Compilation from existing materials

The preparation of derived maps—i.e., maps that are compiled from other maps or existing data—involves the search for, and evaluation of, all extant data pertaining to the subject area. Depending on the nature of the map to be compiled, thoroughgoing research includes boundary references, historical records, name derivations, and other materials. Selection of the most authentic items, on the frequent occasions when some ambiguities are detected, requires careful study and references to related materials. The sources finally selected may require some adjustment or compromise in order to fit properly with adjacent data. When it becomes evident that some sources are of questionable reliability, the cartographer explains this in the margin of his compilation. Sometimes this is placed in the body of the map where the doubtful features or delineations are located.

When selected materials have been assembled they are reduced to a common scale and copied on the compilation base, often in differentiating colours for the respective features. Reductions to a common scale are usually made by photography but may be made by projection and traced directly on the drawing. Minor adjustments may have to be made during compilation even though the source materials are of good quality. In particular, the need to make appropriate generalizations, omitting some details in smaller scale maps, requires much study and judgment.

Except for the new methods of preparing final colour-separation plates by scribing (described below), rather than by drafting or copperplate engraving, compilation processes have changed little over the years. Automatic-focusing projectors and better illumination have made the tracing of selected data at compilation scale easier. Better and more extensive facilities for photoreduction and copying, improved light tables, and a wider choice of drafting materials and instruments have served to facilitate compilation. The basic chores of research, selection of best data, and adjustment of these into the compilation, however, remain essentially the same.

The preparation of small-scale maps from large ones is sometimes simpler than the process just described, which pertains to compilation from a miscellany of differing sources. The relatively straightforward preparation of 1:62,500-scale maps from those of the 1:24,000-scale series, for example, may require little more than photoreduction and colour-separation drafting, or scribing. Even in this case some generalizations, as well as omission of a few of the least important details, are in order. To avoid the considerable expense involved in such scale conversions, straight photoreduction of colour-separation plates appears to be a promising procedure.

Larger reductions from one map series to another—1:62,500 to 1:250,000 for instance—are more of a problem, since the need for generalization is greater and the omission of many details is involved. The considerable differences in road and other symbol sizes also create displacement problems.

The component maps are reduced, and the negatives are cut and assembled into a mosaic on a clear sheet of plastic, the master negative of which provides guide copy for the several colour-separation plates required, which are then completed for reproduction. More often, however, it is necessary to make an intermediate compilation rather than burden the draftsman with too many adjustments to be made while following copy on the colour-separation plates. The intermediate scale for initial reduction of the component maps provides better legibility than direct photography to reproduction scale. This negative mosaic is copied on a metal-mounted drafting board. A compiler then inks the whole map, usually in three or more contrasting colours. He also draws roads and other symbols at the intermediate size, so that they will reduce to proper dimensions at reproduction scale, and makes the necessary displacement adjustments. Minor features and terrain details to be omitted on the new map are not inked in. The drawing is now ready for photoreduction to the final colour-separation plates, providing much better copy for the draftsman or engraver than direct reduction in one step would have produced.

Most smaller scale map series are prepared from large-scale maps as described above. In earlier days original reconnaissance surveys were made at small scales such as 1:192,000 for publication at 1:250,000. Ideally, the small-scale series of maps should be compiled progressively from those of larger scale and greater detail. Most countries, however, started their mapping programs with relatively small-scale reconnaissance surveys because of economic considerations. Later, affluence and technical competence permitted mapping at larger scales with better accuracy.

Geologic, soil, and other thematic maps usually have a topographic base from which woodland tints and road classification printings have been omitted. Such a map, therefore, has a topographic background printed in subdued colours on which the geologic or soil patterns are overprinted in prominent colours. Small-scale thematic maps showing weather patterns, vegetation types, and a large amount of economic and other information are of similar origin. Backgrounds are drawn from appropriate outline maps of provinces, countries, or regions of the world, while overlaying subject matters are compiled from specialized sources of information.

Generalization of detail

The generalization of detail is a problem that frequently confronts the cartographer in original mapping and in reducing the scale of existing maps. There are two principal reasons for taking such liberties (or topographic license in the case of the original mapping). The primary purpose is to avoid overcrowding and the resulting poor legibility. In addition, the degree of generalization or detail should be as consistent as possible throughout the map. Generalizations in some parts and excessive detail in others confuse the user and make the map’s reliability suspect. Effective generalization requires good judgment based on seasoned knowledge and experience.

In approaching such problems as the thousands of islets in the Stockholm archipelago or the thousands of small lakes in the Alaskan tundra areas, when the map scale will accommodate only a small number, the cartographer may decide to draw the features in groupings that reflect the patterns shown in the large-scale source maps or aerial photos. This is difficult and at best presents the nature of the respective areas rather than a literal portrayal. There is also the possibility that the source maps may already have been generalized by some omissions to accommodate to their own scales. Another device is to note, in appropriate text or marginal references, that many minor lakes or islets are omitted because of scale. Such areas may also be symbolized and explained. The “pattern” representation noted above is actually a form of symbolization.

Intricate coastlines are also extremely difficult to generalize consistently. Here again, the purpose is to omit minor details while retaining the main features and their distinguishing characteristics. These and many equally perplexing questions arise in preparing maps of very small scale from any source. The problems of equalization of detail are also present in such cases. The topographer of earlier days had the equalization problem between areas close at hand and those viewed distantly. In addition, the topographer had to deal with terrain on the far sides of obscuring features.

Photogrammetrists—that is, persons who compile original maps from aerial photos—have similar problems when, for example, one side of a ridge is seen in more detail than the opposite side. Indeed, in steep terrain, parts of the far sides of some mountains are not seen at all. Appropriate steps must be taken in such cases to avoid differing renditions on opposite sides of the mountain. This may be accomplished by adding, in field completion of the manuscript map, the segments not seen by the photogrammetrist; or additional aerial photography, patterned to cover the obscured sectors, may be requested.

Map production from original surveys

The instrumentation, procedures, and standards involved in making original surveys have improved remarkably in recent years. Geodetic, topographic, hydrographic, and cadastral surveys have been facilitated by the application of electronics and computer sciences. At the same time, superior optics and more refined instruments, in general, have enhanced the precision of observations and accuracies of the end products.

The improved quality of surveys has increased the reliability of maps and charts based on them. In turn, the greater output of basic data has accelerated the production of maps and charts, while parallel improvements in processing steps have increased the volume and improved the final product. In a sense the production of maps from original surveys parallels the process steps after a compilation is made from derived sources. This phase is sometimes referred to as map finishing and involves editing, colour separation, and printing. In original surveys for topographic maps and nautical charts, however, the end products are provided for in all the process steps leading to the completed basic manuscript. The manuscript scale is, for example, selected to accommodate the plotting instruments involved as well as the final rendering for printing. In early years it was usual to choose a manuscript scale somewhat larger than that prescribed for publication. This was to allow for some generalization and line refinement in the final reduction. Thus, maps to be published at 1:62,500 scale were plotted in the field at 1:48,000 or thereabouts. With modern photogrammetric instruments, plotting is usually at reproduction scale.

Maps are not directly derived from geodetic surveys, and only land-line plats are produced from cadastral surveys. Accordingly, the primary original map and chart productions are those from topographic and hydrographic surveys. The surveys are somewhat similar as the nautical chart is, in effect, a topographic map of the coast with generalized offshore topography interpolated from depth soundings.

A variety of electronic devices are used to determine a survey ship’s precise location while taking soundings, which are also made with electronic equipment. Both hydrographic and topographic surveys now employ aerial photography and precise plotting instruments to develop the base map. In order to simplify the description of modern mapmaking techniques, the process developed for topographic mapping will be described below, with comments where procedures for nautical charts differ significantly. Both processes start by expanding upon the basic control previously established from geodetic surveys.

Surveying, in which the facts are discovered and recorded, must precede mapping, in which the facts are presented in graphic form. Surveying involves (1) global positioning, in which the area to be mapped is located on the Earth’s surface, usually by fixing a number of points in the area by astronomical observations or, after the techniques became available, by satellite or radar procedures; (2) establishing the framework, in which these points, and commonly many others connected by some combination of distance and angle measurements, are integrated into an accurately defined structure—like the steel framework of a modern building—on which the detail survey is based; and (3) making the detail survey, which establishes by less accurate (and therefore cheaper) methods the relative positions and shapes of the features being mapped. Constant reference to the framework prevents the errors in the detail survey from accumulating and growing unacceptably large.

Mapping also consists of three steps: (1) fair drawing, in which the accurate but not publishable surveyor’s plot is redrawn by a skilled cartographer with uniform lines and lettering and, if a multicoloured map is being produced, is separated into several drawings, one for each colour; (2) reproduction, in which a negative is prepared from each of the fair-drawn originals and special colouring (to represent areas of vegetation, for example) is added; and (3) printing, in which a printing plate is made from each negative, the plates are mounted on a press, proofs (a few trial copies) are made to facilitate correction of errors and blemishes, and the final maps are produced.

Final steps in map preparation

After all the features visible in the aerial photographs have been mapped, the manuscripts are contact printed on coated plastic sheets for review by the field engineer. He examines the whole map, adding such details as houses, trails, and fences that were not visible or were overlooked by the photogrammetrist. Political lines such as state, county, and township limits are located, as are geographic and other names in local use. Roads are classified, and woodland outlines are checked.

Contour accuracy is tested if the operator has noted areas that may be weak. The determination of names involves extensive local inquiry, as do political lines, and both may require research of records.

In remote areas it is more efficient to combine the above activities with supplemental control survey to avoid the extra field phase. Then the photos must be carefully examined and annotated for the compiler, while buildings must be encircled or pricked. Roads are classified and political lines located in the usual manner and noted on the photos or overlays.

Field corrections are applied to the original manuscripts. They must be scribed (engraved) on the originals so that guide copies can be prepared by contact printing for final colour-separation scribing. At this time all factual detail is carefully checked. Editing may proceed, to conserve time, while the colour-separation scribing is in progress. The editor reviews all names, boundaries, and related data, comparing them to information thereon that may be available from other sources. The editor’s function is to see that the map conforms to standard conventions and is clear, legible, and free of errors.

Controversial names, or those found to be in confused or ambiguous spelling or usage, are documented and referred to an appropriate official body. The designation of type styles and sizes as well as placement of lettering is another function of the map editor.

Because modern topographic maps are printed in several colours, separate plates must be prepared for each. Some of the earliest maps were printed from woodcuts, usually in a single colour. Various hand processes were developed through the years, culminating in the fine rendering of copperplate engraving, which dominated the map production industry for many years. The process became obsolete, however, with increased production demands and the development of efficient printing presses. After World War II engraving on glass, and later on coated plastic sheets, was developed to a point that recovered the fineness of copper engraving. These methods of engraving have become firmly established in map production throughout the world.

Scribing

In the negative engraving or scribing process, guide copy is printed on several sheets of plastic coated with an opaque paint, usually yellow. The scriber follows copy on the respective plates by engraving through the coating. Because arc light can pass only through the engraving scratches, the completed engravings are, in effect, negatives from which the press plates are made. The finest lines (0.002 inch, or 0.05 millimetre, wide), such as intermediate contours, are engraved freehand. Heavier lines, such as index contours, engraved at 0.007 inch, may require a small tripod to assure that the scriber is perfectly vertical. Gravers for double-lined roads, others for buildings, and templates, or patterns, for a variety of symbols are used. Woodland and similar boundaries and shorelines are contact printed and etched on their respective coated sheets, and the areas of the woodland or water are then peeled off, leaving open windows for their respective features. If portions of scrub, orchard, or vineyard are contained in the “woodland” plate, negative sections for these are stripped into their respective locations. Press plates are then processed from the negatives.

A combined-colour proof is then made by successively printing the several completed negatives on a sensitized white plastic sheet that serves for the final checking and review of all aspects of the map. After all corrections have been made, the negatives are ready for the reproduction process.

Nearly all maps are now printed by rotary offset presses, using flexible aluminum-alloy printing plates. The system uses surface plates (very slightly raised or recessed) as opposed to the letterpress and intaglio processes, which involve greater image heights and depths respectively. In the printing sequence, ink goes from the plate to a rubber blanket to the paper. Thus, the printing plate is positive, or right-reading, as is the printed map. The negatives from which the printing plates are prepared are accordingly wrong-reading. This is the process for so-called surface plates. To retain fineness of line on very long runs (10,000 or more impressions), some map printers prefer “etched” plates, prepared from film positives. Both may be considered essentially surface plates, however, since the respective raise or recess is quite small.

Presses are of many varieties and makes. Huge multicolour types are used in large plants, printing several colours at a time. In effect, a multicolour press is several presses built into one. Each unit has three cylinders for plate, rubber blanket, and paper as well as rollers for water and ink. Presses with automatic feed may produce as many as 6,000 impressions per hour, while hand-fed types are limited to about 2,500 per hour.

Nautical charts

Nautical charts are commonly large, 28 by 40 inches (70 centimetres by 1 metre) being an internationally accepted maximum size. In order that a navigator may work with them efficiently, charts must be kept with a minimum of folding in drawers in a large chart table in a compartment of the ship having ready access to the navigating bridge, known as the chart room or chart house. Such structures are not possible in small craft, which therefore require charts of a more convenient size. With the recognition that there are many more small boats in the world, particularly recreational craft, than there are ships and that they are navigated primarily by piloting rather than by celestial or electronic means, many hydrographic offices have given attention to the production of special chart series in a small format for yacht navigators.

A typical series is that produced by the U.S. Coast and Geodetic Survey with the designation SC (for small craft). Such charts are only 15 inches (38 centimetres) in the vertical dimension and thus need to be folded only in the vertical direction. Printed on both sides of the sheet, they are oriented along the most probable route rather than by parallels and meridians. Several are stapled together into a stiff cardboard folder for protection. Along with the ordinary chart information, they contain a year of tide tables and information on small-craft facilities in the area. New editions are produced annually.

Practical uses of charts impose some constraints on the selection of colour. Red, for example, would logically be chosen as the color in which to print warnings of navigational hazards. But navigators, who must work at night, prefer to retain the darkness adaptation of their eyes by viewing their charts under red light. Under such illumination, red, orange, and buff are invisible. Hence these colours have been superseded by magenta, purple, and gray.

Charts are working instruments, and, since ships often voyage far from where replacement charts are readily obtainable, hydrographic offices give attention to the quality of the paper on which charts are printed. A ship’s reckoning is kept in pencil and erased after each voyage. Thus, printing stock that permits multiple erasures is chosen. In view of the environment where charts are used, another quality commonly sought is high wet strength.

Automation in mapping

During the past few decades, there was much interest in the automation of mapping processes, and considerable progress was made in this area. Achievements in the fields of electronics, high-speed digital computers, and related technologies provided a favourable period for such progress. In Great Britain, development of a set of procedures utilizing automatic elements, known as the Oxford System, was begun.

Some success was also achieved in the difficult area of automatic plotting. Instruments now available can automatically scan a stereo model and generate approximate profiles from which contours may be interpolated. Some steps, however, must be closely monitored or else performed completely by the operator. Contouring interpolated from a profile scan is inferior to an operator’s delineation. This contouring meets some less exacting requirements for elevation data, and refinements in the system are improving its precision. The need for human intervention when automatic devices get “lost” is not a decisive drawback, as one operator can monitor several machines. The reduction of tedious and repetitive steps for stereo-operators offers a significant advancement.

Coordinatographs with high repeat accuracies facilitate the automatic plotting of control points and projection intersections. Line work can also be drafted or scribed automatically by the same process, but the respective features must first be coded to provide the necessary input tapes. Automatic colour scanning and discrimination is operational but has not become widely used; it is still necessary for an operator to trace the various features on the manuscript to code them. Obviously, little is to be gained by automatic scribing until the input can be provided automatically. Coded line work can be displayed on a cathode-ray tube and corrected with a light pen, but it is much simpler to check and correct the manuscript or finished drawing. Systems of automatic type placement at present offer only marginal advantages over conventional methods. In short, automation has made substantial advances but has not become fully operational in a practical sense.

An aspect of automation that is developing rapidly concerns graphic data acquisition, storage, and retrieval. Data banks are being accumulated by specialized users of topographic information, often to produce thematic maps showing soil types, vegetation classifications, and a variety of other information. Such data banks are usually organized in two parts: one for line work, such as boundaries, and the other for descriptive information or classifications. Assuming that the necessary inputs have been made to the data bank, special plats can be generated speedily. Examples of such graphics include profiles showing elevations along a selected radio propagation path and cross sections for earthwork on roads and other construction.