction of errors in range and azimuth beyond that probable in good
artillery practice.
4.05. Unit of Measure. Three general systems of linear measure are
commonly encountered, on maps and in grids: the metric system, the so-called
English system, and the nautical system. Mixtures of these systems unfortunately
are prevalent. This matter is further complicated by the fact that three
differing elements are involved - map quantities, grid quantities, and the
quantities employed by using arms and weapons.
a. Map quantities include azimuths, horizontal distances, contour
intervals, and underwater depths. To the map user, the unit of measure
in which horizontal distances on a map are expresses is not particularly
important, as the conversion from map distances to ground distances is frequently
done graphically against either an appropriately graduated bar scale on the
map or range scale. However, the unit of linear measure used in the basic
survey of the map may complicate the computation and compilation of trig lists
for fire control. This latter operation is already quite complex due to the
differing origins of longitude, datum planes, spheroids and schemes of projection,
and other variations encountered in the native surveys of the world.
Thus, the conversion from one unit of linear measure to another incommensurable
unit adds an operation subject to mistakes and affecting final accuracies.
b. Contour intervals and spot elevations should, but may not, be
in the same unit of measure as the horizontal distances, in order to provide
for the ready calculation of true slant ranges, defilade, mask, profile, etc.
c. Underwater depths are generally expressed either in meters
or in fathoms, although shoal water depths may be also expressed in feet.
It is highly desirable that those units of measure be the same as the horizontal
unit in order to be readily useful in the computation fo underwater profiles
and beach gradients.
d. The military grid, while essentially concerned with angles
and horizontal distances, must be precisely related to the computed geographic
positions. The necessary correlation between the vertical unit of measure and
the horizontal unit of measure on the grid as indicated in b above, is essential
for the quick solution of problems involving defilade, mask and true gun target
distance.
e. The using arms and weapons are not entirely coordinated in
the units of measure employed in the laying of the piece. The Coast Artillery
measures azimuths in degrees and hundredths of a degree from grid south as the
origin. The Field Artillery measures angles in mils, with filed orientation
of base circles. The Coast Artillery measures its range in yards, while the
Field Artillery may measure it in yards or meters. Due to the tangent relationship
of the mil, Field Artillery can readily transpose from angular measure to
linear distance in either meters or yards. Each artillery weapon is served
and laid by employing a multiplicity of tabular information, plotting tools,
and gunner's instruments. All these things must be related to the unit of
measure selected for map and grid quantities. At the present moment, due to
the recent tremendous concentration of field artillery weapons in the European
campaigns and to the use of the metric system throughout in that area, our
Field Artillery is well equipped and trained in the use of the metric system.
The Seacoast Artillery of the United States, including the Panama canal and
Oahu, are not so equipped or trained. They use the yard-hundredths of a degree
system, except in the coast defenses of San Diego where the coast Defense
grid is graduated in feet rather than in yards.
f. Existing Conditions. The majority of the large scale maps
of the world are made on the metric system. Exceptions to this rule are the
United States, Canada, Australia, United Kingdom, Union of South Africa,
India, Melanesia, and Middle East. The following table shows the unit of
measure of native maps and military grids employed in operational areas of
the recent campaigns.
MAP UNITS GRID UNITS
------------------------------------------- -------------------
AREA HORIZONTAL UNIT VERTICAL DEPTH UNIT GRID GRID TYPE
(Map Bar Scale) UNIT (Map (Bathymetric UNIT
Contour) Contour)
-----------------------------------------------------------------------------
1. France Meter Meter Meter Meter Lambert
2. Germany Meter Meter Meter Meter Trans. Merc.
3. Italy Meter Meter Meter ----- -----
4. Tunisia Meter Meter Meter ----- -----
5. Libya Meter Meter Meter Meter Trans. Merc.
6. Egypt Meter Meter Fathon Meter Trans. Merc.
7. Okinawa Meter,Cho Meter Meter Meter Unknown
8. Burma Miles Foot Fathom Yard Lambert
9. China Li,meter Meter Fathom Yard Lambert
10. Russia Meter Meter Foot Meter Trans. Merc.
11. Hawaii Mile Foot Fathom Yard Polyconic
12. Philipines Mile Foot Fathom Yard Polyconic
13. Poland Meter Meter Meter Meter Stereographic
14. Holland Meter Meter Meter Meter Stereographic
15. Belgium Meter Meter Meter Meter Bonne
16. Japan Meter,Cho Meter Meter Meter Trans. Merc.
4.06. Width of Zone. An inherent fault of any system of plane coordinates
applied to the surface of a spheroid is the fact that inaccuracies increase
as the zone is extended east and west or in other cases, as the belt is extended
north and south. It is also desirable, although not entirely necessary, to keep
at a minimum the deviation of grid north from true north. This deviation likewise
increases materially as the distance from the central meridian increases.
These inaccuracies can be kept within reasonable limits by the adoption of narrow
grid zones. It has been stated only semi-facetiously that there are three
military engineering axioms:
a. It always rains in war.
b. It's always too cloudy to get aerial mapping photographs.
c. Battles are always fought on grid junctions.
This Inst axiom is spoken from the depths of bitter experience, rendering it
obvious that a grid zone should include as much area as possible so as to
obviate too frequent junctures between grid zones on the battlefield. However,
this desirable criterion cannot be widely applied without including intolerable
inaccuracies of the grid. Elimination of the undesirable features consequent
upon fighting a battle on a grid juncture can be partially accomplished
by providing for overlap between grid zones. The 9 degree width of the military
grid system presently prescribed for use in the United States introduces
appreciable inaccuracies near the edges. A reduction in width to 6 degrees betters
this situation materially. The computation of ranges and azimuths where the
gun position is located in one grid zone and the target in another can be
provided for by a half degree overlap between grid zones, enabling the coordinates
of the gun zone to be extended a half a degree into the grid zone in which the
target is located.
5. Comparison of Grids
5.01. Polyconic Grid. For military purposes, a grid may be regarded as
a set of perfect squares ruled on a plane map, scale 1:1, and then transferred
to the earth's surface. Evidently after being transferred to the earth's
surface the squares will no longer be perfect; and distortions they will have
received in being put on the surface of the earth will reflect the distortions
of the projection used for the map.
a. The polyconic projection is defined as one which the central
meridian and all parallels are mapped to scale and with true curvature. All
other lines are stretched, the amount increasing as the square of the distance
from the central meridian, and being greatest for north-south lines. Angular
errors also appear, increasing with distance from the central meridian. Of
course it is possible to compute these errors, at least roughly, and to allow
for them, and this is regularly done by engineer survey troops. But the
corrections are generally considered beyond what is to be expected of artillery
units in the field, and for that reason all mention of them is omitted in
artillery technical manuals, even when survey procedures are discussed.
It is proper, therefore, to compare the errors of the projection as shown in the
following table, directly with the errors of the guns (Section 3.03).
Errors of U.S.Military Grid at 4°30'
from Central Meridian, and 30° Latitude
-------------------------------------------------------------------------------------------------
| | | True
| True Azimuth 0° | True Azimuth 45° | Azimuth 90°
True |----------------------------|-----------------------------|----------------------------
Range | Range Rela- | Defl. Rela- | Range Rela- | Defl. Rela- | Range Rela- | Defl. Rela-
(yds) | Error tive | Error tive | Error Defl. | Error tive | Error tive | Error tive
| (yds) Error | (yds) Error | (yds) Error | (yds) Error | (yds) Error | (yds) Error
---------|--------------|-------------|--------------|--------------|--------------|-------------
10,000 | 23 1:435 | 0 0 | 12 1:836 | 12 1:836 | 0 0 | 0 0
30,000 | 70 1:428 | 4 1:7500 | 35 1:859 | 35 1:859 | 0 0 | 0 0
40,000 | 70 1:428 | 4 1:5000 | 47 1:850 | 47 1:850 | 0 0 | 0 0
It is evident that at 40,000 yards, and an azimuth of 45°, the error of the grid
in deflection is almost five times the probable error of a 16-inch gun. Errors
of the above amount are characteristic of the s0-called non-conformal projections,
i.e., those in which the shaped, as well as the scale of small areas is distorted.
Such non-conformal grids were widely used prior to World war I, but most of them
have been abandoned in recent years, chiefly, no doubt, because of the breakdown
of the old French Bonne grid during the War. In addition to its lack of comformality,
the polyconic projection possesses the disadvantage that it has not been
studied thoroughly from a mathematical standpoint. Hence the small corrections
needed for precise surveys are not known; the transformation from other
grids to be polyconic is not known; not even the transformation from one
polyconic belt to another has been studied.
b. So far as grid junctions are concerned, the polyconic
is theoretically excellent; it can be extended indefinitely both north
and south, so that the world can be divided up into meridianal strips.
In practice, the present World Polyconic has two latitudinal junctions,
one near 24°, due to failure to put the origin of the old U. S. grid
sufficiently far south; the other is near 49°, and is due to inaccuracy in
the old tables which amounts to 1.1 meters.
c. The polyconic grid is not well suited for foreign maps
due to its lack of conformality. By the older laborious hand methods
of grid plotting, this introduced no difficulties other than laborious
calculations; but the use of the coordinatograph, for rapid plotting of
grids and projections, requires a conformal grid.
5.02. Cassini-Soldner Grid. This is a non-conformal grid, very
similar to the polyconic, and open to the same objections. It differs
only in that the grid east-west lines, rather than the parallels are
represented to true scale and with true curvature.
5.03. The Bonne Grid. Both meridians and parallels are represented
to true scale on the Bonne projection; the error shows up on lines which
run NE to SW, or NW to SE. It is just as bad as in the case of the other
non-conformal projections; and this projection has been generally abandoned.
5.04. The Stereographic Grid. The stereographic grid may be briefly
defined as a conformal grid (that is, one having zero angular distortions
for small distances, and very small angular distortions for any distance)
in which the scale error is Zero on a standard circle.
This grid is conformal insures that within about 200 miles of the center-
point, the error in deflection will be less than the error of the most
accurate guns. The range error is considerably larger than the deflection
error , but is considerably less than the range error of permanently
emplaced artillery. Unfortunately, this grid cannot be extended more
than 300 miles from its center point and grid zones are circular in
Shape. It is quite suitable for small roughly circular countries such
as Poland, Holland or Romania, which use it ; but on a world-wide basis
It would lead to a great multiplicity of grid junctions, and points where
three or more grids meet could not be avoided.
5.05. Lambert Grid. Like the stereographic, the lambert is a
conformal projection. It may be defined as a conformal projection in which
the scale errors are zero along two parallels . It is well suited to the
mapping of moderately large areas , and has been extensively used by the
British and French, especially in recent years. The numerical values
of the errors are similar to those for the Transverse Mercator given in
the next paragraph. The errors are very small in deflection, as in necessary
for artillery grid. It is readily adaptable to gridding maps
On other projections. In general a slight change of scale only would be
Required. On the other hand, the rapidly increasing grid declination
makes it impracticable to extend the grid more than 15° from the
central meridian (except along the Equator , where grid declination is
always zero). For this reason, the British and French were forced to
introduce numerous zones, and to permit junctions of three grids at the
same point. These numerous grid zones necessary, both north-south and
east-west, make the Lambert undesirable for extensive coverage
5.06. Transverse Mercator Grid. The Transverse Mercator grid may
be defined as a conformal grid in which the central meridian is represented
by a straight line at true scale. it is well suited to largo areas, and
is being used by the Germans, Russians, British, and Japanese. The errors
at a given point vary little in all azimuths and average values for
different ranges are given in the following table for 30° north latitude and
for 3 1/2° from the central meridian.
-------------------------------------------------------------------
| | Transverse | Transverse |
| True | Mercator Grid | Mercator Grid |
| Range | Range Relative | Deflection Relative |
| (yds) | Error Error | Error (yds) Error |
| | (yds) | |
|--------|-----------------------------|--------------------------|
| 10,000 | 10 1:1000 | 0 0 |
| 30,000 | 30 1:1000 | 3 1:10000 |
| 40,000 | 40 1:1000 | 5.8 1:6900 |
|--------|-----------------------------|--------------------------|
Obviously, the grid is well suited to artillery purposes since the
inevitable errors ere thrown into range rather than deflection. The grid
can be extended indefinitely in latitude like the polyconic. Hence it
never necessary to have a grid junction involving more than two grids
(except, of course, near the Poles). Transformation of coordinates from
one bolt to another can be done by a formula already worked out. The
formula is always the same, and is very simple in character. The grid
declination will remain moderate throughout the belt. The grid can be
readily adapted to use on other projections. Much theoretical work has
already been done on this subject by a large group of mathematicians,
including especially Professor W. K, Hristow. Extensive computations,
especially for the Balkan countries, were done by the German High
Command (O. K. H. ) during World War II which could, in an emergency, be
promptly utilized if the proposed projection is adopted. In addition,
much geodetic data of foreign areas on file at Army Map Service are on
this system. The projection is well suited for converting data on
various spheroids to a common basis. Transverse and lower orders of
triangulation may be computed and adjusted directly on the grid due to its
conformality. This feature, which is a large saving in field and office,
is not practicable where a non-conformal projection such as the present
Polyconic is used.
5.07. Tabular Comparison of Grids.
____________________________
--------------------------------------------------------------------------------------
GRID SYSTEM APPLICABILITY GRID MAXIMUM MAXIMUM
TO FOREIGN MAPS** JUNCTIONS RELATIVE RELATIVE
RANGE ERRORS* DEFLECTION
(45,000 yds) ERRORS*
(45,000 yds)
--------------------------------------------------------------------------------------
Polyconic Very Poor Few & simple 1/1228 1/2454
Cassini-Soldner Very Poor Few & simple 1/1228 1/2454
Bonne Very poor Many & complex 1/2454 1/615
Stereographic Good Many & complex 1/2454 1/15,360
Lambert Good Many & complex 1/2416 1/7,663
Transverse Mercator Excellent;
much work
already done Few & simple 1/2416 1/7,663
---------------------------------------------------------------------------------------
** A grid system is considered applicable to a foreign map if it can be put on
most maps without changing map or grid except in scale.
* Range and deflection errors are maximum values within 160 miles from the center
of the projection, whether the center is a line, as in the Transverse Mercator,
Lambert, Cassini-Soidner, and Polyconic, or a point as with the other two. The
figures are based on GSGS "Survey Computation". 160 miles is the approximate
distance (at 40° of latitude) from the center of the proposed Transverse
Mercator zones to the junctions, about 3° of longitude. The actual maximum errors
of the present world Polyconic grid are considerably larger, since the grid
zones are 9° in, width.
6. Conclusions as to System
to be Adopted.
6.01. The Lambert Orthomorphic projection is conformal but is not
suitably as it requires grid zone junctures both north and south and east
and west. The polyconic grid system now prescribed for use as military
grid on all maps of IJ. S. is inaccurate in both azimuth and distance.
The greater inaccuracy is in azimuth and is more than the probable error
in deflection of permanently placed guns. The transverse mercator grid
is conformal and is immediately applicable without plottable error, to
the majority of the map projections commonly encountered on the native
maps of the world. The transverse mercator grid reduces inaccuracies
to a point where they are compatible with the accuracies required by all
modern artillery weapons. This grid is sufficiently accurate to
eliminate the necessity for a special Coast Artillery grid in the
vicinity of coast defense locations.
6.02. In view of the foregoing, a military grid system based on.
the transverse mercator projection applied to the local spheroid and
measured in meters, or in the standard unit of the country concerned,
should be applied in zones running from Latitude 80° N to Latitude 80°
S, 6 degrees of longitude wide, with one degree of overlap (1/2 degree
each side). The latitude of the origin is the equator. (Incl #2).
The false casting to be applied for each zone would be 500,000 meters
or yards. Scale factor on the C. M. should be 0.9996. The zones should
be numbered, commencing with Zone 1, with its western edge at 180°
longitude, running east to 174° west longitude. Consecutively numbered zones
continue eastward by successive steps of 6 degrees until reaching the
point of beginning; these number designations being identical to the
I.M.W 1:1,000,000 layout. It will be noted that in certain countries
where the native maps use the English units in elevations. and contours,
such as the U. S., Canada, Australia, India, the proposed grid system
should be graduated in yards rather than meters. In certain training
areas in the U. S., both metric and yard grids will be required for
training purposes. Where the metric Grid is used in the domestic U.S.,
the spot elevations should be in meters and the contours should be
converted to a metric interval, provided that such conversion of contours
shall be limited to those maps to be used for metric training purposes.
7. Proposed Specifications
Projection: Transverse Mercator.
Spheroid: Same as that used to compute the triangulation
of the area.
Unit: Meter in most areas; yard in U. S. and other
areas where the English system is firmly
established.
Central Meridians: 3°E (or W) of Greenwich and every 6° thereafter.
Latitude of Origins: 0°
False Easting: 500,000 meters (or yards) .
False Northing: 0 for Northern hemisphere
10,000,000 for southern hemisphere.
Scale Factor: 0.9996
Zone Width: 6° of longitude (plus 1/2° overlap at each edge)
Limits of Zones: North 80° latitude
South: 80° latitude
Zone Numbering; Commencing with Zone 1 east of 180° longitude, and
continuing easterly around the earth. (Identical
to I. M. W. system designation)
Limits of Tables: North: -80° latitude
South: -80° latitude
8. Implementing Actions and Costs
8.01. It should be noted that the application of the transverse
mercator grid system should be progressive rather than instantaneous.
Priorities for conversion are indicated as follows:
a. Military areas in the United States,
b. Mapping and map revision of foreign areas embraced in
the 20-year strategic mapping plan approved by the War Department.
c. General areas of the United States as stocked by the Army
Map Service.
d. Other maps of foreign areas as reissued.
Upon the issue of the new map in any area, the new grid will normally
be shown in full, but to safeguard against the event of the occurrences
of an emergency while a series is still in a state of partial conversion,
the maps will carry marginal marks to permit easy plotting and
overprinting of the old grid.
8.02. Since Amy Map Service and map depots in the theaters now
hold extensive stocks of maps carrying the expedient war-time grids, It
is to be expected that the proposed standardization of the transverse
mercator grid automatically renders obsolescent these stocks. The cost
of a conversion in this respect is estimated as follows:
a. Cost of conversion of points and correction of drafting copy
(1) United States areas $99,500
(2) European areas $103,000
(3) Other overseas areas $22,000
b. Cost of replacement of stocks to be retired
(1) United States areas $ -0-
(2) Overseas areas $10,000
Since all United States maps are to ho converted to military scales
(1/25,000, 1/50,000), new stocks are to be prepared in any event.
Overseas accumulation of war time maps, it has been reported, are being
salvaged except for small reserves which will probably not be replaced. The
$10,000 figure should be ample to provide for all requisitions directly
attributable to the change in grid. The cost of now authorizing the
proposed conversion is properly to be weighed against the much greater
cost that would be borne should circumstances require the conversion
ten years hence. The conversion must ultimately be made in view of the
inadequacies of the present medley of grids to suffice for the anticipated
requirements of another war. It is considered unquestionable
that the cost of the conversion should be accepted now if it be agreed
By the General Staff that the transverse mercator grid is in fact the
correct design for the future, based on what can now be discerned as to
future characteristics requirements.
8.03. It is suggested that the views of the Navy Department be
obtained prior to final standardization in view of the application of the
military grid to maps and charts for amphibious operations. It would
also be desirable to coordinate the design so far as possible with the
Director of Military Survey, War Office, London, who has expressed his
general views in an informal letter (Incl. #3).
ARMY MAP SERVICES
Recommendations for Military Grids
6 December 1946
A link to a pdf of the original document
ARMY MAP SERVICES
Corps of Engineers, U.S. Army
6 December 1946
MEMORANDUM TO: Chief of Engineers
SUBJECT: Recommendations for Military Grids
1. Included herein are certain recommendations for grid numbers and
grid references. A pertinent discussion and background information follows
each recommendation.
2. The Universal Transverse Mercator Grid supersedes approximately
85 previously used grids which made up an undesirable heterogeneous system.
This modernization should be further expanded to revise the outmoded standards
for grid numbers and grid references which were designed primarily for fire
control purposes only. The inadequacy of the present system for use in making
general grid references became apparent during the past war. To satisfy
their needs it was necessary for the individual theater commanders to devise
new methods to fill this deficiency. There was an unfortunate lack of
consistency for the systems used varied with the theater. Before formulating
these recommendations an exhaustive study was made of all the war-time
provisional systems. The best of each is incorporated within these recommendations,
which if adopted would assure a standard but simple fool-proof system, designed,
at the same time, to accommodate changing techniques in warfare.
3. GRID INTERVALS
a. RECOMMENDATION - It is recommended that grid intervals be:
Maps 1:5,000 and larger 1,000 yards (or meters) with
grid lines ticked at 100 yard
(or meter) intervals.
Maps 1:10,000)
1:25,000) 1,000 yards (or meters)
1:50,000)
Maps 1:100,000)
1:250,000) 1,0000 yards (or meters)
b. The grid intervals authorized at present are:
Maps larger than 1:5,000 100 yards
Maps 1:5,000 to 1:63,360
inclusive 1,000 yards
Maps smaller than 1:63,360 to
larger than 1:100,000 5,000 yards
Maps 1:100,000 to larger than
1:400,000 10,000 yards
Maps 1:400,000 to 1:500,000
inclusive 50,000 yards
c. The intervals of British grids are:
Maps larger than 1:5,000 100 meters (or yards)
Maps 1:5,000 to 1:100,000
inclusive* 1000 meters (or yards)
Maps smaller than 1:100,000* to 10,000 meters (or yards)
1:500,000
d. It is noted that according to the new edition of AR 300-15, authorized
map scales are:
Small scale (1:1,000,000
Medium Scale (1:250,000
(1:100,000
Large scale (1:50,000
(1:25,000
(1:10,000
(1:5,000
e. The war proved that generally the British grid intervals were
superior. The U.S 5,000 yard interval was awkward and confusing
in as much as the abbreviated reference for a common point on maps
of different scales would be dissimilar in all instances. Authorization
should be granted to revise the grid intervals to overcome
this defect and to make them more compatible to the revised
authorized map scales.
4. GRID REFERENCES
a. RECOMMENDATION - To satisfy particular needs two types of grid references
should be made standard: general references, and fire control references.
(1) General reference - Such a reference should generally consist
of the grid zone designation by a group of numbers
expressing the E and N coordinates of the referred point;
examples:
30 NC 80432864 27 SF 69143872 (1,000 unit reference)
30 NC 804286 27 SF 691387 (10,000 unit reference)
-----------
*In all except Europe, AMS sheets of 1:100,000 falling in British Grid
areas were gridded at 10,000 meter (or yard) intervals
(a) Grid zones
1. Zones for the Universal Transverse Mercator grid are
identified with the IMW column (6° E-W) numbers,
starting at the international date line (180° meridian)
and reading 1 to 60 in an easterly direction. (See
attached index). To prevent similar references for
points 1,000,000 units apart (north-south) the IMT
row letters preceded by N (for north) or S (for south)
should be incorporated within the system and added to
the zone number designation. Under the IMW plan, each
row (4° N-S) is assigned a letter of the alphabet
starting from the equator, preceding in both directions.
2. To assure proper identification each sheet should carry
in its grid reference box its complete zone identification.
3. Within an area assigned to an army the grid zone
designation may be deleted at the discretion of the
Commanding General for reporting within the grid zone
providing sender and receiver are not more than 500
miles apart in a N-S direction. (The zone designation
is necessary in such a case since the numerical reference
is the same at 1,000,000 units in a N-S direction).
For reports to higher headquarters, however, the complete
reference must be given.
(b) Numerical reference - To facilitate making such references
from a 1,000 unit grid, a reference should be simply an
eight digit number; for example: 80472866. The "804"
represents the 100,000, 10,000 and 1,000 digits of the
easting grid line to the west of the referenced point,
the "7" represents the estimated tenths from the easting
grid line to the point, the "286" represents the 100,000,
10,000 and 1,000 digits of the northing grid line south
of the referenced point, and the "6" represents the
estimated tenths from the northing grid line to the point.
To maintain a relationship between similar grid references
from different scale maps, a reference from a 10,000 unit
grid should be a six digit number; for example: 804286.
The "80" represents the 100,000 and 10,000 digits of the
easting grid line to the west of the referenced point,
the "4" represents the estimated tenths from the easting
grid line to the point, the "28" represents the 100,000
and 10,000 digits of the northing grid line south of the
referenced point, and the "6" represents the estimated
tenths from the northing grid line to the point.
(2) Fire control references to be used within the sphere of the
equivalent of one adjacent 1:50,000 sheet in all directions
(two 1:25,000 sheets, four 1:10,000 sheets, etc.) - Existing
methods for determining grid references for fire control
as outlined in FM 6-40, Part Four, Chapter 2, should be
retained with but one modification: Sheet name designations
should never be used. When a grid reference is being sent
to a station outside the sphere of the equivalent of one
adjacent 1:50,000 sheet in all directions (for example: a
long range gun), then the full grid reference should be
sent preceded by the zone designation (see (1) above); for
example:
30 NC (804.72-1286.68)
b. Existing regulations (See FM 6-40, Part Four, Chapter 2) designate
the following methods for reading a grid reference:
(1) Designation of sheet, parenthesis, X coordinate, decimal,
location to nearest yard, a dash, Y coordinate, decimal,
location to nearest yard, parenthesis.
Example: Annapolis (804.729-1286.684)
(2) When the map is definitely understood, its designation may be
omitted.
Example: Annapolis (804.729-1286.684)
(3) If the location to the nearest 10 or nearest 100 yards only
is desired, or if the measurements cannot be made with greater
accuracy, the digits indicating units or tens may be omitted.
Examples: (a) (804.72-1286.68) to nearest 10 yards
(b) (804.7-1286.7) to nearest 100 yards
(4) For expediency it is permissible to include only two digits to
the left of the decimal point (10,000 and 1,000 digits),
omitting any preceding digits.
Examples: (a) (04.729-86.684)
(b) (04.72-86.68)
(c) (04.7-86.6)
(5) If the point is fixed within an area 10,000 yards square, only
one digit need be given before the decimal point of each
coordinate.
Examples: (a) (4.729-6.684)
(b) (4.72-6.68)
(c) (4.7-6.7)
(6) If a large number of points are being designated by the
abbreviated coordinates shown in example (c), the decimals
and dashes may be omitted and the reference given as (4767).
c. References for British Grids are read according to the following
methods:
(1) Maps bearing a 10,000 unit interval (1:100,000 to 1:500,000);
letter of 500,000 unit square (written as a small capital
letter), letter of 100,000 unit square (written as a large
capital letter), 10,000 digit of easting line to the left of
the point, estimated tenths (1,000 units) eastward to point,
10,000 digit of northing line south of the point, estimated
tenths (1,000 units) northward to the point.
Example: cA1428
This locates point to nearest 1,000 units.
(2) Maps bearing a 1,000 unit interval (1:5,000 to 1:100,000 inclusive):
Letter of 100,000 units square, 10,000 and 1,000 digits of easting
line to left of point, estimated tenths (100 units) eastward to
point, 10,000 and 1,000 digits of northing line south of the
point, estimated tenths (100 units) northward to the point.
Example : A143286
This location point to nearest 100 units.
d. During the war, the Pacific and Southwest Pacific Commands found it
feasible to use a system for reading general grid references
similarly to that used with British Grids. Apparently, a broad
interpretation of existing regulations was made to find authority
for the change. The name of the map is not mentioned (authority:
see 4 b (2) above); digits to the left of the 10,000 and 1,000
unit digits are omitted (authority: see 4 b (4) above); decimals
and dashes are omitted (authority: see 4 b (6) above).
(1) To read a reference point on a map employing a 1,000 unit
interval read: the 10,000 and 1,000 digits of the casting
line to left of point, estimated tenths (100 units) eastward
to point, the 10,000 and 1,000 digits of the northing line
south of the point, estimated tenths (100 units), northward
to the point. Write as a 6 digit continuous number.
Example: 143286
(2) A similar procedure is followed in reading a reference on a
map using a 10,000 unit interval, except that the digits for
the grid lines are for the 100,000 and 10,000 units (the
last four digits being omitted) and the estimated tenths
represent 1,000 units. Thus, a reference for the same point
cited in (1) above, might read: 214128
e. Discussion of recommendation (par. 4 a (1), above) for: General
references.
(1) Experience in the Pacific Theater proved that general grid
references were frequently used. Usually it was unnecessary
that these general grid references possess the same accuracy
as that required for fire control purposes. It was deemed
sufficient to identify any general grid reference to the
nearest 1/10th of the grid interval (i.e., 100 units at a
1,000 unit grid interval; 1,000 units at a 10,000 unit grid
interval). The system which employed a continuous six
digit number as a grid reference (example: 143286) proved
highly successful. Its principal merits wore simplicity and
intelligibility. The standard method for reading grid
references )see par. 4 b (1), above) was primarily designed for
fire control purposes and when used for general purposes becomes
very awkward. This was the experience in the Pacific Theater
which found that sheet name designations, parentheses, decimal
points and hyphens were superfluous and only increased the time
necessary to reading and sending general grid references.
(2) There were two faults with the Pacific system: danger of
confusion between a reference taken from a map bearing a 1,000
unit interval and from one of a 10,000 unit interval, in as much
as both were six digit numbers; and lack of connection between
references for a common point taken from a 1,000 unit grid and
from a 10,000 unit grid, in as much as in reading a reference from
a 1,000 unit grid the principal digits were the 10,000 and 1,000
ones and for a reference from a 10,000 unit grid the principal
digits were the 100,000 and 10,000 ones. Thus, references for
a common point might read: 047866 (from a 1,000 unit grid) and
804286 (from a 10,000 unit grid).
(3) Under the ANS proposal these faults would be eliminated. An
eight digit reference would immediately be recognized as
being from a 1,000 unit grid, and a six digit as being from a
10,000 unit grid. Further, a coordination would exist between
references from different unit grids for common points, as:
80432863 (reference from 1,000 unit grid)
804286 (reference from 10,000 unit grid)
(4) Normally, under the system as proposed it is required that
in referring a point the entire reference be given - grid zone
designation and numerical coordinates. In reporting in a
single grid zone between points not more than 500 miles apart
in a N-S direction designation of the zone is unnecessary.
Consequently, if he is certain that no confusion will result,
the theater commander should be permitted to issue instructions
to omit the zone designation from grid references. However,
in reporting to higher headquarters, between grid zones, and
between points in the same grid zone, more that 500 miles apart
in a N-S direction, the grid zone designation should never
be omitted. The 500 mile rule is required since a reference
will read the same for points which are 1,000,000 units apart
in the same grid zone.
(5) With the new type of warfare in which activities are far-flung
it is important that references given in communications identify
the area. The use of a sheet name as presently required by
regulations is inadequate for the receiver would generally
expand too much time searching map catalogues and indices to
identify the locale of the sheet. To introduce such a reference
with only the designation number of the grid zone would
require the use of 1,000,000 digits in the numbers. This is
not desirable as it would mean that the numerical reference
would differ from that used for a local general reference,
and would also require the use of decimal points and hyphens
since the 1,000,000 digit might occur only with one coordinate.
The solution is to introduce such a grid reference with the
designation number of the grid zone followed by a sub-zone
letter designation. (See paragraph 4 a (1) above). This
makes an absolute identification. Its use would simplify
the overall grid reference system in that the numerical
reference would be the same for both an abbreviated general local
reference and for a reference used in official communications
to higher headquarters. The use of the sub-zone letter
designation does not create a new system but makes complete
utilization of the entire IMW numbering system whose row
numbers are the basis of the numbering of the zones of the
Universal Transverse Mercator Grid. (See attached diagram).
f. Discussion of recommendation (par. 4 a (2), above) for: Fire
control references - The system presently in use is generally
quite adequate for its purpose. However, it is deemed more
desirable to use grid zone designations instead of sheet
identifications.
(For arguments see paragraph 4 e above).
5.GRID REFERENCE BOX
a. RECOMMENDATION - It is recommended that the grid reference boxes
used in foreign areas on AMS maps be made standard practice for
use on all maps including areas in the United Stats. The grid
reference box should contain instructions for determining a general
reference.
b. The inclusion of a grid reference box in the margin will assure
standard renditions of grid references, eliminating any reference
to military manuals by personnel unfamiliar with grids.
c. It is felt that grid references for fire control come within the
category of special purpose and are not as widely used as general
grid references as they generally are limited to artillery use.
Consequently, the method of determining such references need not
appear in the grid reference box but should be explained in proper
military manuals (see par. 7). If considered necessary, reference
to such manuals could be included in the grid reference box.
6. GRID NUMBERS
a. RECOMMENDATION - It is proposed that a modification of the so-called
Canadian Grid Numbering System be made standard on all maps published
by the Army Map Service. Under the system, grid numbers would appear
on all four sides of a sheet labelling each grid line, and "principal
digits" would be shown on the face of the map labelling each grid
line, appearing east or north of every accentuated grid line (every
even tenth line - 10,000 on a 1,000 unit grid and 100,000 on a 10,000
unit grid). On a 1,000 unit interval grid except for the values
shown in the southwest corner the last three digits of each grid
number are omitted and the principal digits (100,000, 10,000 and
1,000) appear larger than the 1,000,000 digit; for example:
(corner)
1 276 000 yds. 1 277 1 278
Numbers for a 10,000 unit grid appear in a similar manner, except
that the last four digits odf wach number are omitted; for example:
(corner)
1 27 0000 yds. 1 28 1 29
b. Existing standard practice requires that on grids of intervals of
1,000 units the last three digits be omitted and that on grids of
10,000 units the last four digits be omitted. Regulations do not
specifically limit the appearance, frequency or location of the numbers.
c. The advantages of this numbering system are apparent: the numbers
on the face of the map materially aid the map user in reading the
grid and in determining references; the use of superior type around
the border accentuates the principal digits (100,000, 10,000 and
1,000) materially aiding the map user in making general grid
references.
d. Under the system used in the Pacific for 1,000 unit grids, numbers
appeared on all four sides of the sheet labelling each grid line.
The last three digits of each number were omitted and the principal
digits (10,000 and 1,000) appear larger than the 1,000,000 and
100,000 digits, as for example:
(corner)
12 76 000 yds. 12 77 12 78
The principal digits also appeared on the face of the map labelling
each grid line appearing at 10,000 unit intervals east or north of
every 10,000 unit grid line (which are accentuated in weight).
The modification to this system recommended in a, above, is
necessary since the 100,000, 10,000 and 1,000 digits would appear
in a grid reference as recommended in par. 4 a (1), above.
7. GRID MANUALS
a. RECOMMENDATION - Subject to approval of the recommendations appear-
ing in the preceding paragraphs, a recommendation is made that
the Army Map Service be directed to prepare a new military manual
covering the subject of grids, and to prepare the text necessary
for any revisions to existing manuals.
b. Investigation reveals that no military manual covers the subject
of military grids completely. This is a serious omission and
should be remedied.
8. It is felt that the above recommendation will materially improve the
use of our grid system. The needs of the various grid users are provided
for; a standardization is effected; and full use is made of knowledge gained
through experience during the past war.
W.H. MILLS
Colonel, Corps of Engineers
Commanding Officer.
Chairman of JMPC Ad-hoc Committee on Universal Military Grid Referencing System
A link to a pdf of the original document
27 February 1948
FROM: Chairman of JMPC Ad-hoc Committee on Universal Military Grid
Referencing System
To: Colonel Northrup, U.S. Army
Captain Hobbs, U.S. Navy
Colonel Tison, U.S. Air Force
Reference: (a) JIC 410/M of 23 January 1948
Enclosure: (A)
1. In accordance with directive contained in par. 1 of reference (a),
a committee consisting of Colonel Mills, Commanding Officer, Army Map
Service (Chairman), Mr. Bloom of the Aeronautical Chart Service and Mr. Medina
of the U. S. Hydrographic Office has devised a universal military grid
referencing system which it considers suitable for the Armed Forces.
2. This is submitted for consideration. Enclosure (A) contains complete
information on the system. It is requested that this proposal be
transmitted to the operating forces for study.
3. Should the Army, Navy or Air Force believe that there is another
system more suitable than that described in Enclosure (A), complete information
regarding such a system should be submitted to this ad-hoc committee
for investigation and consideration.
4. In view of the urgency for an early decision, action should be
expedited. Approval, or any recommended changes, are desired by 15 April
1948.
W. H. MILLS
Colonel, Corps of Engineers
Chairman, JMPC Ad-hoc Committee
on Universal Military Grid
Referencing System
Copies:
Navy: 25
Army: 50
AF: 25
Lt. J. R. Phillips, JMPC: 1
Joint Mapping Photography Committee
Ad-hoc Committee on
Universal Military Grid Referencing System
A PROPOSED STANDARD UNIVERSAL MILITARY GRID REFERENCING SYSTEM
1. INTRODUCTION
There is a mandatory and urgent need for a standard referencing system,
for use by the Armed Forces.Such a system must have the following
outstanding characteristics:
a. It must meet the individual and collective requirements of the Army,
Navy and Air Forces in such a way that all services speak the same
language.
b. It must insure positive identification of any point in the world,
without danger of ambiguity,particularly at spheroid and datum junctions.
c. It must be simple of understand, brief and capable of being abbreviated
either for large-scale or for local operations; and adapted
to the requirements of fighter pilots traveling at high speed.
d. It must be suitable for rapid computations of range and azimuth,
with the required accuracy.
e. It must be usable in Polar areas.
f. It must avoid difficulties arising from the reversal of sign at
the 0 and 180th meridians and Eqautor.
2. PRESENT SITUATION
a. Section IV, W.D.Circular No.33, 5 February 1937, establishes the
Universal Transverse Mercator Grid as the official grid for the
Department of the Army. Accordingly, all maps published by the
Army since 5 February 1947 bear the UTM Grid. In the interests of
consistency, the U. S. Hydrographic Office has also surprinted the
UTM Grid on all approach and bombardment charts, and on all
hydrographic charts designed for use in amphibious training exercises.
The UTM Grid supersedes an undersirable heterogeneous system which
included the World Polyconic Grid, the U. S. Polyconic Grid, the
Panama Grid and approximately 80 so-called British Grids. Since the
projection of the grid is conformal, the grid can normally be
applied to map constructed on any the standard projections. (I)
----------------
I
The merits of the UTM Grid and deficiencies of the others are
discussed in Appendix I.
b. The following are the grid referencing systems which are authorized
and/or presently in use by the various armed forces(2); none can
satisfactorily fill the requirements for a universal system as outlined
in paragraph 1, above:
(1) Fire Control Referencing (FM 6-40, 1 June 1945, Chapter 2, Section
1 , pp.163-164: also, see Appendix II-I): This is the system
presently authorized for use by the Army.
(a) Advantages: The system is suitable for rapid computations
of range and azimuth;no difficulties arise from the
reversal of sign at the 0° and 180° meridians and at the
Equator.
(b) Deficiencies: Designed originally for fire control use by
the Army, it does not fill the needs of the Air Force and
the Navy; it is not adaptable to world wide use; it does
not provide for spheroid junctions; while it allows for
abbreviation for large scale activities; it does not
provide for referencing in the polar areas; it is lengthy
and awkward due to use of sheet-names, decimals, hyphens
and parentheses.
(2) Pacific World War II Practice (see Appendix II-2): Finding the
authorized referencing system inadequate, the commanders of
the Pacific Theaters devised a new system to fit their needs.
This was a compromise between the Fire control and British
Grid referencing systems. To a certain degree, this alleviated
but did not eliminate the deficiencies of the authorized
system.
(a) Advantages: The system is suitable for rapid computations
of range and azimuth;no difficulties arise from the
reversal of sign at the 0° and 180° meridians and at the
Equator; a reference is brief and not complicated with
decimals,hyphens and parentheses.
(b) Deficiencies: While fairly satisfactory for the local needs
of the Army an Navy, it did not fill the needs of the Air
Force: it is not adaptable to world wide use; it does not
provide for spheroid junctions; it does not provide for
abbreviation; it does not provide for referencing in polar
areas; while the awkwardness of the authorized system was
eliminated, ambiguities in reporting outside grid zones
could result.
(3) British Grid Referencing (TM 44-225, section IX, pp. 70-72; also,
see Appendix II-3): By agreement, U. S. Forces used the British
Grids wherever they existed.
______________
2
Regulations also provide for the Point Designation Grid (TM 44-225, 30
June 1944, Section V, pp. 63-64), the Jan grid (TM 44-222, Section VI,
pp. 65-66), and the Thrust Line Method (TM 44-222, Section VII, p. 67).
These are special purpose systems and their continued use will not be
affected by the adaption of a standard referencing system.
(a) Advantages: The system is suitable for rapid computations of
range and azimuth; no difficulties arise from the reversal of
sign at the 0° and 180° meridians and at the Equator; a
reference is brief and not complicated with decimals, hyphens and
parentheses.
(b) Disadvantages: It does not fill the needs of the Air Force
and Navy; it is not easily adaptable to world wide use due
to lack of order and homogenity; it does not provide for
spheroid junctions; while it permits slight abbreviation in
local areas, no provision is made for similar abbreviation
for large scale activities; it does not allow for referencing
in polar areas; the number of grid and datum junctions was
excessive resulting in frequent difficulty in rapid computation
of range and distance when origin and destination were in
different grid areas.
(4) Air Defense Grid (TM 44-225, Section X, pp.72-79)
(a) Advantages: The Air Defense Grid referencing partially
satisfies large scale use of the Air Force.
(b) Disadvantages: The system does not satisfy the requirements of
the Army and Navy; the system does not lend itself to the use
of the pilot in fighter support; while it is fairly satisfactory
in reporting areas in world wide activities it is not
satisfactory for reporting spot positions; while it provides
for abbreviation, in certain areas if the first letter of
the reference is dropped, the nearest similar reference is
near enough to create ambiguity; it is not suitable for rapid
computation of range and azimuth; it does not provide for
polar areas; it is ambiguous at datum junctions.
3. PROPOSED SYSTEM
a. General
The proposed system is based on the Universal Transeverse Mercator Grid
between 80° N and 80° s; in the polar areas it is based on polar
Stereographic grids.
(1) In conjunction with these grids, the reference system meets all
the requirements for a referencing system as outlined in
paragraph 1, above.
(a) It provides for the individual and collective needs of the
Army, Navy and Air Force. The basic principles followed by
any of the armed forces in giving a reference are alike; a
reference cited by any one branch will be readily recognized
by all and will permit easy conversion when required.
(b) It provides a positive and unambiguous identification for any
point on the globe especially at spheroid and datum junctions.
(c) Since the referencing in all cases follows the basic and simple
system of reading "right-up", a reference is easily understood.
A reference is brief and is capable of being abbreviated
either for local or large scale activities, thus satisfying
any and all needs.
(d) It is well suited for rapid computation of range and azimuth. (3)
(e) It is suitable for polar areas. Azimuth will agree with that
now in use by polar aviators; this avoids the difficulties
of meridian convergence which are a necessary feature of
computations in latitude and longitude.
(f) There is no reversal of sign at any meridian or any other
junction of zones.
(2) The system has the following additional merits:
(a) The coordinates always increase to the east and to the north,
except in the polar regions where the direction of positive
coordinates is umambiguously indicated.
(b) The system conforms to the spherical surface of the earth.
The grid line can be considered as a system of coordinates
whose position on the earth's surface is as exactly fixed
as meridians and parallels.
(c) As the grid is conformal, it can easily be applied to a map
constructed on any of the conventional projections, unless
they are so extended as to have unusually large distortions.
b. First division - 8° NS X 6° EW rectangles
(1) Between 80° N and 80° S the world is divided into rectangles 8°
NS X 6° EW (See Exhibit E-H). The columns (6° wide )are identified
by the Universal Transverse Mercator(UTM) Zone numbers --
that is,starting at the 180° meridian the columns are numbered
from 1 to 60 consecutively proceeding easterly. The rows (8° NS)
are identified by letters; starting from 80° south and proceeding
northerly the rows are lettered consecutively from C to X (I and
O omitted). The designation (called the grid zone designation)
of such an 8° NS X 6° EW rectangle is determined by reading
(right-up) first the column designation (as 54) and second row
designation (as U); as: 54U
(2) The north polar area above the 80° parallel is divided into two
parts by the 0° and 180° meridians; the half beginning west of
0° is identified as Y; the half beginning east of 0° is
identified as Z
__________
3
See Appendix I.
(See Exhibit I). Similarly,the south polar area below 80° is
divided into two halves by the 0° and 180° meridians; these halves
are identified by A and B respectively.
c. Second division - 100,000 meter squares
(1) Between 80° N and 80° S each 8° X 6° rectangle is divided into
100,000 meter squares based on the UTM grid for zone. Every
column of squares is identified by a letter; likewise, every row
of squares is identified by a letter. (See Exhibits A and B). On
the equator, starting at the 180° meridian, and proceeding easterly
for 18°,the 10,000 meter columns,including partial columns
(caused by convergence), are lettered A to Z (I and O omitted)
consecutively. The 100,000 meter rows are labelled from A to V
(I and O omitted) reading from south to north, with the partial
alphabet being repeated every 2,000,000 meters. Every odd numbered
6° wide UTM zone will have the alphabet of the 100,000 meter row
letters beginning at the equator; the even numbered 6° wide UTM
Zones will have the alphabet of the 100,000 meter row letters
beginning at the 500,000 meter northing grid line north of the
equator. This staggering will considerably lengthen the distance
between duplicating letters and will permit necessary manipulation
along spheroid junctions. Below the equator the 100,000
meter row letters will continue consecutively following the plan
of the letters above in the same zone. The designation of a
100,000 meter grid square is determined by reading (right-up)
first its column designation (as X) and second its row
designation (as Q): as: XQ.
(2) Under this system a 100,000 meter square designation will not be
repeated in an aera 18° NS 18° EW. This will normally eliminate
the necessity of preceding grid references within such an area by
the grid zone designation (54U in b (1) above) even though report
is being made from as many as two grid zones away.
(3) In the polar areas the 100,000 meter columns at the right angles to the
90° - 90° meridians are lettered from J to Z in zone designation Y
and A to R in zone designation Z (I and O omitted; also omitted are
D, E, M, N, V and W to avoid confusion with 100,000 meter
squares in adjoining UTM zones.Starting at the 80° line the
100,000 meter rows at right angles to the 0° - 180° meridians
are labelled A to Z cosecutively (I and O omitted). The identification
of a 100,000 meter square consists of two letters,
reading right-up (See Exhibit I).
d. Grid references for U.S. Army
(1) A U.S. Army reference shall consist of a number and a letter (the
grid zone designation) followed by two letters (identifying the
100,000 meter square in which the point of reference lies),
followed by a group of numbers expressing to the required
accuracy the E and N coordinates of the referred point within
the 100,000 meter square; examples:
(a) 54UXQ (locating a point within 100,000 meters)
(b) 54UXQ55 (locating a point within 10,000 meters)
(c) 54UXQ5354 (locating a point within 1,000 meters)
(d) 54UXQ539544 (locating a point within 100 meters)
(2) Normally, a general reference is seldom located to an accuracy of
more than the closest 100 meters. To provide for the needs of
surveying and to anticipate any contingency, the following
references are provided:
(a) 54UXQ53925443 (locating a point within 10 meters)
(b) 54UXQ5392354432 (locating a point within 1 meter)
(c) 54UXQ539234544321 (locating a point within 0.1 meters)
(3) Normally,all elements of a grid reference shall not be used. Those
to be omitted will depend upon the size of the area of activities.
Thus:
(a) If activities are confined to an area not exceeding 18° EW X
18° NS,the grid zone designation (54U) usually will be
omitted. In such an area, a reference to the closest hundred
meters usually will read: XQ539544
(b) If activities are confined to an area not exceeding 100,000
meters NS X 10,000 meters EW,in addition to omitting the
grid zone designation, the 100,000 meter square identification
(XQ) will be omitted; the point will be referenced only by
numbers. Thus, in such an area, a reference to the closest
hundred meters will read: 539544
(c) Numerical reference: The numerical part of a reference taken
from a 1,000 meter grid, will be a six digit number; for
example: 539544. The "53" represents the 10,000 and 1,000
digits of the easting grid line to the west of the referenced
point, the "9" represents the estimated tenths from the easting
grid line to the point, the "54" represents the 10,000
and 1,000 digits of the northing grid line south of the
referenced point, and the "4" represents the estimated
tenths from the norhthing grid line to the point. (See
Exhibit C). To maintain a relationship between similar grid
references from different scale maps, the numerical part of
a reference taken from a 10,000 meter grid will be a four
digit number; for example: 5354. The "5" represents the
10,000 digit of the easting grid line to the west of the
referenced point, the "3" represents the estimated tenths
from the easting grid line to the point, the second "5"
represents the 10,000 digit of the northing grid line south
of the referenced point, and the "4" represents the estimated
tenths from the northing grid line to the point. (See Exhibit
D).
e. Tad Grid Reference
(1) A TAD reference shall consist of two letters (identifying the
100,000 meter square containing the point of reference (see
paragraph 3 c, above), followed by four numerals (the 10,000
and 1,000 digits of the northing grid line south of the point
and the 10,000 and 1,000 digits of the easting grid line west
of the point, (the identification of the 1,000 meter grid
square containing the point), followed by a letter (identifying
the 200 meter grid square containing the point). (See Exhibit
F). Example: XQ5354J
(2) The first two letters (XQ) normally shall be omitted; they will not
be used unless the reference is being reported more than 100,000
meters away. Thus, the usual TAD reference will be written simply as: 5354J
(3) The p1an for lettering the 200 meter squares follows:
A B C D E
F G H I J
K L M N O
P Q R S T
U V W X Y
f. Air Defense references
(1) A reference shall consist of the grid zone designation (as 54U),
followed, if necessary, by two letters (identifying the 100,000
meter square containing the point), followed, if necessary, by a
group of numbers expressing to the required accuracy the E and N
coordinates of the referred point within the 100,000 meter square;
(See Exhibit G). Examples:
(a) 54U (locating a point within a 8° NS X 6° EW)
(b) 54UXQ (locating a point within 100,000 meters)
(c) 54UXQ55 (locating a point within 10,000 meters)
(c) 54UXQ5354 (locating a point within 1,000 meters)
(2) If reports are being confined to an area not exceeding 18° NS X
18° EW, the grid zone designation may be omitted and the
reference read as:
(a) XQ (locating a point within the 100,000 meters)
(b) XQ55 (locating a point within 10,000 meters)
(c) XQ5354 (locating a point within 1,000 meters)
4. CONCLUSION
It is felt that the system outlined above is a homogeneous solution of
the problem and will satisfactory fill the needs of the Navy, the Air
Force and Army. The needs of the various grid users are provided
for and a standardization is effected.
APPENDIX 1
STUDY AND DISCUSSION OF MILITARY GRIDS
by
Army Map Service
and
Military Intelligence Division, Office, Chief of Engineers, U.S. Army
1. Purpose of Study
The purpose of the following study is to determine the characteristics
required in a military grid and to select a system most nearly answering these
requirements. Marked disadvantages are inherent in most grid systems now used.
These disadvantages are complicated by the existence of many incompatible systems.
2. Existing Conditions
2.01. The polyconic military grid is prescribed by Section VII, AR 300-15,
for use on all military maps of the United States. This system is laid out in
zones 9 degrees wide in longitude with 1 degree of overlap between zones. It
is so inaccurate at long ranges in certain directions that it cannot be used
satisfactorily for the control of the fire of coast artillery weapons or heavy
field artillery.
2.02. Because of this inaccuracy, the coast artillery harbor defense grid
for area in the neighbourhood of the harbor defences in the continental United
States is also authorized in paragraph 28, Change No.4, AR 300-15. This harbor
defense grid system is a Lambert conic conformal designed particularly to
serve the guns of the harbor defense concerned. It is not only not connected
to the military grid system in the same area but is incompatible therewith.
2.03. Many other grid systems are in use not only in the Unites States
but also in the rest of the world. Twenty states of the Unites States have
adopted the state plane coordinate systems measured in feet and especially
designed to serve the particular state in question. Each such system is hardly
extensible beyond the borders of the state without the introduction of mater-
ial inaccuracies. The enclosed map shows the overall picture of the grid
systems used in allied military operations during the recent campaigns. (Incl. 1)
2.04. It is obvious that the presence of so many systems complicates map
preparation and impose material confusing handicaps on actual combat operations.
The presence of more than one grid system covering one area presents no particular
problem in peace time or on manoeuvres involving but one arm. When the
fog of war confuses men's minds, the presence of several coordinate systems in
one area for use of different arms in fraught with potent opportunity for
disasters resulting from uncoordinations attributed to mistakes in using the
military grid. Involved in this matter are branch pride and branch stubbornness,
each branch feeling justified in having a special grid system designed to the
particular capabilities and needs of that branch. An example of this occurred
in the United Kingdom where the British coast artillery, Navy and Air Force covered
the coastal area with three incompatible, incommensurable grid systems.
Intolerable confusion which resulted from the use of these grids during the
numerous German air raids in the Battle of Britain makes it highly probable that
these conflicting systems would have led to at least a few local disasters had
there been an invasion of Great Britain.
3. Basic Requirements
3.01. Primary Purpose of a Grid: The primary purpose served by the
military grid on a map is to provide quick solutions to problems of
distance and azimuth for the firing of weapons. It provides a
quick simple means for referring to spot locations and for designating
targets. It is an essential tool in coordination of military
operations.
3.02. The coordination of the efforts of the many arms used on land, sea,
and air, is a problem so complex as to make mandatory a single
simple solution for problems of target designation and determination
of range and azimuth. This requirement is believed to be so
important in war that the use of a single system of limited but
adequate accuracy is held to be better than the simultaneous use
of two incompatible but otherwise more accurate systems.
3.03. The characteristics of the using arms and weapons which affect the
design of the system to be adopted involve relatively little
research. As a general rule, it has been assumed that permanently
emplaced batteries will be more accurate in their fire than
batteries temporarily emplaced in the fields. Therefore, the
following table appears to provide sufficient criteria to determine
the desirable accuracies of the grid system adopted.
Probable Errors of Different Caliber
Permanently Emplaced Guns at Ranges Shown
------------------------------------------------------------------------
Minimum Probable
Probable Errors in Yards Relative Errors
--------------------------------------------------------------
Range in 6" Gun 8" Gun 16" Gun
Yards Range Defl. Range Defl. Range Defl. Range Deflection
10,000 22 2 68 3 18 3 1:555 1:5,000
15,000 35 4 70 5 28 4 1:535 1:3,750
20,000 52 6 73 8 40 6 1:500 1:3,333
25,000 68 8 77 13 52 7 1:480 1:3,571
30,000 83 19 63 9 1:476 1:3,333
35,000 73 10 1:479 1:3,500
40,000 80 10 1:500 1:4,000
45,000 *77 *7 *1:584 *1:6,428
------------------------------------------------------------------------
* These values are appear unusual
4. Desirable Characteristics
4.01. Primarily a grid system should be accurate enough for all weapons
and all military uses other than for very long distance missiles,
should be quickly applicable to any previously ungridded native
map, should yield readily to simple computing methods and should
provide simple numerical designators for location of targets.
4.02. Plane System. The system of coordinates desired is one with which
all computations for the most accurate artillery firing can be
simply yet accurately performed and especially one in which the
integrity of angles is preserved. A mathematically exact graticule,
such as that presented by the meridians and parallels, requires
the use of geodetic functions to solve the spherical triangles
involved, and entails a long, time-consuming complicated computation.
Moreover, due to the convergence of the meridians, the arc of the
latitude increases. Complicated geodetic formulas would be
necessary in the computation of any distance except one along a
meridian. Complex fire control instruments would be needed,
manned by personnel highly trained in a branch of advanced
mathematics. Neither the personnel, the instruments, nor the time are
normally available. As a consequence, the system adopted should
be one in which plane trigonometry can be employed in the solution
of triangles. In such a plane system for general application to
large areas, the simplest and quickest computations can be secured
through use of a grid network of equally spaced parallel and
mutually perpendicular lines.
4.03. Grid Accuracy. A high degree of accuracy is, of course, desirable.
However, grid accuracies which are greatly in excess of the
accuracies for the most precise weapon using the grid appear to
be neither necessary nor practicable. By reference to paragraph
3.03 above, it will be noted that the probable errors of artillery
weapons are much greater in range than are their probable errors
in deflection. The minimum probable error of permanently emplaced
guns rarely is less than 1/555 in range and 1/5,000 in deflection.
Consequently, a suitable military grid should be one designed to
conform to these minimum probable errors.
4.04. Adaptability to Various Projections. The grid system selected
should be adaptable for use on native maps without complicated
recomputation or redrafting of that map. There are many map
projections used in the making of large scale maps throughout the
world. It is desirable to be able to overprint the adopted grid
system on any or all of these projections without the introduction
of errors in range and azimuth beyond that probable in good
artillery practice.
4.05. Unit of Measure. Three general systems of linear measure are
commonly encountered on maps and in grids: the metric system, the
so-called English system, the nautical system. Mixtures of these
systems unfortunately are prevalent. This matter is further
complicated by the fact that three differing elements are involved -
map quantities, grid quantities, and the quantities employed by
the using arms and weapons.
a. Map quantities include azimuths, horizontal. distances, contour
intervals, and underwater depths. To the map user, the unit
of measure in which horizontal distances on a map are expressed
is not particularly important, as the conversion from map
distances to ground distances is frequently done graphically
against either an appropriately graduated bar scale on the
map or range scale. However, the unit of linear measure used
in the basic survey of the map may complicate the computation
and compilation trig lists for fire control. This latter
operation is already quite complex due to the differing origins
of longitude, datum planes, spheroids and schemes of projection,
and other variations encountered in the native surveys of the
world. Thus, the conversion from one unit of linear measure
to another incommensurable. unit adds an operation subject to
mistakes and affecting final accuracies.
b. Contour intervals and spot elevations should, but may not, be
in the same unit of measure as the horizontal distances, in
order to provide for the ready calculation of true slant
ranges, defilade, mask, profile, etc.
c. Underwater depths are generally expressed either in meters or
in fathoms, although shoal water depths may be also expressed
in feet. It is highly desirable that these units of measure
be the same as the horizontal unit in order to be readily
useful in the computation fo underwater profiles and beach
gradients.
d. The military grid, while essentially concerned with angles and
horizontal distances, must be precisely related to the
computed geographic positions. The necessary correlation between
the vertical unit of measure and the horizontal unit if measure
on the grid as indicated in b above, is essential for the
quick solution of problems involving defliade.,mask and true
gun target distance.
e. The using arms and weapons are not entirely coordinated in the
units of measure employed in the laying of the piece. The
Coast Artillery measures azimuths in degrees and hundredths
of a degree from grid south as the origin. The Field Artillery
measures angles in mils, with field orientation of base circles.
The Coast Artillery measures its range in yards ,while the Field
Artillery may measure it in yards or meters. Due to the
tangent relationship of the mil, Field Artillery can readily
transpose from angular measure to linear distance in either
meters or yards. Each artillery weapon is served and laid by
employing a multiplicity of tabular information, plotting
tools, and gunner's instruments. All these things must be
related to the unit of measure selected for map and grid
quantities. At the present moment, due to the recent
tremendous concentration of field artillery weapons in the
European campaigns and to the use of the metric system
throughout in that area, our Field Artillery is well equipped and
trained in the use of the metric system. The Seacoast
Artillery of the United States, including the Panama Canal and
Oahu, are not so equipped or trained. They use the yard-
hundredths of a degree system, except in the coast defenses
of Sen Diego where coast Defense grid graduated in
feet rather than in yards.
f. Existing Conditions. The majority of the large scale maps of
the world are made on the metric system.Exceptions to this
rule are the United States, Canada, Australia, United Kingdom,
Union of South Africa, India, Melanesia and Middle East. The
following table shows teh unit of measure of native maps and
military grids employed in operational areas of the recent
campaigns.
MAP UNITS GRID UNITS
------------------------------------------- -------------------
AREA HORIZONTAL UNIT VERTICAL DEPTH UNIT GRID GRID TYPE
(Map Bar Scale) UNIT (Map (Bathymetric UNIT
Contour) Contour)
-----------------------------------------------------------------------------
1. France Meter Meter Meter Meter Lambert
2. Germany Meter Meter Meter Meter Trans. Merc.
3. Italy Meter Meter Meter ----- -----
4. Tunisia Meter Meter Meter ----- -----
5. Libya Meter Meter Meter Meter Trans. Merc.
6. Egypt Meter Meter Fathon Meter Trans. Merc.
7. Okinawa Meter,Cho Meter Meter Meter Unknown
8. Burma Miles Foot Fathom Yard Lambert
9. China Li,meter Meter Fathom Yard Lambert
10. Russia Meter Meter Foot Meter Trans. Merc.
11. Hawaii Mile Foot Fathom Yard Polyconic
12. Philipines Mile Foot Fathom Yard Polyconic
13. Poland Meter Meter Meter Meter Stereographic
14. Holland Meter Meter Meter Meter Stereographic
15. Belgium Meter Meter Meter Meter Bonne
16. Japan Meter,Cho Meter Meter Meter Trans. Merc.
4.06. Width of Zone. An inherent fault of any system of plane coordinates
applied to the surface of a spheroid is the fact that inaccuracies
increase as the zone is extended east and west or in other
cases, as the belt is extended north and south. It is also
desirable, although not entirely necessary, to keep at a minimum
the deviation of grid north from true north. This deviation
likewise increases materially as the distance from the central
meridian increases. These inaccuracies can be kept within reasonable
limits by the adoption of narrow grid zones. It has been
stated only semi-facetiously that there are three military
engineering axioms:
a. It always rains in war.
b. It is always too cloudy to get aerial mapping photographs.
c. Battles are always fought on grid junctions.
This last axiom is spoken from the depths of bitter experience,
rendering it obvious that a grid zone should include as much
area as possible so as to obviate too frequent junctures between
grid zones on the battlefield.However,this desirable criterion
cannot be widely applied without including intolerable
inaccuracies of the grid. Elimination of the undesirable features
consequent upon fighting a battle on a grid juncture can be
partially accomplished by providing for overlap between grid zones,
the 9 degree width of the military grid system presently prescribed
for use in the United States introduces appreciable inaccuracies
near the edges. A reduction in width to 6 degrees betters this
situation materially. The computation of ranges and azimuths
where the gun position is located in one grid zone and target
in another can be provided for by a half degree overlap between
grid zones, enabling the coordinates of the gun zone to extended
a half a degree into the grid zone in which the target is located.
5. Comparison of Grids
5.01. Polyconic Grid. For military Purposes, a grid may be regarded as
a set of perfect squares ruled on a plane map, scale 1:1, and then
transferred to the earth's surface. Evidently after being transferred
to the earth's surface the squares will no longer be perfect;
and distortions they will have received in being put on
the surface of the earth will reflect the distortions of the
projection used for the map.
a. The polyconic projections is defined as one which the central
meridian and all parallels are mapped to scale and with true
curvature. All other lines are stretched, the amount increasing
as the square of the distance from the central meridian, and
being greatest for north-south lines. Angular errors also
appear, increasing with distance from the central meridian.
Of course it is possible to commute these errors, at least
roughly, and to allow for them, and this is regularly done by
engineer survey troops. But the corrections are generally
considered beyond what is to be expected of artillery units
in the field, and for that reason all mention of them is omitted
in artillery technical manuals, even when survey procedures are
discussed. It is proper, therefore, to compare the errors of the
projection as shown in the following table, directly with the
errors of the guns (Section 3.03).
Errors of U.S.Military Grid at 4°30'
from Central Meridian, and 30° Latitude
-------------------------------------------------------------------------------------------------
| | | True
| True Azimuth 0° | True Azimuth 45° | Azimuth 90°
True |----------------------------|-----------------------------|----------------------------
Range | Range Rela- | Defl. Rela- | Range Rela- | Defl. Rela- | Range Rela- | Defl. Rela-
(yds) | Error tive | Error tive | Error Defl. | Error tive | Error tive | Error tive
| (yds) Error | (yds) Error | (yds) Error | (yds) Error | (yds) Error | (yds) Error
---------|--------------|-------------|--------------|--------------|--------------|-------------
10,000 | 23 1:435 | 0 0 | 12 1:836 | 12 1:836 | 0 0 | 0 0
30,000 | 70 1:428 | 4 1:7500 | 35 1:859 | 35 1:859 | 0 0 | 0 0
40,000 | 70 1:428 | 4 1:5000 | 47 1:850 | 47 1:850 | 0 0 | 0 0
It is evident that at 40,000 yards, and an azimuth of 45°, the error
of the grid in deflection is almost five times the probable error
of a 16-inch gun, Errors of the above amount are characteristic
of the so-called non conformal projections, i.e. those in which
the shapes, as well as the scale of small areas is distorted.
Such non-conformal grids were widely used prior to World War I;
but most of them have been abandoned in recent years, chiefly,
no doubt, because of the breakdown of the old French Bonne grid
during the War. In addition to its lack of conformality, the
polyconic projection possesses the disadvantages that it has not
been studied thoroughly from a mathematical standpoint. Hence
the small corrections needed for precise surveys are not known;
the transformation from other grids to the polyconic is not
known; not even the transformation from one polyconic belt to
another has been studied.
b. So far as grid junctions are concerned,the polyconic is theoretically
excellent; it can be extended indefinitely both north and
south, so that the world can be divided up into meridional strips.
In practice, the present World Polyconic has two latitudinal
junctions, one near 24°, due to failure to put the origin of the
old U. S. grid sufficiently far south; the other is near 49°,
and is due to inaccuracy in the old tables which amounts to 1.1
meters.
c. The polyconic grid is not well suited for foreign maps due to
its lack of conformality. By the older laborious hand methods
of grid plotting, this introduced no difficulties other than
laborious calculations; but the use of the coordinatograph, for
rapid plotting of grids and projections, requires a conformal
grid.
5.02. Cassini-Soldner Grid. This is a non-conformal grid,very similar to
the polyconic, and open to the same objections. It differs only in
that the grid east-west lines, rather than the parallels are represented
to true scale and with true curvature.
5.03. The Bonne Grid. Both meridians and parallels are represented to true
scale on the Bonne projection; the error shows up on lines which run
NE to SW, or NW to SE, It is just as bad as in the case of the other
non-conformal projections; and this projection has been generally
abandoned.
5.04. The Stereographic Grid. The stereographic grid may be briefly defined
as a conformal grid (that is, one having zero angular distortions
for small distances, and very small angular distortions for any
distance) in which the scale error is zero on a standard circle.
The fact that this grid is conformal insures that within about 200
miles of the center-point, the error in deflection will be less than
the error of the most accurate guns. The range error is considerably
larger than the deflection error, but is considerably less than the
range error of permanently emplaced artillery. Unfortunately, this
grid cannot be extended more than 300 miles from its center point
and grid zones are circular in shape. It is quite suitable for small
roughly circular countries such as Poland, Holland, or Romania, which
use it; but on a world-wide basis it would lead to great multiplicity
of grid junctions, and points where three or more grids meet could
not be avoided.
5.05. Lambert Grid. Like the Stereographic, the Lambert is a conformal
projection. It may not defined as a conformal projection in
which the scale errors are zero along two parallels. It is well
suited to the mapping of moderately large areas, and has been
extensively used by the British and French, especially in
recent years. The numerical values of the errors are similar
to those for the Transverse Mercator given in the next paragraph.
The errors are very small in deflection, as is necessary
for an artillery grid. It is readily adaptable to gridding
maps on other projections. In general, a slight change
of scale only would be required. On the other hand, the
rapidly increasing grid declination makes it impracticable
to extend the grid more than about 15° from the central
meridian (except along the Equator, where grid declination
is always zero). For this reason, the British and French were
forced to introduce numerous zones, and to permit junctions
of three grids at the same point. These numerous grid zones
necessary, both north-south and east-west, make the Lambert
undesirable for extensive world coverage.
5.06. Transverse Mercator Grid. The Transverse Mercator grid may
be defined as a conformal grid in which the central meridian
is represented by straight line at the true scale. It is well
suited to large areas, and is being used by the Germans,
Russians, British, and Japanese. The errors at a given point
vary little in all azimuths end average values for different
ranges are given in the following table for 30° north latitude
and for 3 1/2° from the central meridian.
-------------------------------------------------------------------
| | Transverse | Transverse |
| True | Mercator Grid | Mercator Grid |
| Range | Range Relative | Deflection Relative |
| (yds) | Error Error | Error (yds) Error |
| | (yds) | |
|--------|-----------------------------|--------------------------|
| 10,000 | 10 1:1000 | 0 0 |
| 30,000 | 30 1:1000 | 3 1:10000 |
| 40,000 | 40 1:1000 | 5.8 1:6900 |
|--------|-----------------------------|--------------------------|
Obviously,t he grid is well suited to artillery purposes since the
inevitable errors are thrown into range rather than deflection.
The grid can be extended indefinitely in latitude like the polyconic.
Hence it is never necessary to have a grid junction
involving more than two grids (except, of course, near the
Poles). Transformation of coordinates from one belt to another
can be done by a formula already worked out. The formula is
always the same, and is very simple in character. The grid
declination will remain moderate throughout the belt. The grid
can be readily adapted to use on other projections. Much
theoretical work has already been done on this subject by a
large group of mathematicians, including especially Professor
W. K. Hristow. Extensive computations, especially for the
Balkan Countries, were done by the German High Command (O.
K. H.)during World War II which could, in an emergency, be
promptly utilized if the proposed projection is adopted.
In addition, much geodetic data of foreign areas on file at
Army Map Service are on this system. The projection is well
suited for converting data on various spheroids to a common
basis. Traverse and lower orders of triangulation may be
computed and adjusted directly on the grid due to its
conformality. This feature, which is a large saving in field
and office, is not practicable where a non-conformal
projection such as the present Polyconic is used.
5.07. Tabular Comparison of Grids.
--------------------------------------------------------------------------------------
GRID SYSTEM APPLICABILITY GRID MAXIMUM MAXIMUM
TO FOREIGN MAPS** JUNCTIONS RELATIVE RELATIVE
RANGE ERRORS* DEFLECTION
(45,000 yds) ERRORS*
(45,000 yds)
--------------------------------------------------------------------------------------
Polyconic Very Poor Few & simple 1/1228 1/2454
Cassini-Soldner Very Poor Few & simple 1/1228 1/2454
Bonne Very poor Many & complex 1/2454 1/615
Stereographic Good Many & complex 1/2454 1/15,360
Lambert Good Many & complex 1/2416 1/7,663
Transverse Mercator Excellent;
much work
already done Few & simple 1/2416 1/7,663
---------------------------------------------------------------------------------------
* Range and deflection errors are maximum values within 160 miles from the
center of the projection, whether the center is a line, as in the Transverse
Mercator, Lambert, Cassini-Soldner, and Polyconic, or a point as with the
other two. The figures are based on GSGS "Survey Computation". 160 miles
is the approximate distance (at 40° of latitude) from the center of the
proposed Transverse Mercator Zones to the junctions, about 3° of longitude.
The actual maximum errors of the present world Polyconic grid are
considerably larger, since the grid zones are 9° in width.
6. Conclusions as to System to be Adopted
6.01. The Lambert Orthomorphic projection is conformal but is not
suitable as it requires grid zone junctures both north and south
and east and west. The ployconic grid system now prescribed for
use as military grid on all maps of U. S. is inaccurate in both
azimuth and distance. The greater inaccuracy is in azimuth and
is more than the probable error in deflection of permanently
emplaced guns. The transverse mercator grid is conformal and is
-----------------
**A grid system is considered applicable to a foreign map if it can be put on most maps without changing map or grid except is scale.
immediately applicable without plottable error, to the majority
of the map projections commonly encountered on the native
maps of the world. The transverse mercator grid reduces
inaccurecies to a point where they are compatible with the
accuracies required by all modern artillery weapons. This
grid is sufficiently accurate to eliminate the necessity for
a special Coast Artillery grid in the vicinity of coast
defense locations.
APPENDIX 2
1. FIRE CONTROL REFERENCING
Existing regulations of the Department of the Army (See FM 6-40, Part
Four, Chapter 2) designate the following method for reading a gird
reference.
a. Designatiuon of sheet, parenthesis, X coordinate, decimal, location
to nearest yard, a dash, Y coordinate, decimal, location to nearest
yard, parenthesis.
Example: Annapolis (804.729-1286.684)
b. When the map is definitely understood, its designation may be omitted.
Example: (804.729-1286.684)
c. If the location to the nearest 10 or nearest 100 yards only
is desired, or if the measurements can not be made with greater accuracy,
the digits indicating units or tens may be omitted.
Examples: (1) (804.72.1286.68) to nearest 10 yards.
(2) (804.7.1286.7) to nearest 100 yards.
d. For expediency it is permissible to include only two digits to
the left of decimal point (10,000 and 1,000 digits), omitting
any preceding digits.
Examples: (1) (04.729-86.684)
(2) (04.72-86.68)
(3) (04.7-86.6)
e. If the point is fixed within an area 10,000 yards square, only one
digit need be given before the decimal point of each coordinate.
Examples: (1) (4.729-6.684)
(2)(4.72-6.68)
(3)(4.7-6.7)
f. If a large number of points are being designated by the
abbreviated coordinates shown in example (3), the decimals
and dashes may be omitted and the reference given as (4767).
2. PACIFIC WORLD WAR II PRACTICE
During the war, the Pacific and Southwest Pacific Commands found it
feasible to use a system for reading general grid references
similar to that used with British Girds. Apparently, a broad
interpretation of existing regulations was made to find authority for the
change. The name of the map was not mentioned (authority: see 1 b
above) digits to the left of the 10,000 and 1,000 unit digits were
omitted (authority see 1 d above) decimals and dashes were omitted
(authority see 1 f above).
a. A reference on a map employing a 1,000 unit interval was determined
by reading the 10,000 and 1,000 digits of the easting line to
left of point, estimated tenths(100 units) eastward to point,
the 10,000 and 1,000 digits of the northing line south of the point,
estimated tenths (100 units), northward to the point. This was
written as a 6 digit continuous number.
Example: 143286
b. A similar procedure was follows in reading a reference on a
map using a 10,000 unit interval, except that the digits for the
gird lines were for the 100,000 and 10,000 units (the last four
digits being omitted) and the estimated tenths represented 1,000
units. Thus, a reference for the same point cited in (1) above,
was as: 214128
c. There were two major faults with the pacific system: danger of
confusion between a reference taken from a map bearing a 1,000
unit interval and from one of a 10,000 unit interval, inasmuch
as both were six digit numbers; and lack of connection between
references for a common point taken from a 1,000 unit gird and
from a 10,000 unit gird, inasmuch as in reading a reference from
a 1,000 unit gird the principal digits were the 10,000 and 1,000
ones and for a reference from a 10,000 unit grid the principle
digits were the 100,000 and 10,000 ones. Thus, references for
a common point might read: 047866 (from a 1,000 unit gird) and
804286 (from a 10,000 unit gird).
3. BRITISH GRID REFERENCING
References for british grids are read according to the following methods:
a. Maps bearing a 10,000 unit interval (1:100,000 to 1:500,000):
letter of 500,000 unit source (written as a small capital
letter), letter of 100,000 unit square (written as a large
capital letter), 10,000 digit of easting line to the left of
the point, estimated tenths (1,000 units) eastward to point,
10,000 digit of northing line south of the point, estimated
tenths (1,000 units) northward to the point.
Example: cA1428
This locates point to nearest 1,000 units.
b. Maps bearing a 1,000 unit interval (1:5,000 to 1:100,000 inclusive):
Letter of 100,000 unit square, 10,000 and 1,000 digits of easting
line to left of point, estimated tenths (100 units) eastward to
point, 10,000 and 1,000 digits for northing line south of the
point, estimated tenths (100 units) northward to the point.
Example:A143286
The locates point to nearest 100 units.