Robinson projection
The Robinson projection is a pseudocylindrical compromise map projection developed in the early 1960s by American cartographer Arthur H. Robinson, originally termed the "orthophanic" projection and designed specifically for thematic world maps to provide a visually balanced representation of the globe by minimizing distortions in area, shape, distance, and direction without adhering to equal-area or conformal properties.[1] In this projection, all parallels are represented as straight horizontal lines of varying length, while meridians are curved and unequally spaced, converging toward the poles to create an oval-shaped graticule that enhances the map's aesthetic appeal and perceived naturalness.[2][3] Unlike traditional projections derived from mathematical formulas, Robinson selected its coordinates empirically through iterative testing and visual evaluation, prioritizing overall harmony over precise preservation of any single cartographic property.[2][4] The projection was commissioned by the atlas publisher Rand McNally and later gained widespread use in general-purpose cartography due to its ability to convey global relationships effectively while avoiding the extreme polar exaggeration seen in other cylindrical projections like Mercator.[1][5][6]History and Development
Creation by Arthur H. Robinson
Arthur H. Robinson, a prominent American cartographer and professor of geography at the University of Wisconsin-Madison since 1946, developed the Robinson projection in 1963.[4][7] The projection originated from a commission by Rand McNally, a leading atlas publisher, which sought a new world map design that prioritized aesthetic appeal over strict adherence to traditional projection properties.[2][8] Robinson's design philosophy centered on achieving visual balance and a pleasing overall appearance for global maps, rather than mathematical perfection in preserving angles, areas, or distances. He employed a trial-and-error approach using graphical methods, iteratively sketching graticules, plotting landmasses, and adjusting parameters to minimize perceived distortions until the map "looked right" to the eye.[8] This process involved estimating parallel lengths and spacings by hand, followed by repeated refinements based on visual assessment, resulting in a compromise projection that avoided the extreme elongations at high latitudes seen in projections like Mercator while forgoing equal-area preservation unlike the Mollweide.[5][8] The initial conception and testing occurred in 1963, with Robinson compiling coordinate tables through these empirical iterations to facilitate map production.[2] He later detailed the projection's development in a 1974 publication.[8]Publication and Initial Adoption
The Robinson projection was formally detailed in Arthur H. Robinson's 1974 article, "A New Map Projection: Its Development and Characteristics," published in the International Yearbook of Cartography.[9] In this work, Robinson described the projection's iterative design process, emphasizing its compromise approach to balancing distortions in area, shape, and distance for world mapping purposes.[9] The publication marked the projection's introduction to the broader cartographic community, building on its earlier experimental use. Although developed in 1963, the projection saw initial adoption by Rand McNally, which commissioned Robinson and began incorporating it into their atlases in the late 1960s.[10] This early commercial integration helped establish its practicality for general-purpose world maps, with Rand McNally utilizing it extensively in publications like The Rand McNally Cosmopolitan World Atlas.[4] The projection gained wider recognition through Robinson's academic influence at the University of Wisconsin-Madison, where his teaching and research on cartographic design promoted its advantages over traditional projections.[4] A significant evolution occurred in 1988 when the National Geographic Society adopted the Robinson projection as its standard for general-purpose world maps, replacing the Van der Grinten projection to achieve a more visually balanced representation of global landmasses.[11] This decision was influenced by the projection's ability to minimize perceptual distortions, as noted by National Geographic's chief cartographer at the time.[11] The society continued using it until 1998, when it transitioned to the Winkel tripel projection.[12] The adoption received public attention through a 1988 New York Times article, "The Impossible Quest for the Perfect Map," which discussed the challenges of map projections and highlighted Robinson's contribution as a pragmatic solution to the trade-offs inherent in projecting a spherical Earth onto a flat surface.[11]Projection Characteristics
Type and Geometric Features
The Robinson projection is classified as a pseudocylindrical compromise projection, meaning it combines elements of cylindrical projections with curved meridians but does not preserve either angles (conformality) or areas (equal-area property), making it suitable for thematic world maps where visual balance is prioritized over strict metric accuracy.[1][13] Developed by Arthur H. Robinson in 1963, it represents a deliberate design choice to flatten the spherical Earth onto a plane through iterative aesthetic adjustments rather than rigid mathematical derivation.[1][2] In terms of its geometric features, the projection displays the central meridian as a straight vertical line approximately 0.51 times the length of the equator, providing a symmetrical reference axis.[1] Other meridians appear as regularly distributed, smoothly curved lines that bulge gently outward and are concave toward the central meridian, mimicking elliptical arcs without intersecting the parallels at right angles.[1][2] The equator and all parallels are rendered as straight, horizontal lines that are unequally spaced, with the spacing increasing progressively toward the poles to better approximate the spherical distribution of latitudes.[1][13] The poles in the Robinson projection are depicted not as points but as straight horizontal lines, each with a length roughly half that of the equator, which contributes to the overall elliptical shape of the map and avoids the extreme compression seen in some cylindrical projections.[1][2] This configuration of widening latitudinal bands and curved longitudes ensures a visually coherent representation of global extent, assuming a basic familiarity with map projections as techniques for transforming three-dimensional globes into two-dimensional surfaces.[1][13]Strengths and Weaknesses
The Robinson projection provides a visually realistic and aesthetically pleasing representation of the world, balancing moderate distortions in area and shape to create an effective tool for general reference maps.[2] Its design minimizes east-west stretching, offering a more proportionate view of continental extents compared to cylindrical projections like Mercator, which exhibit extreme elongation at higher latitudes.[1] This compromise approach results in relatively low overall distortion within 45° of the equator and central meridian, enhancing its utility for thematic world mapping where perceptual accuracy is prioritized over precision.[2] Despite these advantages, the projection suffers from significant angular distortion near the poles, where scale factors drop to approximately 0.53 relative to the equator, introducing up to roughly 47% scale error in polar regions.[1] As a non-conformal and non-equal-area map, it distorts shapes, areas, distances, and directions, rendering it unsuitable for navigation, surveying, or applications requiring exact measurements.[2] Distortions increase markedly beyond 45° latitude or longitude from the central meridian, with shearing and compression becoming more pronounced at the map's edges.[1] The projection's compromise nature intentionally tolerates these distortions to achieve a systematic balance across the globe, with scale factors varying from 1 at the equator to 0.53 at the poles, rather than adhering to any single distortion-free property.[1] True scale occurs along the 38° N and S parallels, but deviations elsewhere underscore its focus on holistic visual harmony over mathematical exactitude.[14] Arthur H. Robinson developed it through trial-and-error computer simulations, prioritizing an appealing "look" for world maps over geometric purity.[2]Mathematical Formulation
Defining Equations
The Robinson projection operates on spherical coordinates, where latitude φ ranges from -π/2 to π/2 radians and longitude λ from -π to π radians, relative to a central meridian λ₀ (often 0). These coordinates represent positions on a sphere of radius R, typically normalized to R = 1 for computational purposes. The projection maps these to plane coordinates (x, y) using the following defining equations: \begin{align*} x &= 0.8487 \times R \times X(\phi) \times (\lambda - \lambda_0), \\ y &= 1.3523 \times R \times Y(\phi), \end{align*} where φ and λ - λ₀ are in radians, and the sign of y follows that of φ to ensure symmetry across the equator.[15] The functions X(φ) and Y(φ) represent precomputed ratios that define the projection's pseudocylindrical nature: X(φ) scales the length of parallels relative to the equator, while Y(φ) determines the meridional distance from the equator along the central meridian. These functions are not derived from closed-form analytic expressions but are tabulated at discrete latitudes (typically every 5° from 0° to 90°, with symmetry for the Southern Hemisphere), making the Robinson projection table-driven rather than purely formulaic. This empirical approach allows for customized distortion minimization across the globe.[15] The scaling constants 0.8487 and 1.3523 are empirically determined values from Robinson's design process, resulting in an unscaled map (for R=1 and full longitude range) with approximate width 5.33 units and height 2.70 units, yielding an aspect ratio of about 1.97:1 (width:height). Maps are often rescaled to fit desired frames while preserving the relative proportions.[15]Standard Tables and Interpolation
The Robinson projection relies on predefined lookup tables for the functions X(φ) and Y(φ), given at 19 latitude points spaced every 5° from -90° to 90°. The projection is symmetric about the equator, so X(φ) = X(-φ) and Y(-φ) = -Y(φ). The tables originate from Arthur H. Robinson's 1963 graphical design process, where he iteratively adjusted curves to achieve balanced visual appearance and minimal average distortion, rather than deriving from analytical equations.[15] The standard table for northern latitudes (0° to 90° N) is as follows; southern latitudes use the same X values with negated Y values:| Latitude φ (°) | X(φ) | Y(φ) |
|---|---|---|
| 0 | 1.0000 | 0.0000 |
| 5 | 0.9986 | 0.0620 |
| 10 | 0.9954 | 0.1240 |
| 15 | 0.9900 | 0.1860 |
| 20 | 0.9822 | 0.2480 |
| 25 | 0.9730 | 0.3100 |
| 30 | 0.9600 | 0.3720 |
| 35 | 0.9427 | 0.4340 |
| 40 | 0.9216 | 0.4968 |
| 45 | 0.8962 | 0.5571 |
| 50 | 0.8679 | 0.6176 |
| 55 | 0.8350 | 0.6769 |
| 60 | 0.7986 | 0.7346 |
| 65 | 0.7597 | 0.7903 |
| 70 | 0.7186 | 0.8435 |
| 75 | 0.6732 | 0.8936 |
| 80 | 0.6213 | 0.9394 |
| 85 | 0.5722 | 0.9761 |
| 90 | 0.5322 | 1.0000 |