Assessing the accuracy of historical maps – the Survey of Western Palestine as a case study
ISBN 978-85-88783-11-9
Authors
1Levin, N.; 2Schaffer, G.
1HEBREW UNIVERSITY OF JERUSALEM Email: noamlevin@mscc.huji.ac.il
2HEBREW UNIVERSITY OF JERUSALEM Email: gad.schaffer@mail.huji.ac.il
Abstract
The use of scientific based methods for field surveying and mapping started in Western Europe in the 18th century, with the creation of some of the world’s first large scale topographic mapping. With the increasing accessibility of digital archives of map collections, there is now a wealth of historical maps that can be examined using geographic information systems. Historical maps can be very useful for studying land cover changes, extending our time frame to the past 200-300 years. However, the accuracy and completeness of historical maps needs to be determined, so as to assess the reliability of reconstructing past landscapes. Here, we present a thorough examination of the geometric and thematic accuracy of the Survey of Western Palestine. Surveyed between the years 1871 and 1877 and finally published in 1881, the Palestine Exploration Fund (PEF) is composed of 26 sheets at the scale of one inch to the mile (1:63,360) covering half of Ottoman Palestine. The geometric accuracy of the PEF map was examined by comparing the locations of triangulation points on present-day topographic maps (RMSE = 74.4 m). The entire map was digitized to reconstruct Pales-tine’s past land cover. To examine map completeness, the digitized land cover map was compared to two sources: an 1870 French 1:100,000 topographic map covering the Galilee region (northern Palestine), and World War I aerial photos. The overall correspondence between land cover of the PEF map and the other historical sources was above 50% in most test sites, and was not affected much by the georeferencing method applied. Having several contemporary spatial historical sources, allowed us to assess their reliability and to identify land cover classes that were mapped with a lower certainty and completeness. We were also able to identify some of the causes that led to observed differences between the maps, including mapping scale, time of year and the interests of the surveyors.
Keywords
Historical GIS; map accuracy; map completeness