A map of global rainfall from a 1922 British atlas. The starting point is again point data from individual weather stations, but the shapes are drawn by hand, without algorithms. Compared to present-day maps, the obviousness here of certain coastal areas and mountain ranges shows a preference for realism – a memorable, causal relationship between regional topography and precipitation. These relationships are much less visible on strictly accurate maps. From J.G. Bartholomew, The Times Survey Atlas of the World (London: The Times, 1922), plate 3.

Have you ever wondered how the brightly coloured environment maps we’re all familiar with are actually produced? If you read William Rankin’s article ‘The Accuracy Trap’ you’ll never look at these maps in quite the same way again. 

Published last year as a fast track article in Environment and History, ‘The Accuracy Trap’ has just been awarded the 2021 Joel Tarr Envirotech Article Prize. To celebrate, we have made it free to download for one month:


The prize committee wrote:

‘How are environmental maps created, and what do they mean? Bill Rankin’s compelling article focuses on a fundamental but easily overlooked component of environmental mapping: the underlying algorithm. Weaving together insights from environmental history, visual studies, and critical cartography, Rankin traces the development and spread of now-ubiquitous mapping algorithms from their Cold War origins in the fields of mineral extraction (in postcolonial Africa) and military weather forecasting (in the USSR). Although their use was limited at first, such algorithms sparked intense philosophical disagreement among specialists over the goals of environmental visualization – between what Rankin calls accuracy and realism. However, epistemological tensions around the purpose of visualization have largely been forgotten as advances in computing made algorithms more accessible to non-specialists. In an increasing retreat from human subjectivity, mapping has become a technique of data-processing, hollowing out the meaning of accuracy. By probing the limits and possibilities of accuracy in environmental mapping, Rankin offers a welcome reminder of the contingent and contested nature of visual representation, and a model for linking the histories of environment and technology to questions of epistemology and visual culture.’

Tarr Prize Selection Committee:
Camille Cole, University of Cambridge
David Fedman, University of California-Irvine
Aristotle Tympas, University of Athens

Global temperature change, 1880–1985, with data from 2,645 weather stations (on land only) interpolated to a coarse grid. Stations within 1,200 kilometres were combined at each point, with nearby stations given more weight; using this technique, coverage in 1900 was not much less than in 1950, despite a drastic change in the number of stations. The overall conclusion was that global temperature had risen between 0.5° and 0.7°C in a hundred years. From James Hansen and Sergej Lebedeff, ‘Global trends of measured surface air temperature’, Journal of Geophysical Research: Atmospheres 92 (Nov. 1987): 13356.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s