This is a chapter excerpt from Guilford Publications.

Transportation: A Geographical Analysis, William R. Black, Copyright © 2003

Chapter 1

An Introduction
to Transportation Geography

Geography is that branch of science that seeks to explain, predict, and otherwise analyze
the location and magnitude of phenomena at or near the earth’s surface. These phenom-
ena may be actual surface patterns (e.g., cities or river networks), or they may be phe-
nomena that for a given time may be mapped (e.g., peak hour traffic in London during an
average day in 2000, or migration flows between states of the United States between
1995 and 2000). Where sufficiently detailed information is available, the geographer can
examine the processes that have led to the formation of a given spatial pattern.
Transportation is the movement of goods or people from one location to another
location. Broadly defined, this includes just about every type of movement possible, so we
need to limit the definition somewhat. Thus, transportation is the movement of goods
beyond their local production area and the movement of people between different geo-
graphical locations. Such a definition enables us to disregard the movement of goods
within a plant or within an industrial complex, of people within an establishment, and of
the migration of populations, topics that are examined by other disciplines.
The transportation geographer’s interests center on both the location and geographic
pattern of transport systems and the magnitude of the movement or spatial interaction
over the elements of such systems.
Defining geography as that branch of science that seeks to explain, predict, and oth-
erwise analyze the location and magnitude of phenomena at or near the earth’s surface
does have implications for transportation. Implicit in this type of definition is the exis-
tence of some type of differentiation; that is, the surface of the earth differs from place to
place. If this were not the case, there would be no variation. All places would be much the
same and as a result there would be no need for transport and travel beyond local areas
since there would be nothing in the rest of the world different from where we are at pres-
ent. Everyone and every place would have the same resources and capabilities.
However, the surface of the earth is different from place to place. Places differ in

3
4 FOUNDATIONS

terms of population size, language, resources, environmental factors, industrial special-

ization, local history, and human activities. It is these differences from place to place
that generate the demand for transportation. People want commodities that are not
produced locally—for example, fruits, vegetables, televisions, shoes, paper, and thou-
sands of other goods—and this desire generates demand for transport. Businesses need
resources that are often not available locally, and this need generates a demand for
transportation. We vacation in areas that have different histories (battlefields, medieval
castles, colonial origins), different environments (coastal areas, the mountains, wilder-
ness areas, the tropics) and different attractions (largest structures, major amusement
parks, casinos, etc.). We work, shop, worship, and seek entertainment, for the most
part, outside our own neighborhoods, necessitating a journey to work, to shop, to go
to church, to visit the multiplex or video store. Transport enables us to carry out all
these activities. One can see how geography influences transport and transport enables
us to cope with geography.
Transport has other ties to the study of geography. A major portion of the field of
geography has as its focus human interactions with the environment. Transport is one of
the major ways in which humans impact their environment. The atmospheric impacts are
among the best known of these; they include air pollution, greenhouse gas emissions, acid
deposition, and ozone layer destruction. However, there are many other ways in which
transport impacts the environment. It is capable of altering soils, geomorphology, waters,
and plant and animal life. Although these impacts are primarily negative, it goes without
saying that transport brings many good things to our lives as well.
Many years ago transport geographer Edward L. Ullman (1954a) published a paper
entitled “Geography as Spatial Interaction.” Ullman saw the field of transport geography
as the definitive center of the field of geography. He believed that it was the manner in
which humans interact with their world—their spatial interaction—that defined the field.
Some might not go so far today, but Ullman’s views are illustrative of the importance of
this field to the overall discipline of geography.
Geographers do not have a monopoly on the spatial approach or the examination of
spatial relationships. Engineers and other social scientists contributed to the development
of the more scientific aspects of the field recognized here as transport geography. One of
the earliest contributions focusing on the importance of transport and transport costs is
an 1826 work by J. H. von Thunen (1875), Der Isolietre Staat. In this study the author
clearly shows the manner in which distance and transport costs affect land rents. Dunn
(1955) made many of these ideas available to geographers.
Another early contribution to this field came from the work of Leon Lalanne (1863),
a French civil engineer. His paper sought to point out certain fundamental relationships
between the distribution of places, their sizes, and the number of transport routes serving
them. These ideas were to be advanced further during the 20th century with the work of
the geographer Walter Christaller (1933). Another engineer who made a fundamental
contribution to our understanding of transport network development was A. M.
Wellington (1887). Wellington’s concern was, which places should be connected to a rail-
road route that was to pass between two known end points. He recognized a financial
desire to pick up additional traffic, if any was to be had, but not if deviations from a least
cost route were excessive.
Of course, one should not overlook the work of historians in advancing our under-
standing of transport development. In order to understand the development of transport
 An Introduction to Transportation Geography 5

routes and networks, one must take a historical approach. Much of what we know today
about transport route location is the result of research by historians and historical geog-
raphers; moreover, many of the academic mistakes made in modeling are attributable to
ignoring the work of these scholars. The works of Ringwalt (1888) and Dunbar (1915)
are critical to understanding transport development in the United States. Vance (1986)
prepared a more recent treatment of transport development in the United States and
Europe. In the case of British road transport, one should be aware of the works of Ander-
son (1932) and Parnell (1838). Construction of the National Road in the United States
used the ideas and plans of Thomas Telford, an engineer who designed many of the roads
in England during the first and second decade of the 1800s, which were available (Gibb,
1935). Of course, a thorough understanding of world transport history, even for a single
mode, would consume an entire career, but an awareness of at least some of this litera-
ture is advisable for any transport researcher.
Systematic treatments of what could be called transport geography are difficult to
come by prior to the middle of the 20th century. Among the few notable pieces is Mark
Jefferson’s (1928) “The Civilizing Rails,” a cartographic study of world railroad develop-
ment. One can only imagine the time and effort it took for Jefferson to prepare the maps
of that study in comparison to the use of geographic information systems (GIS) and trans-
port databases today that could prepare the same in only minutes (or less). Pearcy and
Alexander’s (1951, 1953) work on airline development is also notable for its complete-
ness on a subject that was still evolving at the time their work was completed.
Ullman’s (1957) American Commodity Flow represents the first systematic attempt
to explain what was happening with commodity movement in the United States. It is
proper to view it as the beginning of scientific treatment of the subject in the United
States.
Ullman, with Harold Mayer (1954b), surveyed the field of transportation geography
in their volume American Geography: Inventory and Prospect. Their paper on transpor-
tation geography examined the field and identified 11 areas of interest to transport geog-
raphers at the time. These were:

1. Examination and mapping of transportation as a measure of relations between

areas.
2. The study of ports, their traffic, and hinterlands.
3. Comparisons of transportation systems and the identification of criteria for such
comparisons.
4. The patterns and selection of transportation routes.
5. The problem of securing adequate data on flows.
6. Estimating and forecasting the demand for transportation.
7. Examination of existing or given traffic flows and evaluation of their efficiency.
8. Analysis of the components of freight rates.
9. Studies of the relationship between transport routes and environmental condi-
tions.
10. Studies of the impact of technology on transport costs and flows.
11. The gravity model and social physics.

This volume examines most of these topics. It does not examine port cities with their
facilities, traffic, and hinterlands as such, since these seem to be more of interest to urban
6 FOUNDATIONS

geographers today. It will, however, examine ports as nodes of a network that both pro-
duce and attract flows from other areas.

THE NETWORK A PPROACH

Most of the other areas of interest in the 1954 Ullman and Mayer survey continue to be
of concern today, although major changes in both transportation research and geograph-
ical research have altered them since that time. Geography underwent some major meth-
odological and philosophical changes during the 1950s. Transportation geography
proved to be a focus of much of this change. Exactly when the effects of this change were
felt in transportation geography is difficult to determine, but, based on the literature, it
occurred between 1956 and 1960. The prime innovators of change were E. L. Ullman, W.
L. Garrison and W. Bunge, all of whom were colleagues at the University of Washington
at the time.
Ullman’s primary contributions were his review of the field (1954b) and his mono-
graph on American Commodity Flow (1957). Although a review is rarely recognized as a
contribution, this one was in that it pointed out several areas where research was needed.
In his subsequent monograph Ullman succinctly specified the three basic determinants of
trade and spatial interaction: complementarity, intervening opportunities, and transfer-
ability—all of which are discussed later in this book. In addition, Ullman was one of the
first geographers to suggest that the gravity model should be applicable to the analysis of
flows between regions.
Bunge’s (1966) Theoretical Geography with its emphasis on spatial variables and
location helped to fill a partial void in transportation geography research created by a
paucity of studies concerned with the location of transport routes and networks. A possi-
ble explanation for this lack of development was that locations of these facilities were
examined only if they affected the pattern of circulation (movement). Bunge suggested
that the location patterns of the facilities were worth examining in and of themselves, and
he ventured outside the geographical literature in his search for theory relevant to the
locations of these facilities as well as for theory that would increase our understanding of
movement and spatial interaction.
Garrison’s influence on transportation geography began in the late 1950s with the
publication of three review papers on the spatial structure of the economy (Garrison,
1959a, 1959b, 1960a). The first two papers concerned location theory and transporta-
tion costs as a determinant of the optimal location of production. The third paper in the
series noted the importance of the direct study of transportation networks as a research
area rather than as an explanatory variable in the study of spatial interaction. At about
the same time, Garrison, often working alone and often with Ullman and Brian Berry,
tested the gravity model in an attempt to verify the concepts of interaction (Garrison,
1956b), and made research contributions in the area of network impact and network
enlargement that are regarded as classics today (Garrison & Berry, 1957; Garrison &
Marble, 1958; Garrison, Berry, Marble, Nystuen, & Morrill, 1959).
The association of Garrison and Marble (1965) also resulted in the Prolegomenon to
the Forecasting of Transportation Development. This research report and the preliminary
papers from which it was formed had a significant impact on the scope and methods uti-
lized in transportation geography for more than two decades, and, it continues to be a
 An Introduction to Transportation Geography 7

frequently cited source in any discussion of the network approach to transport geogra-
phy. Of equal importance at the time was Haggett’s (1966) Locational Analysis in
Human Geography, which was followed a couple years later by Haggett and Chorley’s
(1969) Network Analysis in Geography. These were major textbooks that promoted and
spread new methods into all areas of the discipline that were known only to a small
group of scholars before this time. An excellent summary of the network literature will be
found in Leinbach (1976).

EXTERNAL INFLUENCES ON THE FIELD

In addition to the work of the individuals discussed so far, several external factors influ-
enced the direction that transportation geography research in the last half of the 20th
century would take. There were three major stimuli. One of the most significant was the
post-World War II recognition by city planners, governments, and so on of the transpor-
tation problems facing urban areas. Initially, geographers avoided these problems primar-
ily because the field was very descriptive at the time and had not developed ways to
analyze them. But they couldn’t avoid them forever, because the problems were inherently
geographical in nature. The problems had to do with areal variation in traffic produced
in subareas of a city (trip generation); the examination and prediction of flows between
subareas (trip distribution or spatial interaction); and the identification of minimum dis-
tance or time paths for these flows over a network (trip assignment). Perhaps no other
subfield of geography has problems that are so inherently geographical.
A second external influence was the development of research technologies capable of
handling the complex problems encountered in transportation research. High-speed com-
puter hardware and the development of linear programming are prime examples of these
technological changes. The field of operations research, which is itself a product of these
innovations, served as a stimulus to both Garrison and Bunge in their previously noted
works.
A third stimulus to occur at this time was an apparent decrease in the importance of
transportation economics as a subfield of economics. In addition, economists throughout
the United States and elsewhere ignored transportation and its costs when discussing the
theory of the firm. Isard (1956) referred to this as the “anglo-saxon bias.” The few trans-
port economics programs that did exist were being absorbed by courses in public utility
economics or logistics. Prior to this absorption, this branch of economics was concerned
with the historical development of transportation networks, transport geography, the
location of transport terminals, and traffic in urban areas. There was also a major con-
cern for spatial variation in freight rates and the role of government regulation in the
transportation industries. These latter topics dominated textbooks of the 1960s and
1970s. As early as 1955, the economist Troxel had noted that transportation was primar-
ily a geographical problem. His text should be of interest to most transportation geogra-
phers today since it emphasizes the importance of location and space on flows, as well as
the influence of flows on route location.
The network approach was the first major paradigm to emphasize a rigorous sci-
entific approach to research in transport geography. It continues to be influential even
today, with a growing interest in the use of GIS and remote sensing of networks. It
also is a major component of the research of several scholars today (see Black, 1992a;
8 FOUNDATIONS

Black & Thomas, 1998; Ivy, 1993; Kuby & Gray, 1993; O’Kelly & Miller, 1994;
Shaw, 1993).

OTHER RESE A RCH PA R A DIGMS

During the late 1960s and early 1970s there were major concerns regarding the way in
which transportation was affecting society. Attempts were underway to save public tran-
sit. Studies of ridership by the young, the elderly, and various other groups were under-
taken. Equity issues became a concern and studies of accessibility and mobility were
undertaken. Rimmer (1978) tried to redirect the field to what he called a “humanistic
transport geography” which resulted in studies of choice of mode, choice of route, and so
forth. Much of this work was driven in the United States by concerns over relevance, but
as noted elsewhere “it did little to advance the field, since it tended to ignore theory
development” (Black, 1989, p. 317).
The development of this social-behavioral research paradigm had slowed by 1975. It
was followed by a concern with regional transport systems brought on in large part by
the move toward transport deregulation as well as by energy and environmental con-
cerns. The earliest federal deregulation legislation in the United States was enacted in
1973 for the East and the Midwest. It was enacted in response to the 1971 bankruptcy of
the Penn Central and other railroads. General railroad deregulation legislation followed
in the Staggers Rail Act of 1980. In between these dates the airline passenger industry
was deregulated in 1978. Deregulation of motor carriers and intercity bus transport
occurred in 1982. Numerous theses and dissertations examined the impacts of these legis-
lative actions. This examination continues today, as some believe deregulation has gone
too far, resulting in some areas of the United States losing mobility. The same is true in
parts of Great Britain, which has seen its own series of deregulation actions. These have
prompted suggestions of the need for re-regulation. Numerous transport geographers
became involved in policy questions during this period (see O’Sullivan, Holtzclaw, Bar-
ber, 1979; O’Sullivan, 1980; Black, 1986). This interest in government policy continues
today for many transport geographers.
The 1990s saw a growing interest by transport geographers in transport technology
(Garrison, 1984; Garrison & Ward, 2000), infrastructure, intelligent transport systems
(see, e.g., Haynes & Qiangsheng, 1993; Stough, 2001), and sustainable transportation.
Probably the bulk of the transport geography research during that decade fell under the
last of these themes. Transport infrastructure consists of all the facilities that are used to
provide transport services and includes, among other things, highways, streets, and
roads; railroad tracks and trestles; runways, terminals, and transit facilities. Creating
“intelligent transport systems” involves the use of advanced technologies to increase the
efficiency and safety of our transport operations and networks. A “sustainable transport
system” is one that meets today’s transport needs and is expected to meet the needs of
future generations. For various reasons our current transport systems are not sustainable.
Transport geography research being undertaken today is not confined to the last
three areas cited. Indeed, it would probably be possible to find research continuing in all
of the major areas mentioned here. This is desirable. It also means that our interests are
continuing to develop, which is good for the field of transport geography.
It should be recognized that each of the research areas pursued in the field of trans-
 An Introduction to Transportation Geography 9

port geography from the 1950s to the end of the century was in response to an initial
public concern that led to federal research funding. This is not unique to transport geog-
raphy; the same pattern can also be found in physics, medicine, and many other fields. It
is not necessarily bad either. There is considerable satisfaction in seeing a transit system
begin that one has planned, or subsidies provided to rail lines identified by your research
as worthy of support, or participating in the Penn Central reorganization—the largest
corporate reorganization ever undertaken up to that time. But these activities do seem to
take us away from further development of the field of transport geography, or do they?
Perhaps it is best to have a field whose members have skills that can be applied to con-
temporary problems, even if such applications slow the field’s conceptual and theoretical
development.

OV ERV IEW

The remainder of this part includes two more introductory chapters. In Chapter 2 we
briefly examine the history of transportation development. This will place much of what
has occurred in transport geography in a context that makes it more understandable and
will also clarify how we got to our current position. Our emphasis is on the United States,
but we also make passing references to developments in Europe and elsewhere. In the
U.S. case, development has occurred in the last 400 years and most of this development is
documented; developments elsewhere in the world occurred during an era when docu-
mentation was unusual, and as a result it is difficult to follow. The exceptions to this are
the development of some of the roads of the Roman Empire and of the Incas of South
America. Our knowledge of some of these systems is due to archaeological research.
Chapter 3 examines current transport systems and trends in travel and transport. In
this chapter we take stock of what is occurring today, which will delimit the nature of
current transport problems in the world. Again, the primary focus will be on the United
States and Europe, but we also examine some problems and trends elsewhere in the
world.
The chapters that form the remainder of this text fall into five parts. In Part II, on
network analysis, we examine the basic elements of networks, their nodes and links, and
examine how their attributes can be measured. In Chapter 5, we give some attention to
how networks can be placed in a form that allows analysis. Basic notions of connectivity,
accessibility, and matrix representation are treated, as are measures of these and the
structure of these networks. Part II concludes with an examination of the location of
transportation routes (Chapter 6). Although location is fundamental to geographic stud-
ies, it has not received the attention it deserves in transport geography.
Part III, on flow analysis, begins with Chapter 7, which examines trade and com-
modity flows between different areas using the ideas of Ullman and moves on to discuss
some elementary models of trade that have appeared in the literature. Spatial price equi-
librium and some ideas regarding the stability of flows over time are also presented here.
But there are other methods of flow analysis, which are the focus of Chapter 8. Included
are discussions of transaction flow analysis, optimal flow systems, factor analysis of flow
matrices, and network autocorrelation analysis.
Often we don’t know what the “demand” for movement is, so it is necessary to esti-
mate this demand. This is referred to here as the “prediction problem of flow genera-
10 FOUNDATIONS

tion,” the subject of Chapter 9. This basic question must be answered if we are to look at
flows within a nation or a metropolitan area. In some cases travel demand can be easily
estimated, but in other cases the problem becomes very complicated—as we will see.
Chapter 10 of the flow analysis part examines spatial interaction—for example, the
movement of people between places. It begins with a model where distance plays no role
and moves on to the case where distance influences flow. The use of basic gravity poten-
tial models applicable to problems having a single origin or a single destination are exam-
ined, as is the use of gravity models for estimating all possible flows between a set of
places, states, or countries. Several different gravity models are discussed in terms of their
accuracy and logic. Also discussed here are the use of neural network flow models and
some other flow models that are worthy of consideration.
Of course, understanding the amount of flow between places does not necessarily tell
us how it will move between these places. This is the subject of Chapter 11, the last chap-
ter in the flow analysis part of this book. Here we introduce the use of choice models,
models that predict the modes or routes that flows will take. These are basic geographic
questions wherein we try to understand the manner in which people and shippers evalu-
ate their “costs” of spatial separation.
Part IV addresses the reality that many of the problems examined by transport geog-
raphers today have an impact on policy or are themselves impacted by policy (Chapter
12). Many of the goals of public policy in the transport sector create impacts that are spa-
tial. The intent is to have positive impacts, but this is not always the result. In order to
minimize adverse impacts we get involved in transport planning, which we also examine
in this part (Chapter 13). In some cases planning actions may result in negative impacts.
Methods of analyzing these negative impacts are reviewed here (Chapter 14), as are spe-
cific impacts on the environment (Chapter 15).
A final topic covered in this part is the interrelationship between transportation and
development (Chapter 16). In the developed world transport investments are often used
to stimulate the development of specific areas. In some cases these efforts are successful,
but in other cases they fail. What determines success and failure in these cases? Is it rea-
sonable to think that a transport investment can change the economic future of an area?
These and other questions are examined for both the developed world and the developing
world.
Part V (Chapter 17) addresses the use of geographic information systems (GIS) in the
field of transport geography. GIS makes possible sophisticated analyses by a researcher
that would have required substantial research funding two decades ago. It gives the
researcher the capability of handling extremely large problems and examining numerous
alternatives and outcomes. Although one cannot learn GIS in a chapter, the intent here is
to introduce the subject and at least provide a platform for further work. Also examined
here are some applications of remote sensing in the transportation area and their integra-
tion into GIS.
Current societal trends, congestion, and sustainability form the focus of Part VI. Sig-
nificant changes occurring in today’s global society have impacts on transport (Chapter
18). We have an increasing number of single-parent households, which nearly doubles
urban travel demand; larger numbers of women are joining the work force, thereby add-
ing to the vehicles that congest our highways; the elderly are becoming a larger propor-
tion of the population with travel needs they often cannot meet. The United Kingdom
and the United States have undertaken welfare reform without any clear recognition of
 An Introduction to Transportation Geography 11

the transportation implications of these actions. Also included in this general area are
moves toward deregulation of the transport sector, privatization of that sector, and the
controversial moves toward globalization. Globalization of industry results in longer and
longer shipment lengths for products. Trading partners are no longer contiguous, which
results in additional transport inputs to the production–consumption process. We exam-
ine these and other changes and their implications for trade and transport.
Congestion (Chapter 19) is a significant problem throughout the developed and
developing worlds. There is a growing recognition in the United States that highway con-
struction has its limitations, and that therefore actions must be taken to control the con-
gestion on existing facilities. In the developing world we are seeing rapid motorization.
The use of these vehicles exceeds the capacity of the existing road network, resulting in
significant congestion levels. Congestion is the subject of the second chapter of this part.
Sustainable transportation (Chapter 20) is a notion that slipped into the literature of
transportation geography in the late 1980s and has remained. Its origins were in the envi-
ronmental area, but it is viewed in a broader manner than that today. A chapter on this
topic defines the term and its components and suggests how it could be measured. It is
apparent that although we would like to see transport systems that are sustainable for
this and future generations, that may very well mean we must be less mobile. Unfortu-
nately, most countries of the world do not want to give up any of their mobility and
many more countries are interested in increasing their mobility. The chapter develops an
index of sustainable transport that is coupled with the notion of potential mobility and
applies this to more than 100 areas of the world. The chapter illustrates not only how to
go about constructing an index, but also what such an index can tell us about this notion
of sustainability and what it means for future mobility.
Part VII (Chapter 21) looks ahead to what can be expected for transportation in the
coming decades. Chapter 21 examines different scenarios of the form that transportation
will take within the next two to three decades in the developed world. The transportation
systems of the future may be simply an extension of what we currently have, with all of
the problems simply worse than they are at present. Or we may see some significant
actions taken to improve the current situation. It has been said that all our transportation
problems are solvable, and this is true. The problem is that we often do not have the will
or the resources to change. Two people in every car would nearly solve congestion prob-
lems, and expensive hydrogen fuel would eliminate pollution, but people are not anxious
to do the former and the price of the latter makes us hesitate. The future transport world
with its problems (and solutions) is rarely a subject of transport geography textbooks.
Admittedly, the subject is often full of conjecture and colored by the author’s biases, but
even this is relevant if it promotes discussion. These are problems the readers of this vol-
ume will experience through most of their lives.
Just as important as these ideas are questions of what will remain as significant
research problem areas in transportation geography in two or three decades. My final
chapter will make some suggestions in this area.

Guilford Publications
Copyright © 2003 The Guilford Press. All rights reserved under International Copyright 72 Spring Street
Convention. No part of this text may be reproduced, transmitted, downloaded, or stored New York, NY 10012
in or introduced into any information storage or retrieval system, in any form or by any 212-431-9800
means, whether electronic or mechanical, now known or hereinafter invented, without 800-365-7006
the written permission of The Guilford Press. www.guilford.com