51:11.2

3—,.'
“711*

{v :
5.3%.???
:51: y
r.

:vvvrv

x 311.: a...“
.1 an: .1.
hair: 3
783.33% Tu

,...

:2: L .._...n.
3. ..........:...£
1'15? I gyf ll

3... E.

.- . 1 II ‘

2.. ti
b.m.f.v§.§.u .4:

3th .2... .3: ..

1:. t. x z... a.

5?...“ . 5...; 1.

:11, v .

n .. bunny."

v.3... um!!!

.«
.1?"

4r 8...:
3:33....rilx
193...“: .13

v. 1., \ 1 3.1.153...
. ant-Tait...) L
11:511.!20! r

 

 

 

t l. ‘9‘) t5
, . .55 ..3 .2 X

.57, v :1

‘ t. 11:27.
| 6:11.33. xmh

..l..r!cr\4 u
:karla. . 11!):

.21.. h\!17?{. .

LN...
\5.‘ v

R. I. 1. 3:1 1|"...l‘fd? t (:

345‘!!!ch .1. :1‘0. .1 )3.

‘L’llllahiil’ £11.; zurv“.¥..v(
, Vistvsil“ .

7i... .1

.2. 5.x:

1.4»! I! 1

 

 

 

 

 

MICHIGAN STATE UNIVERSITY LIBRARIES
q t

31293 01563 5166

This is to certify that the

dissertation entitled

CONCEPTUALIZING SPACE AND TIME: A CLASSIFICATION OF
‘ GEOGRAPHIC MOVEMENT

presented by

Nancy J. Yattaw

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Geography

 

 

 

Date 7%93/47

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES retum on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

#0? n 3 ”car:-

 

 

V

 

w

 

 

 

"" 2C :33

\rvl—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
c:\clvc\ddeduo.nm3—p. 1

CONCEPTUALIZING SPACE AND TIME: A CLASSIFICATION OF
GEOGRAPHIC MOVEMENT

By

Nancy J. Yattaw

A DISSEKFATI O N

Submitted to
Michigan State University
in partial Fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Geography

1997

Copyright by
NANCY J. YATI‘AW
1997

ABSTRACT

CONCEPTUALIZING SPACE AND TIME: A CLASSIFICATION OF
GEOGRAPHIC MOVEMENT

By

Nancy]. Yattaw

Movement, in a spatial and temporal context, is a significant and unifying theme in
geographical understanding. Geographic movement, or spatial change of a phenomenon, is
associated with virtually all geographic description, explanation, and analysis at some
scale. Because geographic description details the spatial component of a phenomenon, and
since most geographic explanation strives to uncover patterns and processes that are, for the
most part, inherently linked with change, much discussion of geographic phenomena, in one
way or another, considers movement. Conceptualizing the dynamic nature of a
phenomenon as it evolves through space and time is a grounding principle that broadens our
understanding of geography. In an increasingly complex and interactive world, the need to
study, manage, and understand the critical interface between physical and human
environments is becoming more urgent and has precipitated technological advancements in
the dynamic spatiotemporal capabilities of geographic information systems (GIS) and
associated visualization and representational techniques. Such efforts require a thorough
knowledge of the intrinsic characteristics associated with changes in space, coupled with
changes in time. A comprehensive understanding of the underlying dynamic structure
associated with movement is necessary before accurate, integrative, and insightful analyses,
explanations, and representations of complex, geographically dynamic phenomena can be
performed.

Using the logical subdivision method of classification, geographic movement is
conceptualized by its abstract spatial and temporal components creating a matrix of
twelve unique classes characterizing the form of movement. Based on three temporal

conditions identified as continuous, cyclical, and intermittent, in combination with the

four fimdamental features of space represented by points, lines, areas, and volumes, a
classification of geographic movement is presented. The organization is a simplified
framework that oflers a movement language with which to describe the movements
associated with diverse geographic phenomena that includes all physical, social, economic,
cultural, and political realms of geography. Classifying geographic movement by its
changes in time and space serves as a useful construct that facilitates the comprehension of
the general ideas and principles intrinsic to geographical patterns and processes and
provides a solid foundation for greater explanation and understanding of the complex

interactions among diverse geographic phenomena.

7?) my amily and fiiends,
who have traveled t is evolutionary journey with me.

7?) Zach and Tohes,
who spent countless hours with me on the computer.

And especially to I/Vill,
whose enduring patience, encouragement, and understanding
I will not soon forget.

ACKNOWLEDGEMENTS

I am grateful for an advisory committee that supported a dissertation that was
broadly—based and uncharacteristic of the department’s usual doctoral research approach.
Specifically, I thank my major advisor, Dr. Richard Groop, for his invaluable knack of
adding structure and clarification to my ideas and thoughts, Dr. Judy Olson for her
inquisitiveness on the subject matter of geographic movement, Dr. David Campbell for his
encouraging and enthusiastic nature, and Dr. Gary Manson for his helpful thoughts on where
my research fits into geography and science.

I ’d also like to thank Dr. William Lovis from the Department of Anthropology for
serving as Dean’s Representative and for his insightful comments on my work’s relevance
outside the realm of geography.

Finally, my sincere appreciation is extended to those friends and colleagues with

whom I shared innumerable conversations with regarding my academic pursuit.

vi

 

.. movement is a very significant and central element in geographical understanding...
understanding movement should lead directly to a better understanding of what is happening,
a necessary accompaniment to the understanding and study of what is there.

Roger Robinson
Ways to Move, The Geography of Networks and Accessibility (1976, preface)

We currently possess in geography a sound hasic methodology for classification. Let us ensure...
that we make an eflicient use of [classifications] in pursuing a deeper understanding of the

phenomena we are concerned with.

David Harvey
Explanation in Geography(1969, 348)

it is conceivable that a synthesizer may someday, in flat, suggest a schema according to which

a variety of forms of movement and phenomena can he viewed together

John C. Lowe and S. Moryadas
The Geography of Movement (I975, 226)

 

TABLE OF CONTENTS

LIST OF FIGURES

Chapter 1
MOVEMENT AND GEOGRAPHY

 

Research Problem

 

Chapterz
CLASSIFICATION IN SCIENCE AND GEOGRAPHY

 

Classification Define-cl
The Role of Classification in Scientific Explanation

 

Evaluating a Classification

Classification Purposes

Procedures for Classifying
Logical Subdiv' '
Agglomerarinn

The Fundamentals of Classifying

 

Chapter 3
GEOGRAPHIC MOVEMENT

 

 

Properties of Geographic Me.

 

Snare
r

Time

Scale

 

Treatments of Movement in Geography
Movement as a Central Theme

 

The Who or What of Geographic Movement

 

The Where of Geographic Movement

 

 

The When of Geographic Mov
The Why of Geographic Me . ‘

 

The How of Geographic Mv .

AAAAAA'XN

 

 

The Importance of Movement in Geography...

 

Chapter4
A CLASSIFICATION OF GEOGRAPHIC MOVEMENT
Class 1 Continuous Point Movements

 

 

Class 2 Cyclical Point Me.

 

 

Class 3 Intermittent Point My.

 

 

Class 4 Continuous Linear Movements

 

Class 5 Cyclical Linear Mov

 

Class 6 Intermittent Linear Movements

 

Class 7 Continuous Areal Movements

 

viii

Class 8 Cyclical Areal Mu \
Class 9 Intermittent Areal Movements
Class 10 Continuous Volumetric Mov

 

 

Class 11 Cyclical Volumetric MW

 

Class 12 Intermittent Volumetric Mov

 

Multi—Type Phenomena

 

Chapter 5
CONCLUSIONS

LIST OF REFERENCFS

LIST OF FIGURES

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2.1 An Example of the Logical Subdivision Method of Classification ................. 2C
Figure 2.2 An Example of the Agglomerative Method of Classificarinn 97
Figure 4.1 A Classification of Geographic Mu. 72
Figure 4.2 Scale Paradigm of Continuous Point M... 78
Figure 4.3 Scale Paradigm of Cyclical Point Me. RI
Figure 4.4 Scale Paradigm of Intermittent Point Mu. * 87
Figure 4.5 Scale Paradigm of Continuous Linear Movements 92
Figure 4.6 Scale Paradigm of Cyclical Linear Mov 97
Figure 4.7 Scale Paradigm of Intermittent Linear My - .s 102
Figure 4.8 Scale Paradigm of Continuous Areal Movements 104
Figure 4.9 Scale Paradigm of Cyclical Areal M... r 116
Figure 4.10 Scale Paradigm of Intermittent Areal Movements ........................................ 120
Figure 4.11 Scale Paradigm of Continuous Volumetric Movements ................................ 123
Figure 4.12 Scale Paradigm of Cyclical Volumetric Movements .................................... 132
Figure 4.13 Scale Paradigm of Intermittent Volumetric Me. 136

 

Chapter 1
MOVEMENT AND GEOGRAPHY

To understand geographic phenomena, one must often comprehend the
interrelationships between various geographic elements, including space and time.
Understanding the dynamic nature of these inherent structures is necessary for the
description, explanation, and analysis of the interactions of geographic phenomena with
earth, humans, and each other. Using the traditional positivist approach to science,
phenomena from all branches of geography are generalized by abstracting the movements
associated with these events, occurrences, and processes. This conceptualization is presented
as a paradigmatic classification that facilitates the comparative study of all geographic
movements by summarizing their similarities, differences, and relationships as they occur
through space and time and across all scales. As such, the conceptual framework bridges the
gap between the various scientific approaches and fields of study not only within the
discipline of geography, but within other disciplines concerned with spatioremporal
structures.

Movement is an integral part of geography. Spatially dynamic geographic phenomena
are those that refer to motion, the elements and connections of which consrantly change
Spatial location (Griffith and Lea 1983). Geographic movement, or spatial change of an
Object, is associated with virtually any geographic description, explanation, or analysis at
some scale. From transportation routing and commodity exchanges to the theories of
diffusion, migration, and desertification, movement over time is an inherent feature in
g€0graphic understanding. Sometimes geographical analyses involve the direct examination
of the motion itself, such as daily trip pattern studies where personal use of time and space in
an urban environment are examined. With others, spatial change may not be the subject of

1

#4.!

2

direct inquiry, but rather may be viewed historically in the examination of a phenomenon
over time, as with the case of ‘moving regions.’ In describing a region such as the corn belt
in the United States, the boundaries of the area may be examined at a particular point in
time and are considered static. At this point, several things may be analyzed such as what
attributes comprise that region, how this region is similar to and different from other
regions, and what eflcect the region has on the population and the landscape. However, an
even greater understanding of the properties and influences of the region can be gained by
examining the corn belt over several years. With the passage of time, the composition and
boundaries of the region change, causing consideration of the dynamics of a region and
leading to new insights with regards to patterns related to that area.

This inclusion of the time variable, which permits a description of change in the spatial
location of the variables represented, is what separates a dynamic process from a static model
(Martin, Thrift, and Bennett 1978). Geographic explanation may begin with static
descriptions, but dynamic processes are eventually factored into analysis as change over time
is observed, analyzed, and interpreted (Graf and Gober 1992). As Hardwick and Holtgrieve
(1990, 310) note, “static views of the earth are useful for descriptive purposes, but true
understanding of movement and change is necessary for analysis.” Conceptualizing the
dynamic nature of a phenomenon as it evolves and interacts through space and time is a
crucial principle in geographical understanding.

Traditionally however, movement has not been regarded as a significant integrating
concept within the geographic discipline. Rather it has been viewed or interpreted as a
consequence or byproduct of something larger that is taking place. Theories of movement
for the most part have been lodged under different headings, such as spatial interaction and
migration theory. Focus has usually centered on the overall process of which motion is a
part; that is, the ‘what,’ ‘why,’ and ‘where from’ and ‘where to’ of a subject that is moving.
Movement has often been relegated to the last chapters of regional geography books, a “mere

appendage” to the meat of regional description, even though it lies at the core of many

3

geographical analyses (Robinson 1976, v). However, in recent years geographers have
sought to understand the activities and movements on the earth’s surface which help to
explain the distributions of humans and their works (Robinson 1976).

In 1984, the Association of American Geographers and the National Council for
Geographic Education developed a set of guidelines for geographic education for elements
and secondary schools across the United States. The joint committee felt that in order to
improve geographic education, the very definition of what constitutes geographic knowled
had to be clearly articulated (Knight 1994). In support of the interdisciplinary nature of
geography, the guidelines define five central geographic themes: location, place,
relationships within places (human—environment interaction), regions, and movement (Joir
Committee on Geographic Education 1984). The framework, published in the landmark
Guidelines for Geographic Education, is instrumental in demonstrating that geography is no
mere place names (e.g., capitals, countries, and rivers), but rather is the whole science of ti
significance of location. Other frameworks for school curriculum planning can be found i
the geographic literature because, as Hill (1989, 4) explains, “changing perspectives are, in
fact, to be expected in a dynamic discipline.” However, the five fundamental themes, as
defined, serve as a useful mechanism by which geography can be clearly understood.
Acknowledging movement as a central theme within the geographic discipline is an
important step towards greater and more comprehensive understanding of geographic
phenomena.

With the technological advancements of geographic information systems (GIS), spat.
decision support systems, and visualization techniques, coupled with the exponential
accumulation of spatiotemporal data and the expanding data handling capabilities of
modern computers, the time has come when the ideas of temporal geography and comple)
dynamic process models developed in various fields can begin to be tested and refined in a
integrated environment (Peuquet 1994). Currently, representations typically used within

GIS assume a world that exists only in the present. Information contained within a spatia

4

database may be added to or modified over time, but a sense of dynamics, or change thro
time, is not typically maintained (Peuquet and Duan 1995). Enhancement of the
spatiotemporal capabilities of geographic information systems is one area of research that
receiving much attention. A key element in this work is how to store historic and
anticipated geographic data effectively so that geographic information systems are capablt
tracing and analyzing changes in an area and ultimately of allowing temporal modeling a.
simulation of geographic processes (Langran 1992; Peuquet and Duan 1995). Possible
fiiture scenarios of implementing spatiotemporal GIS could include the forecasting of dis
spread in forest resource management, gauging urban development rates, and monitoring
and evaluating long—term land use and cover changes for conservation and planning purp
(Langran 1989a).

Researching dynamic geographic processes has also begun to occur with visualizatio.
techniques. Visualization, as an act of cognition, is the human ability to develop mental
representations by manipulating, rotating, twisting, and inverting two— or three—dimensio
pictorially presented visual stimuli (MacEachren 1992; DiBiase et al. 1992). This ability
identify patterns and create or impose order involves recognition, retention, and recall of
multidimensional structures in which change among the spatial, temporal, and attribute
dimensions is depicted (Golledge, Dougherty, and Bell 1995). Visualization in science
serves many purposes: exploring unfamiliar data sets, identifying meaningful patterns or
anomalies in data, and communicating insight revealed by visual analyses (DiBiase et al.
1992). Visualization techniques use graphic representations such as maps, remotely sense
images, graphs, sketches, diagrams, and photographs to make spatial contexts and problei
visible by bringing three—dimensional objects to life in our visual imagination (MacEachri
et al. 1992; Dorling 1992). The development of new and more powerful visualization to
such as computer animation and three—dimensional modeling has become a major
preoccupation in current scientific visualization research. Computer animation allows

geographers to better understand dynamic phenomena by incorporating change in space :

 

5

through time with the use of movement, perspective shading, and shadows to com
for the lack of an actual third spatial dimension on the flat computer screen (DiBi:

1992).

Research Problem
Geographic explanation revolves around the spatial element of the phenomer
investigation since space is an inherent feature in geography. Space is also naturall
with time; the dimensionalities of these inherent geographic properties interact wiI
another as they evolve and change across the landscape. Such transformations in 3
through time occur across all scales and refer to the movement associated with geo
phenomena. Identifying geographic movement by the spatial and temporal config
leads to the heart of geographic inquiry: “much can be learned about the world by
its forms and patterns, even when such observation does not contribute directly to
explanation” (Goodchild 1992, 159). Whether spatial movement is a primary foc
geographical explanation about a spatially dynamic phenomenon (e.g., transportat
planning), or whether it is indirectly involved in an examination over time where I
and interactions among phenomena on the landscape are revealed, (e.g., analyzing
Changes), the principles are necessary components of geographical understanding.
Interpreted as a dynamic function of many geographical analyses, movement
rCCeiving increasing attention in the research community. In addition to the conti
examinations of social processes such as daily activity patterns and urban growth,
advancements in the understanding of the effects of human activities on the natur:
environment are now being explored with increasing urgency, especially in researcl
involving geographic information systems, visualization, and cartography. Improv
these computerized tools in their spatiotemporal capabilities have aided researcher:
end by facilitating greater understanding of complex dynamic processes at multiph

geographical scales. However, before comprehensive conceptualizations of moverr

 

6

associated with these processes can be made, the individual components of motion
be abstracted. Once the underlying attributes of geographic movement are more I
understood, then significant progress towards performing integrative and insightft
and visualizations of complex, spatially dynamic processes can be achieved.

The key to improved understanding of geographic movement is to transcend
examples and to establish the significance of some general principles and concepts
sense out of seemingly different facts (i.e., diverse geographic phenomena) (Hartsl.
1959). The acquisition of concepts and categorical structures is a fundamental pa.
learning process and leads to the development of “a new set of precepts” with Wh1(
underStand and communicate about the world (Patton 1997; Hardwick and Holt;
1996, 273). Classification, the first and most important step in the development i
process—based model, makes it possible to trace a certain order in the study of pher
and to uncover the concrete content of generic concepts of different scope (W illm
Gaile 1992). As an important and useful generalization step in conceptualizing a:
matter, geographers have long relied on categorization to place order on an often c
World. Classifying geographic movement allows the comprehension of the general
Principles underlying geographical patterns and processes and provides a foundatir
underStanding the complex interactions among diverse geographic phenomena.

Using the logical subdivision method of classification (classification occurs fr
general to the particulars) and adopting a morphometric analytical approach to ex
(inquiries dealing with the spatial structure and form of phenomena), this dissertai
Presents a classification scheme of geographic movement (Harvey 1969). Geograr
movement involves a change of spatial locations over time. As such, both space at
identified as the important inherent properties of any and all geographic movemer
arranged conceptually into a two—dimensional matrix. Spatially, geographic objec1
composed abstractly of points, lines, and areas, with areas sometimes extending in'

three—dimensional space containing volumes. These four spatial characteristics of

 

 

7

dynamic phenomena represent the first subdivision of the classification. The secoi
inherent property of geographic movement is time, which has associated fundamei
features such as duration and periodicity of phenomena. A movement is classified
occurring nonstop, occurring repeatedly in a regular fashion, or occurring infreque
no definitive temporal pattern. Based on these three temporal conditions identifie
continuous, cyclical, and intermittent, respectively, in conjunction with the four 51
features of movement represented by points, lines, areas, and volumes, geographic
movement is organized into twelve unique classes. Each class is described in terms
abstract spatial and temporal components and, with attention given to widely rang
of interpretation, geographic phenomena exemplifying each movement category at
discussed.

The premise of this classification is that any spatially dynamic geographic phi
can be conceptualized by its patterns and processes of movement through space an
and can be characterized by at least one of the twelve movement classes. The pres:
Classification serves as a useful and integrative framework for modeling, interpretin
analyzing, associating, and explaining dimensionally complex, if not distinct, patte
arrangements, and locales that are created on the landscape when human and phys
Phenomena interact amongst each other. By abstracting spatially dynamic geograt

Phenomena into their natural structures of motion, the inherent spatiotemporal

dimensionalities are discovered, thereby leading to a more complete interpretation
Cognition of the interplay of space and time not only with one another, but with o
fundamental principles important in the explanation of phenomena such as spatial
interaction. Understanding these complexities will facilitate a greater understandii
diverse and integrated dynamism of the geographical factors that create regional m

the earth’s surface.

 

Chapter 2
CLASSIFICATION IN SCIENCE AND GEOGRAPHY

“... we would never have learned anything if we had never
thought how objects resemble each other, and whether they
manifest the same properties. If every object in the world
were taken as distinct and unique, our perception of the
world would disintegrate into complete meaninglessness.
The purpose of classification is to give order to t e things
we experience. We classify things so that we may learn
more about them.”

(Berry 1972, 1)

As the simplest form of analysis, classification is the act of placing events, object
or observations into categories based upon their similarity with each other. Classificatl
are constructed as a means for better understanding the world in which we live.
Classification plays a necessary and crucial role in scientific explanation; the function c
positivist science is to develop generalizations and conceptual frameworks about realit
and the classification of phenomena is often the first step taken towards scientific
CXplanation in the diverse branches of science, including geography. Every classificatiOJ
Should have a useful purpose, the most important being that inductive generalizations ca
be. made about the objects under study. The classification method used, the number an
ICpresentative archetype of categories to be constructed, and the identification of possi
relationships among categories requires decisions that will reflect the analyst’s view of
World. As knowledge about the objects to be classified becomes more complete, the
Classification should be modified to reflect newly acquired information. The
determination of class structures is based on prior understanding and theoretical concet
held by the classifier. Although typology building should follow certain principles in
logic, the development of a classification scheme is largely subjective. A classification

may be deemed adequate if it serves its purpose well. The following sections will

8

 

 

9

examine each of these important features of classification in more detail beginning wi

what is classification and why is it an important part of explanation in science.

Classification Defined

Classification consists of the grouping of events, objects, or observations (hereaft
referred to collectively as objects) into a class or group because of certain attributes or
relationships they have in common. Classification is a logical operation in which a ge
concept is divided into classes based on the criterion of similarity of the objects enter
into a class and on their differences from objects entering into other classes. Berry (19
300) defines this principle of classification as “the similarity of members of the class
the differentiation of members from nonmembers.” Matching and categorizing objec
exhibiting similar characteristics is the process by which types or ‘typologies’ are
produced (Johnston 1970). As a “filter” for ordering data (Harvey 1969, 327),
classification can be built on various principles: morphologic, genetic, temporal, spa-
qualitative, etc. However, all must follow certain general and unalterable laws of log
(Armand 1965).

As a type of ‘cognitive description’ that asks “how may the phenomena being stu
be ordered and grouped?” (Harvey 1969, 82), classifications may range from simple
Primary observations through to sophisticated descriptive statements. At one end of t
Scale, the act of classification may involve the simple giving ofa name to a group of
Objects, whereby all objects are automatically classified into two classes, those posse:
the attributes connoted by the name and those not possessing them (e.g., houses and nor
houses). At the other end of the scale, the term classification may be used to describe
conscious act of devising a formal scheme for the grouping of a definite set of objects
an elaborate hierarchy of classes (e.g., land use classification) (Gilmour and Walters 1
Regardless of the level at which considered, classification is an important operation i1

empirical positivist science.

 

10

The Role of Classification in Scientific Explanation
Classification is the first step the scientist takes on his or her route to expl
(Harvey 1969; Abler, Adams, and Gould 1971). Classification is an organizat
process where classes are grouped according to constructs, which are ideas about
experiences that are products of, and answers to, our semantic questions (e.g., a
flower or plant) (Abler, Adams, and Gould 1971). As the foundation of positi
science, classification defines the concepts to be studied (Iohnston 1986). The fi
scientific concepts is to mark the categories which will tell us more about a sul
than any other categorical sets (Kaplan 1964). A concept is scientifically valid
identifies properties which are causes of many other properties. This process 01
formal conceptual structures in science is referred to as theory. Theories are stru
composed of laws and the rules by which those laws are put together. As Kaplar
53—54) explains, the formation of concepts and theories in science go hand in has
knowledge of a particular subject matter grows, our conception of that subject 1
changes; as our concepts become more fitting, we learn more and more... the bet
concepts, the better the theory we can formulate with them, and in turn, the betti

concepts available for the next, improved theory.”

Theory building is operational at every level of the process of scientific e)
The objective of science is to devise theories or laws that explain patterns and b
the natural and social worlds (Goodchild 1992). A theorist is someone who ob
Portion ofthe world and seeks to “find order in the booming, bustling confusior
realm of experience” (Dubin 1969, 5). The idea of order, and the tools utilize(
the sense of order, are in the mind of the theorist. Classifiers, as theorists, take
information generated by empiricists and chroniclers and order it within the p:
framework (Iohnston 1991). The classifier begins to ask questions about the con
which he or she has put in the foreground of attention and thus begins to sort pan

experience into classes of objects. In doing so, the significant criteria to be user

 

11

classifying objects are defined and the process may also involve arranging the cri
some order of importance. This implies some a priori notions about structures, Si
what properties or attributes are important for difierentiating objects in a given a
This notion that the proper concepts are needed to formulate a good theory, butt
good theory is needed to arrive at the proper concepts is known as the paradox of
conceptualization (Kaplan 1964). Hence, a strong link binds classification and th,
type of theory which will emerge is partly conditional upon the nature of the que
asked and on the nature of the explanatory framework chosen. Although these de
typically made subjectively on the basis of preconceived ideas (johnston 1986),
classifications are constructed based on the abstraction and generalization of the
around us.

This abstraction and generalization of the relationships among facts, which .
one to see patterns that are much more difficult to discover from the data [hCm81
perhaps the most important purpose of classification systems; that is to say it peg
inductive generalizations about the objectives we want to study and understand (A
Adams, and Gould 1971). Induction is the presentation of information about thi
a way that suggests explanation (Goodchild 1992). Generalization is the orderin
Observations and experiences and as Gilmour and Walters (1964, 3) claim, is the
method by which humans cope with the “otherwise chaotic multiplicity of indivi
Objects” in the world around them. Others have expressed similar views on the e
humans to deal with their surroundings: Chorley ( 1968, 42) refers to classificatic
accumulation of a backlog of information which is then dissected and categorize
convenient manner;” Harvey (1969, 348) secs classification as a means to an end:
grouping sense-perception data into classes or ‘sets’ we transform a mass of unwii
information so that it may be more easily comprehended and more easily manij
Abler, Adams, and Gould (1971, 150) call classification “a form of intellectual

shorthand” which makes the “transmission of knowledge easier.”

 

12

In essence, creating classifications is a way in which we summarize data and
into a form with which we can comprehend. Not only do typologies necessarily
propositions but, by isolating crucial distinctions between diverse phenomena, a g
typology will facilitate the creation of further propositions (Clark 1972). Grigg
470) uses regionalization to explain this point: “many generalizations in geograpl
from the comparison of two different regional systems; thus, for example, if in an
country an areal classification on the basis of soil type is compared with a classifi
according to land use, then a number of generalizations about the relationship betvs
two may be inferred.” If similar parts of the earth’s surface are grouped together,
statements can be made about the areal classifications which are applicable to all
parts of the classification. Classifications are not fruitful ends in themselves, but 1
enable the ordering of unorganized data or facts, thus producing valuable insights :
relationships between classified phenomena (Clark 1972; Abler, Adams, and G0t
1971)

Thus, classification occupies a key position in the development of every scier
scholarly discipline can function without a systematic approach to the objects und
(Armand 1965). Some scientists have argued that the state of classification in a s
a measure of its level of development (Abler, Adams, and Gould 1971). In the e
Stages of a discipline’s development, theoretical assumptions may be vague and il
defined. In the later stages, classification procedures tend to become part of exp,
design where measurement and classification of data may be derived directly fro:
(Harvey 1969). Significant classifications have dominated such disciplines as ch
geology, zoology, and entomology; the hierarchical system of biological taxonor
perhaps the best known scientific classification (Goodchild 1992). The most volu
literature devoted to classification has been by or for librarians. While such world
dealt with the organization of printed matter, it can be said that classifications in

librarianship represent extensions of knowledge as a whole (Gilmour and Walters

 

13
In geography, classifications are widely used in all branches of the discipline (Armand
1965). Virtually all phenomena geographers deal with have, at some time or other, been
classified such as land uses, towns, climates, soils, coast lines, rivers, economics, and above

all, regions (Harvey 1969).

Evaluating a Classification

Broadly speaking, the preference for one classification, as a theoretical model, over
another is a matter of consensus. What is meant by consensus is that a group of people
sharing an interest in some set of observations (i.e., ‘experts’) come to agree that one
theoretical model hest provides understanding or permits accurate predictions about the
observational set. For such agreement to exist there must be some common agreement on
the boundary containing the observations, that the framework of logic employed in
building the theory be shared by all experts, and that a “proof” of a theory is performed
through empirical and/or statistical tests (Dubin 1969, 13). Whenever people agree
among themselves that understanding has been more or less satisfactorily achieved or that
predictions have proved accurate within agreed—to units of error, then the theoretical
models will continue to be favored. Thus, the continuing viability of a theory rests on
human consensus (Dubin 1969).

The argument about the adequacy of a theoretical model is often an argument about
the logic employed in constructing it (Dubin 1969). A theoretical model starts with
things or variables (i.e., units) whose interactions constitute the subject matter of attention.
The model then specifies the manner in which these units interact with each other (i.e., laws
0f interaction). Since theoretical models are usually of limited portions of the world, the
limits (i.e., boundaries) must be set forth within which the theory is expected to hold.
Most theoretical models are presumed to represent a complex position of the real world,
part of whose complexity is revealed by the fact that there are various system states in each

of which the units interact differently with each other. Once these four basic features of a

 

 

 

14

theoretical model are set forth, the theorist is in a position to derive conclusions that
represent logical and true deductions about the model in operation, or the propositions
the model.

In classification, scientific laws are presented as functional laws, which are statem
of the interrelationships among two or more classes of phenomena. However, prior to ‘
creation of such statements, it is necessary to define the properties or memhership laws
allocating phenomena to their relevant classes. The determination of the relevant
properties is essentially a subjective issue, and no general rules can therefore be appliet
except to say that the most important properties should be given the greatest weight.
objective guides exist to which an investigator can refer for assistance in classifying.
Drawing up membership laws is not a straightforward task; much debate may take pla
(e.g., when does drizzle become rain or youth become adulthood?) Unless the indivit
form definite groups, there are likely to be minor alterations in the classification (Johr
1986). Errors in drawing up the membership laws can lead to failures in the search for
functional laws, because the classes are not internally homogeneous. The membership
must then be reconsidered in order to allow scientific progress.

Classifications are not absolute and as such, they must be modified as more
knowledge is gained about the objects under study. As theory develops, new relations]
Cmerge, making it possible to classify what previously seemed to be very disparate ev
and objects into some unified system (Harvey 1969). The belief, however, that therei
one unified system or true classification of a subject matter has had “unfortunate
Consequences” (Grigg 1965, 471). According to Gilmour and Walters (1964, 20), wh.
Clearly emerges from the literature on library classification is how naive the nineteentl'
Century materialists were. Interested in objectively determining facts as a basis for tl‘
Classification, these taxonomists accepted the idealistic view that a single, true, naturz
Classification “must exist and awaits discovery.” The authors suggest that this quest to

discover a unique and accurate classification is too optimistic and is a widely and SCI!

 

 

 

 

15

held “delusion” of many scientists, particularly biologists. Abler, Adams, and Goulc
(1971, 158) believe geographers are no exception: “in the development of geographic
thought, the whole idea of the natural region... has had a debilitating stranglehold on 1
geographers’ drinking.” Johnston (1986, 471) agrees that geographers, in their attemp
establish regions, often perpetuate the misconception that a single classification exists
nature: “even among those who recognize that regional systems are not a classification
entities that exist in nature there is still a tendency to forget that lines on a map are ra
real and that any given classification or regional system is but one way of looking at t
world.” Well-established regional systems tend to become, in the mind of some tear
the reality rather than simply a device for representing sections of reality and this,
Johnston (1986, 471) says is unfortunate: “if regions were real entities then there coulc
correct regional system.”

Modern logicians and classifiers would agree that there can be no natural
classification in the sense that there is one and only one classification which will serve
purposes (Grigg 1965). The idea of a single, true classification suiting a subject man
does not consider the subjective nature of classificatory procedures. Classifications a
constructed for a particular reason; there can be as many schemes as there are inquiries
abour a subject. To some, classification is a means of data exploration, either to
determine convenient ways of summarizing information, to find new and potentially
useful hypotheses, or to produce a universally true typology. To others classification
Provides a means of facilitating hypothesis—testing or model—fitting, while still other
Concerned with developing improved modes of production (Berry 1972). Harvey (15
326) suggests that since different classes are required for different purposes, “we possr
therefore, no means of assessing the adequacy or efficiency of a given classification
independently of the job it is designed to do.”

How well a classification fulfills a particular inquiry is largely dependent upon

purpose for which it was created. To ask whether one classification is “better” than an

 

 

 

16

without considering “better for what purpose?” is an unanswerable question; the util
given system of classification cannot be answered independently of its purpose (Gil
and Walters 1964, 3). No method has been devised by which it is possible to com
classifications and decide which is the best (Johnston 1968); instead, a classificatim
be judged as “good” if it has an aim beyond the classification alone (Abler, Adams
Gould 1971, 151). There can be no use in placing an object in a scheme unless somi
more than the fact of being in the classification is implied. Grigg (1965, 471-472
that for some classifiers such as geographers, classification becomes no more than a
of identification; once the object has been satisfactorily classed, it is forgotten: “m
regional systems seem to be devised as an end in themselves... the definition of a re
regions becomes important in itself rather than as a means to understanding the area
question.” A classification can thus only be understood not as an end in itself, but a

background of purposes for which it has been made (Gilmour and Walters 1964).

Classification Purposes

Classifications should be designed for a distinct purpose; they rarely serve twi
different purposes equally well (Grigg 1965). The many purposes of classification
however, be grouped into two fundamental types: natural and artificial. Gilmour a1
Walters (1964) have suggested that the most useful and informative way of distingu
bctween purposes for classifications and appreciating how classifications deal with
Objects around us is to use the terms ‘general—purpose for natural classifications and
‘Special—purpose’ for artificial classifications. Grigg (1965) agrees with the usage c
terms because of the confusion that has resulted in the interpretation of the distinctit
between natural and artificial classifications. Natural classifications assign objects
Classes whereby the variation among individuals within each class is minimized ant
Variation between each group and members of other groups is maximized (Abler, t

and Gould 1971). A classification may be considered natural if the attributes chos

 

17
the basis of classification are significantly related to the attributes conceptualized
elsewhere in our thinking (Kaplan 1964). The possibility of making a natural
classification of a set of objects depends on the existence of a powerful influence opera
on them which causes a number of attributes to “hang together” and be highly correlatet
their occurrence (Gilmour and Walters 1964, 4). If there is one influence more powerfi
than any Other, then there will be one classification that is more ‘natural’ than any other,
whereas if several influences are equally powerful, there will be several more or less
equally natural structures. Using an example of furniture classification, Gilmour and
Walters (1964, 4) demonstrate this point: “the influence of intended use is paramount, .
a classification into chairs (sitting on), tables (for placing things on), etc., is a more nat
one than any other (e.g., date or place of manufacturer)”

A natural classification may be designed to serve many purposes, but it is unlike];
serve all purposes with any but a low level of efficiency (Harvey 1969). As a rule, gene
purpose classifications are developed to give names to things as members of groups ant
transmit information, but they may not offer any explanations about the objects classif
(Grigg 1965). Most modern views tend to regard general purpose classifications as
Performing the function of data—storage systems, thereby providing a “comprehensive
nomenclature” for working on empirical problems (Harvey 1969, 331).

Artificial or special purpose classifications serve as the first formal step in scieno
(Abler, Adams, and Gould 1971) and may be developed to test or deal with specific
hypotheses or problems. The problems themselves define the criteria and method to I:
Used in such cases (Harvey 1969). Artificial classifications are developed for a special
PUIpose and we cannot do more with it than was first intended (Kaplan 1964).

The thematic disciplines, such as economics and psychology in the social sciences
Physics and chemistry in the natural sciences, use a general focus upon comparatively
narrow classes of phenomena. In contrast, the integrative disciplines such as history ant

geology generally address broad classes of phenomena, including those that fall under

 

 

 

 

 

18

specialized focus of the thematic disciplines (Casetti 1993). However, some discip
such as geography, have moved to bridge the gap between the general and the specific
methodologies and conceptual frameworks pertaining to scientific explanation. In

particular, classifications of regional systems by geographers have been both general

specific; types of places, regardless of proximity to one another, are classified accor
to a certain set of attributes, such as climate, language, cultural heritage, human use 0
earth, etc. Specific regional systems differ from general regional systems in that all
parts of a homogeneous region must be spatially contiguous; regions are defined not
by combinations of intrinsic attributes, but by location as well (Abler, Adams, and

1971).

Regardless of whether a classification is considered general, specific, or perhap
culmination of the two, it is the purpose itself that drives the classification process.
primary goal of classification must be thought out explicitly before attempting its
construction (Grigg 1965; Rees 1972). Once a purpose for a classification has been
defined, the classifier, on the basis of preconceived ideas, must then make subjective
decisions concerning the choice of classification technique and the definition of a grc
Particular care must be exercised in the choice of relevant properties on which to cla

before proceeding with analyses.

Procedures for Classifying

Classification can be reached by two distinct approaches, logical suhdivision and
flgglomeration (Grigg 1965). Logical subdivision divides a population, using a series
Steps, into groups according to stated criteria, often the presence or absence of two 0
attributes. Agglomeration takes a number of individuals and, by one of a large num
techniques based on discriminatory analysis, assembles them into classes according 1
some grouping or clustering procedure. Grouping similar observations is simply the

side of the coin from subdividing a population; only the perspective at the starting 1:

 

19
different (Abler, Adams, and Gould 1971). With the agglomerative approach to
classification, individuals are grouped together on the basis of some observable and
measurable properties. No assumptions are made, or need to be made, about the cause of
the differences and similarities between the individuals grouped in the same class. But in
subdivision the genus is divided on some principle and thus, a logical subdivision
presupposes that there is some understanding of the system being constructed (Grigg
1965). Using this distinction, Harvey (1969, 334) suggests that given a purpose, and given
adequate information regarding the criteria to be used in classifying, objects can be
assigned using two fundamentally different kinds of classification procedures:
“classification from above” which refers to logical subdivision and is considered to be
deductive classification, and “classification from below,” which refers to agglomeration or

inductive classification.

Logical Subdivision

Sometimes referred to as as the ‘definitional’ method of classification, logical
subdivision involves a conscious laying down of certain attributes that an object must
possess in order to belong to a schema and is a good method to use when avoiding
misunderstandings is important (Gilmour and Walters 1964). The divisive approach to
Classification treats the whole population as a single type and then divides it into
SUbdivisions on the basis of some principle (Johnston 1970). The initial class to be
divided is called the genus, which is divided into its constituent species, and each species
can be further subdivided on the same principle, sometimes referred to as the fitndamentis
divisionis. For example, the universe of land parcels can be subdivided based on the
Principle 0f its land use (Figure 2.1, a). The first order of subdivision yields six major
categories of land use: building lots, farm land, forest, public utilities, recreation, and
was“? (Figure 2.1, h). Each of these major land use categories can be further subdivided

. . ' . ' and
leldlng second order subdivisions such as farm land into pasture, cultivated land,

 

 

 

 

20
a) lst order, h) 2nd order, c) 3rd order, d) 4th order,
Building lots (B)
Farm land (F) pasture (P)
land parcels Forest (FR) cultivated (c)
Public Utilities (U) orchard (0) bush fruit (b)
Recreation (R) —E tree fruit (r)
Waste (W) hops (h)
Fp F Fp F
F c c
—j Kl

 

 

 

 

 

F9 F0 FR Fo, FR
Foh
FR 8 FR B B

FR
[T chu w ”(all IT

Adaptedfi‘om Abler; Adams, and GouM (1971, 155), Figure 6—8.

 

 

 

 

 

 

FU

 

 

 

 

Figure 2.1 An Example of the Logical Subdivision Method of Classification.

orchards (Figure 2.1, c). Continuing with the hierarchical subdivision, third order land use
classes of orchard farmland yield bush fruit, tree fruit, or hops (Figure 2.1, d). As this
land use example demonstrates, the logical subdivision of a classification should proceed
at every stage, and as far as possible throughout the division, upon one principle (Grigg
1965). The first and most important rule in logical subdivision is that a division must be
exhaustive; the purpose of logical subdivision is to distinguish all the species within a
genus.

Techniques used in logical subdivision include the creation of Venn diagrams and
stepwise disaggregation into a hierarchy of classes (Abler, Adams, and Gould 1971), the
latter based on the notion that different criteria may be adopted at different levels of the
Process (Rees 1972). One particular form of hierarchical subdivision is dichotomous
division, the purpose of which is to isolate a single species by dividing upon the principles
that a class must either possess a given characteristics or not possess it. For example, all
the members of a university can be divided into teachers and non—teachers. Teachers can be

subdivided into professors and non—professors, professors can be classified further as

tenured or non—tenured professors. Non—tenured professors are identified as either assistant

 

 

 

 

 

 

21

or non—assistant, non—assistant professors are categorized as lecturers or non—lecturers, and so
forth (Grigg 1965).

With logical subdivision, the division of the universal set takes place in a series of
steps, and at cach step one property or set of properties is used to differentiate between
classes. The kind of classification which emerges, however, is very much affected by the
criterion selected at each step and the order in which the criteria are employed. In
developing classifications by this route, criteria need to be placed in order of significance.
This assumes that much is known about the phenomena being classified; the classifier uses
an adequate theory about structures to deductively identify classes. Logical division is
best suited as a method for classifying when the classification is viewed as the culmination

of scientific theory.

Agglomeration
Agglomeration, the most popular approach to classification by far among

geographers, begins by treating each object as a separate unit and proceeds to group

similar units into types (Johnston 1970). Combining individual observations on the basis
of observable and/or measurable similarities forms a smaller number of clusters (Grigg
1965). Census geography is a good example of how detail is lost in order to gain ease in
interpretation each time observations are grouped into classes (Abler, Adams, and Gould
1971). Census geography is a collective term referring to the geographic entities used by
the Census Bureau in its decennial collection and tabulation operations. Census Statistical
Area Committees (CSACs), in cooperation with the Census Bureau, identify such
Statistical entities. The grouping is designed to create relatively homogeneous units with
respect to population characteristics, economic status, and living conditions at the time the
CSAC established them (US Census Bureau 1997). Households can be described
according to socioeconomic attributes like income, educational level, and age of housing

unit. Blocks, the smallest geographic entity for which the Census Bureau reports decennial

 

 

22

census information, are formed by grouping homogeneous, spatially—contiguous households

(Figure 2.2, a, [7). Each block is assigned a unique three digit number with similarly

 

 

 

a) lst order, h) 2nd order, c) 3rd order, d) 4th order,
grou ing grou ing grouping
into blocks into lockgroups into census tracts
BC 1
BC 3 Tract
133.5
BC 2

 

 

 

 

 

 

 

 

Figure 2.2 An Example of the Agglomerative Method of Classification.

homogeneous adjacent blocks given the same first digit in its census reference number.
From these, combinations of similar blocks called blockgroups are created (Figure 2.2, c),
from which census tracts (or block numbering areas) are formed (Figure 2.2, d). The
grouping procedure for each of these levels uses the same principles of spatial contiguity
and socioeconomic homogeneity, with each higher order grouping employing more

broadly defined criteria.

The agglomerative method of classification by which individuals are compared may
use dichotomous attributes such as male versus female; alternatively, observations may be
measured in terms of gradient variables such as income or height. In using continuous
variables, the limits of each class must be defined before grouping of observations can
begin (Abler, Adams, and Gould 1971). Multivariate measures utilizing continuous data
may also be employed in qualitative approaches for comparing individuals, which may
also use either the presence or absence of certain attributes of phenomena (Johnston 1968).
In situations where a good deal of uncertainty presides, groupings will likely yield much

more realistic classifications. Hence, grouping is often regarded as an inductive procedure

by which the phenomena being examined are searched for regularities and for significant

 

 

 

 

23

interrelationships. As such, grouping is appropriate to use in situations in which the
significant properties are not known.

A type of agglomerative classification uses the typological method, which Gilmour
and Walters (1964, 7) claim is a satisfactory type of method for the “general run of
practical human purposes.” This type of agglomerative approach to classification is
exemplified in the semantic development of the words used in every day language. With
the typological method of classification, a name is given to objects which resemble each
other in certain rather ill—defined ways, thus forming a generalized ‘type picture.’ As each
new object is considered, a decision is made as to whether or not the object resembles this
type-picture sufficiently to be given the same name. No one or more attributes are
necessarily possessed by all the objects in a particular class, but rather these objects show a
‘family resemblance’ to an imaginary ‘type—representative’ of the class. The authors
contend that the typological method of agglomeration has developed and persisted as a
satisfactory method in classification.

What distinguishes agglomeration from logical subdivision, from a philosophical
point of view, lies in the specification of the universal set, or the boundaries imposed on
the subject matter of attention. With logical subdivision, the universal set is specified by
definition and many classes may not have members (Harvey 1969). In grouping
procedures, boundaries are determined out of necessity, whereby all classes will have
members and any general conclusions made from grouping must proceed by induction.
Neither method is better for all purposes; each is adopted for a particular kind of purpose
(Gilmour and Walters 1964). The choice of classification method depends on the purpose
for which the classification is being made; not everyone may agree with the appropriate
method. Harvey (1969) regards the divisive approach to classification as the only
effective and logical way of classifying phenomena. Much of the work done in the

nineteenth and early twentieth century resorted to logical subdivision. Abler, Adams, and

Gould (1971) contend that grouping of like individuals is operationally far superior than

 

 

 

 

24

logically subdividing a population because it is faster, easier, and all classes defined will
have at least one member. Grigg (1965) argues that in practical classifications, the
scientist will use both procedures to establish his or her system. Regardless of the
approach taken, each classification system represents an attempt at placing order on a

complex and dynamic world and is a necessary component of positivist science.

The Fundamentals of Classifying

Logical procedures should be observed when classifying objects into groups. Such
fundamentals of formal logic are designed to prevent internal inconsistencies arising
within an organizational scheme (Grigg 1965). In classification, phenomena are grouped
by reference to their attributes, i.e., on the basis of properties they have in common (Harvey
1969). The objects which are to be classed are called individuals and the total number of
individuals considered in any given classification system is the universe or population
(Grigg 1965). Once the purpose of the classification has been thought out explicitly (the
most important fundamental of any classification system), the first stage of classification
involves the definition of the individual or the objects to be classed and is referred to in
classification theory as the Operational Yaxonomic Unit (OTU) (Rees 1972; Abler, Adams,
and Gould 1971). At the same time, the universe or population of individuals to be
classified must also be delimited.

Several important fundamentals are involved in the second stage of classification
during which objects are grouped together into classes on the basis of a property they have
in common. One such logical principle surrounds the selection of the diflferentiating
characteristic, a property which is possessed in some degree by all the individuals. The
differentiating characteristic, or principle of division, must be a property of all the
objects grouped and must be important for the purpose of the classification (Grigg 1965).

The selection of the differentiating characteristic or criteria is usually based on theory or

prior understanding (Rees 1972). If the differentiating characteristic is chosen carefully,

 

25
then other properties of the individuals will change as the differentiating characteristic
changes. Such properties are called “accessory characteristics” and when two properties
change in such a manner they are often said to display covariance (Grigg 1965, 466). Rees
(1972) identifies three types of differentiating characteristics that may be used:
structural properties which are those involving only the OTU itself, relational properties
where at least one other object other than the OTU in question is involved, and dynamic
properties, or those involving properties of the first or second type at two or more time
periods.

Classifications based on one or just a few differentiating characteristics are called
monothetic classifications in which all the items allocated to one class must share the trait
under consideration. However, the selection of one property and the formation of only one
category of classes may not give sufficient insight into the things being classified.
Classifications based on several characteristics are called polythetic and membership in a
class does not depend on just one property or characteristic shared by all members but
rather, a class or category is established in terms of a set of attributes. The classes of the
first category may then be grouped on the basis of a second differentiating characteristic
into a second category. The classes of the first category are included within the subclasses
of the second category’s classes. This procedure can be carried further until all the classes
are included within one ‘superclass’ or class of all classes (e.g., the United Kingdom, as the
superclass, can be broken down into nations, which then can be broken down into counties,
and further into parishes) (Grigg 1965). Such hierarchical classifications are often called
taxonomies, whereby the sets of classes or categories are referred to as ‘classes of the first
order,’ ‘classes of the second order,’ and so forth (Grigg 1965). Perhaps the most renowned
taxonomies are those pertaining to the biological sciences, such as the Linnean
classification system of plant communities into hierarchical orders of families, genera,

species, etc. (Gilmour and Walters 1964).

There is no limit to the number of dimensions that can be incorporated into a

 

 

26

classification scheme; two, three, or more variables can be used to define a categorization
scheme of two, three, or more dimensions (Abler, Adams, and Gould 1971). The classes
that are crcated will vary greatly in the number of attributes which members possess in
common (Gilmour and Walters 1964). With polythetic classifications, some of the
members of the class may share most of the appropriate attributes for a class, but not all of
them (Abler, Adams, and Gould 1971).

Establishing classifications based on many attributes or characteristics is, as Abler,
Adams, and Gould (1971, 158) suggest, “a complicated business.” The use of more than
one differentiating characteristic or principle of division may produce a hierarchy of
classes and the tabulation and processing of the many terms associated with the many
attributes may be difficult for the observer. Therefore, traditionally each science’s
classifiers have favored one aspect over another in their classification efforts. Properties
which are used to divide or classify in the higher categories must be more important for
the purposes of the division than those used in the lower categories (Grigg 1965). A
classification must not skip logical levels; any violation of this rule makes it difficult for
the reader to orient him or herself in the classification as a whole and in the sequence of
subordination of the classification’s objects (Armand 1965). A special form of
classification uses an intersecting lattice device in which the generic concept is broken
down first by one, then by another criterion. The classes are arranged in the form of a
table; the classification according to one criterion is listed along the abscissa axis and
according to the other criterion along the ordinate axis. Classifications using intersecting
lattices may break down one or both main classes further on the basis of additional
criteria, thus yielding third— and fourth-order classes (Armand 1965).

Besides the determination of the differentiating characteristic, some important rules
concerning the second stage of classification should be upheld. First, a classification must
be exhaustive; every individual must fit into one of the classes (Grigg 1965; Abler,

Adams, and Gould 1971). Armand (I965, 22) uses the following example to support this

 

 

27
principle that the sum of classes must be equal to the scope of the classified generic
concept: “a classification of seas into marginal, inland, and inter—island does not go
beyond the scope of the concept ‘sea’ and, at the same time, includes all possible classes of
seas.” The second rule of classification states that the classes must be designed so that an
individual cannot fall into more than one class; i.e., the classes must be ‘mutually
exclusive’ (Abler, Adams, and Gould 1971). To the questions concerning how many
classes should be specified and what should the class limits in each class be depends on the
aim of the classification. Where properties used as differentiating characteristics can be
measured, the problems of deciding class limits are minimized and classes are necessarily
mutually exclusive (Grigg 1965). Quantitative techniques can be used to devise class
breakdowns by referring to the measurements made on an attribute. In numerical
taxonomy, classification proceeds on the basis of interval measurements of similarity
among objects (Abler, Adams, and Gould 1971). Numerical taxonomic structures are
used to reduce classification to an objective process by minimizing Within class variation
(Goodchild 1992). Many quantitative approaches to identifying classes are possible,
including a generalized distance statistic, principle components and factor analysis,
grouping procedures, and discriminate analysis (Harvey 1969).

In classifications of phenomena where differentiating characteristics are qualitatively
assessed, it is impossible to base a classification on measurable properties simply because
the data do nor exist. As such, the possibility of overlap in classes is considerable and is
difficult to avoid (Grigg 1965). Abler, Adams, and Gould (1971, 157) conclude on the
definition of classes with the following: “we classify to clarify... each time we group
observations into classes we lose detail in order to gain ease in interpretation... if by trial
and error we are able to produce classifications of phenomena which suggest or expose
important relationships, then our choices are good choices... our classifications have served
their purpose.” Nedobity (1986, 349) refers to this principle as “generality in specificity”

and defines it as “the most specific event can serve as a general sample of a class of events.”

 

 

28
Classification as concept formation is the distinction between categories and individuals
as well as between classes and their members.

Making such distinctions is the crux of any classification system. Various integrating
concepts such as singularity, uniqueness, and similarity of objects must be accounted for if
a classification system is to be logical. Establishing generalizations assumes that there are
classes of phenomena about which generalizations can be made. Generalizations would not
be possible if all things were considered singular since there would be no two examples of
any concept. The concept of singularity denies the possibility of a recurrence. Because
laws are general, the positivist conception of science does not accept the empirical
possibility of singular events (Johnston 1986). However, it can be argued that logically,
everything is unique (Lewis 1965). The concept of uniqueness implies a single empirical
example but not a single causal mechanism producing that example. For instance, a
particular set of interactions where two or more laws are operating together may occur
only once, thereby producing a unique outcome which is not replicated. But that same set
of interactions could recur, resulting in the same outcome based on general laws and
producing a non—singular event (Johnston 1986).

The basic task in scientific explanation is to search for similarities, especially
functional similarities, between unique things (Lewis 1965). General statements can then
be made from these similarities and expressed as formal laws. Such laws do not imply
that one event is always followed by another, but rather that a functional relationship exists
between given determinants (Lewis 1965). Determining whether a phenomenon is unique
or general depends on the point of view or on the inherent property of the phenomenon
itself. According to Grigg (1965), nineteenth century classifiers were primarily concerned
with properties which were inherent in the objects classified. This widely used form of
classification involves grouping objects into classes on the hasis ofsimilarity between
objects and is called ‘association by similarity.’ The nature of similarity is fundamental

in classification problems (Abler, Adams, and Gould 1971). In all branches of science the

 

29
ability to establish generic concepts and scientific laws is dependent on, among other
things, the number of identical or essentially similar cases available for examination and
classification (Hartshorne 1959). In studying complex interactions, a small number of
essentially similar specimens may be found. Scientists often attempt to overcome this
difficulty by recognizing categories or types within which differences appear less marked
than the similarities.

In the classification of phenomena where a host of independent or semi—independent
elements is typically involved, specimens that are similar in all essential elements do not
exist. Rather they are similar only in the particular categories chosen and may differ
notably in other respects. Hartshorne (I959, 151) notes that a difficult dilemma arises
from this issue: “in order to study a significant number of [objects] as similar, we must
define the category so broadly as to include individual variations sufficiently great to
upset the validity of generalizations based on the assumptions of identical character; if the
types are defined sufficiently closely to avoid this danger, we may have but one specimen
of each type.” Conversely, grouping objects into classes on the basis of a relationship
between connected and different objects, called ‘associarion by contiguity,’ is what
modern biologists have used to classify plants and animals. Whatever the method chosen
to measure the similarity between individuals, the aim of any classification is to produce
groups of individuals in which within group distance or variance is minimized and
between group variance is maximized. Groups are commonly understood as
“constellations” of similar objects so that the important concept in their definition is
proximity (Johnston 1968, 578). Thus, groups should be homogeneous units and different
grouping processes should produce similar results for the same similarity matrix.

In studies of regions, geographers avoid too much uniqueness and too much generality
by assembling information into manageable spatial aggregates called regional systems,
which are areal classifications lying somewhere between maximum uniqueness and

maximum generality. Searching for similarities between unique things is something that

 

 

30

geographers have done for quite some time, the importance and implications of which have
been accentuated by the attempts to formalize these relationships mathematically (Lewis
1965). In grouping land parcels, farms, census tracts, neighborhoods, or large areas,
individual observations must be combined on the basis of similarities to form a smaller
number of clusters. Each small place on the earth is unique, yet it has combinations of
attributes that resemble combinations at other places: “although every place is unique, we
can know about places only according to their resemblance or lack of resemblance to other
areas” (Abler, Adams, and Gould 1971, 183). Even in the study of particular categories of
phenomena such as mountains, geographers have been concerned not merely to study
different types but to present full description and maximum interpretation of individual
cases (Hartshorne 1959). This concern for individual cases, particularly in regionalization,
is geography’s own distinctive contribution to science. Although simple in concept,
geographers from Hartshorne to I-Iart have celebrated the power of regional classification
to yield insights into complex geographic patterns and processes (Goodchild 1992).
Classification, as the first step in scientific explanation and as an important part of
the generalization process, helps us to come to terms with the issues related to a subject
matter under study. In our attempts to rationalize why an event is taking place or what the
implications resulting from a particular phenomenon are, we must first arrange each
observation in context with other events or objects with which it interacrs. This activity
involves the comparison of unique phenomena and the identification of the fundamental
principles and characteristics inherent with the objects. In other words, we organize the
phenomena we are trying to understand better in terms of the things with which we are
already familiar. Building conceptual frameworks not only allows the identification of
characteristics intrinsic to individual phenomena and of similarities and differences

among diverse objects, but may facilitate greater understanding of the relationships

among diverse phenomena and thus, of their interactions across space and through time.

 

Chapter 3
GEOGRAPHIC MOVEMENT

Geography is a descriptive science that deals with the surface of the earth, including
the climates, plants, animals, natural resources, inhabitants, and industries (Guralnik 1984).
All of these events, experiences, processes, and occurrences that can be scientifically
described or appraised are defined collectively for the purposes of this dissertation as
geographic phenomena. Change, or the act of substitution, alteration, or variation (Guralnik
1984), involves dynamic aspects (Allen 1984). Change to an entity causes it to be
transformed and can involve one or more attributes (e.g., shape or location) (Guralnik
1984). Langran (1989a, 218) provides an example of how some attributes never change,
some attributes change over time, and yet others are measured in the time domain: “The
attribute ‘address’ may be constant to a house but time—varying to a person.” Change
representing variation of some attribute of a place over time (MacEachren et al. 1992)
refers to ‘aspatial data’ (Langran 1992). If the attribute or value changes continuously over
time, it is said to be ‘fluent’ (McDermott 1982). However, when location ofa
phenomenon changes, it is said to be spatial/y dynamic.

Spatially dynamic phenomena are phenomena that move. All geographic phenomena
can be referred to by means of a specific address in three—dimensional earth space (Lowe
and Moryadas 1975). When the characteristics associated with a geographic phenomenon
change location, geographic movement has occurred. Most geographic explanation must
contend with the dynamic nature of earth surface phenomena. Physical processes are
expressions of movement, such as glaciation and change of seasons (Boehm and Petersen
1994). With human geographic phenomena, movement is purposeful and is often
explained in terms of spatial interaction (Lowe and Moryadas 1975). The movement of

31

32

people, goods, information, and ideas is typically discussed in such terms as migration,
trade and transportation, communication, and diffusion, respectively (Hardwick and
Holtgrieve 1996). Regions and places are connected by movement or human interaction:
“People everywhere interact. They travel from place to place, they communicate, and they
depend on others in distant places for products, ideas, and information” (Hardwick and
Holtgrieve 1996, 273). Whether a geographic movement involves people, the physical
earth, or a combination of the two, it is composed of change in both the temporal and
spatial dimensions. The following sections will discuss the inherent properties associated
with spatially dynamic phenomena and will look at how, as a concept, movement has been

explained within the geographic discipline.

Properties of Geographic Movement

Movement takes time and occurs in space (Tuan 1978); it is predicated by change
(Prince 1978). Both human and physical systems change over time and whereas
geographers may begin with static descriptions, their explanations inevitably incorporate
dynamic phenomena. These dynamic geographic phenomena are characterized by changes
in the distributions of its elements and connections (Graf and Gober 1992). The inherent
structure of all geographic observations, especially movement, are multidimensional and
include theme, location, and time (Sinton 1978). Peuquet (1994) describes these three
basic components in her spatiotemporal framework as the what, where, and when of stored
information. Space and time dimensions are the essential framework for any geographical
topology of movement, particularly topologies of human mobility (Gould and Prothero
1973). The importance or significance each of these attributes has in the understanding and
comprehension of the movement will vary with each geographic phenomenon, as well as

with scale.

 

33

Space

Space is the basic organizing concept of the geographer (Blaut 1972); the spatial
point of view is the common bond that cements the geographic discipline (Meentemeyer
1989). Space is many—faceted, ranging from the space of direct experience (relative
space), to the space of abstract geometric and mathematical thought (absolute space)
(Parkes and Thrift 1980). The notion of absolute space arose from the attempts of Greek
scholars to reduce everything to permanent bodies drifting and colliding in empty space
(Blaut 1972). In absolute terms, space is the concomitant of atoms, corpuscles, and
absolute things which enables them to be different and to become different (Lucas 1973).
Absolute space is a distinct, eminently real, and empirical entity in and of itself; space is
purely physical and not abstract or mental. Based on a defined grid system, absolute
space is seen as being composed of points, and a time composed of instants, which has an
existence independent of the bodies and events that occupy them (Blaut 1972). The
location of elements within the grid of the region under scrutiny is that which is
considered critical. This is the point of view typically held in conventional cartography,
remote sensing, and the mapping sciences. Viewed as a “container” for elements of the
earth’s surface, absolute space is defined by Euclidian measurements of length, depth, and
thickness (Meentemeyer 1989, 164). Euclidian geometry states that a location, or point in
space, can be specified by three numbers, or coordinates: latitude, longitude, and altitude
(Hawking 1988). These three spatial measurements describe dimensions, which can be
imagined as direcrions at right angles to one another. According to Gleick (1987, 97), the
process of abstraction that allowed Euclid to conceive of one— or two—dimensional objects
“spills over easily into our use of everyday objects.” For example, a road map, for all
practical purposes is two—dimensional. It uses two dimensions to carry information of a
precisely two—dimensional kind. In reality, road maps are as three—dimensional as

everything else, but their thickness is so slight and so irrelevant to their purpose that it can

be forgotten.

 

34

‘Spatial data’ are defined as data where each element, or discrete entity, has
associated with it a location or series of locations in two—, three-, or N—dimensional space.
These locations can be expressed implicitly or explicitly using any system of coordinates.
For geographers, as well as professionals in many other disciplines, these locations
correspond to points, lines, areas, or a combination of these on the earth’s surface (Peuquet
1977). Distance (or length) is the fundamental geographic dimension and spatial objects
may be classified according to the number of length dimensions they possess (Martin
1991): zero dimensions (point), one dimension (line), two dimensions (area or plane),
three dimensions (surface or space) (Peuquet 1984; Gleick 1987). These four categories or
types of spatial data portray the spatial locations and configurations of individual entities
(Peuquet 1984). Point data refer to a single data element that is associated with a single
location, such as the locations of cities of the United States. The location of line data is
described by a string of spatial coordinates which can represent either unconnected
isolated lines (e.g., fault lines), tree structures (e.g., river systems), or network structures
(e.g., road systems). Polygon data are associated with areas over a defined space and the
location of a data element is represented by a closed string of spatial coordinates.
Peuquet (1977) discusses three types of polygon data: ‘isolated polygons,’ where the
boundary of each polygon is not shared in any part by any other polygon (e.g., standard
metropolitan statistical areas), ‘adjacent polygons,’ where each polygon boundary is shared
with at least one other polygon (e.g., state boundaries), and ‘nested polygons,’ where one or
more polygons lie within another polygon (e.g., contour lines on a topographic map). A
fourth category of spatial data is some mixture of the above types, such as line structures
mixed with polygon structures (e.g., a state bounded by a river which is both a boundary
between adjacent polygons as well as part of the tree structure of a river network).

Spatial entities have individual, unique definitions which reflect the locations of
entities in multidimensional space. These definitions are commonly very complex given

the tendency of natural phenomena to occur in irregular, complex patterns. Dynamic

 

 

35

geographic phenomena may be modeled two—dimensionally, as with the surface of the
earth as a plane, or three—dimensionally to describe subsurface or atmospheric phenomena.
A fourth dimension could be added for time series data (Peuquet 1984). Physical
geographic processes such as tornadoes and hurricanes are good examples of how the
movement of an activity may be spatially defined in absolute terms; the succession of
spatial coordinates, indicating location and direction, are measured and aid in tracking an
entity’s movement through space. Viewed in the Euclidian sense, defining the spatial
extent of an object’s movement can be highly precise and mathematical, especially with
the recent emergence of global positioning systems.

Opposing the absolutist doctrine, the relative view of space rejects the idea that space
is an entity by itself. Rather, space is viewed as relative or as merely a relation between
events or the aspect of events. Concern lies with the conceptual role of space and not with
its metrical properties (Lucas 1973). Older than the belief of absolute space, the relativist
doctrine of space derives from more pragmatic concepts of measurement. Space is
rationalized as bounded to time and process (Blaut 1972) and is seen as a necessary
concomitant of other fundamental features such as substance, time, change, and motion, as
well as a concomitant to consciousness (Lucas 1973). Like time, relative space is
perceived through ordinary visual or tactile experience: “if we have any concept of time,
then we must have some concept of space” (Lucas 1973, 99). Relative space is inseparable
from time, the two forming what is called the “space—time manifold,” or simply process
(Blaut 1972, 43): “pure space is relegated to pure mathematics, and every empirical
concept of space must be reducible by a chain of definitions to a concept of process.” No
phenomenon in the physical world is purely spatial or temporal; rather everything is
process. The temporal dimension may be neglected sometimes, but it is always implied.

Geography is a science of certain types of “space—time integrations” (Blaut 1972, 48).

In a geographical context, relative space is not defined by Euclidian terms, but rather by

the spatial elements and processes under consideration (e.g., migration and community

 

36

patterns, watersheds, and diffusion of ideas) (Meentemeyer 1989). For example, two
areas separated by a barrier may be close in absolute space and very distant in relative
space when time, rates, and interactions are considered. Thus, a functional (spatial)
process region may be difficult to map in terms of absolute space (Meentemeyer 1989).
The spatial extent in which a phenomenon is occurring is not necessarily characterized by
its Euclidian distance, but may be measured more on the internal strucmres operating
within that space. For example, in daily trip pattern studies, the ‘activity space,’ or the
area which contains the colleCtion or majority of desrinations of a particular individual
(Iakle, Brunn, and Roseman 1976), is measurable only to the extent of an individual’s
perception of space. It is a subspace within the mental map, and frequently tends to be
discontinuous (Lowe and Moryadas 1975). In this way, space is measured or defined
relatively, where one object’s spatial location is defined by that of another. Measurements
of space become synonymous with terms such as “near” and “far,” or “left” and “right”
(Peuquet 1984, 74).

Some views of space incorporate both absolute and relative notions of space.
AriStOtle theorized space as metaphysical reality, a fission of empirical perception and an
a priori form of understanding. Kant viewed all processes not as psychological, but as
transcendental where spatial extension is intuitively associated with temporality (Blaut
1972). However, the ‘relativiStic revolution’ had been set in motion long before Einstein’s
special theory of relativity, which refuted Newton’s absolute spatial reference frame.
Scientific verification proved that there is no absolute frame of reference with which one
may orient oneself in relation to Other reference frames (VI/eissert 1991). By the beginning
of the present century, absolute space had been discarded throughout most of science and
philosophy (Blaur 1972). Most modern work in geography involves a relative view of
space (Meentemeyer 1989).

Measuring the space a phenomenon travels through as it moves in time is an important

part in understanding the dynamics of the aetivity. Space can be measured in several ways,

 

37

such as metrically with rulers, temporally with clocks (i.e., ‘drive time’), or by adapting
an arbitrary system of assigning some distance such as degrees of longitude (Lucas 1973).
Perhaps an even more basic knowledge about a movement comes from undersranding the
spatial form or pattern that the phenomenon takes as it evolves and changes with time.
Structure and process are not two different things (Blaut 1972); rather “spatial
arrangements are reciprocally tied to movement processes” (Abler, Adams, and Gould
1971, 237). According to Meentemeyer (1989), it is from spatial form that most
processes are discovered. Spatial forms are represented by arrangements of points, lines,
or areas (Graf and Gober 1992). As Haggett (1967) suggests, the mathematical approach
to geography allows the conversion of different geographical phenomena into the same
class (i.e., streams and highways are treated as line sets), as well as the transformation of
geographical phenomena from one class to another (i.e., line sets transformed to density
surfaces). Surfaces, like point sets and line sets, form a distinct family of geometrical
forms. As we have seen, space can be described in many different ways, but geometrical
features are those general features of space that are the same however space is described
(Lucas 1973). For example, in defining the spatial extent of an individual parcel of land,
the space described in relative terms might include landmarks and relevant locations
while the description of the area in absolute terms would probably involve real spatial
coordinates and measured planar distances. However, both methods of defining the space
yield a description of the physical area as a polygon, comprised ofa set of points and
lines on a plane.

Interpreted in this fashion, different kinds of spatial phenomena can be seen as
basically identical in their spatial structure. Adopting this notion, Abler, Adams, and
Gould (1971, 238) devised a classification of movements based upon the “movement
geometry” or spatial configuration of the sources and destinations of the phenomena.

Ignoring whether or not the movements refer to the human world, the physical world, or a

combination of the two, their movement classification displays the basic spatial

 

38

dimension of moves in sharp relief to one one another where different kinds of moves are
seen as basically identical in their spatial structure (e.g., the discharging of people at a
subway station is the same kind of line—to—point delivery pattern as water pipes and
drinking fountains around the floor in a building). The framework provides a useful

spatial or geographic viewpoint with which to describe and understand how things move in
earth space. However, to fiilly comprehend the dynamics of a phenomenon, the influence of
time must be accounted for as well. Space and time are “universally and inseparably wed
to one another” (Pred 1977, 218), and unless they are taken together, “the geographer’s

world will retain an air of unreality” (Tuan 1978, 16).

Time

Like space, time is a prerequisite of geographic change (Lucas 1973). The idea of
change through time is believed to have been introduced into geography during the
nineteenth century largely through the impact of Darwinism (Bird 1989). As Hartshorne
(159, 106) notes, “some extent of time is necessary in the primary description of existing
interrelationships and rates of change.” Anderson (1982, 273) agrees: “Time is a universal
attribute of the real world and, hence, of information about the real world.” No single
approach can reveal the whole nature of time; different approaches bring different aspects
of time into prominence (Lucas 1973). In geography, time is logically involved in at least
four different ways: as a way of dealing with the present, as a way of describing the
direction and frequency of change, as a way of providing historical context in the
explanation of existing features on the landscape, and as a way of explaining the causes of
development of land surface features. As Hartshorne (1959, 106) stresses, “the recognition
that the constant concern of geography is to study phenomena not in themselves nor in their
separate variations over the earth, but in the areal variation of the phenomena as interrelated
with each other” is the key to a proper understanding of the function of time in geography.

The temporal dimension is fundamental in geographical studies if they are to include the

 

 

39

dynamic (Bird 1989). Presumably, time of some sort is the “primary arbiter” of what
constitutes a dynamic model (Martin, Thrift, and Bennett 1978, 1). Time becomes
“more interesting” when it is used to elucidate ‘how’ rather than just ‘when’ questions (Bird
1 9 89, 34).

Haggett (1990) describes four types of temporal change in geography: constants,
trends, cycles, and shifts. Both constants and trends are long—term changes; constants move
monotonously in one direction, while trends move in a more general direction. Cycles
reverse direction repetitively. Shifts or discontinuities occur when there is a sudden abrupt
change in direction and, like cycles, represent short—term change. The concept of time can
also be used in at least four distinct ways in the study of social change: as a social factor in
clue explanation of changes, as a causal link between other elements, as a qualitative measure
0 f time, or as a quantitative measure of their interplay (Heirich 1964). According to
Heirich, an acceptable explanation of change should tell 1) what happened, 2) how it
happened, 3) why it happened at that time, and 4) why something else did not happen.
Time is relevant to each of these requirements. Time is not a distinct phenomenon, but

“ - g l o l n c
eths irrevocably as an element of the tripartite relationship of mass, space, and time”

 

(Barrera, Frank, and Al—Taha 1991, 2). Simply put, we cannot separate time from space
(Thrift 1977).

But what is time? Lucas (1973, 3) describes it as “not only the concomitant of
C0 nSciousness, but the process of actualization and the dimension of change.” This
”dimension of change’ as Lucas suggests, is the transformation of objects as they occur in
SpaCe over time and their movements in relation to one another (Peuquet 1994). The time
dirrlension, with its clear past, present, and future, is more accurately regionalized than the
SpaCe continuum of existence (Abler, Adams, and Gould 1971). Time is a major

ditTlension along which all events occur and around which human life cycles evolve

 

(Kellerman 1989). This serves as a framework by which events, chains of events, and

developments can be ordered. This relationship between time and events is what

I -

 

40

distinguishes absolute time from relative time. The doctrine of absolute time, which
views time as similar to space in that all are “containers for occurrences” (Langran 1992,
27), argues that time is a dimension uncorrelated with the spatial dimension: times are
logically prior to events because events can only be differentiated and identified with
reference to times. Proposing that time is the quantity of movement, both Aristotle and
Newton felt that one could unambiguously measure the interval of time between two
events, and that this time would be the same whoever measured it, “provided they used a
good clock” (Hawking 1988, 18). Believing that time is absolute and space is not,
Newton theorized that if a pulse of light is sent from one place to another, different
0 bservers would agree on the time that the journey took, but will not always agree on how
far the light traveled (Hawking 1988).

However, Einstein’s theory of relativity, which states that the laws of science should be
the same for all freely moving observers regardless of their speed, revolutionized
Newron’s ideas about space and time. Relativity stated that all observers must agree on
l'low fast light travels, but will not agree on the distance the light traveled and, therefore,
Will also disagree over the time it has taken (Hawking 1988). Einstein showed that time
is affected by motion and that time arises from events, rather than that time exists before
events (Schlegel 1968). By the 19205, most scientists had accepted Einstein’s premise that
time is not absolute, but rather is a fourth dimension that interacts with space (Langran
1 9392). Lucas (1973, 5) explains: “Time is related to change, and through change to the
th ings that change and the space in which they change.” This relativist approach to time sees

eveI‘lts defining the movement of time (Rescher and Urquhart 1971); events are triggered
by Other events, not by time (Lightman 1993). Campbell and Olson (1991, 21) further
this argument by conceptualizing time as being made up of different strands or “braids”
intertwining between entities across scales and over space. So if times do not exist of

theInselves, but rather are constructed (Parkes and Thrift 1980), how then is time

measured?

 

41
The problem of quantifying time has tormented the minds of great philosophers far
more than its correlative, space (Kellerman 1989). For some, measuring time seems
“mysterious” and “suspect to the scientific mind” (Lucas 1973, 4). For others, as Lightman
(1 993, 127) facetiously suggests, the task can be fatal: “their bodies stand frozen on street
corners... in time, these statues are taken to the quarryman, who cuts them up evenly in
equal sections and sells them for houses when he needs the money.” While the temporal
and spatial dimensions can be generally viewed as similar and inseparable, measurements
of each are not. Whereas spatial measures are of changes in state between two entities over
s pace at a single time, temporal measures are of changes in state, over time, for a single
entity. All time measurements involve movements in space (Kellerman 1989). This
f—Lljidamental distinction between duration and distance is a proven difficulty in building
models of spatiotemporal topology (Barrera, Frank, and Al—Taha 1991). Although time
exists, it cannot be measured tangibly like space; rather time is perceived. As Abler,
Adams, and Gould (1971, 9) explain, time is related to consciousness: “Perhaps our
insensitivity to space is related to the fact that movement in space is voluntary whereas
movement in time is wholly involuntary. Our inability to control time makes us more
Sensitive to it.” Lucas (1973, 7) agrees: “I cannot conceive of a mind being conscious of
so mething about whom the question ‘when?’ does not arise, whereas there are many states of
C0 nsciousness for which the question ‘where?’ does not naturally arise.” Rescher and
Urquhart (1971, 152) explain this notion of‘eternal recurrence’ as “ [going] through
eVeI‘ything in exactly the same way again and again and again in a cosmic history that is
inf“ll‘nite in both directions (temporally forward and backward).” Essentially, time may be
Co I‘lcseptualized or perceived in one of two ways: as continuous or “mechanical time” and as
disCrete or “body time” (Baker 1989; Lightman 1993, 23).
We are all familiar with mechanical time. the continuous, every day time of clocks,
calendars, and seasons. Here, standard or universal time is predicated on the natural

thYthms of the earth revolving around the sun and the movement of time is predictable

 

 

42

(Lightman 1993). Mechanical references to time in today’s society often are in units of
hours, minutes, and seconds. Decades and generations are also familiar reference points in
the chronologies of personal histories. However, units of time that refer beyond these
limits such as centuries, eras, and epochs become more difficult to perceive since their
significance at a personal level is often obscured by day—to—day events that take precedence
(Hardwick and Holtgrieve 1996). Morse (1996, 29) defines such lengthy periods of
1 0,000 years or more as “only a tick in time for earth.” Fernand Braudel, a twentieth
century French scholar, referred to these scales of time as ‘geologic’ whose “passage is
almost imperceptible” and also as a kind of historical time in which “all change is slow, a
hisrory of constant repetition and ever returning cycles” (Marwick 1984, 36). In
geography, landforms are the most nearly static among all phenomena of areal variation in
which the process of change is so slow that little error results from assuming they are static
(I—Iartshorne 1959). Other types of mechanical time ‘clocks’ are world time, database
time, and institutional time. With respect to data, world time is the beginning, discovery,
Change, and extinction of some mappable entity. Database time refers to the measured
transactions of the data object, such as insertion, mutation, and deletion. Institutional
time is the chronometric dimension with respect to some institution that has a known
direct relation to the mappable entity. Barrera, Frank, and Al—Taha (1991, 4) use the
example of a bank deposit to show the interplay of these three dimensions of time:

“A customer might make a deposit of funds to a bank’s

after—hours de ository at, say, 8:00 pm. The bank would

consider the eposit to have been made at 10:00 am, the

opening of business, on the following day, while the deposit

may be entered in to the bank’s databases b an employee at

get another time, perhaps 9:15 am. Thus t e same event has
een measured at three different times in each of the three

contexts (dimensions).”

Feb\rre, a leading French scholar of the Annales School on history, observed that sixteenth
Century conceptions of space and time were extremely imprecise and felt that measured or
Clock” time was less significant than “body” or “experienced” time (Marwick 1984, 29).

Body time refers to time being internally perceived as discontinuous. Leach (1961,

 

43

133) offers this explanation of nonmechanical time: “The regularity of time is not an
intrinsic part of nature... it is a man made notion which we have projected into our
environment for our own particular purposes.” For example, time can be related to human
activity, a form of which is composed of segments of movement. Here, social time is
measured by the frequency, duration, and sequence among socially relevant items such as
birthdays and anniversaries (Parkes and Thrift 1980). Leach (1961, 135) describes this
philosophy: “Without [celebrations]... all order would go out of social life... we talk of
measuring time, as if time were a concrete thing waiting to be measured, but in fact we
create time by creating intervals in social life... until we have done this there is no time to
be measured.” As Allen (1984, 127) theorizes, we cannot characterize events and actions
Without first specifying a temporal logic: “the only time we can identify are times of
occurrences and properties.”

As another type of nonmechanical time, lifetime is biological and psychological
because it provides a basis for experimental orientation by facilitating the sequential
Ordering of times (activities) into the past and into the future (Parkes and Thrift 1980).
Essentially, it provides us with a ‘sense of time’ which may seem long or short, depending
on the background of contrasting events (“for children, time moves too slowly already, yet
fol‘ the elderly, time darts by much too quickly”) (Lightman 1993, 175). The
approximate unit ofa lifetime is understood by all (Hardwick and Holtgrieve 1996);
I<'Etllerman (1989, 7) describes this form of inner perception of time as “a basic tool to
orgamize a series of differentiated occurrences along past, present, and future.” In
geography, the past is used as an aid in understanding present geography. The manner and

degree to which time is important for this purpose varies notably within the discipline, but
the “historical dimension of time” is usually combined with the dimensions of space
(Hartshorne 1959, 84).
Regardless of what temporal mechanism by which we account for our activities and

eVents that occur around us, there are several characteristics of time with which we

l _I, s,

 

 

 

44

internalize the progression of events. Temporal properties attempt to measure the time
involved with a particular event or occurrence, either with mechanical quantifiers of time
such as minutes, hours, or years, or with discrete measures such as quick, sometimes, long,
or always (Rescher and Urquhart 1971). One such property of time is direction. Time
always has a direction and as Lucas (1973, 43) explains, to conceptualize time without
direction is not time at all: “If we could not order events according as they occurred
before or after one another, we could not regard them as readily temporal.” Kellerman

( 1989) likens this notion of time’s linearity to water, always present and flowing onward.
Time, although theoretically possible at microscopic levels in such disciplines as physics
and chemistry, is irreversible at the level of our senses (Porush 1991). Using an allegory of
verifying a spatial measurement, Lucas (1973, 4) refers to this irreversible nature of time:
“We can measure space by moving a ruler... and checking that all the readings are
repeatable... but we cannot take a standard second around, and lay it off sixty times
against a minute yesterday, and then against a minute today, and back to check yesterday’s
minute again.” Another characteristic of time is duration. Duration, or ‘how long,’
describes the continuance of an event in a space with some associated change (necessary so
as to know that time is passing). Duration tackles the continuous, linear nature of change or
movement (Vasiliev 1993); it is the ‘thematic continuity’ in the temporal dimension

( I)iarkes and Thrift 1980); it. As a quantifier of duration, instants correspond to real
r"~lrribers and mark the beginning and end of isochronous intervals, another important
Characteristic of time (Lucas 1973).

Temporal intervals are what distinguishes properties, processes, and events; intervals
attempt to quantify the lengths of time between events (Peuquet 1994). Time intervals
an(i durations always begin and end with ‘the same thing’ — a pulse beat, a clock strike,

New Year’s Day (Leach 1961). Some see discrete time intervals as a way of modeling

Chaos, which does not denote true randomness but rather the “orderly disorder

Characteristic” of complex systems (Hayles 1991, 1). Measuring change on a continuum,

 

 

 

45

deterministic models may produce what looks like random behavior. However,
measuring those same changes from year to year will often uncover behavior that has “an
exquisite fine structure” (Gleick 1987, 79). Such is the case with determining next spring’s
gypsy moth population; many insects have a single breeding season. In a simple model
like this one, following a population through time is only a matter of taking a starting
figure and applying the same function again and again, year after year. Time intervals can
also be used in epidemiology where epidemics such as measles, polio, and rubella tend to
come in cycles, regular or irregular. In analyzing records of measles epidemics in New
York City over a two hundred year time period, deterministic chaos was found in
fluctuations of the Canadian lynx population (Gleick 1987). As seen here, viewing time as
a discrete variable instead of as a continuum may sometimes yield greater understanding
of a dynamic process in a large complex system by allowing the discovery of patterns in
change. Intervals or periods refer to frequency or ‘how often’ an event occurs repeatedly
over some longer time interval (Parkes and Thrift 1980), and it is frequency that defines
the temporal pattern of events and processes.

Several disringuishing types of temporal patterns occur at various time intervals:
homogeneous, majoritative, occasional, and wholistic (Rescher and Urquhart 1971).
Homogeneous processes occur at all times throughout the interval. With this type of
temporal pattern, time is linear and continuous. The frequency is one: the event takes
Place only once, but is continuous throughout. This idea of ‘steady—state’ (Peuquet 1994) is
apparent in recurrent wave models. For example, certain trends of diffusion phenomena
are essentially continuous in time; they are constantly with us and merely oscillate in their
.Pl‘cvalence, each oscillation being marked by a distinct peak which may occur regularly on
a cyclical basis (Haggett 1978). Kellerman (1989, 41) asserts that the source of linear
tinie is “a unique event not subject to repetition.” This concept of non—repetition is

described by Leach (1961, 125) in his proposition that life change is irrevocable: “all

1iVing things are born, grow old and die, and this is an irreversible process.”

 

46

Majoritative, occasional, and wholistic temporal patterns, which all describe
processes and activities that are discontinuous in time, deal with recurrences. If the
frequency of events is recurring or oscillating, it is said to be ‘chaotic’ and ‘tandom’
(Peuquet 1994). Majoritative and occasional temporal patterns describe such frequencies as
activities that occur at most and at some times throughout an interval, respectively. \With
these two types of temporal patterns, the frequency is not continuous, but does create
intermittent periods of varying frequencies which are unpredictable (Rescher and Urquhart
1971). Majoritative and occasional temporal patterns are noncyclical processes in which a
system is continuously being carried into states which are different from all previous states
of the system (Schlegel 1968). With these two types of intermittent processes there are
recurrences, but there is no cycle and no closure of a sequence (Parkes and Thrift 1980).

Wholistic temporal processes are those in which the interval is treated as a whole with
the event not possibly transpiring during any subperiod (e.g., flying from New York to Los
Angeles) (Rescher and Urquhart 1971). Cyclical activities are examples of wholistic
processes. Cycles are activities in which the system repeatedly grows from some initial
State, through intermediate states, and then again to the original state (Schlegel 1968);
they are “a string of events known as a sequence, which repeats itself and is represented by a
Closed loop” (Parkes and Thrift 1980, 13). Here time is considered as “a continuous
recurring rhythm” (Kellerman 1989, 39). Circadian or daily rhythms are believed to be
genetically based (Orme 1978) and occur in response to the alternation of day and night.
The rotation of the earth on its axis produces equal, repetitive fluctuations between day
and night, one season to the next, phases of the moon, etc. (Doob 1978). This natural
re[Detition serves as a kind of “metronome” (Leach 1961, 125) by which we measure the
Iloncyclical changes of nature (Schlegel 1968). Lightman (1993, 8) also views time as
Q’Clic: “suppose time is a circle, bending back and forth on itself... the world repeats
itSelf, precisely, endlessly.” Others note that it is change that is perfectly periodic, not

t“Tie that is cyclic: “If time really were cyclic, there would not be a recurrence of events

 

 

 

 

47

that were qualitatively identical though numerically distinct but, rather, the events would
be numerically as well as qualitatively identical... it would be the selfsame event — not the
same sort of event all over again, but the very same event just once” (Lucas 1973, 59).
However time is perceived and measured, the inclusion of the temporal element in
the spatial process is necessary so that ultimately an understanding of the nature and
structure of the processes involved can be advanced (Peuquet 1994). Understanding the
dynamics of a spatial process under scrutiny, which may lead to greater insight on the
interactions among diverse geographic phenomena, involves recognizing the interplay of
the spatial and temporal properties of a phenomenon as it changes through space and time.
Of equal importance in interpreting a geographic movement is scale. Variation in scale in
either the spatial or temporal dimensions, or a combination of the two, will greatly

influence the comprehension of the geographic change that is occurring.

Scale

Movement requires change in both the temporal and spatial dimensions. To register
change and morion we need some fixed reference points, b0th in space and in time. As
Prince (1978, 21) suggests, time and space are intertwined with each other, and with scale:
“In all events, many different kinds of movement are occurring simultaneously. While the
earth r0tates and day passes into night, people move from one place to anOther and they
also move within the carriage of the stream that is conveying them.” Scale is a geographic
variable and is considered as “sacred” to the discipline as disrance (Meentemeyer 1989,
164). Webster’s New World Dictionary defines scale as “a system of grouping or classifying
in a series of steps or degrees according to a Standard of relative size, amount, rank, etc.”
(Guralnik 1984, 1269). Scale may refer to the spatial or temporal dimension of an object
or process, characterized by both “grain” and “extent” (Tumet, Dale, and Gardner 1989,
246). Grain refers to the finest level of spatial clarity possible with a given data set (e.g.,

pixel size for raster data). Extent refers to the size of the study area or the duration of

48

time under consideration. Both of these may be referred to as the resolution, or precision
of the measurement, and should be labeled that of the coarsest element(s) in the study
(Tobler 1988).

‘Scale’ is a word with more than one meaning (Bird 1989); it is used in many
contexts and often connotes different aspects of space and time (Carlile et al. 1989). As
with space and time, scale can be of an absolute or relative nature. Absolute scale refers to
the actual, physical distance, direction, shape, or geometry. Relative scale refers to a
transformation of absolute scale to a scale that describes distance, direction, or geometry
based on some functional relationship (e.g., the relative distance between two locations
based on the effort required by an organism to move between them). Cartographic or map
scale is the degree of spatial reduction indicating the length used to represent a larger unit
of measure (Turner, Dale, and Gardner 1989). Cartographic scale is usually expressed as a
ratio of distance on a map to the corresponding distance on the earth (e.g., 1210,000)
(Martin 1991; Hudson 1992). As Bird (1989, 22) suggests, cartographically, the
definition of scale size may be somewhat confusing: “As geographers, we should be
nervous about using expressions like ‘large—scale’ and ‘small—scale,’ because large—scale
map sheets cover small areas, and to map phenomena on a small scale is to cover a wide
area.” In most other areas of analysis, fine scale refers to minute resolutions on small study
areas, and broad scale refers to coarse resolutions on large study areas (Turner, Dale, and
Gardner 1989). Regardless of its definition, the concept of scale size helps us to
understand a geographic phenomenon and its causes and effects more readily.

Scale affects virtually every analytical thought and process, regardless of subject
matter; The range of spatial and temporal scales used in all scientific analysis is
considerable and varies from discipline to discipline. For example, scale can range in
space from the subatomic in chemistry to the intergalactic in astronomy, and in time from
millionths of a second in physics to billions of years in geology. Geography, viewed

comparatively with other such sciences, focuses more on the intermediate range of

49

temporal and spatial scales because, as Hudson (1992, 283) notes, “its subject matter is
focused on place—to—place relationships involving the human use of the earth.” The range of
spatial and temporal scales used in geography does vary significantly, however, and spans
fifteen orders of magnitude or more (Meyer et al. 1992). Using scale to distinguish
geographic data from all other spatial data, Martin (1991, 45) refers to ‘spatial data’ as
measurements which relate to objects existing in space at any scale and ‘geographic data’
as spatial phenomena occurring in space bounded by the size of a building to that of the
earth, or from “the architectural up to global scales.” Abler, Adams, and Gould (1971,
12) suggest that the lower end of geography’s spatial scale should extend to the room
within a building where they assert “important activities occur and where competition for
space is just as real as at larger scales.” The temporal scale can vary in geographical
analysis from minutes and hours, such as in journey to work studies, to tens of thousands of
years, as used to study continental drift. However, as Hudson (1992, 283) notes,
geographers’ time scales are “strongly skewed toward the present“ and are “more compact
than a geologist’s, broader than a historian’s, and much broader than that of an economist
who focuses on contemporary trends.”

Each discipline adopts its own scale paradigm to describe the range of scales it
typically uses. A scale paradigm is a nested series of spatiotemporal configurations, each
bounded by the next larger scales and each integrating all the patterns and processes
ongoing at lower levels within the hierarchy (Delcourt and Delcourt 1988, 27). Although
subjective, the spatial and temporal extents of each of these scale levels are generally
referred to, in order from small scale to large scale, as micro, meso, macro, and/or mega.
Micro, meaning a millionth of a specified unit (Guralnik 1984), refers to scales at the
lowest levels of administrative units. Geographical analysis conducted at micro or very
small spatial scales are those that examine patterns and movements Within and around a

room, house, or neighborhood and community (Kellerman 1989). Movement behavior at

this scale has been studied extensively, particularly short distance and pedestrian

 

 

50

movements (Jakle, Brunn, and Roseman 1976). Changes occurring at micro temporal
sales, those that make up the history of day—to—day incidents and are usually analyzed in
minutes, hours, and days, are often referred to as short—term or daily events. Micro
temporal scales are often used in meteorology, where measurements are rarely conducted
for more than a few hours or days (Meentemeyer 1989). However, micro scale analysis in
geography has also been suggested to include phenomena occurring with durations lasting
up to 500 years and spatial extents up to about 100 square hectares. Generally speaking,
micro scale analysis is the domain of interest to the landscape manager, the process
geomorphologist, and to some human geographers (Delcourt and Delcourt 1988). Typical
earth surface processes operating at the micro scale include tidal currents, frost action, soil
creep, movement of sand dunes, and fluvial transport and deposition (Delcourt and
Delcourt 1988; Bird 1989).

Mesa, meaning middle or intermediate (Guralnik 1984), refers to the scale level
between micro and macro/mega in the hierarchy of a scale paradigm. According to
Meentemeyer (1989), meso scales are usually the most difficult to define and model of
all scales in a paradigm since some phenomena show distinct scale thresholds, such as
watersheds in temperate zones which display a very peaked discharge response. Delcourt
and Delcourt (1988) suggest meso scale extends in time from 500 years to 10,000 years
and in space from one square kilometer to 100 square kilometers. This domain may
encompass events occurring over the last interglacial interval and could include plate
tectonics, eustatic movements, and cyclones (Bird 1989; Delcourt and Delcourt 1988).
Meyer et al. (1991) define the spatial extent of meso scale studies in geography to range
from less than a square kilometer (biogeographic analysis of a forest stand), to several
square kilometers (geomorphic study of a watershed), to thousands of square kilometers
(analysis of a regional drought). Generally, meso scale analysis in geography occurs at
regional and intranational spatial resolutions and in temporal domains measured in

months and years. Meso scale or medium—term changes involve geographic analysis at the

 

 

51
weekly, monthly, or yearly temporal scale and may include life cycle phenomena such as
migration. Many geographic phenomena are examined at meso spatial scales including
examinations of human geographic movement at the city or regional spatial scale (e.g.,
trips using mechanized transport, including work, social, multipurpose, and long distance
trips) (I akle, Brunn, and Roseman 1976). Transportation planners, as well as urban and
political geographers, may also engage in meso scale analysis.

The highest or largest levels in a scale hierarchy are macro and mega scales. Often, the
two are used interchangeably and mean very large scales within a paradigm. Natural
phenomena operating at macro scales may range temporally from one to many glacial—
interglacial cycles (10,000 to 1,000,000 years) and in spatial extents averaging in width
from 100 kilometers up to 1,000 kilometers such as the physiographic region of the Blue
Ridge Mountains of eastern North America. Bird (1989) suggests macro scale levels
occur at the national and international spatial levels and include such earth surface processes
as solar radiation, water resources, and flood problems. Mega, meaning a million of a
specified unit (Guralnik 1984), generally refers to processes occurring in spaces ranging
from continental to hemispheric and global and in time periods of over a million years.
This scale domain includes the majority of geologic time during which plate tectonics
have changed the configurations of continents and ocean basins (Delcourt and Delcourt
1988). Physical geographic phenomena examined at increasingly large spatial and/or
temporal scales are long—term changes occurring in geologic time and include atmospheric
processes, earthquake cycles, soil development, glaciation, and global climatological
processes such as carbon dioxide variations (Prince 1978).

Kellerman (1989) proposes a simple scale paradigm where three dimensions (time,
space, and society) yield micro, meso, and macro levels of reference. According to
Kellerman, the most studied inter—connections among time, space, and people seem to
have taken place at one or more of the extreme edges of the three continuous dimensions.

That is to say that the spatial and temporal scales are “loosely associated” and are

Lm

 

52

typically engaged in many geographic societal studies (Langran 1989b). For example,
urban planning deals with relatively large areas and long time frames (e.g., communities
and years), while behavioral geography generally focuses on smaller temporal and spatial
human units (e.g., hours and city blocks). Activity space studies have shown that human
activities which are the most routine involve the smallest spaces and are correlated with the
shortest periods of time. Conversely, rare unroutine activities, such as changing careers,
often involve movement over large spaces or distances and can be so rare as to recur only
once or twice in a lifetime (Meentemeyer 1989). Here, time is equated with distance
where temporal and spatial scales are paired into distances that could be traversed in time
spans (e.g., minute/room, day/city, week/region, year/nation, and life/world) (Langran
1989b).

The linearity between spatial scale and temporal scale is also supported by some
earth system processes. As Kelmelis (1991) suggests, physical processes that only occur
within minutes also occur at a spatial scale of a few kilometers, such as atmospheric
turbulence, while processes such as plate tectonics occur over thousands of kilometers, as
well as over millions to billions of years. Processes and phenomena such as glaciation and
continental drift involve broad spatial scales and appear to be changing so slowly that very
long time scales are needed to observe and model those entities (Meentemeyer 1989).
Peuquet (1994, 452) offers the following example, further demonstrating the direct
relationship between spatial and temporal scales: “... individual plant growth tends to be
a micro scale process since it occurs over a spatial scale of meters or fractions of a meter as
well as over a temporal span of weeks. Similarly, changes in vegetation patterns tend to be
meso scale processes since these occur over a regional scale... as well as over a temporal
span of decades.”

As seen with these examples, the “coordinate scheme” for locating events changes with
the scale at which the events are analyzed (Holly 1978, 12). It would seem inconsistent to

attempt an analysis of events recurring at a macro scale with minutes or hours as the

 

53

temporal metric; the use of weeks or months might be more manageable. Although there
are instances where noncomparative levels of scale are used in an analysis, such as in
historical geography where a macro time scale is generally employed while examining the
landscape with a relatively small spatial scale, generally scales at which changes occur are
largely dependent on one another: “In changing one’s spatial perspective by looking at
different geographic scales, we concurrently change temporal scale” (Iakle, Brunn, and
Roseman 1976, 101). Regardless of what level and degree of association among the scales
used, all analyses require the definition or designation of fixed spatial and temporal
reference points by which observations and conclusions can be drawn. Determining which
temporal and spatial scales to use in the examination of a geographic phenomenon is
imperative and will affect the focus, methodology, and results of an analysis.

The scale of analysis is critical in science or any other ordering system (Abler,
Adams, and Gould 1971, 37). As Hudson (1992, 280) describes, a successful research
design first defines the scale at which the phenomenon of interest can be observed and then
selects methods of analysis appropriate for resolving patterns and processes at that spatial—
temporal scale: “All fields of study that make empirical observations must consider
scale. When the scope of any study is set, decisions are made about what is to be studied
in detail and what is to be left in the background. Scale is not a constant in most cases: it
is stretched here or compressed there to emphasize what is important at the expense of
what is peripheral.” Choosing the appropriate scales in space and time in which
observations are made and analysis presented will ensure the appropriate formulation and
answering of meaningful research questions. Scales are determined by observer—dependent
criteria and both analytical (e.g., fractals, Fourier transformation, multiscale ordination)
and empirical methods can be used to determine scale (Turner, Dale, and Gardner 1989).
Choosing the level that is thought a priori best for the problem data may be guided by past
experience of that sort of data, by whether our principal objective is synthesis or analysis,

or by a political view that gives the determining scale. Some concur that for ecological

 

54

phenomena a “natural scale” exists that will yield the most information about the
phenomena (Carlile et al. 1989, 203). Langran (1992) theorizes that ‘common wisdom’
dictates that information should be captured and stored at the finest resolution required.
Agreement on appropriate scale is not always reached: “As many fiitile arguments among
geographers arise out of different opinions as to the scale at which phenomena should be
analyzed as arise in normal discourse because the discussants hold different definitions of
critical terms” (Abler, Adams, and Gould 1971, 37—38). Regardless of the scale chosen to
confront a problem, the choice may not be as error—free or appropriate as supposed (Bird
1989).

Scales may be inappropriately selected due in part to a host of “determinants and
constraints” that economically and logistically restrict the choice of most appropriate
scale, such as the size and speed of a spatial phenomenon or process, scales of exisring
maps, aerial ph0tographs, and remOtely sensed images, data handling thresholds (e.g.,
time, technology, money), mathematical—statistical constraints (e.g., spatial/temporal
autocorrelation), or the size of the spatial units (e.g., census tract) (Carlile et al. 1989;
Meentemeyer 1989, 171). So constrained is the choice of appropriate scale that
sometimes, as Meentemeyer (1989, 171) suggests, “in the end it seems that scales are
unconsciously selecred and therefore may seem to be entirely arbitrary.” Tumer, Dale, and
Gardner (1989) suggest that in developing the appropriate scale for the phenomenon of
interest, it may be necessary to sample processes at several scales in order to identify any
scale dependencies. In determining the most appropriate measurement scales to use for
different processes or physical properties, it may also become necessary to study the
various processes influencing something, thereby introducing different levels of resolution
on the scale of observation (Carlile et al. 1989).

In viewing changes, we apply different scales to different kinds of occurrence. As
with the spatial dimension, the temporal dimension’s scale can range from very short

(micro scale) to very long (macro/mega scale). However, unlike the natural progression of

 

55

increasing spatial scale, scaling time is much more subjective and, as Prince (1978, 22)
comments, the definition of time scale is at the discretion of the examiner: “The
difficulty with delimiting the categories of time has always been to find suitable
descriptions of duration, which could vary according to events while measuring them
against a fixed scale. History has no periodic table of contents and no classification of
types or species; it has only solar time and a few old ways of grouping events, but no
theory of temporal structure.” The choice of time scale not only has important
consequences for analyzing change, but has implications for the kinds of hypotheses one
forms (Heirich 1964). If equal quantitative units such as a month or year are selected,
time can be used as an interval scale, the size of which will affect the kind of changes that
can appropriately be analyzed. For example, a series of individual events can be clustered
into ‘episodes’ and perhaps into ‘cycles.’ Perfectly cyclical disrributions can represent an
important form of steady—state behavior over longer temporal intervals (Peuquet 1994). In
defining a temporal level of duration, the difference between ‘abrupt’ and ‘gradual’ change
is relative depending on the temporal scale used (Peuquet 1994). Furthering this point,
Prince (1978, 22) explains how “revolution,” which he defines as “a fundamental change
accomplished in a short period of time” can describe the “brevity” of the eight month
Russian Revolution in 1917, as well as the four—year French Revolution beginning in 1789.
The ideal temporal scale for viewing phenomena depends on the phenomena. Phenomena
change at different rates, and, ultimately, methods of deriving one temporal scale from
anather is needed (Langran 1989b).

Qualitative and quantitative changes in measurement across spatial and temporal
scales will differ according to how scale is defined; therefore, the definitions and methods
of changing scale must always be explicitly Stated (Turner, Dale, and Gardner 1990).
Sciences dealing mosdy with processes (e.g., meteorology) are better able to switch scales
than those dealing mosrly with phenomena because the size of the phenomena decides the

scales (Meentemeyer 1989). Geographers are especially adept at moving from one scale

56

to anOther. As Bird (1989, 19) claims, “the eyes of a geographer light up when scale is
mentioned” given the discipline’s great advantage in its ability to “slip smoothly”
between scales and compare different levels of approach. This holds true especially for
physin geographers who, given the variety of spatial and temporal scales of landscape
processes (Baker 1989), have always worked at multiple scales (Meyer et al. 1992). The
range of spatial and temporal scales of physical phenomena, such as ecological problems,
has expanded dramatically in recent years and many environmental problems require
landscape, continental, and even global levels of understanding (Turner, Dale, and Gardner
1989). To a lesser extent, this holds true for most human geographers, who have had to
broaden the spatial and temporal extents in analyzing patterns of human interaction given
the rapid technological advancements in communication, information, travel, and trade.
Geographers typically focus on a particular range of scales in their analyses so as to
ensure that conclusions drawn from a study are well matched to the economic, social, and
physical processes known to underlie observed patterns (Hudson 1992). Variations in b0th
temporal and spatial scales largely affect the identification of patterns; trends or patterns
discernible at one scale are often invisible at another (Kellerman 1989). For example,
examining a single volcanic eruption may yield information about that particular volcano
such as the rate and range of the Iava’s movement. But by expanding the spatial scale or the
geographic reference, Other volcanic eruptions elsewhere could be included in the analysis.
Doing so may reveal insights and patterns to volcanic activity in general, thereby allowing
further comparative analysis of causes and effecrs and permitting predictive modeling to
be conducted. Changes in the temporal scale may also reveal very different synopses of an
activity. Looking at the patterns of migration over a period of several years or perhaps
even centuries will reveal identifiable patterns of settlement on the landscape and may
lead to insights on the causes and effects of the migrations. In contrast, examining
residential migration within a small temporal scale of a month or a year will yield an

entirely different set of questions and conclusions about human movements. Simply put,

57

“geographic phenomena look different at different scales” (MacEachren et al. 1992, 131).

Changes in scale also change the important, relevant variables in an analysis.
Parameters and processes important at one scale are frequently n0t important or predictive
at anOther scale (Turner 1990). The value of a phenomenon at a particular place is usually
driven by causal processes which operate at different scales (Meentemeyer 1989).
Interactions between components occur at different scales (Campbell and Olson 1991).
For example, coupling between the surface of the earth and atmosphere occurs across a
wide range of temporal and spatial scales, each influencing each other. Specifically,
evapotranspiration occurs within hours and affects the atmospheric state (i.e., net radiation
at surface, precipitation, temperature, etc.) which in turn affects photosynthetic and
transpiration rates of the surface vegetation and hydrologic cycle, which eventually
(decades to millennia) affects terrestrial landscape Structure and underlying soils (Hall,
Strebel, and Sellers 1988). As the dimensions of time and space change, cause and effect
relationships may be obscured or even reversed, and the system or process itself may be
described differently. For example, in describing the three—fold scalar division under
capitalism, Bird (1989) describes capitalism at the state level as an ‘ideology,’ at the
world system as ‘reality,’ and as ‘experience’ at the local scale.

Generally, fine scales (e.g., individual level) determine cause and effect, while
models for broader scale patterns result in less predictive accuracy at specific points or
places; aggregate scales generate hypotheses and are useful for generalizations. Often,
extrapolation of fine scale measurements for the analysis of broad scale phenomena is
necessary as is in many ecological studies (Turner 1990). As the scale of Study advances
toward the meso and macro/mega scales, more generalization becomes necessary,
involving the judicious selection and omission of detail. Given that geography is
primarily an empirical science, generalizations (e.g., models) are only as good as the
finest-grain spatial data available (Meentemeyer 1989). Bird (1989, 26) compares the

regional geographer to a “painter” or “novelist” who must selectively interpret a vast and

58

complicated landscape from numerous accurate descriptions of every small area that tests
within the landscape. Information is often lost as spatial data are considered at coarser
scales of resolution (Turner 1990). Often times, generalization and problem—solving may
require a focus on the understanding of the context and consequences of a Single event (Bird
1989). In ecology, a variety of processes that have been studied on small areas have
important ramifications on the landscape scale that have nor been well—Studied. For
example, the dieback process in certain forests, which may be caused by anthropogenic
causes such as air pollution and the introduction of exotic pathogens, or by ‘cohort
senescence’ (i.e., aging of particular stands), causes certain alterations such as the forest’s
diversity, age—class Structure, and patch Structure. These alterations, while important in
themselves, also have important ramifications for the spread of subsequent disturbances,
such as decreased habitat for some indigenous species. Thus, better understanding of
landscape processes might come from ‘scaling—up’ from more detailed studies that have
already been completed on smaller land areas (Baker 1989).

Extrapolations from higher levels of resolution to finer scales also occurs in the
scientific method, geography not withstanding (Meentemeyer 1989). As Martin (1991,
59) explains, the key difficulty with this method is that “there are a very large number of
possible areal units which may be defined, even with the imposition of certain Size and
contiguity constraints, and none of these has intrinsic meaning in relation to the underlying
distribution of population.” Referred to as the modifiable area unit problem, once data
transformation has taken place, the characteristics of individuals cannor be retrieved from
the aggregated data. Problems can arise from inferring that statistical patterns at the
macro scale are replicated at the meso and micro scales. Such a “clashing of scales” has
become known as the ecological fallacy where conclusions about individuals from grouped
data are incorrectly deduced (Bird 1989, 26—27). This problem of scale involves
differences in inference and relevant variables caused by different scales or hierarchical

levels (Meentemeyer 1989). Different scales of approach may eventuate in different

59

results. When the scale of mesh used to catch and present the Spatial data (and affecting the
results) is chosen with malice aforethought, situations such as gerrymandering of electoral
districts arise (Bird 1989).

Scaling processes, particularly landscape ecological phenomena, from one level to
another is nor a Simple mathematical progression as with Newtonian mechanics and the
Euclidian geometry on which it is based (e.g., whether the Sides of an isosceles triangle are
two centimeters or two kilometers long, the triangle Still has the same properties) (Golley
1989). Rather, for most complex systems and irregular forms, Statements made about one
scale level do n0t hold true for anOther (Hayles 1991): “We cannOt move as a boat builder
does from a model to the fiill sized craft... scaling up or down requires that the systems at
the two levels be Similar in the properties of interest” (Golley 1989, 72). The Significance
of emergent properties appear when moving from one level to a higher in a hierarchy. Each
level has its own unique properties that are nor derived by mere summations of the
disaggregate parts. Scaling up will not yield these emergent properties; rather, they can be
known only by direct Study of the process of the system at that level (Hayles 1991).
Golley (1989, 71) uses ecological processes such as flows of water, nutrients, and pollutants
and the response of the bi0ta to these flows to demonstrate that although all are linked
with one another, there is great difficulty in moving from one level to another and
comparing results: “...it is not clear how we can move from small scale systems such as
fields and farms to regions and nations, especially when we are considering the impacts of
air and ground water pollution as they interact and affect crop production and human
health.”

The mechanisms that relate phenomena at different scales are poorly underStood
(Langran 1989b). The identification of guiding principles that allow researchers to
combine data and models at different spatial and temporal scales and to extrapolate
information between scales remains a challenge. Turner, Dale, and Gardner (1989)

suggests four guidelines for making predictions across more than one scale: the spatial and

60

temporal scale of the process must be identified, the shift in relative importance of
parameters at different scales must be understood, appropriate methods to translate
predictions from one scale to another must be developed by identifying constraints that
are important at each scale, and finally the methods and predictions must be empirically
tested across multiple scales. The importance of scale in scientific analysis, particularly
in any geographical explanation, cannot be overstated. Accounting for scale in all
dimensions of a phenomenon under investigation will help in the understanding of the

dynamics of that phenomenon as it changes its location in space and time.

Treatments of Movement in Geography

“The essence of geographic inquiry is a sensitivity to the worlds of place and space”
(Graf and Gober 1992, 234). Both human and physical phenomena interact with one
another creating complex, if not distinct, patterns, arrangements, and locales on the
landscape. All geographic explanation attempts to describe in some fashion patterns,
arrangements, and locales in the landscape of interactions among human and physical
phenomena. Geography is fundamentally concerned with asking and answering questions
about phenomena tied to specific locations on the surface of the earth (Martin 1991). Such
explanations always include descriptions of space and, as Hardwick and Holtgrieve (1996,
272) note, “geography occurs not only in space, but in time.” Changes in space and time
often include changes in scale. All three of these dimensionalities surrounding a
geographical phenomenon interact with one another and it is these spatial, temporal, and
scalar changes that compose a phenomenon’s dynamic nature and are at the heart of that
phenomenon’s explanation. Given this natural ‘tendency’ of most geographic explanation
to include, no matter what the intent, aspects of change, how has the discipline as a whole
acknowledged the significance of dynamism in geography? Using the six fundamental

questions of who, what, where, when, why, and how, the combinations and permutations of

which “exhaust our curiosity about our experience” (Abler, Adams, and Gould 1971, 12),

 

61

the role of movement in geography will be examined.

Movement as a Central Theme

Geography is “an eclectic subject” that ranges from the physical sciences through the
social sciences to the arts and humanities (Boehm and Petersen 1994, 211). It is this
“breadth of topics and divergent points of view” that often has been the focus of criticism
of the field of geography (Meentemeyer 1989, 163). Perhaps such criticisms are what have
spawned a national reform movement in defining geography curriculum in this country.
Guidelines fitr Geographic Education (Joint Committee on Geographic Education 1984)
marks the beginning of several conceptual frameworks designed to improve education and
understanding of geographic principles, concepts, and themes. Guidelines did not claim to
represent a ‘new geography,’ a new approach to geography, or a new geographical
taxonomy, but rather served as a vehicle to convey the core ideas of geography to the
general public. Written in 1984, the document contains scope and sequence in geography
with suggested learning outcomes for elementary and secondary grades, outlines basic
social studies skills, and suggests strategies teachers can use to analyze geographic
information that students, in turn, can use to study and eventually find solutions to social
questions or problems. As an alternative to the “detrimental, but unfortunately persistent”
habit of teaching geography through rote memorization, Guidelines stresses higher level
concepts of geography that allow students to learn about the critical interface between
physical and human environments (Boehm and Petersen 1994, 21 1).

The mOSt enduring contribution of Guidelines has been the articulation of five
fundamental themes of geography: location, place, relationships within places, movement,
and regions (Joint Committee on Geographic Education 1984). Although application of
the fundamental themes has been uneven in classroom practice, with some teachers using the

five themes in an overly simplified manner or even erroneously, widespread acceptance of

the framework by teachers has been crucial in reestablishing geography either as a

 

62

free‘standing course or as an integral part of recently revised social studies curricula
(Boehm and Perersen 1994). The five themes, as a tool for organizing the content of
geography in a convenient, widely adoptable format, can accommodate virtually all
conceptual frameworks currently in use by teachers of geography. The five fundamental
themes have become an integral element of social Studies education and have appeared in
most, if nor all, geographic textbooks and social Studies programs as a context for
geographic education.

In 1992, the National Assessment of Educational Progress, a geography consensus
group, identified the principal role of Guidelines’ five themes to be “for teaching” and
devised an alternative framework that organizes the content of geography “more
realistically” into three categories to be used “for assessment” space and place,
environment and society, and spatial dynamics and connections (Boehm and Petersen 1994,
212). Another framework developed during this period of national reform of geography,
Geography for Life lists and outlines eighteen content Standards to divide the discipline
into topics that teachers can use. Movement is involved in twelve of these eighteen national
geography Standards. For example, standard nine States: “the geographically informed
person... knows and understands the characteristics, distribution, and migration of human
populations on earth’s surface” (Boehm and Petersen 1994, 212). Movement is also
associated with such principles and elements as ecosystems, cultural patterns and mosaics,
and geopolitical cooperation and conflict (Hardwick and Holtgrieve 1996). Throughout
the national reform movement in geographic education the concept of movement as an

important, central theme within geography has persisted.

The Who or What of Geographic Movement
‘Who?’ and ‘what?’ are questions of a human and nonhuman phenomenon’s identity,
respecrively. What questions typically deal with semantics (i.e., word definitions,

essence, occurrence, or combinations thereof). Who questions are special cases of what and

 

63

how questions; they may either be disguised what questions or hidden how questions.
After some vocabulary has been acquired, who or what questions are usually how questions
concerned with the uses to which things can be put (Abler, Adams, and Gould 1971). If
movement has been classified or categorically identified within geography, it has been by
who or what is moving across a landscape. Traditionally, theories of movement have been
defined on the basis of their subject matter; as described in the Guidelines fitr Education,
people, ideas, and materials are the basic subjects that often change location over time
(Joint Committee on Geographic Education 1984). Typically these subjects of
geographic movement have distinct process names associated with them, such as migration,
which refers to the movement of people. The movement of ideas, innovations, or disease
are often explained with the theory of diflusion. Fade refers to the movement of goods
and commodities. Besides phenomena that involve humans, movement also occurs in
physical or non—human geographic processes, such as glaciation, volcanism, and hydrology.
Regionalism can be said to also involve movement; the spatial extent of a region (i.e., its
boundaries), and how these change and evolve over time, are at the heart of the description
and explanation of a region.

Lodged within these phenomena are fundamental elements about the movement
itself, such as why, where, when, and how a movement occurs. Explanation within
geography has traditionally centered on one kind or subject of movement with discussion
concentrated on one of these process elements. For example, the subject of people moving
is often discussed under the process heading migration. Within this larger framework lie
several avenues for further explanation, such as the kind of migration, origins and
destinations, length of time for the migration, the nature of the migration, what instigated
it, and what consequences it has on the subject, the landscape, and the interaction between
the two. Geographic concepts associated with this movement and that address some of

these questions are migration fields, from—to tables, push—pull factors, modes of transport,

and distance—decay effects (Jakle, Brunn, and Roseman 1976). The geography of

 

64

movement often becomes the description of people moving themselves, their products,

and their ideas across the earth.

The IVhere of Geographic Movement

Questions of a phenomenon’s occurrence (i.e., ‘where?’ and ‘when?’) are questions
concerning past or future events (Abler, Adams, and Gould 1971). In all geographic
analyses of phenomena involving movement, one fundamental line of inquiry is to ask
where the spatial change is occurring. Sometimes knowing the exact location is a
necessary part in the examination and interpretation of the phenomenon. For example, in
many human migration studies, an important aspect in understanding the movement is
knowing where people moved from and to what location they have moved. Knowing these
spatial points of origin and destination allows closer examination of these locations so as
to offer insight and explanation of the migration itself such as why the phenomenon is
occurring, what spatial patterns have been and are likely to be formed, and what effects
these human movements have on these locations.

In other geographical analyses, particularly involving humans, study usually begins
with the notion that location is space (Parkes and Thrift 1980). Here, the important
element in examining a movement is not necessarily knowing absolute geographic
locations, but rather is understanding the relative spatial structures of the movement, such
as distance affected or characteristics of the landscape. For example, in defining areas
affected by gypsy moth defoliation, it is not imperative to know the exact boundaries of
the region to understand the general pattern of the spread over time. Interpreting the

spatial characteristics of a movement, regardless of the level of specificity, is a crucial and

inherent aspect of understanding a geographic movement.

 

65

The When of Geographic Movement

Spatial change or movement across space requires change or passage of time. Despite
the inherent nature of time and space in geographical movement, examination of time has
nOt been given the same Significance as space (Parkes and Thrift 1980). This may be due
largely to the way time is measured. Whereas Spatial change can be defined in bOth
absolute and relative terms of distance and direction, temporal change measurements are
usually based on relative terms only. Nonetheless, the temporal dimension has been
incorporated into geographical analyses. ‘Time—geography’ is largely the result of
attempts by geographers such as Hagerstrand, “to develop a model Of society in which
constraints on behavior can be formulated in the physical terms (location in space, area
extension, and duration in time)” (Parkes and Thrift 1980, 243). Time—geography accepts
the notion that space and time are universally and inseparably tied together and that
examinations of human organization of the earth’s surface, human ecology, and landscape
evolution cannOt separate the two dimensions (Pred 1977). The basic premise of most
time—geography analyses deals with the time—space ‘choreography’ of an individual’s
existence at daily, yearly, or lifetime scales of observation (Pred 1977). The time—
geography framework defines three kinds Of conStraintS on an individual’s movements:
capability constraints (physiological necessities), coupling constraints (due to the bounded
togetherness of individuals, tools, and materials at given places at given times), and
authority constraints (legal or Other limitations and comfort of access) (Thrift 1977).
Time—geography specifies the necessary conditions for virtually all forms of human
interaction (Pred 1977).

The conceptual base Of time—geography considers bOth Space and time as resources.
Space—time budget Studies are based on this notion Of resource allocation and recognize
that human activities take time that can be measured in physical time units (Thrift 1977).
In analyses of space—time budgets and activity spaces, the time it takes to perform tasks is

empirically Observed in various social and environmental settings. The methodology

 

66

usually involves a selected sample of people who fill diaries detailing their activities and
the length of time required to fulfill them. The activity space, or daily trip patterns of a
typical individual, is usually dominated by movement within and near the home,
movement to and from regular activity locations, and movement in and around those
activity Sites (J akle, Brunn, and Roseman 1976). Some results of these Studies can be
beneficial for planning and decision making by indicating how activities are and can be
performed, and what impact new innovations have on these activities. For example, Forer
and Kivell (1981) analyzed time—budget data of housewives in Christchurch, New

Zealand who did not own a car. By identifying windows Of free time during a housewife’s
day, the actions and potential activity spaces of individuals in four suburbs were delimited
and the variations in access to, and choice between, facilities in these suburbs were
identified. Essentially, despite spatial constraints on busvdependent housewives who could
nOt enjoy large expanses Of the city in which to seek out and visit facilities, in some
instances there was an abundant choice of facilities because of an above—average provision of
facilities in the surrounding neighborhood. The authors concluded that public transport
differentiated access to facilities, yet marginally affected the final determination of a

housewife’s available options.

The thy of Geographic Movement

Another way Of detailing a geographic movement is to ask why it occurs. Why
queStionS can be divided into two categories: true why queStions and how questions in
disguise. According to Abler, Adams, and Gould (1971, 12), “true why questions,
whether asked abour human or physical phenomena, reduce ultimately to theology or
metaphysics.” The authors fiirther claim that “spatial arrangements are reciprocally tied
to movement processes;” that is to say, all movements create spatial Structures, and once
established, such spatial arrangements influence subsequent movements (Abler, Adams, and

Gould 1971, 237). For physical geographic processes, movement or spatial change may

67

occur due to human intervention, such as the redirecting Of a river’s flow, or may be caused
by natural processes or presences such as wind or plate tectonics. However, explaining why
things move in the human world is to refer to the ideas of spatial interacrion.

Many models Of spatial interaction attempt to explain the aggregate flows of
people, goods and services, or information between places (Abler, Adams, and Gould
1971; Jakle, Brunn, and Roseman 1976). The interaction of flow among a set of places is
an integral part of geographical analysis; the concept is broad enough to encompass several
major areas within geography (Morrill 1978). Ullman’s question “what makes objects
move over the earth’s surface?” arguably encompasses all geographic theory because it is
difficult to avoid the notion of movement when explaining how an object acquires its
location (Bunge 1966, 112).

In providing explanation for the conditions Of interaction, particularly material
interaction Of goods or people, Ullman (1956) proposes a three-factor system:
complementarity, intervening opportunity, and transferability. All commodity and
service movements can be evaluated in terms of complementary supply and demands
areas, presence or absence of intervening opportunities, and the transferability of the goods
and services (Abler, Adams, and Gould 1971). Every population movement from room
to room or from continent to continent, ultimately depends on human needs and desires,
and on place tO place complementarities (Abler, Adams, and Gould 1971). Information
(i.e., new facts, data, ideas, and routine communication) flows from place to place,
moving from places of production to demand areas. A notion of minimal movement
underlies the concept of intervening opportunities, as movement tends to be generated only
where no sources of supply are available between the two places or regions. Fansferability
refers to the technology needed to overcome distance friction. Historically, fresh fruit was
far less transferable than manufactured goods; the friction of distance has been greatly
reduced by modern (refrigerated and fast) transport.

Expanding upon Ullman’s spatial interaction model for explaining some forms Of

 

68

human movement, Bunge provides a reclassification of Ullman’s movement theories.
Reducing all movements tO their most abstract level, Bunge (1966, 113) states that the
four traditional types Of movement theories (economic, electric, fluid, and kinetic gas, all
Of which are based on subjecr matters of water, people, electricity, and goods) are an
“unsatisfactory organization because discussion of the subject matter and the mathematics
operating the theory mingle awkwardly.” TO remedy this “traditional confusion,” Bunge
incorporates four additional generic concepts into Ullman’s spatial interaction theory:
duality (real flows, incentives, and pOtentials), conservation of matter (accounting for
quantity of phenomenon), routes, and backhauling (unnecessary and irrational movement).

By relating these seven qualities of movement to the four traditional types Of movement

 

theories, Bunge proposes a reclassification Of all movement theories based on three
fundamental dynamic Structures: routing theories (theories involving finite mathematics),
theories involving two—dimensional phenomena (nonstatistical, areally continuous
theories), and theories involving a combination of the two (statistical theories). Bunge
concludes that since the generality Of these three classes Of movement theory increases from
route to continuous to Statistical, it can be Stated that route theory iS a subset Of continuous
theory, which is a subset of statistical theory. While integrating key elements Of Ullman’s
spatial interaction theory which focuses on explaining why a phenomenon moves, Bunge
(1966, 130) rationalizes all movement theories in terms of the “spatial separations of
techniques and understandings required tO operate these theories.” This change Of focus
from why a movement occurs to one that incorporates how one Studies a movement

introduces a movement’s Spatial configuration or form.

The How of Geographic Movement
How questions are Often considered the most important queStionS asked because they
lead to explanation. We describe how things move in order to understand movement laws,

and because we want to predicr and control social and natural events (Abler, Adams, and

69

Gould 1971). ‘How?’ may be a question Of existence, process, utility, or policy. When
asked about processes themselves, how questions treat sequences of events as a Single event
and are focused on the relationships between events rather than on the events themselves
(Abler, Adams, and Gould (1971). In queStioning the utility of how things move, we may
think of what is it that is assiSting in the mOtion itself, we classify based on different kinds
of carriers (Prince 1978). Carriers for movement in non—human or physical geographic
phenomena are natural processes such as atmospheric pressure, plate tectonics, gravity, and
orbits. These carriers for human movements may be the people themselves, or some mode
of transportation such as automobile, train, or airplane, with which they move themselves,

their goods, and their products. Another mode of transport in human geography is the

 

electronic highway with which ideas, news, and communication are spread via telephone
lines, satellites, and radio Signals. Utility questions Of how things move are Often resolved
in terms of modes of transport, rates of speed, and efficiency.

Asking ‘how?’ about a movement can extend tO include questions about the structure
of spatially dynamic geographic phenomena. Shifting attention from the actors and
activities Of movement, as well as whether they refer to the human world, the physical
world, or both, Abler, Adams, and Gould (1971) devised a classification of all
movements based on the geometric form Of the origins and desrinations. The authors
Stress that the building blocks for movement models are the points, lines, areas, and
volumes among which movements occur. Their classification, which considers the
permutations of these four spatial structures and details sixteen different classes of
movements and spatial interactions, is a useful way Of breaking down the complexity
which exists in the almOSt indefinite number Of movements that we Observe around us. The
basic spatial dimensions Of movements are displayed in Sharp relief to one another and,
interpreted in this fashion, different kinds of moves can be seen as basically identical in
their spatial Structure. Such a framework is important for the description and

understanding Of how things move across three-dimensional Space.

70

The Importance of Movement in Geography

Explanation Of geographic phenomena Often involves the description of change. A
phenomenon may change in regards to its aspatial attributes (e.g., the land use Of a
particular parcel of land is rezoned from agricultural to residential), its spatial attributes
(e.g., annexation of an adjoining land parcel to an existing property), or in both attributes
(e.g., subdivision of an agricultural parcel of land into fifty residential lots). Dynamic
geographic phenomena then, are spatial phenomena that change in attribute or non—spatial
phenomena that change location over time. It iS important to now that both are
‘geographic phenomena’ (involving location) even though the phenomena themselves may
be either Spatial or non—spatial.

Whether movement or spatial change is the primary focus of geographical thinking
about a phenomenon or whether it is Simply part Of a phenomenon’s description, the
principles of movement are vital components Of all geographical explanation. Willmott
and Gaile (1992) suggest that geographers should turn their attention from statistical issues
towards concern about modeling intrinsically geographic systems and processes and to
understanding geographic dynamics. Declaring movement as a theme in geography was
the firSt Step towards this new perspective.

We should now move forward and dissect the underpinnings of geographic
movement. If commonalities could be discovered that govern geographic movements,
such principles would provide a wealth of information about many geographic processes,
and would not only aid in greater understanding of a particular phenomenon, but also of
geography as well. Understanding the temporal and spatial complexities of phenomena
that move over the landscape not only facilitates greater comprehension of an individual
phenomenon under study, but leads to new ideas and insights about, and understanding of,

the interconnections and interactions among other geographic phenomena.

Chapter 4
A CLASSIFICATION OF GEOGRAPHIC MOVEMENT

Given that classification is a method by which geographers provide differentiation
and comprehension of diverse geographic phenomena, and given that “geographers are
interested in the relationship between the space in which the event is occurring and the time
it takes for the event to occur” (Vasiliev 1993, 2—3), a classification of geographic movement
is presented (Figure 4.1). The classification organizes geographic phenomena by how they
move through space and time; that is, by the description of the movement itself. Spatially
dynamic geographic phenomena are placed in categories based on the spatial and temporal
patterns or forms they exhibit as they move. In effect, the classification does not
necessarily describe the pattern of space and time of the objects per se, but rather describes
the patterns of the movement of the phenomena; that is, the interrelationships of the
changing spatial and temporal characteristics.

Geographic movements are broken down by characteristics of change in space and in
time. Spatially, all geographic data can be described by points, lines, areas, and volumes.
Traditionally, geographic classification has involved the subdivision of all subjects into
one of these four spatial object classes and has proved useful as an organizing concept for
discussion of spatial phenomena (Martin 1991). The classification presented incorporates
these four fundamental spatial features and Offers that a geographic movement may be
described as being a point, linear, areal, or volumetric movement. In describing the
movement itself, the motion may be a point movement, where the object is Observed at
two or more locations and perhaps the change between these locations is examined. With
these types of movements, the importance of the spatial change lies in the description of
the relationship between the two or more points of locations; that is, the description of the

71

Spatial Characteristics

linear point

areal

volumetric

72

Temporal Characteristics

continuous

cyclical

intermittent

 

 

 

E Class 1

\‘

 

 

 

 

 

 

 

 

s
‘o
J Q.

electronic communication
revolution of the earth

jet streams, water channels
a hurricane, a tornado

 

hurricane and tornadic activity
mass transit
regubr route services

regular trade and commerce

vehicle navigation wildlife migration irregular human mobility
monitored individual recurrent human mobility volcanic and seismic activity
movements
N Class 4

 

specialized route services
irregubr trade and commerce
apbration

 

Class 7

 

colonization
dtflusion of innovation,
culture, and disease
accuhuration and assimilation

regional boundary changes
fbrestfires

Class 8

 

gypsy moth de oliation

regional bounda changes
dtflusion disease

 

regional boundary changes
desertification
military deployments

 

Class 10

 

continental drift, volcanic
eruption, earthquake, tsunami,
glaciation, rock trans rmation,
soil creep, blowout ormation,

hydrologic cycle, atmospheric

 

motion, rotation of the earth

 

Class 11

 

ocean tides
beach recovery
El Nino
flooding
pollution

 

Class 12

 

pollution
mass wasting
volcanic eruptions
earthquakes

 

Figure 4.1 A Classification of Geographic Movement.

 

73

linkage between locations. Point-to—point movements are interactions that take place
between two locations and occur over non—contiguous space (Campbell and Olson 1991).
Interactions occurring over contiguous space may be described as linear, areal, or
volumetric. With linear movements, movement travels along a path or route such as a road
or river and the morion may be described with a line. Areal movements are those
movements that describe a spread or encroachment. With this spatial classification
category, a phenomenon’s movement occurs across a surface and the affected space results in
a contiguous area or region. Volumetric categories of movement describe motion that
occurs in three spatial dimensions and includes those geographic phenomena that evolve
and change shape in volumes, such as in air or in water.

This classification recognizes that spatially dynamic phenomena are also dynamic
temporally and that all movements may be described in terms of a phenomenon’s
duration, date, and rate. Geographers are generally most concerned with the periodicity
of an event and thus duration is a key factor in understanding a phenomenon’s temporal
characteristics. If a phenomenon moves uninterrupted throughout a period of examination,
then it is said to be continuous. If the movement is discontinuous during examination, then
the morion stops periodically or is said to fluctuate. If such fluctuations are sporadic or
irregular, then the motion may be described as intermittent. With this temporal class, the
movement occurs off and on, the frequency of which is unpredictable and episodic. If the
frequency of movement is predicrable and regular, then the phenomenon may be said to
exhibit cyclical patterns or fluCtuations in its periodicity. Fluctuations in time that are
based on the changes of the day and season are regarded as cyclical temporal classes in this
movement classification.

Using these three different descriptors of time (i.e., continuous, intermittent, and
cyclical), in conjunction with the four spatial descriptors of geographic phenomena (i.e.,
point, linear, areal, and volumetric), the following classification presents twelve unique

categories of geographic movement. Examination of each movement class within this

74

organizational matrix includes detailed discussion of the dimensional characteristics
associated with each category and highlights of particular geographic phenomenon
exemplifying each movement class. For the purposes of this dissertation, variations in
either or bOth the spatial and temporal characterisrics can be accounted for by using four
simple conventions about scale: micro, meso, macro, and mega. Each of these scale levels
are described using “basic hierarchical labels” for the time and space dimensions, such as
‘seconds,’ ‘hours,’ and ‘years’ and ‘neighborhood,’ ‘state,’ and ‘country,’ respectively
(Lloyd, Patton, and Cammack 1996, 182). Geographically, scales of interpretation may
range from seconds and/or feet upwards to include phenomena occurring over millions and
billions of years and/ or involving the entire planet within its solar system. Microscale
analyses use the lowest order of administrative unit and range in time from less than a
second up to a day or so and in space from the size of a room to that of a neighborhood.
The microscale is typically the domain of interest for studies on individual trip pattern
behavior and on highly localized phenomena such as tornadoes. The mesoscale domain
extends temporally from several days up to a couple of years and spatially from areas the
size of towns up to those of metropolitan areas. Seasonal activities generally fall under the
mesoscale domain, which is the typical scale of interest to the landscape ecologist. The
temporal and spatial extents of the macroscale domain range from tens to hundreds of years
(i.e., decades to centuries) and from intranational to national, respectively. Life—cycle
events are often examined within the scope of macroscale analyses. The large5t scale
domain, the megascale, encompasses phenomena occurring over thousands to millions, even
billions of years (i.e., ‘geologic time’). Megascale space ranges from continental and
hemispheric to global and beyond (universe).

Many geographic phenomena are interpreted at only one particular level of scale that
is typically employed over and over again in subsequent analyses. While the acrual precise
levels of scales may alter from one geographic examination to anorher, relatively speaking,

each phenomenon is usually viewed in a similar spatial and temporal context. For

75

example, diffusion studies almost exclusively use areas as their spatial unit of analysis, and
humans are nor usually interpreted geographically as volumetric entities but as point objects
in space. The perspective adopted in this classification considers the general models of
geographic analyses that have traditionally been applied to geographic phenomena. Using
Weibel’s (1995, 259) suggestion that “ideally, spatial data should be analyzed and viewed
at the level of scale at which the modeled phenomena and processes are meaningful and
best understood,” the classification is dependent on the spatial and temporal scales with
which geographic phenomena are typically examined. For example, while some scientific
inquiry may examine gypsy m0th defoliation in ‘real time’ as a phenomenon occurring at
very small spatial and temporal scales, geographic inquiry generally views the phenomenon
at increasingly larger scales, both temporally and spatially.

However, some geographic phenomena may be interpreted using a multiplicity of
perspectives and can, therefore, be classed into more than one movement category. Such
phenomena are considered multi—type and, while generally uncommon, are possible when
different spatial and/or temporal perspecrives can be used to interpret a particular
geographic subject matter. One such multi—type phenomenon involves travel, such as a
plane trip. Spatially, a plane trip may be viewed as a point activity where only the origin
and deStination are of significance. The trip may also be interpreted as a linear
phenomenon, in which case the route of travel is of importance. Different interpretations
of a plane trip also occur in the temporal dimension. The trip may be viewed as an
intermittent phenomenon for many passengers; the event is an unusual ‘or highly irregular
occurrence. The same trip using a different perspective may be considered a cyclical
phenomenon; travel on the route is performed in a highly regular fashion during similar
times of day and on similar days of the week. Yet other perspectives might focus on the
actual trip itself and would consider the trip to be a continuous phenomenon from
beginning to end. Any combination of each of these spatial and temporal perspectives of a

plane trip will yield fundamentally different interpretations of the event. These varying

76

viewpoints in analysis are often distinguished by the names associated with the geographic
phenomenon. For example, analysis of the individuals’ changes in spatial locations is
often referred to recurrent human mobility (e.g., movements of the pilots) and/or irreguhzr
human mobility (e.g., movements of the passengers). Spatial analyses of the periodicity of
the path or travel route of the trip are often interpreted as mass transit, regular or

specialized route services, as well as some forms of trade and commerce.

As the temporal and/or spatial scales of geographical analysis move from micro
scales up through mega scales, the questions posed and views of interpretation may shift to
yield the most appropriate perspecrive of the subject matter under investigation. Often
times, such shifts of scale of analysis will lead to major shifts in the line of inquiry
whereby different branches of science will examine the same subject matter: Studying
tornadoes using subatomic levels will reveal entirely different information about the
subject matter and would be investigated not by geographers, but perhaps by particle
physiciSts. If interesr in a plane trip concentrated on the changing coordinates in three-
dimensional air space the plane travels through, the analysis may be performed by air
traffic control. If interest in the movement of the plane focused on the plane itself (e.g.,
the internal combustion of the engine), the phenomenon may be studied by mechanical
engineers. Studying the individual movements of the people as they travel through air
space might be analyzed by a psychologist or behaviorist, while a physiologist might
analyze the effects of air pressure on the movement of cells within the human body.

As demonstrated with tornadoes and plane travel, virtually any activity can be
dissected and/or expanded, extending from the microscopic to the all—encompassing
universe and beyond, using a wide array of temporal and spatial perspecrives. For the
purposes of this dissertation, only those spatially dynamic phenomena of general interest
to the geographer are examined and classed as geographic movement. Geographers
typically interpret, analyze, and describe dimensionally complex phenomena using

systematic spatial and temporal contexts. Gross deviations from these traditional levels

77

of analyses will usually yield phenomena that are no longer of interest to the geographer
but are to Other scientists. While most geographic phenomena can be interpreted as a
particular movement category, multi—type phenomena such as trips, which may be viewed
with different spatial and temporal perspectives, are exceptions and therefore can be

classed into two or more different movement categories.

 

 

3..

 

 

Class 1 Continuous Point Movements

Continuous point movements occur when an object or phenomenon, as a point in
space, moves uninterrupted through time. The spatial form this type of movement takes is
point—to—point motion. With this category of movement, only the individual points or
locales in space (i.e., its origins and destinations) are of significance as the phenomenon
travels continuously through time. Routes or paths traveled are n0t so important as where
the object is in space as it moves.

Global Positioning System (GPS) technology, which allows highly precise
identification of an object’s geographic location on the earth’s surface, is often linked to
continuous point movements in geographic space. Vehicle navigation and the monitored
individual movements of a vehicle, animal, or person are activities that utilize GPS to
record an object’s different spatial locations as it moves across an area and are considered
continuous point movements (Figure 4.2). With vehicle navigation, a driver can view the
vehicle’s location, as well as associated geographic features and landmarks such as
highways and buildings, on a map display screen mounted on the dashboard. Such
geographic information serves as a navigational guide in reaching the intended destination.
In—vehicle navigation sysrems show real-time movement of the vehicle; the position of the
vehicle while it is moving relative to the area around it is displayed with virtually no

delay. Another application of GPS technology in the transportation industry is the

Spatial Scale (square miles)

78

 

 

megascale

mesoscale - macroscale

10l6 —

universe

1014 __

—u—-

_.,...

1010___

world

.0..- 0‘.”-.-.-... 108 #—
hemisphere

con n'nem

!

l

l

l

l

U

E
y—a
O

O\

l

1

country

state I

104 —4.—

l
l

 

 

microscale

county
102 --
dry

monitored individual
movements
Village ”(hiCk navigation I",
l 00 -' '— "'1'
neighborhood y

l l l l l

 

l l l l I l
10'8 10'6 10“

second minute hour

i l l I l
19'2
day l

.

month year century
I
I

 

 

1
l

—o—
._
—
.—
.—

104 106
10.000 yrs. 1 million yrs.

1 billion ”1.:

 

------------------- microscale MeSOSCdIé"MdCYOSCdle "

---------------- megascale

Temporal Scale (years)

Figure 4.2 Scale Paradigm of Continuous Point Movements.

79

monitored individual movements of a vehicle where a vehicle’s location is observed and
sometimes directed from a central dispatch location (Jordan 1996). Emergency vehicle
services and taxi companies, such as Yellow Cab in Denver, are beginning to integrate GPS
technology not only as a way of optimizing dispatch by automatically assigning each call
to the nearest available vehicle based on pickup and destination locations, but also as a way
of monitoring the movements of its fleet drivers by displaying real-time locations of the
vehicles (Business Geographies 1996). Both vehicle navigation and monitoring activities are
microscale geographic movements which take place in minutes and hours within the
confines of a town or city.

The monitored individual movements of wildlife and people are also considered
continuous point movements and are possible with GPS as well as other wireless
communications such as cellular telephone, radar, and radio waves. As with the use of
radio collars, GPS technology allows the individual movements of endangered or
Otherwise tagged wildlife to be closely monitored for such purposes as habitat
management. Law enforcement officials, who also use portable tracking devices that
interface with global positioning system satellites and a cellular phone network,
continuously oversee the movements and whereabouts of offenders, stalkers, and other
criminals (Business Geographies 1996). At-risk medical patients, including Alzheimer’s
sufferers, can also be observed with wireless receivers. Monitored individual movements
of persons and wildlife generally occurs in the micro to meso scale frames, both
temporally and spatially. As GPS and orher wireless communication technology allow
for more advanced and prevalent monitoring, navigating, and automatic location
capabilities in society, a greater understanding of how objects and individuals move

continuously through space may be gained.

80

  

“ Class 2 Cyclical Point Movements

Cyclical point movements are those in which a phenomenon moves from point to
point with a cyclical or predictable frequency. With this category of geographic
movement, the importance of the phenomenon is the identification of the linkages or
interaction between two or more points or nodes in space (i.e., locations). Many wildlife
migrations, including those exhibited by salmon and Canadian geese, are regular seasonal
movements and are considered cyclical point phenomena at meso to macro scales (Figure
4.3). One such phenomenon, the annual swallow migration to the Mission San Juan
Capistrano, is a good example of how highly cyclical wildlife migratory movements can
be. Every year around March 19 thousands of swallows migrate to the same exact location
in southern California. The individual paths of the birds’ migration are not the primary
focus of examination; rather the unusual temporal aspect and spatial linkages of the
movement are the key to understanding this unique activity. An important element in the
explanation is the spatial and temporal precision of the repetition of the event, nor the
individual events themselves.

Mosr types of recurrent human mobility are also examples of cyclical point
phenomena and include the movements associated with personal action spaces, urban
homelessness, itinerant workers, journey—to—work, multiple residences, and human
migration patterns. Borrowing from Gould and Prothero’s (1973) distinction between
mobility, migration, and circulation as they relate to human movements in tropical
Africa, ‘mobility’ is understood to broadly include all population movements.
Traditionally migration, as a population movement, has been defined as a permanent
change in the usual place of residence between two points in time (Lowe and Moryadas
1975). The phenomenon of human migration has usually been conceptualized in terms of

disrance, gender, age, marital status, unemployment, and education using “standard tools”

Spatial Scale (square miles)

81

 

 

megascale

mesoscale - ma

1010__

world

.---..-._.-.,___-_ 108

hemisphere

con linem

s
i
l
I
i
O
O\

——

human migration
“P patterns

__ wihllife
migration
‘7 multiple

residences

itinerant

 

 

microscale

 

county

city
village

10°
neighborhood

workers
journey—toowork,

.._ personal action Zaces,
urban home 5

l I l l l l I l 1 l J l l l

 

r F I I I I I I I I I I I I I I
10'8 10'6 10 10'2 19° 102 10‘1 106 108 10‘°

second minute hour day j month yt’ar century 10.000 yrs. 1 million yrs. 1 billion yrs.

qt—

 

 

 

; ------------------ microscale ----------------- mesoscaleE -- macroscale -- -------------------- megascale ----------------- -‘
Temporal Scale (years)
Figure 4.3 Scale Paradigm of Cyclical Point Movements.

82

such as ‘life-cycle’ theory, which implies a predetermined set of stages through which all
individuals pass (Vandsemb 1995; Mitchneck and Plane 1995, 18). However, it has
become increasingly evident that many people are exhibiting increasingly diverse histories
such as schooling, cohabitation and living arrangements, marriage, child bearing,
separation, divorce, remarriage, employment and career changes, etc. (McHugh, Hogan,
and Happel 1995); thus, the notion of ‘life—course’ has replaced ‘life—cycle’ in recognition
that nOt all individuals follow the same normative sequence of stages. Individuals are
living and spending time in more than one place, moving between locales on a recurrent
basis.

Therefore, migration is increasingly being defined as a relatively permanent change
of usual residence. This definition seeks to differentiate migration from Other forms of
less permanent mobility that do not involve a change in usual residence, such as commuting
and local trips, temporary employment elsewhere, and leisure travel (McHugh, Hogan and
Happel 1995). Lowe and Moryadas (1975) operationalize this temporal component Of
migration by classifying the movement of persons as being temporary, transient, or
permanent. For the purposes of this research, human mobility is considered to include
relatively permanent or less permanent human movements. Relatively permanent
movements involve one—way moves, where there is no intention of returning to the home
base and are usually quite infrequent in nature. Individual residential migrations are good
examples of these types of irregular movements and are discussed further as intermittent
point movements. Less permanent mobility, which includes all temporary and transient
moves where there is no change in usual residence, includes many types of recurrent human
mobility which are considered cyclical point phenomena.

Roseman (1971) defines recurrent mobility as reciprocal movement, a
morphological classification of human movement. Reciprocal movements are usually
considered circulatory; individuals begin at the home or dwelling, proceed to one or more

specific locations, and eventually return back to the home. Often, these repetitive human

83

movements are subdivided according to the length of their cycles: daily, weekly,
monthly, periodic, and seasonal (Gould and Prothero 1973; Roseman 1971). Microscale
frequencies such as daily, weekly, or monthly movements of individuals are considered
temporary and include consumers shopping for goods and services. Often this highly
cyclical human movement is explained in a geographical context with the concept of
‘action space.’ Lowe and Moryadas (1975, 139) define action space as “the area which
contains the majority of desrinations of a particular individual... [and] is a subspace within
the mental map.” Personal action spaces occur at micro spatial and temporal scales,
especially in automobile-oriented societies such as the United States, because most of the
movement involves personal trips that can be satisfied in short distances and because of the
need to return daily to home base. Numerous impediments are major determinants of the
physical dimensions of action spaces and may include capability, coupling, authority,
time, and budgetary constraints (Adams 1995). Individuals presumably try to maximize
their return on the expenditure of time by adjusting their movement patterns and action
spaces appropriately.

The mobility patterns of the urban homeless, namely where and how often homeless
individuals travel within a city, also exhibit cyclical patterns of movement at the micro
levels in bOth space and time. In studying intraurban mobility among the urban homeless
in the Skid Row area of Los Angeles, Wolch, Rahimian, and Koegel (1993) assert that
daily mobility patterns of homeless persons are subject to severe time—space constraints of
almost every type and that movement is linked to degree of coping success. Coping, in
turn, is tied to homeless social connections, the availability of urban resources, and broader
contextual factors that shape access to Welfare benefits, jobs, housing, and Other critical
human services. Wolch, Rahimian, and Koegel (1993, 163) suggest that homeless people
are extremely mobile within a portion of an urban area and that their daily movements are
highly cyclical: “Thus, most people do similar things every day and meet the same

people at various juncrures throughout the day. They often start and end their day in the

84

same spot.” The authors also suggest cyclical behavior of the homeless in terms of
intraregional mobility, where many homeless persons have a weekly, biweekly, or monthly
routine of longer trips within the greater metropolitan region.

Another type of cyclical human mobility occurring on a daily basis is the journey—
to—work. Geographers have long studied workers’ mobility patterns in terms of the
influence of variables such as gender, household responsibilities, central city versus suburban
residences, and travel mode on trip lengths, measured either by distance or travel time
(Iohnsron-Anumonwo 1992). The time«space structure of this recurrent human mobility
phenomenon is conceptualized at a micro scale, bOth temporally and spatially. Occurring
at slightly larger scales are the cyclical movements of many itinerant workers whose jobs
take them from place to place. Workers such as traveling salespersons, agricultural
laborers, professional athletes, miners, and construction and oilfield workers travel with
varying periodicities and usually return repeatedly to a ‘home’ base (McHugh, Hogan, and
Happel 1995). For example, pastoralists move seasonally; their daily rhythm is similar
to that of nomads, but whose movements have an overall discernible, annual regularity in
space and time (Gould and Prothero 1973). In the United States, agricultural labor
moves seasonally from the Mexican border northward along the West and East Coasts as
well as the Great Plains, following planting and harvesting demands.

Once initiated, cyclical mobility may persist for spells ranging from weeks and
months to many years (McHugh, Hogan, and Happel 1995). Recurrent circulatory human
movements of such large temporal scales (meso to macro) are those of individuals with
multiple residences. Described as ‘transient movers’ (Lowe and Moryadas 1975),
individuals with more than one residence often exhibit a very orderly pattern of movement
between two points in space that can range from weekly or monthly circulation between
residences during a life-course phase (e.g., children shuttled between divorced or separated
parents, students who return home during semester breaks and summers, or military

reassignments) (McHugh, Hogan, and Happel 1995), to weekly circulation between

 

 

 

85

residences across the life—span (e.g., family second homes) (Behr and Gober 1982), to what
Roseman (1992, 38) describes as biannual treks between summer and winter residences
over a lifetime (i.e., elderly “snowbirds”). ‘Snowbirds,’ who are often stereotyped as
healthy, active couples pursuing leisure—related lifestyles in their early retirement years and
living a sort of “foorloose lifesryle,” are, as McHugh and Mings (1996, 539) suggest,
highly cyclical transient movers: “Once established, cycling between summer and winter
places takes on a ‘matter—Of—fact’ quality, it becomes part of an annual space—time routine.”

A final form of human mobility that is considered a cyclical point phenomenon are
some human migration patterns. Using a macroscale reference for both the spatial and
temporal dimensions, some human migration patterns when observed over lengthy time
periods and across large expanses of space are quite cyclical in nature. In studying long—
term global migration patterns, Berry (1993, 389) suggests that transnational urbanward
immigration flows have fluctuated with a “cyclical, pulsating... long-wave rhythmicity”
that mirrors the movement of successive techno—economic systems from innovation to
market saturation.

Regardless of the duration or spatial extent of the movement, several forms of human
mobility can be viewed as cyclical, point—to—point movements and include large scale
migration patterns, as well as individual movements associated with personal action
spaces, homelessness, journey-tO—work, itinerant workers, and multiple residences.
Recurrent cyclical mobility involves oscillatory movements which vary in tempo and
duration and range from frequent back—and—forth movement over a brief period
(temporary job assignment) to biannual treks between summer and winter residences over a
life—course phase (snowbirds) to circulation between residences across the lifespan (family
second homes) (McHugh, Hogan, and Happel 1995). McHugh and Mings (1996, 538)
describe such tempos and rhythms associated with geographical mobility as “place ballet
on a grand scale” where “elegance and grace” exist in the “recurring cycle of separation and

return.” With this movement category, the geographic importance of the phenomenon’s

 

 

 

86

movement are the beginning and ending locations of two or more points in space, the

occurrence of which is cyclical over time.

 

Class 3 Intermittent Point Movements

The third type of geographic point movement is intermittent in its temporal
character. Wth this category of movement, an Object or phenomenon moves infrequently
from location to location, without regular intervals. As with all point—to—point
movements, the spatial emphasis is on the from—to linkages of the two or more locations in
space. Examples of this movement class include forms of irregular human mobility;
including movements associated with residential migration, immigration, refugee and
nomadic behavior, leisure travel, and personal action space, as well as volcanic and seismic
activity (Figure 4.4). Residential migration, defined as “a permanent or semi—permanent
change of residence, usually across some type of administrative boundary” (Wood 1994,
607), can be described spatially as a point—to—point move: “the geographic area that sends
major migration flows to a given place (i.e., a city, metropolitan area, or county), or the
area that receives major flows from a place... in effect, that place is treated as a point in
space” (Jakle, Brunn, and Roseman 1976, 165-166). Whereas many forms of human
mobility are repeated frequently, often at regular intervals (see discussion of class 2,
cyclical point movements), individual residential migrations are less frequent and tend to
be related to such episodic events as graduation, marriage, and retirement. Such irregular
human displacements, where the center of gravity of the weekly movement cycle, the
home, moves to a new location, are essentially one-way and are considered relatively
permanent given that further movement is likely in the future, but neither the time nor the
direction of such movements are known.

Understanding the irregular patterns of residential migrations of individuals requires

Spatial Scale (square miles)

87

 

 

megascale

mesoscale - macroscale

1016_

universe

)—

 

 

microscale

 

county

102 --

dry
village

10° —
neighborhood

1 l

l

l
I ,'

i I

y" refugees, y”

5' immigrants ,t'

residential
migration
i

l
I

kisure
travel

1" nomads,
personal ,5
action .
spaces I:

; volcanic and
" seismic activity

 

 

I I

10°

second

I
10'6

minute

I
10

A 10-2

hour day; month year

------------------- microscale mesoscale --macroscale --

10‘1

century

 

Temporal Scale (years)

106

10,000 yrs. 1 million yrs.

-------------------- megascale

108 1010
1 billion yrs.

 

Figure 4.4 Scale Paradigm of Intermittent Point Movements.

88

a meso to macro temporal scale and a spatial scale varying from micro to mega; an
individual ’s residential moves over a life time may be numerous yet sporadic and can
occur within a same town or city or can extend to the far corners of the world.
Understanding the irregular patterns of residential migrations of aggregates of individuals
is equally as complex. The current interstate migration system in the United States
oscillates between relatively brief periods of abrupt sysrematic change in established
migration patterns followed by longer periods during which new migration equilibria are
established. For example, migration patterns in the early 19803 were different from those
in the mid 19805, which in turn bore little resemblance to migration in the early 19905.
Manson and Groop (1996, 156, 165) describe this phenomenon as “temporally and
spatially transitory” and argue that “it is neither feasible nor necessary to impose a spatial
generalization on such a dynamic system.”

The behavior of immigrants, refiigees and nomads also may exhibit sporadic and
irregular movements (Gould and Prothero 1973). Many quantitative studies attempt to
explain the socioeconomic causes and consequences of such migrations, especially at meso
to macro scales, by describing the sort of societies that are ejecting or attracting certain
types of migrants. Political considerations bOth encourage and inhibit the movement of
people; mosr countries have stringent regulations about who may enter (Lowe and
Moryadas 1975). In examining the undocumented (illegal) Mexican migration to the
United States, immigration has fluctuated with economic cycles in the States during the
first four decades of the twentieth century, going from stages of acrive recruitment to
periods of forced repatriation. In recent times, step migration has come to play a major
role in the flow of undocumented migrants traveling great distances to reach their
destinations in areas in the southwest of the United States. Step migration occurs through
intermediary locations called “redistributive cities” (e.g., Nogales, San Luis, Tijuana)
that serve as “jumping-off—points,” tying internal and international migration (Harner

1995.399)

89

Migrant population patterns are also affected during political and economic tensions,
especially during armed conflict (Mitchneck and Plane 1995). Refugees, individuals who
are forced by a well—founded fear of persecution to migrate involuntarily across an
international boundary (Bascom 1993). may migrate many times, for varied durations,
and across numerous territorial divisions (Wood 1994). Such is the case with Ethiopia,
where the largest and longest forced mass movement from one country in pOStcolonial
Africa has taken place. Beginning in 1967 when 30,000 Ethiopians fled from the northern
province of Eritea into eastern Sudan, the combined “Ethiopian Diaspora” now includes
more than one million people, the majority of whom are sheltered in the neighboring
countries of Somalia, Djibouti, and Sudan (Bascom 1993, 320). Interpreting the
migratory moves Of nomads requires micro and meso scale references; migration is not
dependent on a cyclical or predetermined frequency, but rather moves occur according to
the day—to—day availability of pasture and water for their animals. Where and when their
next move may not be precisely known.

Other forms of irregular human mobility include temporary movements associated
with periodic and intermittent activities outside the home. Such movements can range
substantially in scale, bOth spatially and temporally. Vacations and visits to family or
friends are all leisure travel activities associated with irregular human mobility and can
include hourly visits to locations within one’s neighborhood to vacations lasting several
weeks or more in areas as far away as across the world. Some types Of individual
movements associated with one’s personal action space may also be considered intermittent
point movements. Although many individual activities at micro spatial and temporal
scales are highly regimented, as discussed in class 2, cyclical point movements,
periodically one may travel to different points within one’s action space for the purpose of
satisfying some infrequent need for goods or services. Human movements are purposive to
a degree; an individual has either knowledge or the willingness and ability to search space

to find out the distribution of available opportunities for various purposes. Such is the case

90

with intermittent human movements associated with patient health care access. Often in
rural outlying areas, people will travel great disrances and cross administrative boundaries
to receive specialized health care. The dominant characteristics of this irregular human
movement is that it occurs between rural and urban counties and on an infrequent basis in
mesoscale time (Lowe and Moryadas 1975).

Other examples of intermittent point movement include volcanic and seismic activity.
Whereas a particular volcanic eruption or earthquake is a continuous volumetric movement
and eruptions from the same volcano or earthquakes occurring along the same fault line are
be described as volumetric movement (see discussions of volumetric movement classes 10
and 12, continuous and intermittent, respectively), volcanism, a general term for volcano
building and related forms of extrusive igneous activity, and seismic activity are both
phenomena that, when examined at very large spatial and temporal scales, are considered
to be intermittent point movements. Viewing the entire world as one giant activity space
over thousands to millions of years, individual eruptions and earthquakes, when observed in
relation to one another, can be described as ‘hot spots’ jumping from one point to another
intermittently. Nearly seventy—five percent of the world’s approximate 540 historically
active volcanoes lie within the “Ring of Fire,” a zone running along the western coast of the
Americas from Chile to Alaska, down the easrern coast of Asia from Siberia to New
Guinea, and continuing to New Zealand (Famighetti 1996, 588). Here, active volcanoes
can erupt repeatedly over a period of thousands of years, the frequency and occurrence of
which are unpredictable. Earthquakes occur intermittently and at scattered but relatively
similar locations throughout and beyond the circumePacific belt. During the course of the
last century or more, over 120 major earthquakes have occurred throughout the world, the

frequency and relative location varying significantly (Famighetti 1996).

91

 

 

 

 

N
2 Class 4 Continuous Linear Movements

Continuous linear movements are those associated with an entity that is continuously
changing its spatial location along a path. The geometric shape of the movement as it
travels through space and time is an unbroken line; the linear form representing some
pathway, channel, or route. The temporal aspect is uninterrupted, representing movement
that is continuous, from beginning to end. Of general interesr to the geographer, this linear
kind of movement which occurs among specific origins and desrinations is often described
as the ‘flow’ between locales. A spatial flow is “an observable phenomenon which (an be
measured directly as the specific volume passing a point during a particular time interval”
(Lowe and Moryadas 1975, 158—159). Flows in human and physical sysrems are a focus of
geographic inquiry because “they involve movement across space that connecr origins and
destinations and because they redistribute water, sediment, energy, population, products,
and economic activity” (Graf and Gober 1992, 239). What is significant about this class
of movement is the movement of the path itself; movements occurring within the path or
track are examined as Other spatial forms that are more important to the understanding of
the phenomenon then its linear motion (e.g., with a flood, the important issue to note is not
the expanding borders of the river channel but the depth or volume of the channel).
Geographic phenomena classified in this movement category travel continuously across
space in a linear fashion and include forms of electronic communication, jet streams and
water channels, hurricanes, tornadoes, and the revolution of the earth about the sun (Figure
4.5).

The elecrronic transmission of information, hereafter referred to as electronic
communication, is considered to be a continuous linear movement. Utilizing wire systems
such as telephones, computers, and televisions, elecrronic communication involves the

spatial interaction between two or more places conducted through highly concentrated

Spatial Scale (square miles)

92

 

 

megascale

mesoscale macroscale

1016‘

universe

1014_

101°-

world
-------. _-- -...--- 108
banisphere
continent
-------..----..-. 106
country

“21C

104

1012_

revolution of

‘ the earth

—1—

—

.1

jet streams
a hurricane

electronic
communication

 

 

microscale

 

county
102
city
village
10°
na'ghborhood

room

i water
- channeh

a tornado

l l

1

 

qu—
—.
fl
—1
—

I I
10'8

second

10“1

hour

10°

minute

102 19°
day: month yt’nr century

 

‘—
u-O-

I I I
10‘1 106
l0.000 yrs. I million yrs.

I
10° 101°
1 billion yrs.

 

 

. s i
................... mwmmle mesoscale «macroscale -'

-------------------- megascale

Temporal Scale (years)

Figure 4.5 Scale Paradigm of Continuous Linear Movements.

93

channels, especially with regard to information flows (Mitchelson and Wheeler 1994).
Once a signal is sent, movement of information is continuous, traveling along an electronic
pathway until the connection is broken. Adapting from the Shannon/Weaver mathematical
theory of communication, Jakle, Brunn, and Roseman (1976, 121—122) describe electronic
information transmittal as “bits of information emanating from a source, sent through
some transmitting mechanism such as a voice, printing press, or radio, then through a
channel of some sort, and finally on to a receiver at a given deStination.” The exchange Of
information using electronic wire communication systems generally occurs within seconds
and minutes (micro time frames), but varies considerably in spatial scale, ranging from
local to world—wide connections and links.

AnOther example of a constant linear phenomenon is the revolution of the earth about
the sun. Simultaneously with rOtation on its axis, the earth is in continuous motion in an
orbit around the sun. The geographic view typically used in the description and
explanation of the revolution of the earth describes earth as one of many objects or points
in space that move constantly and steadily in elliptical patterns with respect to the sun.
Spatially, this phenomenon is interpreted on a megascale, where the spatial extent of the
movement occurs within a galaxy. The earth completes its orbit in one year and falls
within the mesoscale temporal domain. This period of the earth’s revolution is important
as it sets the timing for climatic seasons.

jet streams and water channels are also geographic phenomena that move in a
continuous linear fashion. Jet streams are narrow bands of high velocity winds in the upper
level troposphere that follow the wave path near the tropopause at elevations of eight to
fifteen kilometers (Critchfield 1983). This phenomenon consisrs of pulselike movements
of air following broadly curving tracks; a cross section of the jet resembles a Stream of
water moving through a hose (Strahler and Strahler 1984). As part of the earth’s dynamic
atmosphere (see class 10, continuous volumetric movements), jet streams are continuous

phenomena. Frequently shown on television and Other animated weather displays as a

94

dynamic line snaking its way across the earth, the concept of a jet Stream is best
understood using a macro spatial scale and a meso temporal scale, with which areas such as
the continental United States are viewed over the course of several days. Water channels
such as streams, creeks, and rivers are also considered continuous linear phenomena.
Analogous to jet Streams, water channels are comprised of continuous volumetric
movement but whose changing boundaries are often significant. Streams and rivers are
constantly changing their path (i.e., ‘meandering’) and are best interpreted at micro to
meso spatial scales over the course of several months to years.

Additional examples of geographic phenomena included in this movement category
are individual occurrences of some types of severe cyclonic storms such as a hurricane or a
tornado. Traveling cyclones, atmospheric circulations of winds rotating counterclockwise
in the northern hemisphere and clockwise in the southern hemisphere (Famighetti 1996), are
centers of low atmospheric pressure that produce cloudiness and precipitation due to the
convergence of air masses toward the cyclone centers accompanied by lifting of air and
subsequent adiabatic cooling (Critchfield 1983). Dependent for their development on the
coming together of large air masses of contrasting properties, cyclones are typically mild
in intensity, passing with little more than a period of cloud cover and light rain or snow.
However, when pressure gradients are strong and accompanied winds teach gale force, the
disturbance is called a cyclonic storm. As an intense cyclonic vortex Of enormously
powerful winds, or as a tropical cyclone of tropical and subtropical ocean areas, a tornado
and a hurricane, respectively, are best understood geographically as continuous linear
movements.

When these two types of severe cyclonic storms occur, they are part of the earth’s
atmosphere and, meteorologically speaking, are volumetrically dynamic phenomena at
very micro scales (i.e., tremendous volumetric rotation Of the air at or near the center of
these cyclones). At larger micro spatial and temporal scales, individual occurrences of

each of these phenomena could also be considered objects or points in space given the exact

 

 

95

locations these entities are coming in contact with the landscape changing with each new
time frame. However, geographic focus on these phenomena generally employs larger
frameworks. Geographers typically evaluate the relationship between a particular storm
and the landscape with which it is interacting (i.e., the path a disturbance is traveling) for
the major purpose Of improving forecasting and emergency management operations.

Using moderate scale frames, a hurricane falls within the meso temporal scale as,
from creation to dissipation, it can travel for several weeks. Spatially, the movement of a
hurricane is best understood using meso to macro scale references, where storms can cover
tens of degrees of bOth latitude and longitude (Strahler and Strahler 1984). Once formed,
a hurricane typically advances fifteen to thirty kilometers per hour westward through the
trade—wind belt, at which point it may then curve north and northeast where it penetrates
well into the belt of westerlies (Strahler and Strahler 1984). The area of hurricane force
winds (seventy—three miles per hour or higher) takes the form Of a circle or an oval,
sometimes as wide as 300 miles in diameter (Famighetti 1996). The National Weather
Service’s National Hurricane Center (NHC) tracks, among other characteristics, a storm’s
movement or travel path. By making specific forecasts of landfall points and times, the
NHC can aid emergency management officials with advisories and evacuation orders
(Baker 1995).

A tornado’s movement needs micro temporal and spatial scales to be examined.
Tornado paths can vary in length from a few feet to nearly 300 miles (average is five
miles) and in diameter from a few feet to more than three to five kilometers (average is
660 feet) (Famighetti 1996; Critchfield 1983). As a tornado moves across the landscape,
its rate of travel at the ground averages thirty miles per hour but may reach fifty knots or
more and may remain relatively stationary for short periods. Some contact the ground
and then lift, only to strike again at a distance of several kilometers; still Others may
travel only a few meters before rising and dissipating (Critchfield 1983). Regardless of

its actions, geographically a tornado, as with a hurricane, is examined chiefly for its

96

location and subsequent travel path. Locating the important regions of cyclogenesis and
the general tracks taken by these cyclones as a way of understanding the spatial and
temporal trends of tornadoes and hurricanes is also of geographic concern (Raphael and

Mills 1996) and is discussed next in class 5, cyclical linear movements.

  

“ Class 5 Cyclical Linear Movements

Cyclical linear movements are those in which a phenomenon repeatedly moves across
the landscape in a linear or path—like fashion, and does so with a predictable frequency.
The key to understanding phenomena classified as cyclical linear activities is to visualize
a subject or object moving across the surface of the earth along pathways and routes in a
systematic and regular pattern. Hurricane and tornadic activity, as well as acrivities that
utilize transportation networks of various scales for the scheduled movement of people,
goods, and services such as mass transit and regular route services, trade, and commerce, are
examples of geographic phenomena that move along linear paths across the landscape in
regular frequencies (Figure 4.6).

Viewed at meso to macro temporal and spatial scales, hurricane and tornadic
activity are considered cyclical linear phenomena as both exhibit fairly predictable
temporal, as well as spatial, patterns of behavior. Occurrences of both types of cyclonic
Storms are cyclical phenomena in that they are seasonal activities; hurricane season
generally extends from June to September in a given year while tornadoes are most
frequent in spring and summer months. Equally regular are the spatial characteristics of
these storms. Hurricanes only occur in tropical and subtropical zones on earth, form over
warm water, and travel in a westerly then northwesterly fashion, while the tornado is a
typically American phenomenon occurring primarily in the southern High Plains of the

United States (Strahler and Strahler 1984). In studying selected temporal and spatial

Spatial Scale (square miles)

97

 

 

megascale

mesoscale - macroscale

1016_

universe

 

 

microscale

 

county

city

village

10° ~
ndghborhood

hurricane activity

regular trade and commerce

f tornadic activity

mass transit
9

reguhcr

route services

 

 

1 0'3

second

10'6

minute

10“ 10'2

10°

hour dayj month ydar

------------------- microscale mesoscale “771467056415 "

-i'-
_.

century

10,000 yrs. 1 million yrs.

 

Temporal Scale (years)

Figure 4.6 Scale Paradigm of Cyclical Linear Movements.

-------------------- megascale

108 10‘°
1 billion yrs.

 

98

characterisrics of Michigan tornadoes over a fifty year period, Stolle (1987) concludes
that tornadoes in the region do nOt occur at random times or location, but that this
phenomenon generally occurs during late afternoon to early evening hours most
predominately in the spring months of April, May, and June and moves fairly consistently
in a north to northeast direction.

Movement associated with most forms of transportation can also be classified in this
category. Transportation involves the physical movement of people, goods, or services,
between two or more places and can take place with a series of different transportation
modes (e.g., auto, train, bus, airplane, boat, etc.) in a network. In adapting graph theory to
transportation, essentially the topological structure of a route network is a set of
geographic locations (nodes or points) interconnected in a system by a number of routes
(links) (Haggett and Chorley 1969). Movement originates and ends at nodes, and flows
occur on route networks (Lowe and Moryadas 1975). How well a node such as an urban
center or airport is connected to Other nodes is described by connectivity (Lee, Chen, and
Shaw 1994). Connectivity measures “linkage intensity;” the higher the number of routes or
paths connecting two or more points, the greater the linkage intensity and, therefore, the
closer the network comes to a maximally connected structure (Ivy 1995, 283). Network
route structures (i.e., the way that links are connected with each other) (Lee, Chen, and
Shaw 1994) acr as the arterial and nervous system of regional organization (Haggett 1979,
490) and are made up of many links of various size which are hierarchical. This branching
Structure, analogous to river systems, contains a few heavily used channels and many highly
used feeders, or tributary channels (Lowe and Moryadas 1975; Haggett 1979).

The evolution of transportation networks occurs not only at the macro spatial scale of
nations and large regions, but also at the micro scale of urban areas. Perceived and
analyzed as linear forms on the human landscape, transportation is a phenomenon in which
people and cargo are carried from one location to Others along designated routes or links.

These linkages or paths are typically highway, railway, airway, and/or waterway networks

99

that weave across the landscape. Major transportation routes are sometimes referred to as
‘arteries,’ where a transportation system keeps a complex space—economy fimctioning
(Lowe and Moryadas 1975). Impediments and disruptions to established networks such as
accidents, construction, and inclement weather conditions can have paralytic effects and
may alter a chosen route. Route reallocation is perhaps the easiest to accomplish with air
transportation networks. Unlike largely fixed overland infrastructure links such as
railroads, canals, and highways, airline routes can, with governmental approval, be added
or deleted at will (Ivy 1995). Despite such disruptions to a particular transportation
activity, the form of the transporting of people, goods and services as a linear movement
remains unchanged.

Mass transit, as a public transportation phenomenon, is a cyclical geographic
movement largely because the activity occurs rather predictably; the times the transport
begins and ends is similar and predetermined with most occurrences. Each mode Of
transit has fixed schedules and carriers within each transportation network travel in
sysrematic and regular frequencies, which are usually unique in terms of their network
struCture. For example, the flight frequencies of various routes in an airline’s network
depend on the demand and desired service levels, the types of aircraft used, and the degree
of competition from other airlines between different pairs of cities (Lee, Chen, and Shaw
1994). The duration of a transit varies among carriers in this category, but is virtually
identical for each frequency in a given mode within a transportation route. While delays
and disruptions in service are commonplace to the frequent traveler, the movements
associated with mass transit are generally cyclical events. Mass transit can occur over a
wide range of temporal and spatial scales; movements of city buses across part of an urban
landscape are measured in minutes, air and rail transit occurs within and across states and
countries and is measured in minutes, hours, and days, and ship travel, which may occur
over days and weeks, can cover areas as large as the world.

The argument could be made that at fine temporal scales, frequent stops along a mass

100

transit route represent point—to-point activities and that the movement associated with the
transportation is discontinuous from beginning to end. This may be true at the most
personal scale and is discussed as a form of recurrent human mobility in class 2, cyclical
point movements. Geographically, interesr in mass transit sysrems typically lies in the
links berween the nodes (i.e., the routes); the important considerations of mass transit
routing is a service’s relationship between time and space, and how this relates to efficient
route planning including cost—benefit ratios and spatial allocation models.

AnOther cyclical geographic phenomenon that utilizes transportation networks is
regular route services. Routes are channels along which interaction occurs between any two
nodes or locations. “Interestablishment interactions” such as courier mail, commercial
deliveries, many home care services such as Meals—on-Wheels, and waste disposal and
recycling are examples of regular route services (Lowe and Moryadas 1975, 166). Such
delivery and pick—up services can be seen as cyclical linear activities where information,
products, and freight, such as mail, garbage, and commercial goods, are moved from
production locations to consumption locations, all the while traveling along an established
transportation route. These activities are largely cyclical in that services are typically
rendered in a systematic and orderly time frame, usually incorporating highly frequent,
but regimented schedules such as hourly, daily, and weekly patterns. MOSt regular route
services occur across micro scales, both spatially and temporally. However, information,
goods, and services exchanged on a regular schedule may be transported in more complex,
spatially extensive patterns, such as Federal Express deliveries:

“... packages were picked up and taken by van or truck to a
specified nearby airport, placed on a Federal Express
airplane, flown to the central sorting facility in Memphis,
put back on the appropriate aircraft and flown to their
destinations where they would be delivered the next
morning.”

(Wheeler and Mitchelson 1989, 527)

Regular trade and commerce are also examples of cyclical linear geographic

phenomena that utilize transportation networks. A fundamental cause of commodity

101

movement is the existence of spatially concentrated resources, such as corn in the Midwest,
oil in the Persian Gulf, and coal in Appalachia which, in turn, leads to surplus and deficit
areas. Each region exchanges its surplus for other surplus commodities produced in other
specialized regions (Lowe and Moryadas 1975). Regional transportation infrastructure,
especially air transportation, is the ‘lifeline’ for such economic interchange (Ivy 1995).
The flow of capital and commercial and industrial goods across regional and
international boundaries is best understood using meso to macro spatial scale references.
The exchange of commodities Often occurs in regular patterns measured in days, weeks,
and even months and is primarily a mesoscale temporal phenomenon. Some cyclical
commodities exhibit seasonal periodicities such as the shipment of strawberries in the
summer and the mailing of Christmas cards in MidaDecember (Lowe and Moryadas
1975). However, Other kinds of trade and commerce are sometimes irregularly

administered and are discussed further in class 6, intermittent linear movements.

 

Class 6 Intermittent Linear Movements

A third class of linear geographic movement is intermittent in its temporal
frequency. Recapitulating from the discussion of cyclical linear movements, many
geographic phenomena involving transportation are interpreted easily as linear movements
that are considered purposeful; there is one location of supply and anorher of demand.
Occasional home delivery services such as fast food, flowers, and furniture, as well as
public services such as police, fire, and ambulance, represent specialized route services
(Figure 4.7). With delivery of these kinds of goods and services, supplies move from one
location to anOther based on need which is unpredictable and irregular in micro space and
time. Supply and demand economics, which can fluctuate irregularly, govern when these

movements occur and where their origins and destinations will be. This same principle

Spatial Scale (square miles)

102

 

10l6~

univesse

 

megascale
S
:3
l

................. 106 .4

 

county

dry

village

10° ~
neighborhood

 

microscale
5
N
l

room

exploration

irregular trade and commerce

specialized

route services

1

 

 

second

minute

. ------------------ microscale ----------------- mesoscale .. macroscale --

.—
.—
«>—
._

hour day; month yéar century

 

Temporal Scale (years)

106

10,000 yrs. 1 million yrs.

I
10°

1 billion yrs.

Figure 4.7 Scale Paradigm of Intermittent Linear Movements.

‘—

-------------------- megascale

1010

 

103

extends to some intra- and inter—regional irregular commerce and trade. Using meso to
macro spatial and meso temporal scale references, many commercial goods and
imports/exports can be seen to move from one locale to another via any number and
combination of transportation networks, the frequency of which “do not usually experience
periodicities” (Lowe and Moryadas 1975, 166).

Another example of intermittent linear geographic movement is the historical
phenomenon of exploration. Voyages and expeditions of discovery and exploration,
especially to the New World, were most prevalent during the fifteenth and sixteenth
centuries. Such journeys are best understood with macro spatial and temporal scopes;
voyages across oceans and along the coasts of continents often lasted several months to
years. An exploration can be interpreted as a broken linear path extending from point to
point to point as a ship would dock along the way at key ports for necessary supplies and
reinforcements. Exploration, as an intermittent linear movement, is best understood as a
phenomenon moving along a route or pathway from a point Of origin to specific

desrinations, occasionally Stopping for periods of time.

 

Class 7 Continuous Areal Movements

Continuous areal movements involve phenomena that spread out across the landscape
in a radial fashion and do so continuously through time. Imagine a drop of dye spreading
out or ‘bleeding’ on a piece of fabric; the movement is continuous throughout and when the
spread stops, a contiguous area around the point of origin is affected. Exhibiting similar
movement behavior is diffiision, the geographic spread of a phenomenon. The diflitsion of
human settlement (colonization), innovation, culture, and disease, as well as acculturation
and assimibtion, some forms of regional boundary changes, and fbrest fires are examples of

continuous areal phenomena (Figure 4.8).

Spatial Scale (square miles)

104

 

 

megascale

mesoscale - macroscale

--..----....-...-. 103 -

10164
universe

—1

1014_

—I

1012_

1010__

world
hunisphere

continent

0

E

i

I

I

i
p—s
O

O\
L

7 dsficsion of
innovation

diflicsion o
_ culture f

forest fires

 

 

microscale

 

county

102 -

dry

village

10° -
neighborhood

10'2 —
10 “
house

room

10'6 d

colonization

acculturation and assimilation
4

‘.
‘g
a
.a
-

regional boundary changes
.4

I I l I l l

 

—o—
.0—
_
.—

r f I I I I i
10'8 10'6 10'4 10'2 10° 102 104
second minute hour day i

 

 

Temporal Scale (years)

month yéar century 10.000 yrs. 1 million yrs.

------------------- microscale mesoscale «macroscale megascale

1 billion yrs.

 

Figure 4.8 Scale Paradigm of Continuous Areal Movements.

 

105

Diffusion cannOt be described in a static way (Jakle, Brunn, and Roseman 1976);
rather “it is necessary to mesh space and time simultaneously, since the spread of an
innovation over an area is neither geographically uniform nor temporally instantaneous”
(Lowe and Moryadas 1975, 242-243). Spatial diffusion is an important form of place—to-
place movement or spatial interaction; the term itself suggests movement, interaction,
sprcad, contact, change, and growth (Hudson 1972). A principal concern of social
geography, and geography in general, is the way in which new ideas, new inventions, and
Other traits and practices diffiise from individual to individual and group to group across
the landscape (I akle, Brtmn, and Roseman 1976). Diffusion studies have a rich heritage in
geography and can be traced back to the late 1800’s and the German anthropo—geographer
Ratzel who introduced the fisndamental idea of something starting somewhere and
spreading out like ripples on a pond (Lowe and Moryadas 1975). This conceptualization
of diffusion refers to the ‘wavealike’ nature of the phenomenon: “In a true diffusion
process, the loci of action are themselves expanding in space and time, rather like an
impact wave, out from an origin, as from a rock dropped into the water (Morrill 1968, 2).
This description of “waves flowing continuously over space” characterizes the growth
process of diffusion, whereby items become relatively less concentrated over space but
relatively more concentrated over time (Jakle, Brunn, and Roseman 1976, 144).

Parkes and Thrift (1980, 286) define this principle as “diffusion time,” a term
describing the slow acceptance of an idea, culture, etc. at the beginning of a diffusion
process, which then accelerates until the frequency of acceptance is greatesr (i.e., saturation)
after which the rate of acceptance declines until no further acceptance occurs (Morrill
1968). The adopters themselves influence the way in which innovations and ideas spread
through a population over time and may be classed into one of four categories, each
dependent on decreasing individual rates of adoption, respecrively: innovators, early
adopters, late adopters, and laggards. These acceptance rates of adoption influence a

diffusion wave, which has four distinct stages: a primary stage where centers of adoption are

 

106

established, a dtfiitsion stage where the process begins and a powerful centrifugal effect
results, a condensing stage where the relative increase in the number accepting is equal in all
locations, and finally a saturation stage in which the slowing and eventual cessation of the
process occurs.

Essentially two distinct types of diffusion prevail within the geographic discipline:
contagious diflusion and hierarchical difiitsion. Contagious diflitsion depends on direct
personal contacts and is strongly influenced by disrance (Haggett 1979); near potential
adopters receive an innovation before more distant ones (Lowe and Moryadas 1975).
Expansion and relocation diffusion are kinds of contagious diffusion where information,
materials, etc. are spread from one place to another and the spread is strongly influenced
by neighborhood effects (Parkes and Thrift 1980). These two types of contagious
diffusion differ in that with relocation diffusion, the things diffusing actually leave the
area where they originated (e.g., African Americans relocating from rural south to northern
cities) (Haggett 1979). Hierarchical diflicsion involves transmission through a regular
sequence of order, classes, or hierarchies. A common type of hierarchical diffiision is
cascade diffusion, where processes are assumed to move downward, from large centers to
smaller areas (Haggett 1979). Large places, important people, or powerful institutions
tend to get word of new information or innovations first and then transmit this
information, usually selectively, down the hierarchy of places, people, or institutions
(Parkes and Thrift 1980). As a result, early adopters may be located at relatively great
distances from the center of diffusion, while nearer potential adopters are temporarily
bypassed (Lowe and Moryadas 1975).

Perhaps one Of the first diffusion phenomena of geographic study was settlement
patterns. Colonization, comprising a traditional part of human geography, formed such
distinctive features on the landscape that they were viewed as “a fundamental expression on
‘man—land’ relationships” (Garner 1967, 303). Traditional emphasis was placed on form

and patterns which were essentially viewed as a reflection Of physical controls (i.e.,

 

107

mountains, rivers, etc.), supplemented heavily by historical influences. Beginning with the
first Virginia expedition to Chesapeake Bay in 1607 and extending to the turn of the
twentieth century, as more and more migrants came to America the movement of
colonization expanded outward from central ports and activity hubs along the Atlantic
seaboard from central Maine to the Carolinas, eventually expanded westward toward the
Piedmont and Appalachian backcountry, and finally from the Great Plains to the West
Coast during the drought in the 1930’s (Roseman 1977). Between 1650 and 1675 alone,
this settled region (areas with two persons or more per square mile) expanded at a rate of
three percent per year. Colonization of places such as Jamestown, Boston, Philadelphia,
New York, and Baltimore followed a modern urban syStem of settlement; “tiny nuclei
grew overnight into palpable places” (Earle 1992, 484). Viewed at micro and meso
spatial scales and meso and macro temporal scales, colonization can be regarded as a
continuous areal movement where people move from various places to new, largely
uninhabited locations, all the while settling areas in between.

With the spread of settlement across the landscape came the diffusion of innovation,
culture, and disease. An innovation is a new idea or technology that has some advantage
(e.g., it might be more profitable) over those previously in existence and replaces them
(Sugiura 1993). Innovation theory, the process by which an individual, establishment,
settlement, or nation receives and adopts something that was nOt there before (Lowe and
Moryadas 1975), covers several disciplines such as economics, anthropology, and
sociology (Carlstein 1978). In geography, “the salient feature of its innovation theory is
concentrated on the spatial movement of socio—cultural elements, their propagation,
transmission, and communication from place to place” (Carlsrein 1978, 147). Difiiision
of innovation examples include the spread of such diverse things as the printing press before
and after 1471 in England to modern day advancements such as cable television, highway
improvements, and electrical power. As Berry (1993, 395) explains, macro temporal and

spatial scales are necessary to understand the diffusion process of many of these

108

innovations: “Over the past two centuries, the progress of economic hisrory has been
marked by the rise and fall of successive technological systems... with the interval from
takeoff of new innovations to market saturation by the new technologies averaging some
fifty to sixty years.”

A classic example of an innovation spreading continuously and contiguously across
the landscape is the introduction of the gasalight in England in the early nineteenth century.
Referred to as the “colonization of night time” (Parkes and Thrift 1980, 282), the
diffusion of coal gas light began in 1820 where it first appeared in Pall Mall, London and
spread to all major cities in Britain over the next twenty years or so. Electricity, possibly
one of the most influential innovations in modern society, has exhibited hierarchical
spatial diffusion patterns where the innovation spreads from a first adopter city (e.g.,
Japan’s Tokyo Electric Power Company, started in 1887), to lower—ranking cities, towns,
and eventually villages (e.g., Kobe, Osaka, KyOto, and Sapporo) (Sugiura 1993). As with
the colonization of North America, fundamental agrarian innovations also diffused across
the landscape. Between 1680 and 1740, Chesapeake planters, in response to a beleaguered
tobacco economy, introduced agricultural practices such as shifting cultivation, crops such
as corn and wheat, and massive importations of slave labor. The introduction Of wheat,
corn, and wet rice by colonists in Pennsylvania and the Carolinas eventually extended from
Massachusetts to Georgia (Earle 1992). Other examples of innovation diffusion include
some controlled by a governing body, such as the strategic introduction of new sugar cane
varieties to the Caribbean islands around the turn of the twentieth century. Britain’s Royal
Commission of 1897 organized the diffusion of new cane breeding beginning with
Barbados, the center of innovation in the sugar industry. Over the course of fifty years, new
Barbadian varieties of sugar cane were introduced, flourished, and eventually declined on
the islands of the Caribbean (Galloway 1996).

Perhaps the most prominent and dramatic spatiotemporal process is the diflusion of

culture (Morrill 1968). As the first Europeans migrated to and settled areas in the eastern

109

United States, their ideas, customs, and culture became salient features of the landscape as
well. As Hudson (1988, 395) States rather succinctly and effectively, “maps of cultural
diffusion are maps of migration.” Cultural diffusion, which occurs at slightly smaller
scales than innovation diffusion, includes such diverse subjects as the growth of
Protesrantism in England in the mid sixteenth century (Bertin 1983), Rotary clubs, divorce
reform, and rumors (Parkes and Thrift 1980). Communication and interaction, serving as
the catalysts for many of these spreads, have a point—to—area spatial structure, particularly
information transmission by the mass media including radio, television, magazines, and
newspapers, all of which have a distinct spatial hierarchy (J akle, Brunn, and Roseman
1976), as well as by word of mouth (Parkes and Thrift 1980). For example, in 1891 a
rumor spread throughout Japan that a fire in an office building was caused by a short circuit
in the electrical wiring and the public began to regard electricity as dangerous. The
growth of electrical companies slowed for a year or two following this accident after
which the rumor was disproved and the companies gradually spread to smaller towns and
villages (Sugiura 1993). The same spatial pattern was exhibited by the Panic of 1837.
During the month of May in 1837, banks first Stopped converting banknotes into specie
(i.e., gold and silver) in Natchez, Mississippi. Although the information transmission was
word of mouth, and subsequently, newspaper, within two days banks in Tallahassee,
Florida, Montgomery, Alabama, and New York City followed suit.

The concentrations of firms is another example of a cultural diffusion phenomenon.
The number of selected Korean—owned business firms in Koreatown and its outlying areas
in Los Angeles, California experienced a three hundred percent growth rate between 1975
and 1986 with the area experiencing an increased concentration of firms in the central area
and a deconcentration toward a broader local area (Lee 1995). This dualistic pattern of
growth can also be applied to the diffusion of Wal—Mart stores over larger scopes of
analysis. Wal—Mart Stores, Inc., the nation’s largest retailer, has exhibited a pattern of

reverse hierarchical diffusion where the phenomenon spreads from small towns to larger

 

110

cities and metropolitan areas. Wal—Mart has diffused from its first store opening in
northwest Arkansas in 1962 to over a hundred store openings in Arkansas and adjacent
states by 1974, to over five hundred new stores throughout the deep south and southern
plains by 1984, to over eight hundred openings throughout the US by 1990 (Graff and
Ashton 1993).

Some forms of diflitsion of disease are also examples of continuous areal phenomena.
To understand how a disease moves through a human population, comparable scale
references (e.g., households and days or countries and years) should be adopted. The
actual process of communicable disease transmission at micro spatial scales (e.g., among
households and small villages) occurs in a spatially contagious manner (i.e., “neighborhood
diffiision”) and over small time frames such as days and weeks (Wilson 1993, 281).
However, an epidemic (a disease that invades a population and spreads through it at
particular times) is often observed using broader spatial and temporal contexts.

Occurring through social and, therefore, spatial networks and depending on the presence of
critical numbers of susceptible individuals to maintain the chains of transmission (Wilson
1993), epidemics exhibit intricate geographical patterns built upon a series of
interconnected and nested contagious processes occurring at different levels of human
interaction (Wilson 1993). A good example of this spatial and temporal complexity is
the spread of the AIDS virus world wide over the last decade or so.

Acquired Immunodeficiency Syndrome (AIDS) reflects a hierarchical nodal pattern
moving from major urban areas in Africa to other major urban areas such as Paris, New
York, and Los Angeles (Shannon and Pyle 1989). The epidemic has spread in a fashion
characteristic of most sexually transmitted diseases, i.e., spreading outward from larger
to smaller metropolitan areas, and from those cities to adjacent suburban areas. In the case
of the AIDS epidemic, radial diffusion pathways have been guided by concentrations of

populations behaviorally at risk and located at increasing distances from point Of origin to

the infection (Shannon and Pyle 1989). The spread of AIDS within the United States,

 

111

identified by aggregate effects of several years of reporting, has occurred throughout the
northeastern U.S. Megalopolis, beginning with the New York City node in the early
1980’s, and expanding along the Northeastern seaboard and inland (Shannon and Pyle
1989)

Other kinds of geographic phenomena are similar to diffusion processes in their
spatial and temporal movements and include acculturation, the process by which an ethnic
group changes in order to fianction in the host society, and assimilation, the loss of all ethnic
traits and complete blending into the host society (Jordan and Rowntree 1986).
Acculturation is usually the first and easiest stage of assimilation; new immigrants tend to
locate in ethnically concentrated zones characterized by large absolute numbers of
immigrants showing little English language ability, low educational attainment, low rates
of naturalization, and low income levels. After the first few years of assimilation in these
ethnically concentrated areas in the older centers of a city, individuals and groups with
greater economic resources tend to move out of the concentrated zones and into adjacent
areas with less ethnic concentrations called ‘dispersed zones.’ As more years follow and
more cultural and social assimilation occurs, immigrants may move further away into less
central areas with newer, better housing (i.e., highly dispersed zones), where the dominant
population is U.S.—born white Americans. The model of spatial assimilation, as applied
to the United States, focuses on the direct connections between cultural, economic, and
spatial assimilation; immigrants who share a culture, settle together initially for mutual
support, then leave that concentration and find homes among English—speaking US.
residents as they become more familiar with the country’s culture, find better jobs and
earn more money (Allen and Turner 1996). The patterns of spatial assimilation reflecr a
distance gradient: as distance from the ethnic concentration increases, the relative
assimilation of individuals should also increase. These processes of acculturation and
assimilation can be interpreted at micro spatial scales and macro temporal scales with

which the movements of not only the groups of new immigrants but of the adaptation

112

process as well flow across urban then suburban areas over a period of several years, only to
begin the cycle again at the urban center with new groups of immigrants.

Other examples of phenomena moving continuously in an areal fashion are some types
of regional boundary changes and include movements associated with cultural regions, urban
sprawl, and land use and cover changes. Geographically speaking, the landscape is subject
to demarcation and division by its culture, use, and cover as a means of identifying,
understanding, inventorying, and monitoring the earth. These boundaries or land
divisions, albeit fuzzy and often subjective, comprise and identify areas or regions of
space and Often change with the passage of time. Spatially, changes to a cultural region’s
boundaries can occur in areas as small as neighborhoods (e.g., New York’s Little Italy) to
areas encompassing several countries (e.g., Scandinavia). Generally speaking, movement
of a cultural landscape occurs slowly and may even be considered evolutionary; inhabitants
tend to be unreceptive to radical change and generally strive to preserve a landscape’s
original features (Falah 1996). This rate of cultural transformation is typical for many
regions and is besr understood using a macro temporal scale with which spatial changes to
the region are not necessarily noticeable on a day—to—day basis but rather over a period of
many years. However, changes to a cultural landscape’s spatial extent can occur in a very
rapid and dramatic fashion. For example, Falah (1996) examined the exodus/expulsion
of the Palesrinian people during the 1948 war that preceded the establishment Of the State
of Israel. The Israeli military adopted a strategy of “total war” to systematically
eliminate the PaleStinian’s attachment to their habitat (i.e., a strategy of “dc—signification”)
(Falah 1996, 257). Part of this transformation process where the cultural imprint of the
Palestinians was removed from the landscape involved the annihilation Of place or
“landscape erasure” (Falah 1996, 258). Historic buildings and other treasures of
civilization were destroyed. The depopulation of Palestinians by the Israelis over the
course of a year served as a catalyst for the transformation of Palestine’s cultural landscape.

This example of a region’s movement (in this case, the quick and radical obliteration of

113

cultural space) is understood clearly at meso spatial and temporal scales.

Regional boundary changes occurring in land use and cover are also considered
continuous areal movements, especially when viewed over a long time period. Primate
urban growth (i.e., high rates of natural increase in urban areas and high levels of in—
migration from rural areas) helps create urban sprawl, a spatial phenomenon where new
settlements located around an older urban core (CBD) become more populated and
urbanized as a result of high rates of population growth. Metropolitan landscapes
transform spatially as cities have changed from small, isolated population centers to large
areas of connected economic, physical, and environmental features on the landscape
(Buchanan and Acevedo 1996). This expansion or movement of urbanization into outlying
areas of a city is clearly evident when examining the boundaries of the affected area over a
period of several years. For example, in looking at a series of maps depicting the growth
stages of Cairo, Egypt, over the last ten years, the areal spread of urbanization is evident
by the increasingly larger space demarcated as urban surrounding the city (Stewart 1996).

NOt only is urban sprawl a type of cultural and land use regional boundary change, it
also represents changes to the natural land cover of an area. Changes in a landscape’s land
use, including marginal alterations such as diversifying an agricultural parcel of land to
more dramatic changes such as transforming a fallow field to multiple family residences,
“profoundly influence” the native vegetative cover of a landscape (Medley, McDonnell,
and Pickett 1995, 159). Urbanization causes a loss of natural vegetation and open spaces as
well as a general decline in the spatial extent and connectivity of wetlands, wildlife
habitat, and agricultural lands (Buchanan and Acevedo 1996). Also influencing natural
land cover are anthropogenic processes associated with urbanization such as elevated levels
of atmospheric C02 resulting from deforestation and the combustion of fossil fiiels
(Knapp and Soule 1996). In studying the influence of land use on the ability of a wind—
dispersed tree species to migrate in response to climatic warming, Dyer (1994) modeled

the dispersal of the loblolly pine in two areas in the eastern United States, each

 

114

distinguished by the intensity of the surrounding land use. Simulated migration through
the heavily forested study area was projected to occur in 940 years while 1,740 years are
projected for the loblolly pine to migrate through the metropolitan study area. Similar
research by Knapp and Soule (1996) of a transect extending from New York City to
northwestern Connecticut indicated that as measures of urbanization (i.e., human
population density, highway traffic volume, density of the road system, and urban—
residential land uses) declined directly with distance from the urban core, the forest—
landscape structural characteristics increased with disrance.

Other land use/ cover changes may be due to agricultural practices. Zweifler, Gold,
and Thomas (1994) Studied social survey and air photo time series data spanning several
decades of the Dominican Republic to gain both a temporal and spatial understanding of
the area’s land use evolurion from uninhabited wilderness to one of Latin America’s most
severe examples of hill land degradation. Referring to the region’s continually changing
land use/cover pattern as an “evolutionary process,” the authors describe a land use
continuum involving three distinct stages, each marking changes in the region’s landscape
for such land use classifications as grass/ brush (11% down to 2%), forested canopy (12%
down to 5%), and coffee plantations (39% up to 63%) (Zweifler, Gold, and Thomas
1994, 53). The boundaries of each of these land use categories were seen to have
continuously shrunk or expanded over the time period examined. Regardless of the causes
of land use/cover changes in any plant community, the important thing to now is that
continuous change in vegetation composition and dominance is a normal and ever present
component of ecosystem dynamics (Knapp and Soule 1996). As such, many natural land
cover boundaries are considered continuous areal movements when interpreted over periods
of tens to thousands of years.

Sometimes, however, spatial change to an area’s natural land cover may occur very
rapidly over a short period of time due to unforeseen events. As one such phenomenon,

firrest fires can be described as continuous areal movements; once ignited a fire quickly

115

spreads across a vegetative area, often moving from the point source(s) in an outward
fashion. The spatial and temporal dynamics of this phenomenon are most evident using
micro and meso scale references, where the areal spread is easily identified over tens to
hundreds of miles within several days. Factors such as wind speed and direction, as well as
natural and human features on the landscape, will affect the rate and spatial extent of the
fire’s expansion. Nonetheless, the fire does spread in a continuous areal fashion from its
source(s) and does nOt stop advancing until abated in some way. However, some forms of
spatial expansion are nor always temporally continuous, as discussed next with class 8,

cyclical areal movements.

    

“ Class 8 Cyclical Areal Movements

A second class of geographic phenomena that spreads out across the landscape is
cyclical areal movement. Some regional boundary changes, as well as outbreaks of gypsy
moth defoliation and diseases such as measles, dysentery, typhoid, and foot—and—mouth
epizooric, are all examples of how geographic phenomena move across a population or
landscape, affecting large contiguous areas of space in a cyclical fashion (Figure 4.9).
These areal movements occur with a fair regularity ranging from a small periodicity of
just a few days to longer periods of several to hundreds of years. Regional boundary
changes, as geographic phenomena that are interpreted to move in areal fashion over the
landscape, are sometimes cyclical. A region is a spatially dynamic entity as its spatial
demarcations often change over time. As with the case of administrative or political
boundaries, the delineation of space with borders is somewhat rigid and usually clearly
defined. For example, every ten years the Census Bureau Officially makes amendments to
its political/administrative boundaries such as census blocks, blockgroups, tracts, minor

civil divisions, and Congressional diStricts. Changes to the borders of these regions are

 

Spatial Scale (square miles)

116

 

 

megascale

mesoscale macroscale

10‘6“

universe

1014__
1012_
F

10‘°--

world

hanisphere

, -1
cont] nent f
............... 106 n_
country t”

“at:

10

 

 

microscale

 

'3
gypsy moth defoliation
.‘

county

102 ~-

city ."
village "I!

10° —-

regional
nd ghborhood

boundary changes

10'2 ~—

house

room

10'6 -

I

I I I l l l l

 

I

 

-II—

i 1 1 1 1 I i ‘r I 1
10'“ 10'6 10'4 10‘2 10° 102
second : monlh ye’ar

minute hour day :

century

 

 

Temporal Scale (years)

104

106

10.000 yrs. I million yrs.

------------------- microscale mesoscale-‘macroscale megascale

Figure 4.9 Scale Paradigm Of Cyclical Areal Movements.

fi—

I
108 10‘°
I billion yrs.

 

117

apparent at micro spatial and macro temporal scales.

Cyclical movement of land use boundaries are also noted in many agricultural
regions of the world. Moran’s (1993, 698) study of wine appellation in France suggests
that major wine—producing areas in the region are highly localized, with each variety of
grape having its own distinCtive geographic pattern: “the definition of territory is central
to the French wine appellation... its primary aim is to permit the association of a
particular wine with the territory from which it originates.” However, the author further
notes that the mOSt prestigious wine—grape varieties are beginning to disperse. For
instance, until the 1980’s, Cabernet Sauvignon was overwhelmingly localized in Bordeaux
(southwesr France), but has dispersed and moved into more south central areas of France,
such as the Languedoc—Rousillon, Provence, and Midi regions.

Gypsy moth defoliation and Other pest infestations are also examples of cyclical areal
movements. Gypsy moths, believed to have first evolved in the Far East, were
accidentally introduced into the United States in 1869 when some escaped from a
laboratory experiment in silk worm breeding (Pond 1992). From its now infamous
origination point on Myrtle Street in Bedford, Massachusetts, the insect occurs today in
more than half of the states in America, and is firmly entrenched from Canada south to
North Carolina and from Ohio west to central Michigan, with isolated infestations in a
number of other States. Although large areas Of forested land are altered each year from
gypsy mOth larvae feeding, contiguous patches of Stripped canopy foliage (i.e., defoliation)
form a mosaic pattern at micro spatial scales (Thurber, McClain, and Whitmore 1994).
Gypsy mOth defoliation is a highly cyclical phenomenon at meso and macro temporal
scales; nOt only is the reproducrive cycle of the larvae cyclical (yearly), but the regional
outbreaks occur in a predictable fashion as well. The number of acres infeCted from 1967
to 1991 simulate an “ebb and flow;” gypsy moth populations advance to new areas, build
up to outbreak numbers and then wane for a number of years, only to repeat the cycle over

and over, usually every four to five years (Pond 1992, 24).

 

 

118

Other pest infestations such as the western spruce budworm, fir-engraver, and
Douglas—fir beetle are also highly cyclical, spatially contiguous phenomena. The Blue
Mountains of northeastern Oregon have experienced massive tree mortality and damage
over the past decade due to such pest infestations. A major outbreak peaked in 1986,
declined until 1989 when another outbreak peaked, which then subsided in 1992. As with
any infestation, once the pests are introduced, the close proximity of trees in forested areas
make it easy for the infesrations to spread rapidly from one tree to the next (Green and
Cosentino 1996).

Another geographic phenomenon that sometimes acts as a cyclical areal movement is
the dificsion of disease. Understanding the cyclical nature of disease spreads requires
employment of comparable scale references in both time and space that can vary from
micro to macro scale. The spatial descriptions of epidemics in historical populations are
typically rendered at larger regional and state scales over the course of many, many years.
This approach, as Wilson (1993, 276) suggests, is useful since “it shows the ebb and flow of
epidemics across human space and demonstrates the temporal sequence of infection in this
space.” Many infectious diseases such as measles, dysentery, and typhoid act as ‘recurrent
wave models,’ where the phenomenon may always be present in a population, but the
prevalence of which oscillates 011 a cyclical basis (Carlstein 1978). Time and space
correlograms of weekly measles data in southwest England between 1966 and 1970
showed that the disease did not die out in the country, but rather flared up in major
epidemics approximately every two years. A similar wavelike form was also shown in
thirty—three major measles epidemics in Denmark over a hundred year period from 1868—
1968 (Haggett 1975). Cyclical patterns of disease also occurred with the colonization of
the Chesapeake Bay by the Virginia Company of London beginning in the spring of 1607.
The firsr outbreaks of dysentery and typhoid came in the summer of 1608, and returned

every summer thereafter until the colony’s dissolution by the Crown in 1624. The only

break in this cycle occurred between 1613 and 1616 when Sir Thomas Dale, recognizing

 

119

the causal link between water supply and mortality, dispersed the settlement, but his
actions were overturned by company agents freshly arrived in Virginia.

AnOther disease phenomenon shown to exhibit cyclical areal diffiision was the foot—
and—mouth epizootic in Great Britain between 1967—68. The disease which affected many
livestock including pigs, sheep, and cattle had a much shorter periodicity than that of
measles in the human population. The trend of the epizootic was that of a rapid initial
build-up, a sudden peaking, then a much slower decline. The pattern about this trend of
outbreaks was cyclical with a period of about eight days (Tinline 1972). The outbreak
also exhibited areal spread; the movement of the epizooric was traced by plorring the
daily center of gravity of the outbreaks and a general northeast contiguous movement
across the countryside was nOted. This was due largely to airborne spread with wind

serving as the major mechanism for the dispersal of the disease.

 

Class 9 Intermittent Areal Movements

Intermittent areal movements occur contiguously across an area or region in an
irregular fashion and are associated with such geographic phenomena as regional boundary
changes, desertification and military deployments (Figure 4.10). One type of intermittent
regional boundary change is political or administrative in nature. As Bertin (1983, 342)
states, “the movement of political borders... is a discontinuous movement.” Land
ownership parcels, national borders, and voring, school, and emergency service districts are
all examples of defined regions whose boundaries often shift over time. The boundaries
of Nazi Germany between 1935 and 1939 were altered and solidified by, of all things,
the Autobahn’s development (Rollins 199 5). The highway network which the Nazis
referred to as “an ideological consolidation of the nation,” served as a mechanism by

which the “far flung corners of Germany” were linked into one concrete network (Rollins

120

 

 

megascale
8
5
l

mesoscale - macroscale

1016_

universe

world

108 _
hemisphere
continent

.............. 106 _.
country
SIIIC

104 A

— desertification 5"
4

 

Spatial Scale (square miles)

 

microscale

 

county

102 —

my

village

10° .
neighborhood

10'2 -

u
-
v....
u

i.” I l
t regional

boundary changes

"- I
,1

military 3

d_ deployments

 

10'8 10'6 10'4 10'2 10° 102
second minute hour dayj month ytiar century

 

.—
.—
.-
.—
P
-—h—

10“ 106 108 10‘°
I0.000 yrs. I million yrs. 1 billion yrs.

 

 

................... microscale MeSOSCdle"m46703€dk "

-------------------- megascale

Temporal Scale (years)

Figure 4.10 Scale Paradigm of Intermittent Areal Movements.

121

1995, 494—495). More recently, the “fundamental geopolitical transition” associated with
the collapse of the Soviet Union has resulted in the creation of no less than fifteen new
sovereign states, with the prospect of further disintegration of the former Republic
(Bradshaw and Lynn 1994, 442). The spatial scale of political or administrative regions
varies considerably and may fall within the micro, meso, or macro spatial scales, where
delineated areas may be as small as a quarter of an acre or less (e.g., individual land
parcel) to as large as the Russian Federation, currently over six—and-a—half million square
miles. Temporally, these boundaries may change infrequently and are often identified at
macro scale as moving over a period of years.

Economic regions also have intermittently moving boundaries, though not as tangible
as its administrative counterparts: “Unlike political space, economic space is more
amorphous in that its boundaries tend to be fuzzy and permeable” (Lowe and Moryadas
1975, 9). Generally, economic borders are small (micro to meso) in spatial scale and
move over periods of years. For example, Moore (1994), in examining the spatial
dynamics of Appalachia’s economic development which spans more than twenty—five years
of regional planning strategies, uses core—periphery models to provide a useful conStruct
for explaining changes and contrasts in geographic patterns of income within Appalachia.
This approach allowed a clear spatial illustration of regional dynamics and led to the
suggesred redefinition of the Appalachian Regional Commission (ARC) regions).

AnOther geographic example of an intermittent areal movement is desertification. A
term used to describe the widespread distribution of land degradation in dryland
environments whereby land is rendered sterile with respect to agricultural production in a
manner that is irreversible in practical human terms, desertification is a phenomenon that
involves complex earth—surface systems components (e.g., climate, soils, vegetation, etc.).
As Phillips (1993, 638) nOtes, areas being degraded are changing as unpredictably as the
weather: “these systems are subject to potentially unpredictable, chaotic behavior, in

addition to irregular internal climate forcings.” Desertification is best understood using

 

122

meso spatial extents and macro time frames.

A final geographic phenomenon that exemplifies intermittent areal movement is
military deployment. Generally, deployments occur across relatively small spatial extents
over a course of a day to several weeks. For example, in the numerous battles during the
Civil War in the United States, deployment of ground troops, as a means of overtaking a
fort, city, or other key locale, usually began as linear movement, but often led to an areal
encroachment of troops on all sides (i.e., ‘surrounding the enemy’). In such campaigns as
Chancellorsville and Gettysburg, the movement of troops was intermittent and occurred
over just a few days (McPherson 1988). However, the deployment of troops sometimes
occurred over a much longer time frame, as was in the case of the Vicksburg Campaign

and subsequent siege which lasted over three months (Kennedy 1990).

 

 

Class 10 Continuous Volumetric Movements

The last three classes of geographic movement involve volumetric flows and
movement. Volumetric phenomena take on a three—dimensional form as they move across
space and through time. Continuous volumetric movements (Figure 4.11) encompass
many of the earth systems processes, primarily due to the fact that most volumetric
features in geography consisr of the natural states of matter: solids, water, and/or gas.
Studies of the planet earth fall into broad categories that embrace the solid lithosphere,
water in the liquid hydrosphere and frozen cryosphere, the mainly gaseous atmosphere, and
the life forms of the biosphere. Phenomena caused by lithospheric movements, such as
continental drifi, a volcanic eruption, an earthquake, a tsunami, glaciation, rock
transfirrmation, soil creep, and blowout formation, as well as the rotation of the earth all
involve continuous volumetric movement of the solid earth realm. Biospheric earth

systems, such as the hydrologic cycle, encompass all living organisms of the earth and the

123

 

 

megascale

mesoscale - macroscale

 

 

Spatial Scale (square miles)

 

microscale

 

 

 

 

 

 

10l6 __
universe
1014 __
10”-
1010--
world rotation of hydrologic
8 the earth Ck .
"“"MT'" 10 7" 9' continental
hernssphese .
drrfi
““6““ glaciation
"""""""""" 106 ‘7 atmospheric motion
country 3‘
State ‘—
104 ~— -
a tsunami .
rock transfirmatzon
I
county
102 * an earthquake a volcanic eruption
city
village TT
100 -_ ’
neighborhood “”1
d»- creep
10'2 ....
10“ ~—
house
" blowout firmation
[00m 6
l0” "'
I 41 I I l I I l I I L L l I I I l
I I I I I T I I I I I I I I I I I
10'8 10‘6 10" 10‘2 10° 102 104 106 108 10'°
second minute hour day i month yéar century 10,000 yrs. I million yrs. 1 billion yrs.
------------------- microscale mesoscale-imacroscale megascale

Temporal Scale (years)

Figure 4.11 Scale Paradigm of Continuous Volumetric Movements.

124

environment with which they interact. Atmospheric motion involves movement of air in the
earth’s atmosphere due to processes such as climatic and ocean circulation and global
radiation. Continuous interchanges among each of these ‘spheres’ produce an integrated
environment, with the entire earth realm acting as one closed system, continually moving
in space and time. A closer examination of these naturally-occurring phenomena will
reveal their continuous volumetric nature.

Plate tectonics is the catalyst for many continuous volumetric movements such as
continental drift, a volcanic eruption, an earthquake, and a tsunami. Plate tectonics is the
general theory encompassing lithospheric plates and their relative mOtions and boundary
interactions. The earth’s lithosphere, which includes the upper part of its mantle as well as
the crusr, ranges in thickness from twenty—five to fifty miles deep. The lithosphere is
broken into large fragments called lithospheric plates, which are typically of continental
dimensions and capable of moving independently of the plates that surround it. A long
term effect of tectonic activity is the phenomenon continental drifi, which refers to all
forms of breaking and bending of the entire lithosphere, including the crust (Strahler and
Strahler 1984). As a continuous volumetric movement, continental drift, which refers to
the separation of the continents, “involves a spatial phenomenon that integrates all time, all
space, and all physical processes for an entire planet” (Dobson 1992, 188—189).

Various theories have existed about the movement associated with continental drift.
In the early twentieth century, Alfred Wegener, a German meteorologisr and geophysicist,
reconstructed a super continent called ‘Pangaea’ which existed intaCt about 300 million
years ago in a period of geologic time called the Carboniferous Period (Strahler and
Strahler 1984). Wegener visualized this large land mass slowly rifting apart, until the
continents and oceans, as we know them today, were created. Although Wegener’s theory
was scoffed in the 1920’s and 1930’s, within forty years plate tectonics emerged as a
leading theory. Some have suggested the continents have moved laterally while circulation

models suggest a circular form of movement that could account for repeated collisions

125

and separations of adjoining continents (Dobson 1992). The modern interpretation is that
continental drift involves entire lithospheric plates, much thicker that merely the outer
crusr of either the continents or the ocean basins. Plate movements over a soft plastic
aSthenosphere have allowed the continents to be carried according to the general timetable
posrulated by Wegener (Strahler and Strahler 1984). Regardless of the direction, speed,
and actual paths of the movements, all theories of continental drift agree that the
continents, as macroscale three—dimensional volumetric objects, have moved spatially over
the earth in one continuous geologic time—frame.

Another example of a continuous volumetric movement is a volcanic eruption.
Volcanoes, conical or dome—shaped structures built by the emission of lava and its
contained gases from a constricted vent in the earth’s surface, vary in form and dimension
depending on the type of lava and the presence or absence of tephra, the ejected solid
fragments ranging in size from gravel and sand down to fine silt. The nature of a volcanic
eruption, whether explosive or quiet, depends on the type of magma; felsic lavas have a
high degree of viscosity, hold large amounts of gas under pressure, and result in lavas
producing explosive eruptions while mafic lava is highly fluid, holds little gas, and
produces relatively quiet eruptions with lava traveling over long disrances in thin layers
(Strahler and Strahler 1984).

Spatially, a volcanic eruption can affect microscale areas from a hundred square
miles up to macroscale spatial extents covering the western half of the United States and
beyond. Clouds of incandescent gases, also known as ‘glowing avalanches,’ can soar to
heights of more than 60,000 feet before descending down the flank of the volcanic cone,
such as the 1902 eruption of Mount Pelee (Strahler and Strahler 1984). The descent of
millions of tons of ash, Cinders, and lava can spread our over vast areas of several hundred
acres. Debris flows, called ‘lahars,’ are flash floods of semiliquid mud, rock, and ice that
surge down the slopes with terrifying speed and, as the Colombian farming community

Armero experienced in 1985, can bury the surrounding landscape with over thirty feet of

 

126

debris (Krakauer 1996). Temporally, a volcanic eruption can be analyzed in both the
microscale and the mesoscale domains. While the actual eruption may only last several
minutes, the aftermath associated with the eruption can linger for several days, weeks, even
months, as was the ease with the 1815 eruption of Mount Tambora in Indonesia. The year
following that eruption was known as “the year without a summer” in the United States and
Europe due to the estimated 150 square kilometers of ash that had been ejected into the
atmosphere (Critchfield 1983, 241).

An earthquake is another geographic phenomenon that occurs in a continuous
volumetric fashion. Like ripples produced when a pebble is thrown into a quiet pond,
seismic waves, produced by sudden movements along faults, gradually lose energy as they
travel outward in widening circles from a focus, or a point of sudden energy release.
Strahler and Strahler (1984) stress that an earthquake is not a cyclical or patterned event,
even though it is inevitable. An earthquake falls within the microscale domain, bOth in
time and in space, as the phenomenon typically lasts for a duration measured in seconds
and adversely affects a relatively small geographic area, such as part of a city or county.

A tsunami, a seismic sea wave, is an important environmental hazard associated with
a major earthquake and is interpreted as a continuous volumetric phenomenon. Occurring
over several thousand square miles (meso spatial scale) and lasting less than one day
(micro temporal scale), a tsunami or ‘tidal wave’ is a train of water waves often generated
in the ocean at a point near the earthquake source by a sudden movement of the seafloor.
The waves move over the ocean in ever—widening circles and cause a rise of water level
when it arrives at a distance coastline. Wind—driven waves, superimposed on the
heightened water level, allow the surf to attack places inland that are normally above the
reach of waves. For example, the seismic wave of 1933 in the Pacific Ocean caused waves
to attack ground as high as thirty feet above normal tide level, causing widespread
deStruction and many deaths by drowning in low-lying coastal areas (Strahler and Strahler

1984).

 

127

Other examples of continuous volumetric geographic phenomena include many
geomorphological processes such as glaciation, rock transformation, soil creep, and
blowout formation. Geomorphological analyses all deal with the fiindamental
components of landform evolution and weathering (Gates and Nordstrom 1995).
Weathering, the general term applied to the combined action of all processes causing rock
to be disintegrated physically and decomposed chemically because of exposure at or near
the earth’s surface, involves a continuous agitation in the soil and regolith over very long
time intervals (Pope, Dorn, and Dixon 1995). Glaciation, the t0tal process of glacier
grOWth and landform modification by glaciers, is one example of landform evolution that
occurs progressively (Strahler and Strahler 1984). A glacier may be defined as any large
natural accumulation of land ice affected by present or past m0tion, the size, speed, and
volume of which may vary considerably. Large ice sheets, such as the Greenland Ice Sheet
with an area of 670,000 square miles, generally move just a few centimeters per day while
an active alpine glacier with a much smaller volume may move at a rate of several meters
per day. Some alpine glaciers experience episodes of very rapid movement, described as
‘surges’, and the glacier may develop a sinuous pattern of movement (i.e., “galloping
glaciers”) (Strahler and Strahler 1984, 352—3 53). Regardless of these spatial and
temporal variations, glaciation is best understood as a continuous volumetric phenomenon
when using macro and mega scale frames, bOth spatially and temporally.

Roe/e transformation is another geomorphologic example of a continuous volumetric
phenomenon. The process of rock transformation is a total circuit of rock change in
response to environmental stress where mineral matter is continually recycled through the
three major rock classes (igneous, sedimentary, and metamorphic) (Strahler and Strahler
1984). This ‘rock cycle’ is in continuous operation over some three million years or more
of geographic time (mega time frame) and generally can be said to occur over mesoscale
spatial extents.

Also affected by weathering and contriburing to landform evolution are continuous

 

128

sedimentological processes involving soil erosion and deposition such as soil creep and
blowout formation (Blewett and Winters 1995). Soil, a natural earth surface layer
containing organic and mineral matter, is dynamic as many complex physical and
chemical activities are occurring simultaneously within it (Strahler and Strahler 1984).
Soil creep is the extremely slow downhill movement of soil and regolith as a result of
continued physical and chemical weathering (e.g., frost action, unloading), as well as Other
catalySts such as burrowing animals and earthquakes (Strahler and Strahler 1984). The
process is besr understood using a micro spatial scale (i.e., a particular location) and a
macroscale temporal frame; soil creep is, for the most part, slow—acting and produces
effects that are visible only when accumulated over many years, even centuries. Eroded
sediments may travel anywhere from three or four kilometers in a hundred years or more,
to more than twenty«five kilometers in the same time frame (Beach 1994). The volume of
transported sediment can be remarkable in some highly eroded areas such as the Loess
plateau in China, where an average of one centimeter of loessal soils are moved each year
(Veeck, Zhou, and Ling 1995).

Blowout fimnation involves the development of depressions in dune systems as a result
of deflation, a process where loess sediments lying on the ground surface are lifted in the
air or rolled along the ground by wind (Strahler and Strahler 1984). Like many
geomorphic features, a blowout evolves as a result of process/form interaction representing
adjustments between wind speed and direction, topography, and vegetative cover. Gates
and Nordstrom (1995, 17), in studying three blowouts located within a twelve kilometer
long section of New Jersey’s Island Beach State Park, note that blowouts occur over
microscale spatial extents (approximate averages of fifty square feet in surface area and
five feet deep), and evolve through distinct morphological stages over the microscale time
frame of one to two decades.

Like solid earth matter, water is analyzed and undersrood in three dimensions and is

considered to be a continuous volumetric movement: “the flow of water, measured as

129

cubic feet per second... incorporates both volume and time” (Lowe and Moryadas 1975,
159). The hydrologic cycle, or global water balance, can be treated as a closed material
flow system in which water moves through the atmosphere and oceans and upon or beneath
the earth’s land surfaces in a series of processes such as condensation and precipitation,
infiltration and runoff, and evaporation and transpiration. The hydrologic cycle should be
viewed with a yearly temporal frame and on a global spatial scale where the “water of
oceans, atmosphere, and lands moves in a great series of continuous interchanges of bOth
geographic position and physical state (Strahler and Strahler 1984, 165).

Atmospheric motion is yet anOther example of continuous volumetric movement.
Defined for this discussion to include all contributing and interconnected atmospheric
system processes, atmospheric morion refers to the continuous “ceaseless morion of the
atmosphere,” (Strahler and Strahler 1984, 74). Variable forces keep the atmosphere (the
envelope of gases surrounding the earth) in a state of continual agitation and include glohal
radiation, climatic and ocean circulation, atmospheric pressure, and air masses. Atmospheric
motion is best understood with micro and meso temporal scales and meso and macro
spatial scales (i.e., viewing the earth’s circulation patterns over the mid latitude zone as it
moves over the course of a week).

Glohal radiation, or electromagnetic energy, travels through space at the speed of light
and consists of two flows, one of short—wave radiation from the sun and one of long—wave
radiation from the earth. Similar to the hydrologic cycle, global radiation is a system by
which a state of matter, as a volume, continuously and endlessly moves through space.
Critchfield (1983, 83) describes climatic and ocean circulation as “analogous to the
movement of air in a room by h0t radiators or by forced air ventilation;” as long as
sufficient differences in air density prevail at different parts of the sysrem, air will
continue to flow in response to the pressure gradient. Atmospheric pressure, as the main link
between solar energy and motion, directly affects world patterns of winds (air motion

with respect to the earth’s surface), storms, and related phenomena such as sea and land

130

breezes, which are circulation systems common along coasts due to the rising and sinking
motion of the air and by a weak return flow at higher levels (Critchfield 1983; Strahler
and Strahler 1984). Air masses, bodies of air in which the upward gradients of temperature
and moisture are fairly uniform over a large area, move from one region to another
following the patterns of barometric pressure. The particular air mass properties at a
given place may reflect the composite influence of a travel path covering thousands of
kilometers and passing alternatively over oceans and continents (Strahler and Strahler
1984). As each of these atmospheric forces suggest, all are uniquely dynamic but they
interact with one anOther to produce one large “kaleidoscope of many motion systems,”
i.e., atmospheric motion (Critchfield 1983, 139).

A final example of a continuous volumetric movement looks at the earth as a whole.
The phenomenon of the rotation of the earth involves the planet spinning like a top on an
axis. To understand roration, the earth is viewed as a three—dimensional sphere suspended
in space, continuously turning in the same direction and at a constant speed. Spatially, the
earth falls under the megascale domain with a mass of over six sextillion short tons and a
surface area encompassing close to two hundred million square miles (Famighetti 1996).
Temporally, the rotation of the earth is best understood using a microscale view, where one
earth turn with respect to the sun defines the solar day (Strahler and Strahler 1984).
Phenomena such as sunrises and sunsets provide us with the knowledge that the earth is

consistently and ceaselessly spinning around on itself.

  

\‘A Class 11 Cyclical Volumetric Movements

Discussion of continuous volumetric movement has shown most natural earth system
processes to be continuously evolving and moving through space and time. Cyclical

phenomena often occur in response to these continuously moving earth system processes and

 

131

include phenomena such as ocean tides, beach recovery, El Nino, flooding, and some forms of
pollution (Figure 4.12). Ocean tides are caused by the continual diurnal rhythm imposed
by the earth’s r0tation and are a natural phenomenon involving the alternating rise and fall
in the large fluid bodies of the earth caused by the combined gravitational attraction of
the sun and moon (Strahler and Strahler 1984). The combination of these two forces
produces the complex recurrent cycle of the tides; a sequence of two high tides separated
by two low tides is produced each day (Famighetti 1996). Ocean tide can be viewed as a
cyclical volumetric phenomenon since the rhythmic movement of the ocean water (as a
three-dimensional object) yields tide ranges so predictable that the tidal cycle can be
referred to as a ‘clock’ (Strahler and Strahler 1984). Ocean tides are best understood when
viewed using micro scales, watching the tide rise or fall on a beach over several hours.

Related to tidal activity is the cyclical geographic phenomenon of beach recovery.
Beaches are thick wedge—shaped deposits of sand that absorb energy of breaking waves.
Beach recovery involves the curring back of the beach during short periods of storm and the
subsequent restoration of sand during long periods of calm ocean waters when waves are
weak (Strahler and Strahler 1984). Using standard beach surveying techniques and depth—
of—activity rods to monitor the morphology and volume of sediment evolution changes on
an estuarine barrier on the Delaware Bay, Jackson (1995, 33) nores the temporal regularity
associated with beach recovery: “The results of this investigation support previous findings
that document cyclic beach response and swash bar migration during post storm recovery
in ocean and lacusrrine environments.” As a volumetric geographic phenomenon occurring
regularly in estuarine environments, the dynamics of beach recovery are best understood in
the domain of microscale space and micro to meso scale time.

As described with continuous volumetric movements, largevscale atmospheric and
associated hydrologic systems are always in morion. Often times, these systems produce
effects that occur rather predictably in nature such as El Nino. In response to major changes

in oceanic circulation, El Nino is a disruption of the ocean—atmosphere system in the

 

Spatial Scale (square miles)

132

 

 

megascale

mesoscale macroscale

1012_4

108 —

continent

I

C

l

l

l

l

l

E
p—
O

O\
I

s
O
c

 

 

microscale

 

county
102 -
aw
village
10° «
ndghborhood

10'2 —

104‘

house
.—4

room

10'6 ~

pollution

heach

recovery

ocean tides

I

El Nino

 

10“

hour

10'6

minute

10'8 .
second day; month yt’nr

t
I
t

 

................... microscale 'mesoscalér-macroscale --

century

 

—i
d!—

108 10'°

10.000 yrs. I million yrs. 1 billion yrs.

 

 

-------------------- megascale

Temporal Scale (years)

Figure 4.12 Scale Paradigm of Cyclical Volumetric Movements.

 

133

tropical Pacific where warm water from the western coast of South America spreads over
cold water near the shore, resulting in heavy rainfall in nearby coastal areas (Critchfield
1983). El Nino occurs on a regular basis, usually during February and March, and is best
understood at a macro spatial scale. Among the consequences of El Nino are increased
rainfall across the southern tier of the United States and in Peru which has caused
destructive flooding, an activity that has occurred on such a regular basis that it has become
predictable (Soreide 1996; Graf and Gober 1992). As a cyclical volumetric movement,
flooding occurs quite regularly and can take many forms in nature. ‘Flash flooding’, the
result of copious amounts of rain in a short period of time (Famighetti 1996), can be said
to exhibit regular patterns in spatial and seasonal occurrence (Winkler 1992). ‘CoaStal
flooding’ is also seasonal and occurs when ocean water is driven inland due to winds
generated from tropical Storms, hurricanes, or intense offshore low pressure systems.
During ‘urban flooding,’ streets can become swift moving rivers and basements can become
death traps as they fill with water; urbanization increases runoff two to six times over what
would occur on natural terrain. ‘River flooding’ is a natural process that occurs seasonally
when winter or spring rains are coupled with melting snow, filling river basins with too
much water, too quickly, a process that can be magnified due to accumulations of ice at
natural or artificial obStructions which stop the natural flow of water.

The geographic phenomenon of flooding is generally best understood using a micro
spatial scale and meso and macro temporal scales, where cyclical patterns of the event
become evident over the course of many seasons and years. Regularly inundated by stream
waters about once a year when the discharge of a river cannor be accommodated within the
margins of its normal channel are river floodplains, broad belts of low flat ground
bordering the channels on one or bOth sides (Famighetti 1996; Strahler and Strahler 1984).
Such regular inundation is considered a flood, even though its occurrence is expecred and
does not generally interfere with agriculture or growrh of dense forests (Strahler and

Strahler 1984). Flood plains are named for the general recurrence interval that can be

134

expected for certain magnitude flood events: small flood events (less than five year
recurrence interval), moderate flood events (five to twenty-five year), and large flood
events (greater than a twenty—five year recurrence interval) (VVoltemade 1994). Although
the periodicity of these more rare and disastrous large flood events does nor lend to
highly accurate predictions of the occurrence, taken over hundreds of years, a marked
regularity does persisr.

A final example of a volumetric geographic phenomenon that is cyclical in behavior
are some forms of pollution. Air pollution is a phenomenon that can occur across a wide
array of temporal and spatial scales (micro to macro) and occurs when pollutants (foreign
matter) are injected into the lower atmosphere as particulate matter or as chemical
pollutants (Strahler and Strahler 1984). Blowing dust is a kind of particulate matter
pollutant that is a common phenomenon of dry climates and drought—stricken areas such as
the Great Plains of the United States. Blowing dust events, which begin when horizontal
visibility falls below eleven kilometers and ends when visibility regained exceeds eleven
kilometers (usually within a day), result from the action of strong winds picking up loose
earth material and carrying it to great heights and perhaps for great distances as well
(Critchfield 1983). Studying dust storms occurring in the southern High Plains of the
United States over a forty—two year period, Lee and Tchakarian (1995, 690) found that
although the amount of sediment moved per unit force was nor found to be consistent
among events, a cyclical temporal variability of events was apparently attributed, in large
part, to “the seasonal timing of agricultural practices.” Ozone pollution, another kind of
air pollution, also exhibits cyclical patterns of occurrence. Comrie (1994), in studying the
origins and pathways of air masses controlling ozone concentrations in Pennsylvania’s
Allegheny Plateau forests, found high ozone levels eleven times in a three—year period, but
only during the growing season. Although some forms of pollution are cyclical in nature,
many do n0t occur with any regularity and cannot be predicted, as discussed next in the

final movement class, intermittent volumetric movements.

135

 

Class 12 Intermittent Volumetric Movements

The final class of geographic movement describe intermittent phenomena such as
certain forms of pollution, mass wasting, volcanic eruptions, and earthquakes (Figure 4.13).
Phenomena in this class are complex in bOth their spatial and temporal components and
involve the irregular and unpredictable movement of a volume across space and through
time. As discussed in the previous movement class, some forms of pollution are cyclical
in nature. However, many pollution events are human induced and/or dependent upon
many contributing conditions such as the weather, thereby contributing to the
unpredictability of an event. While a pollution event may acr as a continuous volumetric
movement while it is occurring, it is the phenomenon of pollution as an unplanned,
unpredictable activity that contaminates the earth that is of geographic significance. As a
spatially dynamic phenomenon occurring unpredictably in three dimensions (air, water,
earth matter), pollution can vary significantly in scale, borh spatially and temporally, and
understanding such complexities requires scopes of analysis ranging from micro to macro
scales.

One type of pollution that occurs irregularly is air pollution. Pollution domes form
when pollutants are trapped beneath an inversion lid within the earth’s atmosphere that
takes the form of a broad dome centered over a city when winds are light or a calm
persists (Strahler and Strahler 1984). Pollution domes are fairly common in urban areas
and, although they occur more frequently during the summer, they are unpredictable due to
the fact that “background meteorological conditions vary on a daily, monthly, and annual
basis” (Comrie 1994, 635). Other events of intermittent air pollution may occur as a
result of a single catastrophic episode, such as the burning of the Kuwait oil fields, the
eruption of Mount St. Helens, and radiation leaks from nuclear reactors, such as the 1979

Three Mile Island meltdown and the 1986 Chernobyl accident.

Spatial Scale (square miles)

136

 

 

megascale

mesoscale ma

universe

continent

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

l
'_n
O

0\

country

state

10‘2“

pollution

 

 

microscale

 

county
102
my
village
10°
ndghborhood

room

I
I
I
I
I

landslides

‘7 mudflows

avabnches

earthflows

I
I

earthquakes

volcanic eruptions

 

I
T
10'8

second minute

hour day i

O
.L.
p—o
-. - - -o —I—
M

 

I I 1

10° 102

month yin

century

------------------- microscale mesoscale " ”140035416 "

 

104 106
10.000 yrs. 1 million yrs.

-------------------- megascale

Temporal Scale (years)

108 10‘°
1 billion yrs.

 

Figure 4.13 Scale Paradigm of Intermittent Volumetric Movements.

137

Pollution also occurs in the Other earth systems, such as the hydrosphere and
lithosphere. Like Other forms of environmental contamination, water pollution, although a
somewhat common occurrence, is difficult, if nor impossible, to predict. Generally
affecting less area over shorter periods of time than many forms of air pollution, pollution
of surface water and ground water by sewage, fertilizers, and chemicals occurs irregularly
due to such point and non—point pollution sources as leaching landfills, urban and suburban
areas and industrial plants. Hazardous contamination, like the 1989 Exxon Valdez oil
spill that dumped over ten million gallons of crude oil in Alaska’s Prince William Sound,
pollute not only large volumes of water, but also sand and rock along miles of shoreline.
Since 1967, over twenty notable oil spills have intermittently occurred throughout the
world (Famighetti 1996). Once entered into the water balance system, pollutants may
travel to the water table, where the pollutants then follow the flow paths of ground water.

Other examples of volumetric geographic phenomena that occur intermittently are
some forms of mass wasting, a collective term for the spontaneous downward movement of
soil, regolith, and bedrock under the influence of gravity. Although abundant evidence
shows that on most slopes at least a small amount of downhill movement is going on
continuously, much of this morion is imperceptible (Strahler and Strahler 1984). Rather it
is when the regolith exhibits “rapid colonization over large areas triggered by an acute,
perhaps unique, event” and slides or flows abruptly causing environmental hazards such as
earthflows, mudflows, landslides, and avalanches that should be nored (Parker 1993, 626).
Each of these phenomena involve the sporadic movement of mass wasting volumes that can
vary in both temporal and spatial scale.

Earthflows occur in regions of humid climate and are a mass of water—saturated soil,
regolith, or weak shale that moves down a steep slope. Falling within the spatial and
temporal microscale domain, an earthflow may affect areas from a few square meters up
to several hectares, and usually has a slow flowage rate of a few hours. Mudflows consist of

the downslope flowage of a mixture of water and mineral fragments, varying in

 

138

consistency from a mixture resembling thick viscous concrete to consistencies that are

little different than in turbid stream floods. As with earthflows, mudflows occur in micro
temporal and spatial scales, but at slightly larger spatial extents such as a village.
Landslides also occur sporadically and yield perhaps the quickest and largest volumetric
movement of all mass wasting phenomena (Strahler and Strahler 1984). Usually occurring
in thinly populated mountainous regions, landslides involve the sudden release of millions
of cubic meters of rock down a mountainside, such as the 1959 Madison River gorge
landslide that formed a debris dam over 200 feet high and contained thirty-five million
cubic yards of debris (Strahler and Strahler 1984). Similar to landslides in that they
occur within minutes and can cover tens to hundreds of square miles, avabnches involve the
sudden downward movement of snow. Occurring in alpine environments where climate
and snowpack characteristics vary spatially, avalanches are unpredictable due to variations
in weather during a season which affCCt tendencies of the snowpack toward either failure or
increased stability (Mock and Kay, 1992).

Some final examples of intermittent volumetric geographic phenomena are volcanic
eruptions and earthquakes. While a singular volcanic eruption or earthquake is interpreted
as a continuous volumetric movement at very micro scales, and numerous occurrences of
these geologic cataclysms are interpreted with much larger scale frames (both spatially
and temporally) as intermittent point phenomena, multiple occurrences of these
phenomena at the same location viewed over long time frames can be seen as exhibiting
intermittent volumetric movement. For example, Mount Rainier, the highest volcano peak
in North America’s Cascade Range, has erupted numerous times in the past (most recently,
just 150 years ago) and geologists warn that it will erupt again anytime within the next ten
to 10,000 years. However, given that at least sixty major lahars (flash floods of semi
liquid mud, rock, and ice) have flowed down Rainier over the past 10,000 years,
volcanologisrs suggest that the recurrence interval for major debris flows is somewhere

between 500 and 1,000 years (Krakauer 1996). Similar interpretations can be applied to

 

139

the rather unpredictable occurrence of earthquakes at or near the same location or focus such
as locations along the renowned San Andreas fault in the western United States. Using
micro spatial and mega temporal scale references, the unpredictable seismic activity

related to particular earthquake sites and active volcanoes becomes apparent.

M ulti—Type Phenomena

Twelve individual classes of movement have been presented as a general framework
with which to organize and understand all spatially dynamic phenomena. The
classification of geographic movement is exhaustive; with four spatial categories cross
referenced with three temporal categories, the matrix allows for all possible forms of
geographic motion to be described. Each spatially dynamic geographic phenomenon can
be classed into one or more of the twelve unique movement categories. Some geographic
phenomena are viewed using a wide array of spatial and/ or temporal perspectives, whereby
they may fall into more than one movement class. Examples of such multi—type
phenomena occurring in the human world include trade and commerce, route services,
regional boundary changes, diffiision, and human mobility. Using very small or narrow
scale frames in analyzing a person’s mobility through space and time yields journey—to—
work and personal action space analyses. With this vantage point, human mobility is
considered cyclical point movement; the movement occurs repeatedly over a period of
hours, days, and weeks. If the time of analysis is lengthened to years, the movements of
the same individuals can be interpreted as intermittent, representing residential migration
over the course of a life-time.

Tornadoes, hurricanes, earthquakes, and volcanic eruptions are examples of multi—
type phenomena in the physical realm. As a geographic movement, volcanic eruptions are
classed into three different movement categories: continuous volumetric, intermittent
volumetric, and intermittent point. When a volcano erupts, lava, gases, and solid earth

fragments flow, as various three-dimensional forms, down its side and through the air and

 

 

140

is considered a continuous volumetric phenomenon. Examination Of that same volcano
over a much longer time frame Of thousands to millions Of years describes intermittent
volumetric movement; numerous eruptions may have occurred, the periodicity Of which are
infrequent and largely unprediCtable. By changing scales once again, those same volcanic
eruptions can be examined with an even broader scope that places them in mega time and
space with respect to all Other volcanic eruptions across the world. Viewed with this
perspective, each eruption is interpreted as a point in space that, once in a while, is active
and displays movement.

As human mobility and volcanic eruptions demonstrate, scale can vary in bOth the
temporal and spatial dimensions in interpretations Of phenomena and any variation in one
or bOth Of these components will change the focus Of the geographic inquiry. When the
scope Of any study is determined, decisions are made about what is to be studied in detail
and what is tO be left in the background (Hudson 1992), and depending on the point Of
view of the Observer, different analyses Of the same phenomenon may take place. Despite
the possibility Of scalar variations occurring in approaches to particular geographic
problems, this classification allows the identification and interpretation Of the changes in
the spatial location and/or configurations Of a phenomenon. Describing the individual
components that comprise a geographic phenomenon will lend itself to greater
understanding and appreciation of that phenomenon’s place in the world with which it

interacts.

Chapter 5
CONCLUSIONS

Space is the cornerstone of geography and is the universal characteristic that grounds
the discipline. By definition, or at least by one pervasive philosophical viewpoint in
geography, a geographic phenomena can be viewed in a spatial context (Lowe and Moryadas
1975). Geographic description and/or explanation Often involves examination Of a
phenomenon over time, thereby revealing patterns and interactions among phenomena on
the landscape. Geographic movement involves changes in space and time, which Often vary
with changes in scale. Movement is an inherent part Of virtually all geographic explanation;
most geographic concerns, topics, and problems involve consideration Of a phenomenon’s
changing spatial patterns at some level. Many geographic phenomena can be described as
spatially dynamic because they dO not take place at a particular location and moment frozen
in time, but rather “unfold in a variety Of sequences” (Knowles 1995, 262). Other
phenomena, however, do not exhibit movement through a time period of Observation and
may be considered spatially static. However, long—term or historical analyses Of these
phenomena over extended periods Of time Often reveal a more dynamic spatial nature.

As part Of geographic explanation, movement has many associated forms and stimuli,
many Of which are based on spatial interaction: “Humans are increasing their levels Of
interaction, in communication, travel, and foreign exchange... People migrate and travel out
Of curiosity, economic or social need, as a response to environmental change, or because they
have been forced to move for Other reasons. Physical processes are also expressions Of
movement — e.g., traveling weather patterns, ocean and wind currents, flowing water, plate
tectonics, and volcanism” (Boehm and Petersen 1994, 217). Spatial interaction can be
conceptualized as all forms Of exchange movement involving people, goods, information,

141

142

and/or ideas between two or more places. All human geographic phenomena, regardless Of
whether they involve individuals, establishments, or settlements, are involved in spatial
interaction processes as a necessary part Of their overall functions. The four integrating
concepts Of spatial interaction are Often taken into consideration simultaneously in order tO
provide a “parsimonious” explanation Of why place—tO—place movement occurs (Lowe and
Moryadas 1975, 11). As geographers, we typically invesrigate particular features of a spatial
interaction or process such as why it has occurred or what multiple effects has it imposed on
Other related phenomena. Process, pattern, and behavior are Often separated and
distinguished in geography by what moves, why it moves, and how it relates to various well
defined subjects Of inquiry such as economics, culture, psychology, and human medicine.
Geographically, we try to understand the dynamic spatiOtemporal patterns and processes Of
human and physical phenomena as they interact with one another.

Geography’s study Of phenomena over space and time is neither new or unique. Since
the passage from the paradigm Of ‘geography—as—chorology’ in the 1950’s, a variety Of
approaches for studying space—time phenomena has evolved from early works in historical
geography demonstrating “geographical change,” to autocorrelation, to Hagerstrand’s
models Of diffusion and time geography (Peuquet 1994, 441). Geographers Often generalize
the interrelationships between time and space by asking “what kinds Of order are exhibited
by geographical information and on what scale Of space and time each operates” (Haggett
and Chorley 1967, 20). Greater emphasis has been placed on this approach in current
geographical inquiry and research as urban growth, agricultural impacts, and global warming
have become critical problems on earth (Peuquet 1996). The need to better understand the
effects of human activities on the natural environment by analyzing patterns Of change at all
geographic scales is now viewed with increasing urgency. For example, in natural resource
management within developed nations, the emphasis is shifting from inventory and
exploitation toward maintaining the long—term productivity of the environment. This

requires interaCtive space—time analyses at multiple scales as a way Of understanding the

143

complex interrelationships Of environmental syStems.

Recent developments in technology—oriented tools have helped in the examination and
analysis Of such dimensionally complex geographic problems. Peuquet and Duan (1995)
have incorporated spatiotemporal capabilities in their recently developed geographic
information system called the Event—based SpatiOTemporal Data Model (ESTDM) that will
ultimately allow temporal modeling and simulation Of geographical processes. Other
geographic information systems are currently using Global Circulation Models (GCMs) to
study climate and ocean dynamics, as well as global warming. Verification and refinement
Of these models require sophisticated analysis of large volumes Of multidimensional data,
particularly the study Of change and patterns Of change through time over the earth, in the
oceans, and through the atmosphere. Related to this research Of dynamics in GIS is the
development Of new computerized techniques for visualization Of phenomena as they move
in three spatial dimensions over time. Research in this area is Of particularly high priority
but has only begun tO be explored and involves using tools such as regular polytope
geometry and ObjeCt—Oriented design tO achieve the needed representational schemes specific
for volumetric and dynamic geography (Peuquet 1996). Spatial visualization techniques,
particularly animation, allow geographers tO incorporate change in space and through time
and have released cartography from the confines Of the two—dimensional page in which it has
been “trapped” over the years (DiBiase et al. 1992; Langran 1992, 163). Peuquet (1996)
urges that a medium—term (concrete benefits in thirty—five years) priority area for
visualization research is to develop a space-time data model that can represent dynamic
processes and spatial interactions in an effective manner that includes multiple scales,
multiple dimensions, and a diverse range of data types.

The presented classification of geographic movement represents such a conceptual
model. Classification, as a descriptive model, is a valuable research tool used in the
dissemination and organization Of complex information. As a way Of comprehending the

underlying principles and generic relationships between Objects and processes, classification

144

is a traditional methodological approach used by geographers as a way Of providing
differentiation in geography (Armand 1965; Haggett and Chorley 1967). Over thirty years
ago, Berry (1964) suggested that classification in geography be conceptualized as a matrix,
with phenomena categorized in terms of their type Of activity, their location, and their time.
The classification Of movement adopts a similar abstraction; spatially dynamic geographic
phenomena are generalized in terms Of their temporal and spatial components. The
framework presents the conceptualization Of how things move and in doing so, presents a
description of the geometry Of geographic movement. Serving as a movement language that
enables one to describe and interpret the changing diStribution Of phenomena in space and
through time, the model accounts for all forms Of geographic movement with twelve unique
classes. Spatially, the shape a phenomenon exhibits as it moves through three—dimensional
space can be interpreted as one Of four basic types: point, linear, areal, or volumetric.
Movements in space occur in one Of three temporal fashions representing the duration and
periodicity Of the occurrence: continuous, cyclical, or intermittent. The classification,
which can be applied on a variety Of spatial scales, from local, to international, to global, on
a variety of temporal scales, from seconds and minutes, to days and weeks, to various
groupings of years, and tO a variety of phenomena including physical, human, and human—
environmental relations, has utility far reaching tO all branches Of social sciences and beyond.
Analysts from other disciplines such as history, anthropology, and sociology can apply the
classification’s generalization Of space and time to their analyses Of varied earth phenomena.
The classification Of geographic movement is important at many levels as a vehicle with
which to strengthen explanation and undersranding Of dynamic geographic phenomena.
First, by breaking a complex process, event, or Observation down into its individual inherent
geographic characteristics Of space and time, the basic principles associated with these
fundamental dynamic properties can be Observed and comprehended. Once each
spatiotemporal component’s contribution to the movement is understood, then conclusions

about the relationships between space and time, as well as among other features associated

145

with the phenomenon, can be drawn. For example, if one can relate the movements
conceptualized with a bouncing ball (i.e., the Object is moving from point to point to point)
to the ideas of interaction that is occurring at each Of these individual points with the surface
Of the earth, then one can apply these same fundamental nOtions about space and time to
pointatO—point phenomena such as human mobility. By building a foundation that is
predicated upon characteristics universal to all geographic phenomena, a more clear
understanding of the spatiOtemporal interrelationships among phenomena can be gained.
For instance, to understand the geography of satellites, one must first comprehend the
movements associated with the revolution of the earth about the sun, as well as the rotation
of the earth on its own axis. Once these dynamics Of the earth can be visualized and
conceptualized in terms of space and time, understanding the associated movements Of a
satellite with respect to the earth is possible, thereby facilitating conclusions as to how and
for what purposes this technology is used.

Understanding the fiindamental spatiotemporal characteristics Of a geographic
phenomenon also allows parallels to previously unlinked phenomena to be drawn. In the
case Of spatial diffusion, the areal spread of phenomena involves the locational transfer Of
people, things, and information and is signified with some recognizable pattern and
organized Structure (Lowe and Moryadas 1975). The diffusion Of gypsy mOth defoliation
spreads across a stand in a similar fashion as urban sprawl encroaches land around a
metropolitan center. Traditionally, the two phenomena are not compared because
geographic understanding of gypsy mOth infestations has often been focused on studying the
correlations between defoliation and landscape variables such as tree type, climate, and soil
composition for predictive and management measures. Urban sprawl studies typically focus
on identifying settlement and urbanization patterns for Observation and planning practices.
However, the dynamic geometry Of these two phenomena are similar and parallel
conclusions about the movement patterns associated with each can be made. For example,

the understanding that features in the landscape such as rivers or mountains may serve as

146

barriers to the movement of defoliation can then be applied to study such ‘spillover’
abatements in spreads of urbanization.

Finally, and most importantly, the classification’s utility extends to the research
community where current and fitture enhancements of computerized techniques dealing
with movements associated with geographic phenomena rely on an extensive knowledge of
space—time dynamics. Increasing attention is being paid to the integrative changes, activities,
and interactions taking place in the human and physical landscapes. Spatiotemporal data are
becoming significantly more tangible in research and educational settings, as the means by
which we compile, store, analyze, and represent data have produced an explosion of dynamic
geographic information. However, many geographic problems are approached, analyzed,
described, and explained in very diverse manners. Any modeling Of spatially dynamic
phenomena should consider the spatiOtemporal complexities associated with various,
previously unrelated, phenomena. Incorporating the components and nuances important to
the understanding of all movements will ensure nOt only that dynamic features will be
handled appropriately for a given studied phenomenon but that the dynamics of other
phenomena influential to its movements are adequately accounted for and can easily be
incorporated into an analysis. For example, in developing computer applications for
effective emergency management procedures related to hurricanes, modeling various space—
time configurations related to the phenomenon, such as the changing locations of the
srorm’s eye, the spatial dimensions of the extent or diameter Of the storm, as well as the rate
and direction of the storm’s movement across water and land are of utmost importance.
However, one must also consider and integrate into the model the concurrent movements Of
the earth’s atmosphere at or near the storm’s eye and path, as the dynamic conditions of the
atmosphere most definitely influence the movement of the hurricane. Also important to
incorporate into this analysis are the associated movements that interaction of the hurricane
on the surface of the earth will cause, such as flooding and evacuation procedures of local

residents. An effective and complete emergency management model of hurricanes should

147

therefore account for all the various integrating movements Of phenomena interacting with
the movement Of the hurricane.

Modeling, analyzing, and representing complex phenomena in a computer
environment is important for furthering explanation, education, and understanding of
complex phenomena and their interactions with one anOther. Creating a knowledge—based
system by which the fiindamental properties associated with all geographic movement are
synthesized and organized in a convenient manner in which the abstract nOtions of space
and time are easily conceptualized will help this cause. Just as cartographic research on color
has defined the significance of various core principles (e.g., saturation, hue, and value) for
effective modeling and execution Of various cartographic applications, the identification of
important and distinctive spatial and temporal characteristics of all spatially dynamic
phenomena will provide researchers, educators, and the like a foundation with which to
strengthen geographic description and explanation. Summarizing the conceptualization Of
movement, the classification serves this purpose well by fiinctioning as an aid in the
description and comprehension Of what is happening in a geographical context and as a
method to relate various geographic phenomena based on the form taken as they occur
through space and time. Understanding geographic movement by interpreting the complex
spatial and temporal characteristics associated with various geographic phenomena may lead
to the consciousness of ideas, interrelationships, and conclusions about spatially dynamic

phenomena within the larger realm Of geography.

LIST OF REFERENCES

 

 

LIST OF REFERENCES

Abler, Ronald, John S. Adams, and Peter Gould. 1971. Spatial Organization: The
Geographer} View of the World. Englewood Cliffs, New Jersey: Prentice—Hall.

Adams, Paul C. 1995. A Reconsideration of Personal Boundaries in Space«Time. Annals
of the Association of American Geographers 85(2):267—28 5.

Allen, James F. 1984. Towards a General Theory of Action and Time. Artificial
Intelligence 23(2): 123—1 54.

Allen, James P., and Eugene Turner. 1996. Spatial Patterns Of Immigrant Assimilation.
Professional Geographer 48(2): 140—1 55.

Anderson, T. Lougenia. 1982. Modeling Time at the Conceptual Level. In Improving
Database Usability and Responsiveness, ed. Peter Scheuermann, 273—297. New York:
Academic Press.

Armand, David L. 1965. The Logic Of Geographical Classifications and Regionalization
Schemes. Soviet Geography: Review and Panslation 6(9):20-38.

Baker, Earl J. 1995. Public Response to Hurricane Probability Forecasts. Profissional
Geographer 47(2): 137—147.

Baker, William L. 1989. A Review of Models of Landscape Change. Landscape Ecology
2(2):]11—133.

Barrera, Renato, Andrew Frank, and Khaled Al-Taha. 1991. Iémporal Relations in
Geographic Information Systems: A Workshop at the University of Maine, Orono,
October 12-13, 1990. Technical Paper, 91—94. Orono, Maine: National Center for
Geographic Information and Analysis.

Bascom, Jonathan. 1993. The Peasant Economy of Refugee Resettlement in Eastern Sudan.
Annals of the Association of American Geographers 83(2):320—346.

Beach, TimOthy. 1994. The Fate Of Eroded Soil: Sediment Sinks and Sediment Budgets
of Agrarian Landscapes in Southern Minnesota, 1851-1988. Annals of the Association
of American Geographers 84(1):5—28.

Behr, Michelle, and Patricia Gober. 1982. When a Residence is not a House: Examining
Residence—Based Migration Definitions. Professional Geographer 34(2):178—184.

Berry, Brian J. L. 1958. A Note Concerning Methods of Classification. Annals of the
Association of American Geographers 48:300—303.

Berry, Brian J. L. 1964. Approaches to Re ional Analysis: A Synthesis. Annals of the
Association ofAmerican Geographers 54 1):2'1 1.

148

 

 

 

149

Berry, Brian J. L. 1972. The Goals of City Classification. Introduction in City
Classification Handbook: Methods and Applications, ed. Brian J. L. Berry with the
assistance of Katherine B. Smith, 1—8. New York: John Wiley and Sons.

Berry, Brian J. L. 1993. Transnational Urbanward Migration, 1830-1980. Annals of the
Association of American Geographers 83(3):389-405.

Bertin, Jacques. 1983. Semiology of Graphics: Diagrams, Networks, Maps. Tianslated by
William J. Berg. Madison, Wisconsin: University Of Wisconsin Press.

Bird, James H. 1989. The Changing Worlds of Geography, A Critical Guide to Concepts and
Methods. New Yorle Oxford University Press.

Blaut, James M. 1972. Space and Process. Chap. 2 in The Conceptual Revolution, ed.
Wayne K D. Davies, 4251. London: University of London Press.

Blewett, William L., and Harold A. Winters. 1995. The Importance of Glaciofluvial
Features Within Michigan’s Port Huron Moraine. Annals of the Association of
American Geographers 85(2):306'3 19.

Boehm, Richard G., and James F. Petersen. 1994. An Elaboration of the Fundamental
Themes in Geography. Social Education 58(4):211—218.

Bradshaw, Michael J., and Nicholas J. Lynn. 1994. After the Soviet Union: The Post-
Soviet States in the World System. Professional Geographer 46(4):439—449.

Buchanan, Janis T., and William Acevedo. 1996. Studying Urban Sprawl Using a
Temporal Database. Geo Infi) Systems 6(7):42—44, 46—47.

Bunge, William. 1966. Theoretical Geography. Lund Studies in Geography Set. C. General
and Mathematical Geography NO. 1. Lund, Sweden: C. W K. Gleerup.

Business Geographies. 1996. BG News. 4(8):12—14, 16-17.

Campbell, David J., and Jennifer M. Olson. 1991. Framework firr Environment and
Development: The Kite. Occasional Paper NO. 10. East Lansing, Michigan: Center for
Advanced Study of International Deve opment.

Carlile, David W, John R. Skalski, John E. Barker, John M. Thomas, and V. I. Cullinan.
1989. Determination of Ecological Scale. Landscape Ecology 2(4):203—213.

Carlstein, Tommy. 1978. Innovation, Time Allocation and Time—Space Packing. Cha .
8 in Timing Space and Spacing Time, Vol. 2: Human Activity and Time Geography, ed?
Tommy Carlstein, Don Parkes, and Nigel Thrift, 146-161. New York: John Wiley
8c Sons.

Casetti, Emilio. 1993. Spatial Analysis: Perspectives and Prospects. Urban Geography
14(6):526-537.

Chorley, Richard J. 1968. Geography and Analog Theory. Chap. 4 in Spatial Analysis, A
Reader in Statistical Geography, ed. Brian J. L. Berry and Duane F. Marble, 42—52.
Englewood Cliffs, New Jersey: Prentice—Hall.

 

 

150

Clark, Tet N. 1972. Urban Typologies and Political Output. Chap. 5 in City
Classigi'ation Handbook: Methods and Applications, ed. Brian J. L. Berry with the
assiStance of Katherine B. Smith, 152-178. New York: John Wiley and Sons.

Comrie, Andrew C. 1994. Tracking Ozone: Air—Mass Trajectories and Pollutant Source
Regions Influencing Ozone in Pennsylvania Forests. Annals of the Association of
American Geographers 84(4):63 5-65 1.

Critchfield, Howard J. 1983. General Climatology. 4th ed. Englewood Cliffs, New
Jersey: Prentice—Hall.

Delcourt, Hazel R., and Paul A. Delcourt. 1988. Quaternary Landscape Ecology: Relevant
Scales in Space and Time. Landscape Ecology 2(1):23—44.

DiBiase, David, Alan M. MacEachren, John B. Krygier, and Catherine Reeves. 1992.
Animation and the Role Of Map Design in Scientific Visualization. Cartography and
Geographic Information Systems 19(4):201—214, 265—266.

Dobson, Jerome E. 1992. Spatial Logic in Paleogeography and the Explanation of
Continental Drift. Annals of the Association ofAmerican Geographers 82(2):]87—206.

Doob, Leonard W 1978. Time: Cultural and Social Anthropological Aspects. Chap. 5
in Timing Space and Spacing Time, Vol. 1: Making Sense of Time, ed. Tommy
Carlstein, Don Parkes, and Nigel Thrift, 56—65. New York: John Wiley and Sons.

Dorling, Daniel. 1992. Stretching Space and Splicing Time: From Cartographic
Animation to Interacrive Visualization. Cartography and Geographic Information
Systems 19(4):215—227.

Dubin, Robert. 1969. Theory Building. New Yorlc The Free Press.

Dyer, James M. 1994. Implications of Habitat Fragmentation on Climate Change—
Induced Forest Migration. Profissional Geographer 46(4):449—459.

Earle, Carville. 1992. Pioneers Of Providence: The An lO—American Experience, 1492—
1792. Annals of the Association of American Geograp ers 82(3):478—499.

Falah, Ghazi. 1996. The 1948 Israeli—Palestinian War and its Aftermath: The
Transformation and De-Signification Of Palestine’s Cultural Landscape. Annals of the
Association of American Geographers 86(2):256—285.

Famighetti, Robert. 1996. The World Almanac and Book of Facts, 1996. Mahwah, New
Jersey: Funk and Wagnalls Corporation.

Forer, P. C., and Helen Kivell. 1981. Space-Time Budgets, Public Transport, and Spatial
Choice. Environment and Planning 13:497—509.

Galloway, J. H. 1996. Botany in the Service of Empire: The Barbados Cane—Breeding
Program and the Revival Of the Caribbean Sugar Industry, 18803—19305. Annals of
the Association of American Geographers 86(4):682—706.

 

 

151

Gates, Paul A., and Karl F. Nordstrom. 1995. A Cyclic Model of Foredune Blowout
Evaluation for a Leeward Coast: Island Beach, New Jersey. Annals of the Association
of American Geographers 85(1): 1—20.

Garner, Barry J. 1967. Models of Urban Geography and Settlement Location. In Models
in Geography, ed. Richard]. Chorley and Peter Haggett, 303—360. London: Methuen.

Gilmour, John S. L., and Stuart M. Walters. 1964. Philosophy and Classification. Vistas
in Botany 4: 1—22.

Gleick, James. 1987. Chaos: Makinga New Science. New York: Viking Press.

Golledge, Re inald G., Valerie Dougherty, and Scott Bell. 1995. Acquiring Spatial
Knowletf e: Survey Versus Route—Based Knowledge in Unfamiliar Environments.
Annals of8 the Association of American Geographers 85(1):]34—158.

Golley, Frank B. 1989. A Proper Scale. Landscape Ecology 2(2):71—72.

Goodchild, Michael F. 1992. Analysis. Chap. 7 in Geography} Inner IVorlds, Pervasive
Themes in Contemporary American Geography, ed. Ronald P. Abler, Melvin G. Marcus,
and Judy M. Olson, 138—162. New Brunswick, New Jersey: Rutgers University Press.

Gould, William T. S., and R. Mansell PrOtherO. 1973. Space and Time Dimensions in the
Stud of Popubtion Mobility in Tropical Afiica. African Population Mobility Project
Wor 'ng Paper No. 1. Liverpool, United Kingdom: University of Liverpool.

Graf, William L., and Patricia Gober. 1992. Movements, Cycles, and Systems. Chap. 11
in Geogra hy’s Inner Worlds, Pervasive Themes in Contem orary American Geography,
ed. Ron d F. Abler, Melvin G. Marcus, and Judy M. O son, 234—254. New
Brunswick, New Jersey: Rutgers University Press.

Graff, Thomas O., and Dub Ashton. 1993. S atial Diffusion OfWal—Mart: Contagious
and Reverse Hierarchical Elements. Profiisional Geographer 46(1):19—29.

Green, Kass, and Brian Cosentino. 1996. Using Satellite Imagery to Detect and Monitor
Forest Change. Geo Infir Systems 6(2):22, 24—26, 28—30.

Griffith, Daniel A., and Anthony C. Lea. 1983. Introduction. In Evolving Geographical
Structures, Mathematical Models and Theories for Space— Time Processes. NATO
Advanced Science Institutes Series D: Behavioral and Social Sciences, NO. 15, ed.
Datilielfp. Griffith and Anthony C. Lea, 1—8. The Hague, Netherlands: Martines
Nij o .

Grigg, David. 1965. The Logic of Regional Sysrems. Annals of the Association of
American Geographers 55:465—49 1.

Guralnik, David B. 1984. Webster’s New World Dictionary of the American Language. 2nd
college ed. New York: Simon and Schuster.

Haggett, Peter. 1967. Network Models in Geo raphy. Chap. 15 in Models in Geography,
ed. Richard J. Chorley and Peter Haggett, 09'668. London: Methuen.

 

 

152

Haggett, Peter. 1975. Simple Epidemics in Human Po ulations: Some Geographical
Aspects of the Hamer—Soper Diffusion Models. C ap. 18 in Processes in Physical and
Human Geography, ed. Ronald Peel, Michael Chrisholm, and Peter Haggett, 373-391.
London: Heinemann Educational Books.

Haggett, Pcter. 1978. Regional and Local Components in Elementary Space—Time
Models. Chap. 2 in Timin Space and Spacing Time, Vol 3: Time and Regional
Dynamics, ed. Tommy Carfstein, Don Parkes, and Nigel Thrift, 19-34. New York:
John Wiley BC Sons.

Haggett, Peter. 1979. Geography, A Modern Synthesis. 3rd. ed. New York: Harper and
Row.

Haggett, Perer. 1990. The Geographer} Art. Cambridge, Massachusetts: Basil Blackwell.

Haggett, Peter, and Richard J. Chorley. 1967. Models, Paradigms and the New
Geography. Chap. 1 in Models in Geography, ed. Richard J. Chorley and Peter
Haggett,19—41. London: Methuen.

Haggett, Peter, and Richard J. Chorley. 1969. Network Analysis in Geography. London:
Edward Arnold Publishers Ltd.

Hall, Forrest G., Donald E. Strebel, and Piers J. Sellers. 1988. Linkin Knowle e
Among Spatial and Temporal Scales: Vegetation, Atmosphere, limate an RemOte
Sensing. Landscape Ecology 2(1):3-—22.

Hardwick, Susan Wiley, and Donald G. Holtgrieve. 1990. Patterns on our Planet:
Concepts and Themes in Geography. New York: Macmillan.

Hardwick, Susan Wiley, and Donald G. Holtgrieve. 1996. Geography firr Educators:
Standards, Themes, and Concepts. Upper Saddle River, New Jersey: Prentice Hall.

Harner, John P. 1995. Continuity Amidst Change: Undocumented Mexican Migration to
Arizona. Professional Geographer 47(4):399-411.

Hartshorne, Richard. 1959. Perspective on the Nature of Geography. Chicago: Rand
McNally.

Harvey, David. 1969. Eprnation in Geography. London: Edward Arnold.

Hawking, Stephen W 1988. A Brief History of Time: From the Big Bang to Black Holes.
New York: Bantam Books.

Hayles, N. Katherine. 1991. Complex Dynamics in Literature and Science. IntroduCtion
in Chaos and Order; Complex Dynamics in Literature and Science, ed. N. Katherine
Hayles, 1'33. Chicago: The University of Chicago Press.

Heirich, Max. 1964. The Use of Time in the Study Of Social Change. American
Sociological Review 29(3):386-397.

Hill, A. David. 1989. Rediscovering Geo raphy: It’s Five Fundamental Themes.
National Association of Secondary Schoo Principals 73(521):1—7.

 

 

 

 

 

153

Holly, Brian P. 1978. The Problem of Scale in Time—Space Research. Chap. 1 in Timing
Space and Spacing Time, Vol 3: Time and Regional D namics, ed. Tommy Carlstein,
Don Parkes, and Nigel Thrift, 5—18. New York: JO n Wiley 8c Sons.

Hudson, John C. 1972. Geographical Dijfitsion Theory, Studies in Geography, NO. 19.
Evansron, Illinois: Northwestern University.

Hudson, John C. 1988. North American Origins of Middlewestern Frontier Populations.
Annals of the Association of American Geographers 78(3):395—413.

Hudson, John C. 1992. Scale in Space and Time. Chap. 13 in Geogra hy’s Inner IVorlds,
Pervasive Themes in Contem orary American Geography, ed. Ron d F. Abler, Melvin
G. Marcus, and Judy M. O son, 280—297. New Brunswick, New Jersey: Rutgers
University Press.

Ivy, Russell L. 1995. The Restructuring Of Air Transport Linkages in the New Empire.
Professional Geographer 47(3):280—288.

Jackson, Nancy L. 1995. Wind and Waves: Influence of Local and Non—Local Waves on
Mesoscale Beach Behavior in Estuarine Environments. Annals of the Association of
American Geographers 85(1):21—37.

Jakle, John A., Stanley Brunn, and Curtis C. Roseman. 1976. Human Spatial Behavior: A
Social Geography. North Scituate, Massachusetts: Duxbury Press.

Johnston, Ronald J. 1968. Choice in Classification: The Subjectivity of Objective
Methods. Annals of the Association ofAmerican Geographers 58(3):575-589.

Johnston, Ronald J. 1970. Grouping and Regionalizing: Some Methodological and
Technical Observations. Economic Geography 46(2 Supplement):293—305.

Johnston, Ronald J. 1986. Philosoph and Human Geography: An Introduction to
Contemporary Approaches. B timore, Maryland: Edward Arnold.

Johnston, Ronald J. 1991. Geography and Geographers: Anglo—American Human Geography
Since 1945. New York: Edward Arnold.

Johnston—Anumonwo, Ibipo. 1992. The Influence Of Household T on Gender
Differences in Work Trip Disrance. Profissional Geographer 44 2):161—169.

Joint Committee on Geographic Education. 1984. Guidelines for Geographic Education.
The Association of American Geographers and the National Council for Geographic
Education.

Jordan, Lawrie E. III. 1996. High-Resolution Imagery for Business Decision Support.
Geo Info Systems (Supplement):24‘27.

Jordan, Terry G., and Lester Rowntree. 1986. The Human Mosaic, A Thematic
Introduction to Cultural Geography. 4th ed. New York: Harper and Row, Publishers.

Kaplan, Abraham. 1964. The Conduct of Inquiry: Methodology fir Behavioral Science. San
Francisco: Chandler Publishing.

 

 

154

Kellerman, Aharon. 1989. Time, Space, and Society: Geographical Societal Perspectives.
Netherlands: Kluwer Academic Publishers.

Kelmelis, John A. 1991. Time and Space in Geographic Information: Toward a Four—
Dimensional Spatio— Pmporal Data Model Ph.D. diss., Pennsylvania State University.

Kennedy, Frances H. 1990. The Civil War Battlefield Guide. Boston, Massachusetts:
Houghton Mifliin.

Knapp, Paul A., and Perer. T. Soule. 1996. Vegetation Change and the Role of
Atmospheric C02 Enrichment on a Relict Site in Central Oregon: 1960—1994.
Annals of the Association of American Geographers 86(3):387—41 1.

Knight, Cheryl S. 1994. The Five Fundamental Themes 0 Geography, Geograph Across the
Curriculum: A Iéacher’s Activity Guide (Grades K—5 . Boone, North Caro ina:
Parkway Publishers.

Knowles, Anne K. 1995. Immigration Trajectories Through the Rural—Industrial 'Ii'ansition
in Wales and the United States, 1795-1850. Annals of the Association ofAmerican
Geographers 85(2):246—266.

Krakauer, Jon. 1996. Geologisrs Worry About Dangers of Living ‘Under the Volcano.’
Smithsonian 27(4):33—40.

Langran, Gail. 1989a. A Review Of Temporal Database Research and its Use in GIS
Applications. International journal of Geographical Infirrmation Systems 3(3):215—232.

Langran, Gail. 1989b. Time in Geographic Information Systems. Ph. D. diss., University of
Washington.

Langran, Gail. 1992. Time in Geographic Information Systems. Philadelphia: Taylor 8c
Francis.

Leach, Edmund R 1961. Two Essays Concerning the Symbolic Representation of Time.
Chap. 6 in Rethinking Anthropology. London School of Economics Mono raphs on

Social Anthropology NO. 22, ed. Edmund R. Leach, 124—136. London: he Athlone
Press.

Lee, Dong Ok. 1995. Koreatown and Korean Small Firms in Los Angeles: Locating in the
Ethnic Neighborhoods. Professional Geographer 47(2):184—195.

Lee, Jeffrey A., and Vatche P. Tchakarian. 1995. Ma nitude and Frequency of Blowing
Dust in the Southern High Plains of the United rates, 1947-1989. Annals of the
Association ofAmerican eographers 85(4):684-693.

Lee, Jay, Ligang Chen, and Shih—lung Shaw. 1994. A Method for the Exploratory Analysis
Of Air ine Networks. Professional Geographer 46(4):468—477.

Lewis, Peter W 1965. Three Related Problems in the Formulation Of Laws in Geography.
Professional Geographer 17(5):24—27.

Lightman, Alan. 1993. Einstein’s Dreams. New York: Panthem Books.

 

 

 

 

 

155

Lloyd, Robert, David Patton, and Rex Cammack. 1996. Basic—Level Geographic
Categories. Professional Geographer 48(2): 181—194.

Lowe, John C., and S. Moryadas. 1975. The Geography of Movement. Prospect Heights,
Illinois: Waveland Press.

Lucas, John R. 1973. A Treatise on Time and Space. London: Methuen.

MacEachren, Alan M. 1992. Visualizing Uncertain Information. Cartographic Perspectives
13(Fall):10-19.

MacEachren, Alan M., Barbara J. Buttenfield, James B. Campbell, David W DiBiase, and
Mark Monmonier. 1992. Visualization. Chap. 6 in Geography’s Inner IVorMs,
Pervasive Themes in Contemporary American Geography, ed. Ronald F. Abler, Melvin
G. Marcus, and Judy M. Olson, 99—137. New Brunswick, New Jersey: Rutgers
University Press.

Manson, Gary A., and Richard E. Groop. 1996. Ebbs and Flows in Recent U. S. Interstate
Migration. Profissional Geographer 48(2): 1 56166.

Martin, David. 1991. Geographic Information Systems and Their Socioeconomic Applications.
New York: ROUtledge.

 

Martin, Ron L., Ni el J. Thrift, and Robert]. Bennett. 1978. Introduction. In Towards
the Dynamic nalysis of Spatial Systems, ed. Ron L. Martin, Nigel J. Thrift, and
Robert J. Bennett, 1—6. United Kingdom: Pion.

Marwick, Arthur. 1984. The Nature of History. New York: Garland Publishing.

McDermOtt, Drew. 1982. A Temporal Logic for Reasoning About Processes and Plans.
Cognitive Science 6(2): 101—1 55.

McHugh, Kevin E., TimOthy D. HO an, and Stephen K. Happel. 1995. Multiple Residence
and Cyclical Migration: A Li e Course Perspective. Profissional Geographer
47(3):251—267.

McHu h, Kevin E., and Robert C. Mings. 1996. The Circle of Mi ration: Attachment to
P ace in Aging. Annals of the Association of American Geograp ers 86(3):530-550.

McPherson, James M. 1988. Battle Cry of Freedom: The Civil IVar Era. New York:
Oxford University Press.

Medley, Kimberly E., Mark J. McDonnell, and Steward T. A. Pickett. 1995. Foresr—
Landscape Sgructure Along an Urban-TO-Rural Gradient. Profissional Geographer
47(2):159—1 8.

Meentemeyer, Vernon. 1989. Geographical Perspectives of Space, Time, and Scale.
Landscape Ecology 3(3&4):163«173.

 

156

Meyer, William B., Derek Gregory, Billie L. Turner 11, and Patricia F. McDowell. 1992.
The Local Global Continuum. Cha . 2 in Geography} Inner Worlds, Pervasive Themes
in Contemporary American Geograph , ed. Ron F. Abler, Melvin G. Marcus, and Judy
M. Olson, 25 5279. New Brunswick, New Jersey: Rutgers University Press.

Mitchelson, Ronald L., and James O. Wheeler. 1994. The Flow of Information in a
Global Economy: The Role of the American Urban System in 1990. Annals of the
Association ofAmerican Geographers 84(1):87—107.

Mitchneck, Beth, and David Plane. 1995. Migration Patterns During a Period of Political
and Economic Shocks in the Former Soviet Union: A Case Study of Yatoslavl’
Oblast. Professional Geographer 47(1):17-30.

Mock, CaryJ., and Paul. A. Kay. 1992. Avalanche Climatology Of the Western United
States, with an Emphasis on Alta, Utah. Professional Geographer 44(3):307‘318.

Moore, Tytel G. 1994. Core-Periphery Models, Regional Planning Theory, and
Appalachian Development. Professional Geographer 46(3):316«331.

Moran, Warren. 1993. The Wine Appellation as Territory in France and California.
Annals of the Association of American Geographers 83(4):694—717.

Morrill, Richard L. 1968. Waves of Spatial Diffusion. journal of Regional Science
8(1):] 18.

Morrill, Richard L. 1978. Geography as Spatial Interacrion. In A Man for all Regions:
The Contributions of Edward L. Ullman to Geography. Papers Of the Fourth Carolina
Geographical Sym osium, Studies in Geography NO. 11, ed. John D. Eyre, 1629.
Chapel Hill, Nort Carolina: University Of North Carolina.

Morse, Steans A. 1996. Phenomena, Comment and Notes. Smithsonian 27(1):26, 28—29.

Nedobity, Wolfgang. 1986. Classification, Artificial Intelligence and Cognitive
Psychology. In Classification as a Tool of Research. Proceedings Of the 9th Annual
Meeting of the Classification Society, ed. Wolfgang Gaul and Martin Schader,
345—352. New York: Elsevier Science.

Orme, John E. 1978. Time: Psychological As ecrs. Chap. 6 in Timing Space and Spacing
Time, Vol 1: Making Sense of Time, ed. ommy Car stein, Don Parkes, and Nigel
Thrift, 6675. New York: John Wiley and Sons.

Parker, Albert J. 1993. Strucrutal Variation and Dynamics of Lodgepole Pine Forests in
Lassen Volcanic National Park, California. Annals of the Association ofAmerican
Geographers 83(4):613—629.

Parkes, Don, and Nigel Thrift. 1980. Times, Spaces, and Places: A Chronogeographic
Perspective. New York: John Wiley and Sons.

Patton, David K. 1997. The Effects of Prior Knowledge on the Learning of Categories of
Maps. Professional Geographer 49(1):]26-136.

 

 

157

Peuquet, Donna J. 1977. Raster Processing of Spatial Data in Geo raphic Infbrmation Systems.
Ph.D. diss., State University of New York (SUNY) at Bu alO.

Peuquet, Donna J. 1984. A Conceptual Framework and Comparison Of Spatial Data
Models. Cartographica 21(4):66—113.

Peuquet, Donna J. 1994. It’s About Time: A Conceptual Framework for the
Representation of Temporal Dynamics in Geo taphic Information Systems. Annals
of the Association of American Geographers 84(3 :441—461.

Peuquet, DonnaJ. 1996. Extensions to Geographic Representation. Paper presented as part
of the University Consortium firr Geographic Information Science (UCGIS) Virtual
Seminar discussion. [Online] Available
http://www.ncgia.ucsb.edu/orhcu’ucgl.,,’ ' hrml

Peuquet, Donna J., and Niu Duan. 199 5. An Event—Based Spatio—Temporal Data Model
(ESTDM) for Temporal Analysis Of Geographic Data. International journal of
Geographical Infirmation Systems 9(1):7—24.

Phillips, Jonathan D. 1993. Biophysical Feedbacks and the Risks Of Desertification.
Annals of the Association of American Geographers 83(4):630—640.

Pond, Doris. 1992. Gypsy on the Move. American Forests 98(3&4):21—24, 65.

Pope, Gregory A., Ronald I. Dorn, and John C. Dixon. 1995. A New Conceptual Model
for Understanding Geographical Variations in Weathering. Annals of the Association
of A merican Geographers 85(1):38—64.

Porush, David. 1991. Fictions as Dissipative Structures: Prigogine’s Theory and
Postmodernism’s Roadshow. Cha . 3 in Chaos and Order; Complex D namics in
Literature and Science, ed. N. Ka erine Hayles, 54—84. Chicago: T e University of
Chicago Press.

Pred, Allan. 1977. The Choreography Of Existence: Comments on Hagerstrand’s Time—
Geography and its Usefulness. Economic Geography 53(2):207—221.

Prince, Hugh. 1978. Time and Hisrotical Geography. Chap. 2 in Timing Space and
Spacing Time, Vol 1: Making Sense 0 Time, ed. Tommy Carlstein, Don Parkes, and
Nigel Thrift, 17—37. New York: JO n Wiley and Sons.

Raphael, Marilyn, and Gerald Mills. 1996. The Role of Mid—Latitude Pacific Cyclones
in the Winter Precipitation Of California. Professional Geographer 48(3):251-262.

Rees, Philip H. 1972. Problems Of Classi ing Subateas Within Cities. Chap. 10 in City
Classification Handbook: Methods an Applications, ed. Brian J. L. Berry with the
assistance of Katherine B. Smith, 265—330. New York: John Wiley and Sons.

Rescher, Nicholas, and Alasdair Urquhart. 1971. Pmporal Logic. New York: Springer—
Ver ag.

Robinson, Roget. 1976. Ways to Move, The Geography of Networks and Accessibility. New
York: Cambridge University Press.

 

158

Rollins, William H. 1995. Whose Landscape? Technology, Fascism, and
Environmentalism on the National Socialist Autobahn. Annals of the Association of
American Geographers 85(3): 494— 520

Roseman, Curtis C. 1971. Migration as a S atial and Temporal Process. Annals of the
Association ofAmerican Geographers 61(3 :—589 598.

Roseman, Curtis C. 1977. Changing Migration Patterns Within the United States. Resource
Papers for College Geography No. 77- 2. Washington, DC: Association of Amer1can
Geographers.

Roseman, Curtis C. 1992. Cyclical and Polygonal Migration in a Western Context. In
Community, Society, and Migration Noneconomic Migration in America, ed. Patrick C.
Jobes, Wil1am F. Stinnet, and JohnM. Wardwell, 3345 Lanham, Maryland:
University Press Of America.

Schlegel, Richard. 1968. Time and the Physical World New York: Dover Publications.

Shannon, GaryW, and Gerald F. Pyle. 1989. The Origin and Diffusion of Aids: A View
from Medical Geography Annals of the Association ofAmerican Geographers
79(1): 124.

Sinton, David F. 1978. The Inherent Structure of Information as a Constraint to Analysis:
aped Thematic Data as a Case Study. In Harvard Papers tn Geographic
Ina ormation Systems, vol. 7, ed. Geoffrey Dutton, 1— 17. Redding, Massachusetts:
Wesley Addison.

Soreide, Nancy. 1996. What is an El Nino? [Online] Available
http://www.pmel.noaa.govhoga-taO/el-ninO—storyhtml, U. S. Dept. of Commerce.

Stewart, Dona. J. 1996. Cities 1n the Desert: TheE Egyptian New—Town Program. Annals
of the Association ofAmerican Geographers 86(3): 9 480.

Stolle, HansJ. 1987. Michigan Tornadoes: 1930-1969 and 1970—1980. East Lakes
Geographer 22: 88— 102.

Strahler, Arthur N, and Alan H. Strahler. 1984. Elements of Physical Geography. 3rd ed.
New York. John Wiley and Sons.

Sugiulra, Yoshio.1993Satial Diffusion of Japanese Electric Power Companies, 1887—
9106. A Discrete C oice Modeling. Annals of the Association of American
Geographers 83(4): 641 —.655

Thrift, Nigel. 1977. An Introduction to Time Geography. Concepts and Techniques in
Modern Geography NO. 13. Norwich, United Kingdom: Geo Abstracts.

Thurber, Dale K., William R. McClain, and Robert C. Whitmore. 1994. Indirect Effects
Of Gypsy Moth Defoliation on Nest Predation. The journal of Wildli e Management
58(3): 493— 500.

Tinline, Rowland 1972. A Simulation Study of the 1967- 8 Foot—and— Mouth Epizootic in
Great Britain. Ph. D. diss., University Of Bristol.

 

159

Tobler, Waldo. 1988. Resolution, Resampling, and all that. Chap. in Building Databases
fbr Global Science, ed. Helen Mounsey and Roger Tomlinson, 129-137. Philadelphia:
Taylor 86 Francis.

Tuan, Y1—Fu. 1978. Space, Time, Place: A Humanistic Frame. Chap. 1 in Timing Space
and Spacing Time, Vol 1: Making Sense of Time, ed. Tommy Carlstein, Don Parkes,
and Nigel Thrift, 7—16. New York: John Wdey and Sons.

Tumer, Monica G. 1990. Spatial and Temporal Analysis of Landscape Patterns.
Landscape Ecology 4(1):21—30.

Turner, Monica G., Virginia HnDale, and Robert H. Gardner. 1989. Predicting Across
Scales: Theory Development and Testing. Landscape Ecology 3(3&4):245-252.

Ullman, Edward L. 1956. The Role of Transportation and the Bases for Interacrion. In
Man’s Role in Changing the Face of the Earth, ed. William L. Thomas Jr., 862—880.
Chicago: University of Chicago Press.

US Census Bureau. 1997. TIGER The Coast to Coast Digital Map Database! [Online]
Available http://www.census.gov/geo/www/tiger/.

Vandsemb, Berit H. 1995. The Place of Narrative in the Study of Third World
Migration: The Case of Spontaneous Rural Migration in Sri Lanka. Professional
Geographer 47(4) :41 1—425.

Vasiliev, Irinia. 1993. Categorizing the Cartographic Representation of Time. Geneseo, New
York: State University of New York at Geneseo. Photocopied.

Veeck, Gre ory, Li Zhou, and Gao Ling. 1995. Terrace Consrrucrion and Productivity on
Loess Soils in Zhongyang County, Shanxi Province, PRC. Annals of the Association
of American Geographers 85(3):450—467.

Weibel, Robert. 1995. Introduction: Map Generalization in the Context of Digital
Sysrems. Cartography and Geographic Information Systems 22(4):259—263.

Weissert, Thomas P. 1991. Representation and Bifurcation: Borges’s Garden of Chaos
[gynamics Chap. 10 in C aos and Order; Complex Dynamics in Literature and Science,
e . N. Katherine Hayles, 223—243. Chicago: The University of Chicago Press.

Wheeler, James 0., and Ronald L. Mitchelson. 1989. Information Flows Among Major
Metropolitan Areas in the United States. Annals of the Association of American
Geographers 79 (4): 523—543.

Willmott, Cort J., and Gary L. Gaile. 1992. Modeling. Chap. 8 in Geography} Inner
Worlds, Pervasive Themes in Contem orary American Geography, ed. Ronald F. Abler,
Melvin G. Marcus, and Judy M. 0 son, 163-186. New Brunswick, New Jersey:
Rutgers University Press.

Wilson, James L. 1993. Mapping the Geographical Diffiasion of a Finnish Smallpox
Epidemic from Historical Population Records. Professional Geographer
45(3):276—286.

 

160

Winkler, Julie A. 1992. Regional Patterns of the Diurnal Properties of Heavy Hourly
Precipitation. Professional Geographer 44(2):127—146.

Wolch, Jennifer R., Afsaneh Rahimian, and Paul Koegel. 1993. Daily and Periodic
Mobility Patterns of the Urban Homeless. Profissional Geographer 45(2):]54-169.

Woltemade, Christopher J. 1994. Form and Process: Fluvial Geomorphology and Flood—
F low Interaction, Grant River, Wsconsin. Annals of the Association ofAmerican
Geographers 84(3):462479.

Wood, Wlliam B. 1994. Forced Migration: Local Conflicts and International
Dilemmas. Annals of the Association of American Geographers 84(4):607—634.

Zweifler, Mark 0., Michael A. Gold, and Robert N. Thomas. 1994. Land Use Evolution
in Hill Regions of the Dominican Republic. Professional Geographer 46(1):39—53.

 

 

"lillllllllllll'llllli