Premium
THE MAPPING OF TRAVEL TIME IN EDMONTON, ALBERTA
Author(s) -
Muller JeanClaude
Publication year - 1978
Publication title -
canadian geographer / le géographe canadien
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.35
H-Index - 46
eISSN - 1541-0064
pISSN - 0008-3658
DOI - 10.1111/j.1541-0064.1978.tb01012.x
Subject(s) - geographer , diagonal , matrix (chemical analysis) , column (typography) , distance matrix , combinatorics , mathematics , representation (politics) , order (exchange) , metric (unit) , geography , geometry , cartography , operations management , materials science , finance , connection (principal bundle) , politics , political science , law , economics , composite material
T ime ‐ distance is defined as the time required to travel a specific distance. Consider the time‐distances among n places. They are tabulated in an n by n matrix and it is assumed that each element of the square array represents the minimum travel time to go from one place to another. Usually the matrix will be non‐symmetrical with all diagonal elements equal to zero. Imagine a graphic representation of those places in a two‐dimensional space. Theoretically it is possible to determine two configurations, one arising from trip i to j , the other from trip j to i , such that the locations of all n points approximate the n ( n ‐1)/2 time‐distance relations. 1 In practice, however, the geographer will not be satisfied with a solution that does not preserve geographic neighbourliness. He will argue that geographical order is a necessary ingredient to the understanding of place relations. In most cases, a total solution that reconciles the geography with the metric is unattainable and only segments of the time‐distance matrix are mapped. On polar isochronic maps, for instance, places are located according to their geographic azimuth and their time‐distance with respect to one single origin. Therefore only one row or one column of the original matrix is plotted. The number of maps required for representing the entire data would be twice as large as the number of observations.