Spatial diffusion analysis of commercial land-use changes associated with the Jacksonville, Florida, International Airpo...


Material Information

Spatial diffusion analysis of commercial land-use changes associated with the Jacksonville, Florida, International Airport, 1965-1976
Physical Description:
ix, 111 leaves : ill., maps ; 28 cm.
Adedibu, Afolabi Adegbite, 1945-
Publication Date:


Subjects / Keywords:
Land use -- Mathematical models   ( lcsh )
Airports -- Environmental aspects   ( lcsh )
Sol, Utilisation urbaine du -- Floride -- Jacksonville   ( rvm )
Économie urbaine   ( rvm )
Urbanisme -- Floride -- Jacksonville   ( rvm )
Changement social   ( rvm )
Développement économique -- Aspect social   ( rvm )
Geography thesis Ph. D   ( lcsh )
Dissertations, Academic -- Geography -- UF   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis--University of Florida.
Includes bibliographical references (leaves 103-110).
Additional Physical Form:
Also available online.
General Note:
General Note:
General Note:
Photocopy of typescript. Ann Arbor, Mich. : University Microfilms International, 1978.--22 cm.
Statement of Responsibility:
by Afọlabi Adegbite Adedibu.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 04164737
lcc - HD211.F6 A33 1978
System ID:

Table of Contents
    Title Page
        Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
        Page iv
        Page v
    List of Figures
        Page vi
        Page vii
        Page viii
        Page ix
    Chapter 1. Spatial diffusion as a framework for land-use change analysis
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
    Chapter 2. Analyzing land-use changes: A review of past research directions
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
    Chapter 3. Spatial diffusion theory, land-use changes and airports
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
    Chapter 4. Problem parameters and the methodological design
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
    Chapter 5. A comparison of the simulated and observed land-use changes
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
    Chapter 6. Generalizations, limitations, and conclusions
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
    Biographical sketch
        Page 111
        Page 112
        Page 113
Full Text







No research is ever done without the assistance of other people. This research is no exception. I wish to express my profound gratitude to Dr. C.W. Spurlock, who as my advisor and the chairman of my committee, was always encouraging and directing me on the subject-matter. As a result of his skillful counseling, I was able to narrow the focus of the research to a tractable topic. without his cooperation and guidance, this research would not have been finished in time. Dr. V.H. Hetrick assisted me in preparing and modifying the computer simulation program used in the study; to her many thanks are due.

I am also indebted to the other members of my committee, Professor Shannon McCune, Dr. C. Cross, Dr. T.D. Boswell and Mr. John Alexander for their invaluable corrections and suggestions.

My thanks also go to Mr. Chris Beale, the Assistant

Director of the Jacksonville, Florida, International Airport for his time and for supplying the needed material from his office. I desire also to acknowledge the encouragement of Dr. 5.0. Omotoso during the critical period of writing this report.

The people to whom I owe thanks must also include Dr. D.F. Chichester, who spent many hours editing the manuscripts


and the typed copy of the report.

Finally, I cannot imagine the completion of this

research without the patience, understanding and sacrifice of my family. My thanks to you all.



ACKNOWLEDGEMENTS ........................................ i

LIST OF FIGURES ...............

ABSTRACT ............... ...........
USE CHANGE ANALYSIS ........................

introduction .............................. 1
Purpose of This Study ...................... 4
An Airport as an Initiator of Land-Use 4
Changes ............................. 6
The Research Hypothesis ....................
Chapter Organization ....................... 7
Notes ...................................... 9

PAST RESEARCH DIRECTIONS ................... 12

Introduction ............................... 12
The Land Inventory Method .................. 13
Some Other Approaches in the Study of
Land-Use Changes ....................... 14
Income as a Measure of Land-Use Changes .... 16
Population and Employment Twin Variables
in Land-Use Change Analysis ............ 16
Socio-Economic Classification for LandUse Changes ............................ 18
Distance Factor in Land-Use Changes ........ 19
Land ownership Factor in Land-Use
Changes ................................ 22
Land value as Indicator of Land-Use
Changes ................................. 23
other Variables in Land-Use Change Studies 24 Notes ...................................... 27

AND AIRPORTS ............................... 31

Overview of Chapter ........................ 31
Spatial Diffusion: Its Validity and
Applications ............................ 31




Contagious Diffusion .................... 32
Hierarchical Diffusion ....... 33
Contagious-Hierarchical Diffusion ....... 35
Moderators of the Adoption Process
in Diffusion ......................... 36
Diffusion Axioms Tailored to Fit LandUse Changes ............................. 38
Airports as Initiators of Land-Use Changes 40 Notes ...................................... 48

DESIGN ..................................... 53

Overview of Chapter ........................ 53
Jacksonville International Airport: An
Introduction to the Study Area .......... 53
The Need for a New Airport in
Jacksonville ............................ 55
The Research Hypothesis .................... 61
The Research Methodology ................ 64
The Monte-Carlo Simulation Model ........ 64 Mean Demand Field ....................... 66
Data Collection Procedures .............. 67
Conversion from Nominal to Ordinal Data 68
Comparing the Simulation With
Observed Pattern ..................... 70
Notes ...................................... 71

OBSERVED LAND-USE CHANGES .................. 73

Overview of Chapter ...................... It. 73
The Simulation of Change Over Time ......... 73
Spatial Distribution of the simulated
Commercial Land-Use Changes ............. 82
Comparison of the Observed Spatial
Pattern with the Simulation Spatial
Pattern .................... 88
Some Conclusions from This Chap e'r*-.*.-'.** 91 Notes ...................................... 93




CONCLUSIONS ............................... 94

Generalizations ........................... 94
Limitations in Applying Diffusion Theory
to Land-Use Changes .................... 95
Concluding Statement ...................... 96
Notes ..................................... 98
APPENDICES ............................................ 100

BIBLIOGRAPHY ........................................... 103

BIOGRAPHICAL SKETCH .................................... ill





COUNTY ..................................... 56

STUDY AREA ................................ 60

OF CHANGE, 1965-1976 ...................... 74

1965-1976 ................................. 76

TO COMMERCIAL USE, 1965-1976 .............. 78

CHANGES ................................... 80

WITHOUT CHARACTERISTICS ................... 83

1965-1976 ................................. 84

CHARACTERISTICS, 1965-1976 ................ 85,

1965-1976 ................................. 86

CHANGES, 1965-1976 ........................ 89

IN THE STUDY AREA ......................... 90


Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


Afolabi Adegbite Adedibu

December, 1977

Chairman: Professor Carl W. Spurlock Major Department: Geography

The processes associated with land-use changes are so complex that deterministic-type modeling appears to be overly restrictive, so that simulation in a stochastic framework may be better suited to modeling the complexities involved. This research was designed to test the conceptual and methodological applicability of the spatial diffusion processes for explaining commercial land-use changes that were triggered by the construction of a major airport. The Jacksonville, Florida, International Airport and the surrounding area served as the test case. It was hypothesized that parcels of land would at first change to commercial uses near the airport, then expand outward following the spatial diffusion processes.

In order to evaluate the research hypothesis, a diffusion simulation using the Monte Carlo model was


accomplished. Instead of the traditional Mean Information Field (MIF), a Mean Demand Field (DIDF) served as the probabilistic framework in which the Model operated. The MDF was based on the probability of any parcel of land (not already having a commercial use) chancring to a commercial use. Parcel probabilities were assigned according to the attributes of the parcel, specifically: distance from the airport, accessibility of the parcel to a major route, ownership of the parcel, topography and zoning of the parcel. Seventy-one parcels of land changed to commercial uses during the study period.

The research hypothesis was empirically confirmed in two steps. First, it was shown that land-use changes, simulated over time, were similar to the observed process of land-use changes. Second, through cartographic analysis of the spatial distribution of land-use changes, it was demonstrated that, under given conditions, observed land use changes could be simulated. Since the analysis of the graphs and maps supported the hypothesis of the study, it was concluded that the changes in commercial land-use associated with the Jacksonville, Florida, International Airport can be partially explained through spatial diffusion theory.

Several limitations concerning this application

should be noted. One limitation is that the results cannot be transferred to include all land-use change situations,


because of the dependence upon a growth point to attract complementary land uses. Secondly, the assumption of a closed system, that does not accept input from external factors, can only partially explain the real situation. Further, there is an element of subjectivity in the data conversion, which may create problems for researchers and the use of resistance variables in simulation modeling has not been empirically verified.

Up to this time, very few comprehensive land-use models have been developed. There are many problems in land-use studies, but it is only by integrating parts of related theories, as was done in this study, that researchers can expect to achieve a thorough understanding of the process involved in land-use changes.





Spatial diffusion is a concept used to describe and analyze certain processes by which phenomena spread through space and over time. The concept was introduced to the social sciences by anthropologists in the nineteenth century who were concerned with cultural diffusion across societal boundaries.1 Following its use by anthropologists, the concept of diffusion was adopted by numerous other disciplines these include the medical professions, psychology, sociology and, in particular, geography.

Since the introduction of the diffusion concept to geography by Hqgerstrand, geographers have employed this device to study many types of spatial phenomena.2 Major applications of this concept have centered on innovation, marriage, migration, the spread of diseases and adoption of new crops.3 Such applications of the diffusion process to various subjects of interest by geographers, as well as by researchers in other disciplines, suggest that the diffusion concept could be applied to any research area which involves changes through space and over time. For example, the concept



of spatial diffusion has been successfully transferred to technological changes in the manufacturing sector.4 Diffusion of a new technology over time and geographic space in this case is a process of imitation of the innovation (technology) by potential users, wherein information about the technology is the base for the changes in technology.

If information is the basis for the change, any phenomenon affected by information can change spatially over time and hence follow the diffusion process. The analysis and comprehension of the complexity involved in land-use changes may also require-among other things-information which pertains to the nature of the demand for and characteristics of the activities on land.5

Literature about land-use changes dates back to the Eqyptian civilization, but intensive studies of land-use changes can be traced back only to the nineteenth century.6 Since that time, however, methods have been devised to study land-use changes associated with rural areas, urban areas, various types of establishments and settlements of all kinds. In the majority of these studies the inventory of the past and present land uses was regarded as the basic method of achieving a meaningful understanding of the process of land-use changes.

While such inventories are useful, a notable weakness is that the processes that impel the changes are not identified. In recent years, emphasis has shifted from the inventory method to quantitative analysis (particularly correlation analysis) in which the idea is to link land-use changes statistically with other variables. These current efforts are


not fully satisfactory, however, because they link together only some factors of land-use changes and do not necessarily explain why they are related. Such studies lay emphasis on one core factor like distance, population, land value, employment, accessiblityfor example. Such factors are not the only ones that produce the land-use form; hence they can give only partial explanation of the process.

Some factors which also may have effects on the process of land-use changes, such as information and land attributes, are often neglected. Such studies that lay emphasis on one core factor dealt with the form of land uses and the processes that produce certain forms were often ignored. If the processes that cause land-use changes to occur are not understood, it remains difficult to predict future land-use changes. Alexander realized this deficiency and has introduced an alternative model of land-use analysis which is based on information. 7 The information analysis model of land uses comprises various ideas which center on grouping land uses by their attributes, the amount of information received about the land, and the number of people involved in the decision-makings that cause changes. The model developed by Alexander, however, has a weakness in that the data used in building it are subjective in nature.

Despite such progress in methodology made by Alexander and many other researchers of land-use changes, difficulties in understanding land-use change processes remain. Clearly, alternative models are needed. Such models should be dynamic and should be able to identify the process that creates the spatial form that most of present land-use change studies describe.


Purpose of Study

The purpose of this research is to apply diffusion theory to the analysis of land-use changes in order to evaluate the theoretical and methodological applicability. More specifically, the focus is on modeling commercial land-use changes that are proximal to a major international airport. An extensive search of the literature on land-use studies does not reveal any previous applications of this type, although it has been suggested by researchers in the past.8 If the applicability of diffusion models to understanding land-use changes can be demonstrated, it would provide a valuable tool to use in the quest for more efficient land-use planning.

An Airport as an Initiator of Land-Use Changes

In order for diffusion to occur, there must be a point (or points) of origin from where the process diffuses outward or where it begins to "hop" from one point to another. Such a characteristic, however, is not unique to the diffusion concept. Most of the past land-use studies that have deduced land-use theories also have origins.9 Many land-use theories have used the Central Business District (CBD) as their origin. 10 The CBD is, however, not the only place that can be used. Major industries, major roadsand many other landmarks can serve the purpose.

A major difference between diffusion theory and previous land-use theories is that diffusion theory is dynamic in

operation while the former ones tend to be static. They are. static


in that the variables used are assumed constant over time and space. Also, temporal variables are not often considered. When political variables are considered they are not longitudinally examined. Population, however, varies from place to place and time to time, also topography is not even over a given region and government policies change over time. The changes that may occur over time and in space render the static models unsatisfactory; hence, there is a need for dynamic models which consider changes over time and in space.

For this study, an airport is used as an origin. An

airport is chosen because of its increasing importance in the movement of goods, people, and services within and between cities. Because of these functions airports have become initiators of land-use changes and can be regarded as the origin of a diffusion process. Evidence of the airport's ability to serve as initiator of land-use changes is shown in many studies that have been conducted on the impacts of airports in various communities.11 The ability of the airport to attract changes in land use and its efficiency in generating traffic which may be enough to begin a growth point has lured many cities and communities to demand construction of an airport to the extent that some communities of about 5,000 people now have airports or are planning to have one. For example, there were about 11,000 airports in the United States in 1970, and by 1976, the number had reached over 12,000.12 This addition is an increase of about 7.2 percent in six years.


Not only will airports continue to increase in number,

but construction of such facilities near communities will continue to have impact on the surrounding areas. Many attempts have been made to indicate the impact of airports. Of note is the economic impact which has received considerable attention. Relatively little attention has been paid to the impact of airports as initiators of land-use changes. Whenever reference is made to the impact on land uses, it is always an insignificant portion of the economic impact.13 Because very little attention has been paid to this subject, and because other impacts that airports have on the surrounding areas depend on the changes that occur in land uses, in-depth analysis is needed.

The meager work that has been done on airport land-use changes demonstrates that an airport repels residential construction but attracts commercial establishments.14 Since it has been demonstrated that the airport attracts commercial activities, the focus of the present study is to investigate whether changes in commercial uses of land proximal to an airport follow a spatial diffusion process.

The Research Hypothesis

It is hypothesized that commercial land uses that are attracted to airports follow a diffusion process. According to this view, changes in land uses start with the airport and expand outward, or changes begin with the airport and "hop"


to another area along the major routes serving the airport. The rationale for the hypothesis is given in detail in the following chapters.

In order for the hypothesis to be valid for changes in commercial land use around an airport, a diffusion simulation model must fulfill three objectives: (1) the model should be able to simulate patterns of past and present commercial land uses around an airport; (2) the diffusion process and supportive literature should be able to explain the process that forms the simulated pattern of commercial uses; and (3) extension of the process should be able to generate the pattern of commercial land uses anticipated in the immediate future around an airport.

If these three objectives are achieved in the study, it will enable land-use planners, urban geographers, and regional and land economists to gain a greater understanding of the way in which commercial land uses develop around an airport. In addition, the findings of the study may also enable the decision makers of any given community where an airport is located to forecast the expected impact of such a facility on commercial land uses. Further, the findings may serve as a stepping stone for other research of similar nature.

Chapter Organization

The general background and the purpose of the study have been presented in the above section. Chapter Two of this study is concerned with the problems of analyzing land-use changes.


Chapter Three discusses the Pertinent literature on diffusion, and the airport is justified as an origin for a diffusion process of commercial land-use changes. Chapter Four has two main sections to it. In the first section, the study area and data collection processes are described. The second section is devoted to the methodology used in the research. Chapter Five presents the results of the application of a Monte-Carlo model of Diffusion Theory to commercial land-use changes in the area proximal to the Jacksonville, FloridaInternational Airport. In Chapter Six, the findings of the study are presented with their conclusions.


Note S

1R.V. Bowers, "The Direction of Intra-Societal Diffusion," American Sociological Review, Vol. 2 (1937), pp. 826-836.

2Ha1gerstrand first observed that diffusion of any
phenomenon is a function of communication and anything that can be communicated between people in space can spread by a diffusion process. Hagerstrand applied the diffusion process to adoption of many spatial phenomena like various innovations in farm technology.

3These are some of the studies on the issue:
T. H~gerstrand, Innovation Diffusion as a Spatial Process, (Chicago, The University of Chicago Press, 1973) ; R.L. Morrill, "Marriage, Migration and Mean Information Field: A Study in Uniqueness and Generality," Annals of the Association of American Geographers, Vol. 57 (1967), pp. 401-422; L. Brown, Diffusion Dynamics: A Review and Revision of the Quantitative Theory of the Spatial Diffusion of Innovation
(The Royal University of Lund, Sweden, 1968); G.F. Pyle, "The Diffusion of cholera in the United States in the Nineteenth Century," Geographical Analysis, Vol. 1 (1969), pp. 59-75.

4M.D. Thomas and R.B. Le Heron, "Perspective on
Technological Change and the Process of Diffusion in the Manufacturing Sector," Economic Geography, Vol. 51 (1975), pp. 231-251.

51.C. Alexander, "Multivariate Technique in Land Use
Studies: The Case of Information Analysis," Regional Studies, Vol. 6 (1971), pp. 93-103.

6R.C. Brown, "The Use and Mis-Use of Distance variables in Land-use Analysis," The Professional Geographer, Vol. 20 (Sept. 1968), p. 337.

71.C. Alexander, Op. cit., p. 95 (Note 5).

8B. Harris suggested diffusion theory for land use changes in 1966 in his article, "The Uses of Theory in the Simulation of Urban Phenomena," Highway Research Record, No. 126, pp. 1-16; also W.L. Garrison suggested the same methodology in the same year in "Difficult Decisions in Land Use Model Construction," Highway Research Record, No. 126, pp. 17-22.


9For example, the concentric zone theory published in 1924 by Burgess has the origin in the CBD; the article was reprinted in R.E. Park, E.W. Burgess and R.D. McKenzie, The__City (Chicago, 1925), pp. 47-62. Also the sector theory by H. Hoyt entitled The Structure and Growth of Residential Neighborhoods in American Cities (Washington, D.C., 1939) has its origin in the CBD. On the other hand, the multiple nuclei theory, by C.D. Harris and E.L. Ullman entitled "The Nature of Cities," Annals of the American Academy of Political and Social Science,
Vol. 242 (1945) suggested that more than one origin can be used.

10R.C. Brown listed many studies on land use that have used CBD as their origin in his article "The Use and Mis-Use of Distance Variables in Land-Use Analysis," Op. cit.,
pp. 337-341 (Note 6).llThe following studies, among several conducted by the Federal and various local governments about the impacts of an airport on its surrounding areas, confirm the fact: U.S. Department of Transportation, "Airports and Their Environment: A Guide to Environmental Planning," National Technical Information Service, U.S. Department of Commerce, 1972; R.D. Speas Associates, Gainesville Municipal Airport, Gainesville, Florida, "Master Plan Study: Noise Impact Analysis," Gainesville,
Municipal Airport Commission; and G. J. Bean, "Area Survey of Airport Environs, Economic Reaction--Present and Future," Hillsborough County Aviation Authority (1974).

12 Federal Aviation Administration, Aviation News (August 1972), p. 13.

13The following are some economic impact studies done for various airports. In some, the impacts on land uses are mentioned and in many nothing is said about land uses: G.J. Bean, "Area Survey of Airport Environs, Economic Reaction-Present and Future," Hillsborough County Aviation Authority; North Central Texas Council of Governments, "Dallas/Fort Worth Economic Impact" (1970); Port of Seattle Commission, "Airborne Traffic of Sea-Tac International Airport and Its Impact on the Economy of King County," Seattle, Washington (1974).

14Most of the research done on noise factor of airports indicates that residential uses of land are repelled from the airport, but that other uses, like commercial activities, are attracted. For example, see the following studies: D.V. Hayes, "The Dilemma of Aircraft Noise at Major Airports," Transportation Journal (Spring 1971), p. 5; P.T. McClure, "Indicators of the Effect of Jet Noise on the Value of Real Estate," Rand Corporation, Santa Monica, California (July 1969),

p. 34; A. Cohen, "Physiologic-it] and Psychological Effects of Noise on Man," Boston Society of Civil Engineers (1961), p. 70; "Investigation and Study of Aircraft Noise Problems,"1 First session, 88th Congress, H. R. No. 36.




A cursory examination of land-use research reveals that a thorough review of all literature is an overwhelming task. An almost uncountable number of published materials about some aspects of land use exists. Even, as in this review, one focuses on a narrow category of land-use changes, the number of published studies is still astonishing. Sparks states one reason why so many land use change studies have been accomplished:

The analysis of change is the core of planning.
The diagnoses of past changes, the motivation
underlining current changes, and the attempt
at prognosis of future changes are significant
considerations which the planning profession
can ill afford to ignore.'JPlanning can ill afford to ignore changes because it is the core of human life! It is not only the planning profession, however, that is concerned with future changes. Land economists, sociologists, environmentalists and, in particular, geographers have historically been interested in changes in land use as these relate to their interests. The changes may be economic, social, physical, spatial and cultural in nature.



Why does land-use change occur? Barlowe indicates that the shifts in land use are not new and that shifting of land areas from one use to another is a perfectly natural phenomenon in a dynamic economy. 2 This is especially so in an economy characterized by a growing population, expanding demands and increasing per capita income.

Natural changes that may occur will have different

effects on different people. To some, the ef f ect of a set of changes may be desirable and to others the same changes may be undesirable. Barlowe's reasoning and the fact of different effects on people may be the major reason why various disciplines have to study land-use changes. In the search for workable methods of forecasting changes which may lead to fewer negative effects on people the several disciplines have utilized different approaches as are summarized in the succeeding sections.

The Land Inventory Method

Foremost in the approaches to the study of land-use

changes is the inventory method. This postulates that a parcel by parcel inventory of land activity could show the amount of space used by each activity and would permit comparison among different communities concerning what the future activities on the land will be. 3 Judgement of the individual research is involved in this method and judgement alone is obviously inadequate as a basis for making multiple decisions required in a land-use forecast. 4 Hamburg and Creighton have an objective


view on the issue of land-use forecasting by the inventory method. They suggest that careful inspection of aerial photos and the study of existing development in the area in question should be the base in setting a future for land uses. 5 Gessaman and Sisler also advocate the use of aerial photos. 6 They indicate that data thus compiled have explicit vocational identity and are free of any distortion which may be caused by faulty memories. While there is no doubt about the applicability of aerial photos to research on land use, a notable weakness is in its inability to identify the processes which cause the identified pattern. Without understanding the processes of change identified, the causes cannot be known and the solution to the negative effects of changes will be uncertain. Most land-use researchers have demonstrated a realization of the importance of problem identification but have often ignored the process. To many of them, problemsolving is often less difficult than the formulation and identifying of an adequate problem statement and even the understanding of the process of changes. 7 on the basis of this belief, new approaches which differ from the inventory method have been used to examine anticipated changes.

Some Other Approaches in the Study of Land-Use Changes

There are three major methods considered here that have been used to tackle the problem of land-use analysis. The first is through modeling. In this approach, land-use changes


are linked to a mathematical or logical construct. A good example of this method is the gravity model which postulates a distance-decay function of changes. The second modeling method is carried out through analogy. In this case land-use changes may be compared with other phenomena, for example, an organism. 8 This analogy is, however, scientifically useless unless the phenomena compared are clearly and logically defined. Also the phenomena compared must be theoretically identical. The third approach is correlation analysis which has been widely used, particularly in recent years. This approach identifies a phenomenon that has logical relationships with other changes and, in turn, uses the phenomenon as a surrogate for the change.

Most objectives of studies based on correlation analysis are achieved by subjecting to statistical tests variables that the researchers think are the major ones dictating landuse changes. Among these are such socio-economic variables as income, employment, population structure, and social class of the people in a given community. Other studies have attributed changes in land use to land value, ownership, distance to the Central Business District (CBD) and accessibility by using samples of these variables. The reason for using samples in researching, as Poincare puts it, is to admit that if one had a thorough knowledge of what effect each of the samples had on the changes, one would have a knowledge of all of the changes. 9 The next sections discuss these variables and how they have been used.


Income as a Measure Land-Use Changes

Income is one of the socio-economic variables that have been used widely in land-use change studies. Blumenfeld indicates thatin business, income per worker can be related to worker density by business type and the result of the relationship can be expressed in terms of income per unit of land such as income per acre or as income per square foot. 10 Similarly, a combination of family income can be transposed into changes expected in land uses.

An important factor to consider is that income variables do not take into consideration the relative ability of all activities to acquire space. 11 Before income can be used to improve the land, there must be questions as to what tradeoffs are involved in such a decision process. In more sophisticated terms, one is dealing with difficult problems of public discount rates, collective consumption, spillovers, externalities, the aggregation of utilities and the reconciliation of conflicting interest. 12 With these complications, income can be used as only one of many interwoven variables applicable in explaining land-use changes.

Population and Employment Twin Variables in
Land-Use Change Analysis

Population, another variable often used by land-use

researchers to achieve a reasonably accurate forecast, has the advantage of control which makes it a stronger variable than income. 13 Control results from the constraints that can be


imposed on the population for the available land. For example, if the rate of increase in population is known, and the land available is known, then changes that will occur can be structured to fit the expected population. Although this is a reasonable and good idea, two weaknesses can be pointed out.

In the first place, population projections are seldom accurate because the real world exhibits extreme inequality in population development. 14 Population may be over or under estimated. In either case, forecast of land-use changes will be affected.

Secondly, projecting population does not necessarily

indicate the changes expected in land uses. For example, the emphasis on the use of land may change, which may have no connection with the entire population. Population projections may be useful to forecast land-use change only when residential land-use change is being considered. In this cases population increases, the land use for residences will usually increase with the population. There are, however, many landuse changes that do not depend on population. For instance, commercial land uses may not entirely depend on population but on a myriad of other variables: the type of commercial activities, the ownership of land and availability of the necessary materials may be other factors. Also, in an area with natural resources, land-use changes will depend on the resources and their accessories.

Dickens indicates that businesses are complex adaptive systems operating within and integrating with continually


changing external factors. 15 Thus, the factor of population does not include in itself all other factors and if it does, the factors need to be explained. Such interaction of factors with population is mediated through the coding of the perception mechanism of the business and it is perhaps through the study of perception and information flow that progress can be made in understanding the process whereby commercial activities adapt in their changing environment. 16

Employment has also been used as a variable in explaining land-use changes. However, this variable has not been employed in isolation. In most cases, the employment variable is often grouped with population. Blumenfeld indicates that, in land-use forecasting, future use should be achieved by extrapolation of some independent variables. 17 In his combination of variables, he states that population and employment should be grouped together, because by doing so, one is going to eliminate most of the endless number of variables which influence changes. 18 Blumenfeld combined population and employment because, according to him, the latter depends on the former.

Socio-Economic Classification for LandUse Changes

Socio-economic status has been considered in many landuse changes studies. Most of the popular theories of urban land use are based on this variable. 19 These theories are mainly residentially oriented, but residential land use is just a fraction of the changes that may occur in land uses.


For example, the Burgess theory of urban land use puts the CBD in the center of the city and the domiciles of various social classes are arranged concentrically around the CBD. 20 In recent years, however, business has moved to the suburban areas where people live and the general pattern of land uses has also changed.

The sector theory, a modification of the concentric theory, postulates that land-use changes in the form of a wedge. 21 Although this may be a better approach to the analysis of urban land use changes, the pattern may not necessarily be like a wedge as indicated by the third major theory of urban land use-the multiple nuclei theory. 22 These theories may not answer all the questions that may be asked about urban land-use changes. However, there is a concept in all of them that can be applied in some other ways to answer most questions that may be raised about land-use changes. This is the concept of origin. All of them ascertain that changes started in a certain point or points and spread outward. The factors that produced the outward spread are, however, not necessarily identified by these theories. For these theories to be appropriate in analysis of changes, processes that produced the form of concentric, sector and multiple nuclei should be identified and simulated together.

Distance Factor in Land-Use Changes

Perhaps the most widely used variable in the study of land-use analyses is distance. The argument for using this


variable is that all activities or changes in land use depend on convenient proximity to certain other activities. 23 Browning, who has explored this issue, postulates that the value of distance lies not in itself but in its association with many aspects of urban structure. 24 Distance cannot answer every question about urban land-use changes; it must be considered along with other variables.

Further, distance as a variable may sometimes be misleading because the effect of distance is not smooth on all segments of space at a given distance of equal importance to an individual. 25 Also, superior information about certain uses can overcome the distance variable. 26 Thus the changes that may be anticipated in land use will depend on the amount of information one has on the location of the establishment in question. For example, a decision about locating a wholesale commercial activity may not depend on distance alone but on the spatial distribution of the customers as well as the customers' knowledge about the location. While the argument is not that distance variables are useless in land-use studies, it is believed that too much emphasis has been laid on this variable, and many times variables relating to distance are misused.

Brown, commenting on studies on the use and misuse of distance variables in land use, states that, in many cases, distance has been used as a proxy for social and institutional variables. Such variables, according to him, sometimes indicate blights on land uses and influence extension of urban


services to rural hinterland-a misuse of the distance variable in land-use changes. 27 while distance can be a useful variable in land-use studies, emphasis laid on it has made it monotonous. Brown further states that there is the need to measure distance variables, but one should understand the underlying phenomena which researchers know to be very important but do not always know how to measure. 28 On the same issue of distance, Olsson suggested the need for deeper penetration into non-rational behavior and general motivation of people in land-use studies, rather than depending solely on rational behavior. 29

Warning has also been issued by many other researchers in their work regarding the over-use of distance variables. For example, Brown commented negatively about Yeates' use of distance variables. In his study, "Some Factors Affecting the Spatial Distribution of Chicago Land Values, 1910-1960," Yeates had six of his variables related to distance. 30 Application of distance variables in this form taxes the validity of distance variables and the ability of the variables to identify the spatial structure of land use has been greatly reduced. On the same issue, Browning further states that if people are unsure of the nature of the urban spatial structure, they are even less sure exactly how urban land use changes. 31 More work needs to be done in determining the value of distance variables in land-use studies not only to illuminate past changes, but to serve as guidelines for forecasting. One needs to know more, for example, about what process leads to


the urban fringes filling up, and to what extent are these changes evident. 32 To answer these questions, one has to know the basic phenomena which underlie the process of change.

HHgerstrand states that change does not occur unless one invents the change himself, hears about it, sees it or reads about it. 33 This point, however, does not mean that communication or information tells the whole story of the process, but the point made here is that before any change occurs in land-use, the owner of the land should have heard of, read about, or seen other types of land uses which are better than its current use. What this means, in essence, is that land-use changes, be they in urban or rural areas, are a function of information received about the new use and other factors as previously mentioned. It should be noted, however, that information does not travel by itself in space. It is usually transported from one person to another.

Land Ownership Factor in Land-Use Changes

The fact that land ownership influences change in land use has made ownership of land an important variable in the analysis of land-use changes.

Many studies have recognized the role that land ownership plays in changing land uses. 34 Three types of ownership are identified in the present study: private individual, private corporation and public land ownership. Each of these types exerts different forces on changing land uses. Of all the


three types of ownership, public ownership does not pose much of a problem to researchers because changes anticipated in the public land sector are identifiable in the comprehensive plan for the community in question, and often the uses are as indicated in the plan. Both private individuals and private corporation types of ownership are constantly exposed to external factors and also to changes in use at any point in time--depending, however, on the situation of the owner(s) at that point in
time. Since fifty-nine percent of land in the United States belongs to private individuals and private corporations, it is essential to understand the processes of such changes to have efficient land use planning in the future.

Land Value as Indicator of Land-Use Changes

Land economists have been eagerly concerned with land value and many times they have indicated that land value is a good variable to use in forecasting the land-use changes. 35 Land value, however, does not have the properties to explain itself. For example, one needs to know the factors that influence, increase, or decrease the value of land in certain areas. Also, a decrease in value of a parcel of land does not mean that the land use will remain constant. Therefore, there must be some other phenomena associated with land value. For example, the relationship between land value and transportation can better explain the changes in land use without reducing the efficiency of planning, forecasting, or describing the


land-use evolution. By relating transportation to value, one can show that value increases because of improved transportation and this increase in turn induces changes in land use. The inducement to change occurs because the old use may not be able to pay for the new economic rent, and rationally, the land should be able to pay for its economic rent as well as bring profit to its ownership. Hence, land value by itself cannot stand the test of causal analysis in land-use change.

Other Variables in Land-Use Change Studies

There are many other variables which, when taken in

isolation, cannot give a satisfactory explanation of land-use changes. For example, accessibility and even travel patterns have been used to forecast land-use changes in the same way land value has been used. Tn their study on land-use changes, Hamburg and Creighton indicate that travel can be related to land use changes, since it is the activities of people on land that generate traffic. 37 Row and Jurkat also indicate that the single most important hypothesis in land-use changes is the transportation system. 38 This is because transportation systems have changed the effective distance between all other places in the city and the CBD. By coordinating the importance of each variable discussed so far, a meaningful understanding of land-use changes could be achieved. Coordination of all these variables, however, has not been achieved.


Without an objective consideration of all variables

identified above, questions about land-use changes will remain unanswered. One of the problems may be in the type of data that land uses yield. Nominal data do not easily lend themselves to statistical testing; thus, methodologies that utilize nominal data appear to be necessary before any significant progress can be made in forecasting land-use changes. 39 Endeavors of researchers should now be directed to producing more sophisticated methods than those that have been discussed so far. There is need for methods that are objective in using nominal data by aggregating the effect of all the variables under consideration. One such method may be the application of Diffusion Theory to the study of land-use changes. Although the diffusion method has rarely been applied to land use changes, it has been suggested by several researchers in the past. For example, Lowry recognizes usefulness of diffusion theoryespecially the Monte Carlo generation type, in land-use changes when he states:

The difficulty is that there also exists a particular
class of quantitative models commonly called simulation models; these are most clearly distinguished
from other quantitative and mathematical models of
the same phenomena by the method of solution. If
we think of a set of models of any given phenomenon
or real-world system, as suchidentifiable with a set of simultaneous equations, the simulation
model is one which is meant to be solved by numerical
substitution rather than by analysis. The use of
this brute-force technique is most often associated with large models whose logical closure is not selfevident, and also with Monte Carlo methods of
generating input or intermediate values for certain
of the variables.40


Garrison also notes the possibility of applying diffusion models to land-use changes. 41 He thinks that past models on land uses pointed out a rather chaotic situation resulting from the way behavior questions are included in economic models. Some studies, he states, represent the site selection process strictly in terms of economic choice. On the other hand, there are some other studies that have represented the diffusion or spread of the structure of land uses from the existing urban area. For example, those studies that based their origin in the CBD are typical of studies that can adapt the diffusion model. Studies on diffusion are numerous and the subjects covered by the application of the diffusion theory vary. The following chapter discusses what diffusion means and how the diffusion process can be tailored to fit the analysis of land-use changes.



1R. M. Sparks, "The Case for a Uniform Land-use Classification," Journal of the American Institute of Planners, Vol. 24 (1958) pp. 174-178.

2.Barlowe, "Minimizing Adverse Effects of Major Shifts in Land Use," American Journal of Agricultural Economics, Vol. 40 (Dec. 1958), p. 1339.

3 A. Z. Guttenberg, "A Multiple Land-us'e Classification System," Journal of the American Institute of Planners, Vol. 25 (1959) pp. 143-150.

4 J. R. Hamburg and R. L. Creighton, "Predicting
Chicago's Land-use Pattern," Journal of the American Institute of Planners, Vol. 25 (1959), pp. 67-72.

5 Ibid.

6 .H. Gessaman and D. G. Sisler, "Highways Changing
Land Use and Impact on Rural Life," Growth and Change, Vol. 7
(1976) pp. 3-8.

7 J. R. Hamburg and R. L. Creighton, Op. cit. (Note 4).

8 .Harris, "The Uses of Theory in the Simulation of Urban Phenomena," HighwayResearch Record, No. 126 (1966), pp. 1-16.

9 H. Poincare, Science and Hypothesis (New York, Dover Press, 1952).

10 H. Blumenfeld, "Are Land-use Patterns Predictable?" Journal of the American Institute of Planners, Vol. 25 (1959) pp. 60-64.

1A.Row and E. Jurkat, "The Economic Forces Shaping Land-use Patterns," Journal of the American Institute of Planners, Vol. 25 (1959) pp. 77-81.

1G.Frommn, "Civil Aviation Expenditure: Measuring Benefits of Government Investments," Paper presented at a conference of experts Nov. 7-9 (1963) the Brookings Institution, Washington, D. C.

1H.Blumenfeld, Op. cit. (Note 10).


14 R. L. Morrill and F. R. Pitts, "Marriage, Migration, and the Mean Information Field: A Study in uniqueness and Generality," Annals of the Association of American Geographers, Vol. 57 (1967), pp. 401-422.

15 P. Dickens, "Some Aspects of the Decision-Making Behaviour of Business Organizations," Economic Geography, Vol. 47 (1971), pp. 426-437.

16 Ibid.

17 H. Blumenfeld, Op. cit. (Note 10).

18 Ibid.

19 For example, the concentric zone theory enunciated by Burgess in 1924 utilized socio-economic class to determine in which ring an individual will live. The poor live in the first concentric ring where the jobs are. The richer people live away from the job place and the distance from their home to their job depends on what class they are and how rich they are.

20 E. W. Burgess, "The Growth of the City," Proceedings of the American Sociological Society, Vol. 18 (1' 23), pp. 85-89. 1

21 H. Hoyt, "The Structure and Growth of Residential
Neighbourhoods in American Cities," Federal Housing Administration, Washington, D. C. (1939).

22 C. D. Harris and E. L. Ullman, "The Nature of Cities,"
Annals of the American Academy of Political and Social Science,
Vol. 242 (1945), pp. 7-17.

23 A. Row and E. Jurkat, Op. cit. (Note 11).

24 C. E. Browning, "Selected Aspects of Land-Use and Distance from the City Center: The Case of Chicago," The Southeastern Geographer, Vol. 4 (1964), pp. 29-40.

25 R. L. Morrill and F. R. Pitts, Op. cit. (Note 14).

26 Ibid.

27 R. C. Brown, "The Use and Mis-Use of Distance Variable in Land-use Analysis," The Professional Geogra2her, Vol. 20 (Sept. 1968), pp. 337-340.


28 Ibid.

29 G. Olsson, "Distance and Human Interaction: A Review and Bibliography," Regional Science Research Institute, Bibliography Series, No. 2, Philadelphia, 1965.

30 M. H. Yeates, "Some Factors Affecting the Spatial Distribution of Chicago Land Values, 1910-1960," Economic Geography, Vol. 41 (1965), pp. 57-85.

31 C. E. Browning, Op. cit. (Note 24).

32 R. J. Colenutt, "Linear Diffusion in an Urban Setting: An Example," Geographical Analysis, Vol. 1 (1969), pp. 106114.

33 T. Hdgerstrand, "Aspects of the Spatial Structure of Social Communication and Diffusion of Information," Regional Science Association Pa2ers, Vol. 16 (1965), pp. 27-4f.

34 For example see R. H. Platt, "Land-use Control:
Interface of Law and Geography," Association of American Geographers, Commission on College Geography, Publications, Resource paper No. 75-1 (1976), pp. 1-24. Platt has also laid emphasis on ownerships of land and he insisted that ownership, especially private ownership, has much to do with land-use changes. Of equal importance is the interaction between different ownerships, for example, the relationship of the public ownership to private corporation ownership. Also, the policy of the government in acquiring land is of paramount importance in land-use changes.

35 Ibid.

36 For example, see the following studies: E. Kurnow, "Land Value Trends in the United States," Land Economics, Vol. 36 (1960), pp. 241-248; M. T. Rancich, "Land Value Changes in an Area Undergoing Urbanization," Land Economics, Vol. 46 (1970), pp. 32-43; A. Roth, Jr., "Needed: A Method of Western Mountain Land Valuation," Land Economics, Vol. 24 (1948), pp. 181-185.

37 R. J. Hamburg and R. L. Creighton, Oz.). cit. (Note 4).

38 A. Row and E. Jurkat, Op. cit. (Note 11).

39 W. L. Garrison, "Difficult Decisions in Land-use
Model Construction," Highway Research Record, No. 126 (1966), pp. 17-23.


40 T. S. Lowry, "Discussion On Land-use Forecasting
Concepts," Highway Research Record, No. 126 (1966), pp. 32-35.

41 W. L. Garrison, Op. cit. (Note 39).



overview of Chapter

Chapter III is divided into three major sections. The first section reviews the literature on spatial diffusion, with particular attention to the validity of the application of diffusion theory to various subjects of interest to researchers in the social sciences. Section Two links diffusion theory to the analysis of land-use changes. The linkage is achieved by tailoring the axioms of diffusion theory to such changes. In Section Three, the airport is viewed as an initiator of land-use changes and its suitability as the origin for the diffusion process is discussed.

Spatial Diffusion: Its Validity and Applications

Diffusion is a mechanism in the spread of phenomena

which are necessary for understanding major changes in human life. An unlimited number of phenomena undergo changes. These phenomena can be social, economic, and political in nature. Also, technological changes, as well as changes that occur in the environment, fall under the diffusion subject. In natural sciences and social sciences research has been conducted on



diffusion. Research in natural. sciences on this subject is beyond the scope of this study. Application of the diffusion theory in the social sciences varies with the paradigm of each discipline. In anthropology, emphasis is laid on cultural changes across societal boundaries.1 Sociology stresses social changes over time, while geography, in addition to changes over time, emphasizes space. The spatial emphasis differentiates geographical research on this subject from other social science research, as indicated by Cohen. 2

Since the 1950s when Hdgerstrand initiated the spatial diffusion concept in geography, geographers have employed the concept in various ways to explain the spread of phenomena. Through cursory examination of these studies two common themes can be identified. The first is the contagious spread of phenomena and the second is hierarchical spread. Both contagious and hierarchical diffusion can be independent of each other. However, when they are combined, a third diffusion process called contagious-hierarchical diffusion is formed.3 Each of the three types is discussed in the succeeding sections. Contagious Diffusion

Contagious diffusion, which is sometimes called expansion diffusion, originated in the natural sciences; however, indecent years the process has been widely accepted and used in social sciences studies.4 The process of contagious diffusion postulates that phenomena emerge from one origin and spread outward


to wider areas. Evidence of this type of diffusion has been clearly illustrated in the work of Morrill 5 and Hagerstrand. 6

Morrill illustrates the concept of contagious diffusion by the use of an analogy with waves, in that the farther away a point is from the origin of the wave, the later the process of change begins and ends at that point. Conversely, the closer an area is to the origin of the wave, the more rapidly the effect of the wave is felt. Hcgerstrand, in a similar fashion, postulates that the extent of adoption of an innovation decreases with distance away from the origin. Examples of contagious diffusion application have also been given by Brown, Bowers, Donaldson, and Cohen. 7Their application of the theory is governed by the principle of contagion,or nearest neighbor influences,a principle which implies proximity and physical contact. Distance is a factor in personal contact in the absence of mass media such as the television and radio. Thus contagious diffusion has a "neighborhood effect" characteristic.

Hierachical Diffusion

Hierarchical diffusion, the second major division of

spatial diffusion, postulates that larger places tend to yield to change first, and then transmit the change to the smaller areas. 8Sometimes the process of hiera-rchical diffusion is synonymous with the "trickling down process" which has been widely used in the analysis of growth poles. In his study of the diffusion of cholera in the United States, Pyle indicates


that the infection struck larger cities first before filtering

down to the small cities. 9 In one of his conclusions, however,

Pyle indicates that there are many factors accounting for the

hierarchical movement other than the size of the population.

Besides the population or the size of an area, the direct

access to other centers and the structure of the interaction

of people in bigger cities account for the spread of phenomena.

Hudson supports the above idea:

The explanation of a sudden appearance of an innovation at separate large centers without the existence of intervening adopters is consistent with central-place theory . the
earliest adopters are those most likely to be
first exposed, and they are those having the
greatest volume of interaction: the largest centers. Very small centers are the last to
be exposed since they must wait until the high-order-places in their area have been
exposed, and so on.10

In addition to the above factors, however, many socio-economic

variables contribute to the hierarchical diffusion. Blaikie

illustrates the influence of such factors:

In other words, larger farmers with more liquid capital, with a high socio-economic status, with
stronger aspirations towards modernity tend to
move about more in daily life and have larger
MIF in any case. Their early adoption is often
more determined not as much by their early awareness, but by quicker interest, evaluation and 11
trial, and by the economic and social factors.

Once hierarchical diffusion occurs, however, the contagious effect may start from each hierarchy-an idea leading

to the third category of spatial diffusion.


Contagious-Hierarchical Diffi-ision Although specific research has not been directed to this type of diffusion process, the conclusions of some of the studies that have laid emphasis on one of the two major diffusion process categories mentioned earlier show that a third process exists. This can be denoted as contagious-hierarchical diffusion. HSgerstrand, in one of his pioneer studies, recognizes the possibility of the combination of these two types of diffusion when he states:

A closer analysis shows that the spread along the
initial "frontier" is led through the urban
hierarchy. The point of introduction in a new
country is its primate city, sometimes some other
metropolis. Then centers next in rank follow.
Soon, however, this order is broken up and replaced by one where the neighborhood effects
dominate over the pure size succession.12

In other words, the process may start with hierarchical diffusion but as many large centers are aware of the changes, the people around each center begin to adopt the change. Probably, diffusion processes in the real world situation follow this third type. Both contagious and hierarchical diffusion often occur at one time or another in the entire diffusion process. Thomas and Le Heron support this idea. 13 They conclude that technological changes in a manufacturing sector, i.e., change in technology, first occur in big firms scattered all over a given region. As time goes on, however, smaller firms near large centers tend to adopt the technological change.


Moderators of the Adoption Process in Diffusion Whether it is contagious, hierarchical, or contagioushierarchical diffusion, some factors which modify the spread of phenomena in space and over time exist. These factors have been given different names in the literature such as barriers or resistance. 14 The names given to these factors are not of concern in this research, but it is their activities in diffusion theory, which made the theory superior to others in analyzing changes, that is of concern.

In the present research, the words 'barrier' and 'resistance' will be used interchangeably. Barriers are any variables that help moderate the spread of phenomena. To understand the spread of any phenomena then, there should be a comprehension of the barriers involved. According to Hagerstrand, phenomena can be very different in nature and still their spread tends to show a number of recurring traits. 15 Different phenomena may also run through the diffusion process with different speeds and may have various degrees of irregularities. 16 Evidently, resistance is one of the reasons that can be advanced for such different speeds of spread. The effect associated with each resistance may, however, depend on the type of resistance it is. many types of resistance levels have been mentioned in the literature. 17 Some are absorbing barriers. Some are reflective and yet others are permeable in nature.

In order to more closely approximate the real world conditions in simulations of change with diffusion models, it is necessary to study the effects of these different types of


resistances. 18 If the resistances associated with the spread

of a phenomenon are known, the real world situation should be amenable to simulation. Without a trial and error procedure it may be difficult to ascertain the major factors which can serve as resistances for the spread. Harris sees this difficulty and

suggests the use of other probabilistic models in simulating land use. 19 Another suggestion by Garrison is to observe the cultural and psychological factors which may be handled via assumption concerning the statistical nature of the distribution of the phenomenon in question rather than by identifying the factors. 20 In case the factors can be identified, the result may be a knowledge of the form of the phenomenon, but nothing may be known about the process that produced the form.

Changes in land use do not just occur; they must be

instigated. Hence, there are two components in land-use changesa form component and a process component. 21 The two components, however, are related, for it is the process that produces the form. To understand the relationship better all the factors that serve as barriers should be comprehended. Such factors as land owners, speculators, demand intensity, noise level, governmental regulations, land value, topography, drainage, and accesibility should be aggregated to simulate the changes anticipated in land use. If this can be done, then, it is possible to adapt diffusion theory to the analysis of land-use changes around an airport, while the airport serves as origin for the diffusion process.


Diffusion Axioms Tailored to Fit Land-Use Changes

Diffusion is a dynamic process which can be adapted to

many situations that involve changes over time. As is true for much of scientific reasoning, diffusion theory is based on a number of assumptions. In the case of diffusion, however, the assumptions are not as numerous as for other major theories (central place.theory for example). The few assumptions are general enough to be adapted to various aspects of change over time and through space.

Spatial diffusion theory assumes that adoption of any change at any time does not proceed in sequential order. 22 This assumption can readily be transposed to land-use changes. Changes that may occur cannot be expected to follow a sequential order. In the real world, changes in land use leap over stretches of intervening parcels of land and the results are noted in inventory-type studies of land use.

Hierarchical spatial diffusion theory also assumes that adoption of any change may move in an hierarchical fashionthat is, first moving from higher order to high order before moving to lower levels of the hierarchy. 23 This assumption poses no problem in dealing with land-use changes, as changes in land use often occur in larger land holdings before small holdings, as has been observed by Hzgerstrand. 24

Adoption of any change in a diffusion process evolves slowly at first, then accelerates and eventually levels offanother assumption that can be adapted to land-use changes.


Generally, changes in land use occur slowly at first, but as more and more people are aware of the change process, many parcels of land may change to a new use until no further development is possible. The curve produced by this type of trend may best be likened to a logistic curve. Cassette, when illustrating the same idea with innovation, writes:

The adoption of the innovation takes place only
after "resistance" to change is overcome. The
model . postulates (1) that the adoption of technological innovations by potential users results primarily from "messages" emitted by
adapters; (2) that potential users have different degrees of "resistance" to change; (3) that within
any region there are potential users with different degrees of resistance; (4) that resistance is overcome by sufficiently large repetition of messages.
It can be shown that the dynamic interaction of
these postulates causes the proportion of adopters
to increase slowly at first, then rapid then
slowly again until saturation is reachely 6

Spatial diffusion theory postulates that acceptance of

any change appears in the immediate area of its origin initially because of the neighborhood effect, or it may "hop" to another area with similar attributes as the origin. 27 many studies of land-use changes have used distance variables as their major thrust and have demonstrated the validity of this assumption. 28

The propagation of change is a continual outward shift of the outside boundary in all directions, assuming that no barriers to movement are encountered. 29 Concentric zone theory is probably based on the same assumption.

Diffusion theory also assumes that the developments

associated with polar extremes will be attracted to each other. 30


Land-use changes in regard to low-land-use change activators move towards each other rather than in other directions. For example, if two industries are located close to each other, the land between them may likely be changed before any change occurs in the other directions. Such a situation can be commonly observed in urban areas.

It appears, then, that most of the axioms of diffusion theory are conceptually transferable to land-use changes that are associated with an initiating origin, such as an airport. Even though the diffusion assumptions are conceptually transferable to land-use changes, the theory's applicability to landuse changes around an airport depends on many factors which may serve as barriers to the process of change.

Among these factors for example, is the demand generated by the airport for certain commercial activities near its location. Also, the availability of vacant parcels for such activities in the area most appropriate should be considered. Other factors, such as accessibility, topography, drainage, and noise of the airport environment may influence the process of diffusion. The major thrust of this research is to test the transferability of diffusion theory on land-use changes in a real world situation around an airport by considering those factors that may affect commercial land-use changes.

Airports as Initiators of Land-Use Changes

Airport land use is becoming increasingly important. Land occupied by early airports was so small that the impacts on the


surrounding land uses were not noticeable. As the size of airports increased, however, airport land uses have become a common category of urban land use. 31 The location of such facilities in communities not only consumes a large proportion of the land but also their presence has always been noted to exert a strong impact on the adjacent land uses. Land uses attracted to airports often change the land-use structure of the communities where they are located; and the changing structure of the land use affects the population and economic condition of such communities. One researcher suggests that the impacts resulting from airports can be parallel to those initiated in 1956 with the creation of the interstate and defense highway systems in the United States. 32

By 1946, researchers had become very concerned about the impact of airports on community land-use changes. 33 By the mid to late 1950s, many researchers began to develop methodologies for analyzing airport land-use changes. Perhaps the major attempt was the study of Kirchherr in 1959. 34 He traced the history of airport land uses in the Chicago area, and in the process was able to demonstrate that over the years the location and size of airports have changed. Changes in the size of airports have exerted influences on the adjacent land uses. Regardless of the size or type of airport in question, however, the location will exert some influences in changing the adjacent land uses.

In the early 1960s, efforts directed toward identifying the impacts of airports on the area proximal to their locations


continued. Hanten derived a wl, thodology for measuring the airports'influence and determined the effects on land uses, and arrived at some interesting conclusions. 35 First, he noted that over the years, as commercial air traffic increased, the size of the airports also increased from about 1,600 acres to over 2,500 acres on the average. many airports at present, however, are about 4,000 acres, evidence of increasing importance in land-use changes. Second, Hanten noted in his survey of commercial enterprises adjoining the airport, that the reason business enterprises located where they did was not because of the airport but because of the accessibility to their particular site created by the network of highways built to serve the airport. Further, he indicated that total understanding of the effects of an airport on the surrounding land uses will not be achieved until the area proximal to the airport is thoroughly analyzed. He also suggested some important factors which influence changes. Among these are the ownership of the land, government regulations, the value of the land, accessibility, the noise level around the airport and the topography of land. 36

In generating his methodology, he noted that all these factors worked together to dictate the effect on the surrounding areas. Hanten did not, however, discuss the importance or weight of

these factors on the anticipated changes.

Both Kirchherr's and Hanten's studies lay emphasis on the effect of airports on the population and the use of land for residences. While the use of land for residences is an important aspect of urban land uses, such uses in the area proximal to the


airport are becoming insignificant because of their incompatibility with the airport activities. Stromberg's 1974 study of airport land uses has suggested about four hundred types of commercial land uses attracted to the airport environment. 37 If, in recent years, an increasing percentage of land uses around the airport is devoted to commercial activities Stromberg indicated, there is a strong need to know more about the factors that influence the changes. Also, there is a need to know what type of commercial activities are attracted to the airport.

Rubin states that an airport is a major land use, a concentration of activities, an environmental issue, and an economic necessity. 38 Efforts have been made by researchers to understand how airports fit into each of these categories. Concentration of activities is probably one of the most important effects of airports on their surrounding areas. Such concentration is not instantaneous in nature. However, the process evolves over time. In order to understand the process of the evolution of the-concentration, researchers have laid emphasis on identifying the effects of such by taking inventories of the changes around the airport. 39 From these studies, it can be observed that the impact of an airport on the surrounding area creates three types of multipliers -the income multiplier, employment multiplier, and spatial multiplier. The last of these three has to do with land-use changes which initially

The concept of a multiplier, as used here, means that the location of the airport creates impacts on the community and that these impacts generate other impacts on the surrounding area, and eventually affect the whole community.


occur at, or near the airport, and as time elapses, the geographic impact spreads outward from the airport.

Since the location of an airport creates convenience and prestige value for other types of enterprises locating nearby, an airport cannot be treated in isolation. It must be planned in relation to the whole urban area, or the region in question. 40 In this case the airport can be regarded as a growth point that attracts other activities to, or near, its location in the urban periphery.

Smith states that, although the initial growth pole concept, as developed by Perroux, was non-spatial, the work of Darwent has cleared the confusion associated with the economic concept and the spatial concept of growth points. 41 Basing his argument on Darwent's study, Smith concluded that a growth pole can be referred to in an economic context, while a growth center, or point, can be referred to as a location. 42 Numerous researchers from Perroux through Hirschman and Boudeville, to Penouil, Thomas, and Mansfield have written about various aspects of the growth pole concept. 43

It appears that commercial land-use changes associated with an airport follow the growth pole hypothesis, i.e., the "trickling down process," and the "backwash effect." The trickling down process occurs in those land uses that are initially attracted to the airport and they, in turn, attract other land uses. Lasuen emphasized this positive interdependence linkage between small establishments and big establishments. 44 In this context, the airport can be regarded as the big establishment that attracts other small commercial


activities. If this is so, commercial land-use changes in the area proximal to the airport are a function of the level of activities in the airport itself. 45 Nichols indicated that increased levels of activities in the center raised the demand for other activities in the surrounding area; hence, the activities generated by the airport generate other smaller activities. 46

The backwash effect, on the other hand, occurs because establishing a growth point, such as the airport, within a community involves the identification of the place where its location for the community has a comparative advantage over other locations. The selection of a focal point for investment implies, however, that other areas in the community will get correspondingly less public assistance. 47 This is a backwash effect, since other areas far from the airport may have little future commercial potential, because they could not provide the amenities that the airport environment provides.

Not only do airports create growth benefits to communities, they also create negative environmental issues. Foremost among these is noise pollution. much research has been done on noise associated with aircraft taking off and landing. A complete review of this issue is beyond the scope of this study; however, it will suffice to mention that many court cases have been filed on the environmental issues of the noise factor in airport operation. 48

Noise is not the only environmental issue that can be raised about the airport and its function. Aviation brings about degradation of natural life-an effect that is well


documented.49 In Jamaica Bay, New York, a study conducted to analyze the effect of the expansion of Kennedy Airport to the bay indicated that such expansion would have drastic adverse effects on wildlife, as well as on recreational value of the bay. Similar effects were predicted for a jet port near Miami. 50

various measures have been tried in order to promote a healthy land-use development around airports. Hinkle tried to develop a model which can serve as a guide against the un51
fortunate consequences of airport development. The objective of the Hinkle model is to enhance the identification of alternative, feasible, and complete land-use configuration in areas around the airport. In order to do this, the process includes identification of physical characteristics of the surrounding areas like topography and drainage. The demographic characteristics which consist of population and employment are also identified. Environmental characteristics, which include noise and other types of pollution as well as dangers related to the location of such a facility, are emphasized. While the analysis of airport land uses by this model is valid, the main weakness seen in the model is that it cannot be used for micro-level studies since data used in the model can only be collected at the census tract level. This may be inappropriate for the analysis of land-use changes associated with airports because not all census tracts of the community are affected.


Fromm investigated the situation in another way. He suggested that the decision of airport effect can best be decided by the use of an input and output model. 52 This suggestion is feasible only when one is dealing with economic changes brought about by the airport, but it may be very difficult to measure in terms of land-use changes. It seems then that the suggestion of Lowry and Garrison is the most appropriate to use while dealing with land-use changes associated with an airport. 53 Both researchers suggested a simulation approach to understanding land-use changes. In particular, they suggested the Monte Carlo model for solving land use change problems. 54 The Monte Carlo model can answer a number of questions that may be raised about land-use changes around an airport by simulating the pattern of the changes. Monte Carlo simulation is discussed in the following chapter.



1Y.S. Cohen, Diffusion of an Innovation in An Urban
System: The Spread of Planned Regional Shopping Centers in the United States, 1949-1968. The University of Chicago Department of Geography Research Paper No. 140 (1972), pp. 4-14.

2 Ibid.

3 Samples of studies where these two ideas have been
presented are: S. C. Dodd, "Diffusion is Predictable: Testing Probability Models of Laws of Interaction," American Sociological Review, Vol. 20 (1955), pp. 392-401; L. A. Brown and K. Cox, "Empirical Regularities in the Diffusion of Innovation," Annals of the Association of American Geographers, Vol. 61 (1971), pp. 551-560.

4 The following studies portray that the two types of diffusion can be combined to form the third one: J. C. Hudson, "Diffusion in a Central Place System," Geographical Analysis, Vol. 1 (1969), pp. 45-58; M. D. Thomas and R. B. Le Heron, "Perspective on Technological Change and the Process of Diffusion in the Manufacturing Sector," Economic Geography, Vol. 51 (1975), pp. 231-251.

5 R. L. Morrill, "Wave of Spatial Diffusion," Journal of Regional Science, Vol. 8 (1968), pp. 1-17.

6 T. Hdgerstrand, Innovation Diffusion As a Spatial Process (Chicago, University of Chicago Press, 1967), pp. 132-148.

7 L. Brown, "The Market and Infrastructure Context of
Adoption: A Spatial Perspective on the Diffusion of Innovation," Economic Geography, Vol. 51 (1975), pp. 185-216; R. V. Bowers, "The Direction of Intra-societal Diffusion," American Sociological Review, Vol. 2 (1937), pp. 826-836; B. Donaldson, "An Empirical Investigation into the Concept of Sectorial Bias in the Mental Maps, Search Spaces, and Migration Patterns of Intra-urban Migrants, Geografiska Annaler, Ser. 55B, pp. 13-33; Y. S. Cohen, Op. cite (Note 1).

8 D. Amedeo and R. G. Golledge, An Introduction to
Scientific Reasoning in Geography (Nfew York, John Wiley and Sons, Inc., 1975), pp. 269-271.

9 G. F. Pyle, "The Diffusion of Cholera in the United States in the Nineteeth Century," Geographical Analysis, Vol. 1 (1969), pp. 59-75.


10J. C. Hudson, "Diffusion in a Central Place System," Geografiska Annaler, Ser. 55B, p. 46.

11P. M. Blaikie, "The Spatial Structure of Information Networks and Innovative Behaviour in the Ziz Valley, Southern Morroco," Geografiska Annaler, Ser. 55B, p. 85.

12 T. Hgerstrand, "Aspects of the Spatial Structure of Social Communication and the Diffusion of Information," Regional Science Association Papers, Vol. 16, Cracow Congress (1965), p. 40.

13M. D. Thomas and R. B. Le Heron, "Perspective on Technological Change and the Process of Diffusion in the Manufacturing Sector," Economic Geography, Vol. 51 (1975), pp. 231-251.

14D. Amedeo and R. G. Golledge, Op. cit. (Note 8).

15T. Hdgerstrand, Op. cit. (Note 12).


17D. Amedeo and R. G. Golledge, Op. cit., pp. 270-273 (Note 8); A. G. Wilson and M. J. Kirkby, Mathematics for Geographers and Planners (Oxford, Clarendon Press, 1975), pp. 15 and 30.

18This is a suggestion Harris and Garrison gave. They advised that researchers need to know about the cause-effect relationship.

19B. Harris, "The Use of Theory in Simulation of Urban
Phenomena," Highway Research Record, No. 126 (1966), pp 1-16.

20W. L. Garrison, "Difficult Decisions in Land-Use
Model Construction," Highway Research Record, No. 126 (1966), pp. 17-23.

21W. R. Alves and R. L. Morrill, "Diffusion Theory and Planning, Economic Geography, Vol. 51 (1975), pp. 290-304.

22T. Hagerstrand, Op. cit., pp. 132-148 (Note 6).

23D. Amedeo and R. G. Golledge, Op. cit., p. 279 (Note 8).


2T.Hagerstrand, Op. cit., p. 150 (Note 6).

2L.Brown, Diffusion Dynamics: A Review and Revision of the Quantitative Theory of Spatial Diffusion of Innovation
(The Royal University of Lund, Sweden, 1968), pp. 10-20.

26E. Casetti, "Why Do Diffusion Processes Conform to Logistic Trends?" Geographical Analysis, Vol. 1 (1969), pp. 102-103.

2R.L. Morrill, and F. R. Pitts, "Marriage, Migration and Mean Information Field: A Study in Uniqiueness and Generality," Annals of the Association of American Geographers, Vol. 57 (1967) pp. 401-422.

28 For example, see R. C. Brown, "The Use and Mis-use of Distance Variable in Land-use Analysis," The Professional Geographer, Vol. 20 (Sept. 1968), pp. 337-341.

29T. Hagerstrand Op. cit., pp. 132-148 (Note 6).

30 L. Brown, Op cit., pp. 10-40 (Note 25).

31E. C. Kirchherr, "Airport Land Use in the Chicago Metropolitan Area: A Study of the Historical Development, Characteristics, and Special Problems of Land-use Type Within a Metropolitan Area." (Doctoral Dissertation, Northwestern University, 1959). Dissertation Abstracts International, 1959, Vol. 20, pp. 2224-2225.

32D. Rubin, "Planning an Airport in A Competitive Environment," Paper Submitted for PresentationPlanning '75, Innovation and Action, 57th Annual Conference American Institute of Planners, San Antonio, Texas, 1975, pp. 1-29.

33 L. Vice, "The Location of Airports in the United States," Dissertation, University of Wisconsin (1946).

34 E. Kirchherr, Op. cit. (Note 31).

35 E. W. Hanten, "A method of Measuring Airport Influence and Determining Its Effect on Land Uses and Land Values the Greater Pittsburgh Airport as an Example." (Doctoral Dissertation, University of Pittsburgh, 1962). Dissertation Abstracts International, Vol. 23 (1962) pp. 2089-2090.

36 Ibid.


37 C. H. Stromberg, "How to Keep 30,000 Jobs From Flying Away Through Military Airport Overlay District Zoning," Paper Submitted for Presentation at 57th Annual Conference, Planning '75, Innovation and Action, American Institute of Planners, San Antonio, Texas, 1975, pp. 1-39.

38D. Rubin, Op. cit. (Note 32).

9For example, see B. Higgins, "The Montreal Airport Site, The Spatial multiplier and Other Factors Affecting Its Selection," Growth and Change: A Journal of Regional Development, Vol. 2 (1971), pp. 14-22.

40 A. Jerome and J. Nathason, "Socio-Economic Implications, of Airport Planning," Traffic Quarterly, Vol. 25 (1971), pp. 267-286.

41D. M. Smith, Industrial Location: An Economic
Geographical Analysis (New York, John Wiley and Sons, Inc., 1971) pp. 452-458.

43 F. Perroux, "The Domination Effect and Modern Economic Theory," Social Research (June 1950), pp. 188-206; A. Hirschman, The Strategy of Economic Development (New Haven, Yale University Press, 1958); J. Boudeville, Problems of Regional Economic Planning (Edinburgh, University Press, 1966) ; M. Penouil, "Growth Poles in Underdeveloped Regions and Countries," in Kuklinski and Petrella, eds., Growth Poles and Regional Policy (The Hague, Mouton, 1972); M. Thomas, "Regional Economic Growth: Some Conceptual Aspects," Land Economics, Vol. 45 (1969), pp. 45-46; M. Thomas, "Growth Pole Theory: An Examination of Some of its Basic Concepts," in Hansen, ed., Growth Centers and Regional
Economic Growth (New York, The Free Press 1972) ; E. Mansfield, Industrial Research and Technological Innovations (New York, Norton Press, 1968).

44 J. Lasuen, "On Growth Poles," Urban Studies, Vol. 9 (1969) pp. 137-161.

45 V. Nichols, "Growth Poles: An Evaluation of Their
Propulsive Effects," Environment and Planning, Vol. 1 (1969).

46 Ibid.


47 D. M. Smith, Op. cit., pp. 452-458 (Note 41).

4P.P. Mann, "Aircraft Noise, Recent Developments
Present New Opportunities for Achieving a Compatible Airport Community Relationship," Paper presented at the 57th Annual Conference '75 Innovation and Action, American Institute of Planners, San Antonio, Texas (1975), pp. 1-12; U. S. Department of Housing and Urban and Development, Office of International Affairs, "Major Airports and Their Effects on Regional Planning" (1975), pp. 3-6.

49 D. C. McGrath, Jr., "Multidisciplinary Environmental Analysis: Jamica Bay and Kennedy Airport," Journal of the American Institute of Planners, Vol. 36 (1971), pp. 243-256.

501It has been announced recently that the location for a jet port in the Belle Glade area would have adverse effects on the environment; thus alternative sites are being sought.

51~ J.J Hinkle, "A Community/Airport Economic Development Model, Vol. 1, General Concept and Application," Federal Aviation Administration. Distributed by National Technical Information Service, Springfield, Va. (1972), pp. 1-84.

52 GFrm,"Measuring Benefits of Government Investments: Civil Aviation Expenditure," Paper presented at a conference of experts held Nov. 7-9, 1963, The Brookings Institution, Washington, D. C. (April 1965), pp. 172-229.

53 .L.Garrison, Op. cit. (Note 20); I. S. Lowry, "Discussion of Land Use Forecasting Concepts," Highway Research Record, No. 126 (1966), pp. 32-35.

54 A Monte Carlo Model is the model derived by Hagerstrand, which he used in the research about the adoption of innovation. It utilizes random numbers to assign probabilities to each adopter of the innovation. For more readings on the model see T. Hagerstrand, "A Monte Carlo Approach to Diffusion," Archives Europeennes de Sociologie, Vol. 6 (1965), pp. 43-67.



There are three major sections of this chapter. In the first section the study area in Jacksonville, Florida is described, and rationale for choosing the International Airport as the focus of the research problem is stated. Section Two presents the research hypothesis and the rationale for the hypothesis. In Section Three, the research methodology is described and various procedures used in collecting the data, as well as a rationale for the utilization of such data, are discussed.

Jacksonville International Airport: An
Introduction to the Study Area

In order to test diffusion theory as a methodology for

explaining commercial land-use changes, an area proximal to the Jacksonville, Florida, International Airport was chosen for study (see Figure 1). This particular area was chosen for several reasons: (1) It was thought that the airport chosen should be one which was recently constructed in a rural or a semi-rural area so that the initial attraction of commercial activities could be clearly observed. The Jacksonville International Airport began operation in October 1968, but planning for it


SC ALE ... ......... .

m N






started nine years earlier. Since the area was rural before the location of the airport, it is assumed that any commercial land-use changes there have been influenced to some degree by the location of the airport. (2) The airport is approximately twelve miles from the Central Business District (CBD) and there is a physical separation from the urban expansion. Therefore, commercial activities located close to the airport should not be directly associated with urban expansion, but with either the airport or other nodes serving as commercial land-use initiators, such as highways and industries. (3) For an appropriate micro-scale analysis, data must be available at the parcel level. The Duval County property appraiser has accurate and up-to-date data which were made available for this study. (4) Accessibility of all the airport planning records and Jacksonville's proximity to the University of Florida strengthened the choice.

The Need for a New Airport in Jacksonville

Construction of the International Airport was the culmination of some nine years of planning and analysis. The objective was to provide much-needed aviation facilities for the Jacksonville area because, by 1956, commercial and general aviation in Jacksonville had outgrown the capacity of the existing major commercial airport (Imeson Airport, see Figure
2) After a series of studies from 1958 to 1962, an agreement was reached that the most feasible solution to the problems created by the jet age at the Imeson site was the construction of a new airport.


_ _ Mie5






Reynolds, Smith and Hi. 1.1s (project consultants) pointed

out the disadvantages of Imeson Airport as follows: 3

1. Its location was adjacent to an active U. S. Navy
complex, which includes three major airfields (see Figure 2). It is approximately thirteen
miles from Imeson to NAS Jacksonville and twelve
miles to Mayport. It has been indicated, however,
that there should be at least sixteen miles
(physical separation) between major airports.
Imeson Airport is thus located in an airway
corridor created by surrounding restricted zones and converging airways which results in unsatisfactory air control.

2. Because of the limited straight-in approach path,
the orientation of the instrument runway was highly unsatisfactory and adversely affected the acceptance
rate under poor weather conditions.

3. Residential and industrial construction had encroached on runway clear zones adjacent to the airport property; this made proper control of
aviation rights unfeasible.

4. There was a complete lack of zoning for navigational
purposes in the area of Imeson Airport, because extensive residential, commercial and industrial
development precluded the possibility of obtaining
adequate zoning for airport purposes.

5. Encroachment of all these things mentioned above
coupled with lack of land for buffering had resulted in legal suits being brought against the
city of Jacksonville.

6. Adjacent industrial areas contributed smoke and
haze in the airport area which, under some conditions, adversely affected aircraft operation.

7. The airport facilities were deficient in many ways.
For example, the airport property was inadequate
to permit expansion of airfield facilities. Existing airfield paving was in need of additional
maintenance and in some places complete replacement.
Runways were not designed to support the gross
weights of air carrier aircraft which are now in
service. Muck conditions extended under the south
of the runway, thus additions or improvements in
that direction would require removal of these
materials. The drainage in the field was inadequate and often reduced airport capacity., The
facility did not have capacity for future use and
the terminal building was inadequate.


Reynolds, Smith, and Hills then concluded that, given these disadvantages, it seemed that construction of a new airport was the only feasible solution.

Because of the restrictions placed on Jacksonville by the military airport operations, the only feasible location for a new airport was in the northwest portion of Duval County. The location selected was approximately six miles northwest of Imeson Airport in a relatively undeveloped area. Land surrounding the new location was devoted primarily to agriculture with a few homes scattered throughout.

The topography of the area is flat with a maximum elevation of about 29 feet above mean sea level. Location at the crest of a natural drainage which causes natural run-off in two directions is a further advantage. In addition to local roads, the new site is served by Interstates 95 and 295, which under normal circumstances may reduce traveling time from the CBD at least five minutes.

Despite all these advantages of the new airport's location, its activities, after construction, have decreased over the years. For example, before the new terminal was built, Jacksonville handled 1.3 percent of all air passengers in the United States. Since the new terminal opened, the number of passengers has increased slightly; however, in terms of the percentage of the United States air passengers, Jacksonville handled only 0.45 percent in 1972, a record low. 4

Four major reasons can be advanced for the decline in

the passenger volume of the Jacksonville International Airport: 5


1. Total military employment has remained relatively
stable. Prior to 1960, military employment in
Jacksonville fluctuated and the fluctuations increased passenger traffic at the airport. By
1971, however, only about ten percent of the
total passengers were military personnel.

2. The old airport (Imeson) was used as aconnecting
line to the mid 1960s, but since then it has been
used as a terminating line. In 1962, off-line connections at Jacksonville amounted to 19 percent of the total emplanement, but by 1970 the,
off-line connection had declined to only nine
percent of the total emplanement.

3. Ownership and use of business aircraft by
financial corporations and other business headquartered in Jacksonville has enjoyed rapid
growth and today Winn-Dixie, Publix, and Seaboard Coastline (all having state-wide headquarters in Jacksonville) own and operate their
own multi-engine aircraft.

4. Airline strikes of 1962, 1966, and 1970
drastically affected emplanement. Thus, since the opening of the new terminal, there has not
been a smooth flow of operation.

The lack of steady growth does not disqualify the

Jacksonville International Airport as the study area. On the

Contrary, since the airport was opened, there have been

seventy-one commercial land uses attracted to its location.

Numerous "for-sale" signs on vacant land (see Figure 3a and

b) also show that this area is still undergoing changes.

Specific Area Covered by theStudy

The Jacksonville Tax Appraiser's office has a well-defined

section, township and range system of land division accurately

delimited on maps. 6Because of the availability of these maps,

the decision was made to conduct the analysis at the parcel

level. Parcel land-use data integrated into the section,

township and range system are needed for micro-level research


Figure~~~~~~~~~~-1 3a4hsoe adi oae bu n iees
of th IprS ntemjrrodlaigt h

Figure 3b. This ooed land tsocted not onte irpor ias
alsoporl. The owner walnts to sell i to

developers of any type.

Figure 3


such as this. In all, the s -iidy area covers about 48 square miles (see Map 1). The east-west dimension is eight miles, and the north-south dimension is six miles.

There are four noteworthy features within the study area besides the International Airport. In the south portion is the proposed highway Interstate 295. This highway is near completion and it is expected to have a great influence on the changes that occur in commercial land uses around the airport. Interstate 95 runs north-south, east of the study area. The Seaboard Coastline Railroad and State Road 17 (SR 17) run roughly parallel with each other and parallel to Interstate 95. It is evident that all these features have influenced changes that have occurred in commercial land uses in the study area.

The Research Hypothesis

It is hypothesized that the spatial pattern of commercial land-use changes associated with an airport can be explained by spatial diffusion theory. Stated in another way, it is believed that changes in land use start with the airport and expand outward, or that changes begin with the airport and "hop" to major nodes along transportation routes serving the airport, so that the intensity of land-use changes is greatest close to the airport (or transportation nodes) and decreases with increasing distance.

The rationale for this hypothesis is as follows: (1) Airports tend to be users of very large amounts of land which may


hinder the use of land for other purposes, and any use of land in the domain of the land acquired by the airport authority must be related to the airport. (2) Airports are indispensable parts of a transportation network which specializes in moving people, goods and services from city to city and from nation to nation. The functions that airports perform necessitate that facilitie'be established around them to provide support for those functions. Naturally if those facilities are to serve the functional needs of the airports, they have to be near their locations. (3) The location of airports requires good road systems which allow easy movement to and from the airports. The traffic generated by these road systems also invites commercial activities to serve the people coming in and going out of the airports. Definitely, those commercial activities will have to be close to the airport.

(4) Airports create work forces which are large enough to enhance demand for certain commercial activities, such as gasoline stations, motels and restaurants. (5) Airports exert economic and physical environmental impacts on their surroundings. 7 This means that any establishment that is located around an airport must have a symbiotic relationship to the airport. In other words, the commercial activity that moves close to the airport is complementary to airport activities. Thus, such commercial services as car rental agencies, catering firms and warehousing are found close to airport. On the other hand, those uses that are not complementary will be repelled from the airport.


In the context of the research hypothesis, several

assumptions are made regarding the Jacksonville International Airport and the surrounding area. These assumptions are as follows: (1) Commercial land-use changes in the area proximal to the airport are attracted to the airport and the airport serves as the initiator of all the changes. (2) Commercial land-use changes occur because of demand generated by the airport and associated highways. (3) Characteristics of land, such as topography, accessibility, government regulations, ownership and location of each parcel of land relative to the airport, have unequal effects on each parcel. (4) The land close to the airport meets most requirements for change to commercial uses but resistance created by the law of supply and demand may slow the process. (5) It is possible for most commercial land-use changes to occur at the initial opening of the airport, then to slow down and then to rise again as demand calls for the changes. (6) Hierarchical diffusion may occur in case another node develops during the process of the diffusion. For example, the intersection of an airport road and a major interstate highway can create another origin where commercial land-use changes may be initiated.

In making these assumptions, the concept of the Mean Information Field (MIF) characteristic of a large number of diffusion studies is not considered. Instead, the concept of a Mean Demand Fie ld (MDF) is substituted and the assumptions are needed logically to operationalize the MDF. Thus, the


intensity of commercial land-use changes depends on the demand for such activities generated by the airport and modified by the attributes of each parcel of land. These attributes or characteristics serve as barriers to the demand for commercial land.

The Research Methodology

The development of any change pattern can be regarded as a random or stochastic process. 8 A stochastic process is one that develops over time, according to some probabilistic rules which indicate that the future behavior of phenomena cannot be predicted with certainty. There are a number of stochastic models that can be used to attempt to replicate the change process, when the change process has a random component that is based on the assignment of probabilities. Among these models is the Monte Carlo simulation model which is used in this study.

The Monte Carlo Simulation Model

In the early 1950s, H gerstrand first adapted the Monte Carlo simulation technique to geographic space in order to analyze the diffusion of innovation through time. 9 His application of the model began with a given distribution of units of people who have already accepted an innovation. HZgerstrand then developed rules or assumptions which governed the ensuing spread of that innovation. lie regarded all other factors that


were not in the assumptions to interact randomly in the simulation process. Since the assumptions cannot replicate the innovation process, it is the process of random interaction that replicates reality by suggesting thatwithin the assumptions set, people behave randomly.

To operationalize a model of the diffusion concept, a

series of gaming operations is developed in order to generate the results. Hagerstrand observed the immigration pattern of people in Asby District, Sweden, and he concluded that there is tendency for the number of migrating households per square kilometer to decline rapidly with distance for about 1.5 kilometers and to decrease much less rapidly beyond this distance. He called this type of situation a "distance-decay circumstance." 10 HZgerstrand then transposed this idea to social contacts and affirmed that the probability of the word-of-mouth spread of innovation should be higher for near neighbors than distant neighbors. By rotating the distance decay curve he generated from the observed process through 360', he created the mean information field for the contacts in the Asby area.

The probability of contact of each unit of people then

depends on the location of the unit in the Mean Information Field with respect to the center of the field-hence, the assignment of the probability was cumulated from 0.0 to 0.999. (This cumulation of the probability eases the use of random numbers in the simulation process.) The simulation can be done either of two ways. It can be done by hand in the form of a dice game or the computer can be employed to do the gaming. if


it is done by hand, three sets of random numbers are needed. The first enables the identification of the area to be contacted. The second set of random numbers allows the unit of people in the area to be contacted while the third set indicates whether the contact is successful or not. If the simulation is done by the computer there must be well-defined parameters indicating the contact structure. In the present study the parameters used in defining the structure of the model are given in Appendix B.

Mean Demand Field

A Mean Demand Field (MDF) has the same structure as

the Mean Information Field (MIF) which has been used and explained by many other studies concerning diffusion theory. 11 Instead of defining the field as thousands of contacts which produce a smooth decrease of intensity of contact away from an origin, it is more appropriate for this research to define the probability field as indicating the probability of any parcel of land (not having a commercial use) changing to commercial use.*

In the MIF, as information is passed from a "knower" to a "non-knower," the probability field shifts to include the higher probability of "new knowers" now telling other "nonknowers." In analogous fashion, the MDF probabilities will also shift, once a parcel of land is changed to a commercial use. Since the commercial activities are attracted to the airport, the higher probabilities of land-use change to

This definition differs from that of MIF in that physical contact or information is not involved. The MDF is based on the attributes of each parcel of land.


commercial uses will affect the closest non-used land parcels, ceteris paribus. It should also be emphasized that the MDF contains a random assignment element, just as the MIF. Probabilities become variable because of the site and situation (attributes, including relative location) of the individual land parcels. Those data that were used to measure the attributes are discussed in the following section.

Data Collection Procedures

The data used in this research are derived from three principal sources: tax records, field surveys, and field interviews. The initial data collection was accomplished in the Duval County Property Appraiser's office. Such information as ownership of land, use of each parcel of land and actual

date the land changed ownership are kept up to date in the office's records. The ownership of land and use information for each parcel are derived from the appraiser's record book while the time land changed ownership was derived from deed records.

Questions may arise concerning the validity of using the date land changed ownership as opposed to the date that the commercial facility actually was built (or opened for business). It is argued here that the new land owners know at the time of purchase what the intended use will be and that the land acquisition was actually the first step in the change of land use.

The second major source of data was by field survey. This method was applied for two main reasons: (1) To validate the


data obtained from the tax records about the use of the parcels of land and their locations. (2) To familiarize the researcher with the airport and surrounding area so that an accurate conclusion can be made from the study.

The third main source of data was collected by field

interviews with owners or managers of establishments and parcels of land involved. This procedure was accomplished so that the opinions of owners and managers of land and establishments in the study area could be accurately perceived. The interviews were accomplished by means of a prepared questionnaire (see Appendix A) that was administered individually to all land owners or managers of establishments in the study area. During the interviews, questions were asked orally and the entire interviews were tape-recorded. After integrating and compiling data from the tax records, field survey and field interviews, the data were converted into a common base which will be discussed in the next section.

Conversion from Nominal to Ordinal Data

A portion of the data collected from tax records and all

data obtained from interviews are nominal-scale data which cannot be utilized in the Monte Carlo model.* In the light of this,

Nominal data are those in which a number is used as a
label for a class of category while all numbers in that category have the same defining number. Ordinal data, on the other hand, measure situations where numbers are assigned to phenomena utilizing the rank order system.


those data collected on a nominal scale have to be converted to an ordinal scale. Some of the conversions were based on observation of the location of each parcel after the field survey had been done and individual owners interviewed. For example, the accessibility variable for each parcel was done visually from the base map of the study area. If a parcel devoted to commercial activity was located on a major road serving the airport or was within the airport property, it was assigned a value of 5, and the rating decreases to 0 with no feasible accessibility. All variables used in the study were scaled from 0-5. The higher the score, the better the location for commercial activities.

Ownership in the study area is divided into three types: private individual, private corporation, and public ownership. The public ownership scores the highest since the land is the airport property and any change to commercial use in the land must be related to the airport. The private corporation scores lower than the public ownership but higher than private individuals because the corporation may have capital for investment while the lack of sufficient capital may hinder private individuals in establishing a commercial service. After the data had been converted, simulation of the data was accomplished by using the Monte Carlo model of spatial diffusion (see Appendix B for the parameters in the simulations).


Comparing the Simulation witk observed Pattern

After the simulation there is a need to determine

whether there are any differences between the observed process of change and the simulated process. Various researchers have used numerous types of statistical testing to do this. 12 In this study, however, the differences are noted by comparing the observed changes with simulation changes over time and space.

If the application of diffusion theory to land-use is valid, the curves produced by the observed and simulated processes of change should be logistic curves, and correspond very closely. In this case, the results must be consistent with the gravity exponential theory*. Also, the spatial distributions of the changes should conform to contagious or hierarchical or contagious-hierarchical spread. The computation of the results from these two components of spatial diffusion (time and space) will help in reaching conclusions from the study.

*The gravity concept follows an inversely proportional distance-decay principle. The exponential concept, on the other hand, postulates that the increase in activities at any time period is proportional to the size already attained.



1'Reynolds, Smith, and Hills, Inc., Master Plan Report for Jacksonville International Airport, Jacksonville Port Authority, Jacksonville, Florida (1965).

2 Ibid.


4 Peat, Marwick, Mitchell and Co. in association with Reynolds, Smith, and Hills, Inc. 1992 Airport Master Plan Study, Jacksonville- International Airport, Craig Airport and Herlong Airport. Prepared for Jacksonville Port Authority, Jacksonville, Florida (March 1974).

5 Peat, Marwick, Mitchell and Co. in association with
Reynolds, Smith, and Hills, Inc. Technical Supplement 1992 Airport Master Plan Study, Jacksonville International Airport, Craig Airport, Herlong Airport. Prepared for Jacksonville Port Authority, Jacksonville, Florida (March 1974); Summary Report, 1992, Airport Master PlanStudy, Jacksonville International Airport, Craig Airport, Herlong Airport. Prepared for Jacksonville Port Authority, Jacksonville, Florida (March 1974).

6 Various types of maps could be obtained with different degrees of information. Some are less detailed and others are more detailed than needed for this study, but any type of map on land use can be obtained. The Northeast Regional Planning Board in Jacksonville also has much information in the form of maps. In preparation by this Board is a new map for the land use of the whole of Duval County.

7 Various studies have emphasized such impacts on the environment. For example, see the following studies: P. M. Lynagh, "The Airport and the Environment," High-speed Ground Transportation Journal, Vol. 7 (1973), pp. 53-66; Airports and Their Environment: A Guide to Environmental Planning. Prepared for Department of Transportation (1972), and distributed by National Technical Information Services, U. S. Department of Commerce.

8D. M. Smith, Industrial Location: An Economic Geographical Analysis (New York, John Wiley and Sons, Inc., 1971) p. 270.

9 T. Hagerstrand, "On Monte Carlo Simulation of Diffusion," in W. L. Garrison and D. Marble, Eds., Quantitative Geography, Part 1: -Economic and Cultural Topics (Northwestern University, Evanston, Illinois 1967), pp. 1-32.


10M. H. Yeates, An Introduction to Quantitative Analysis in Economic Geography (New York, McGraw-Hill Book Company, 1968), pp. 53-63.

11For further readings on Mean Information Field see
the following studies: D. F. Marble and J. D. Nystuen, "An Approach to the Direct Measurement of Community Mean Inf ormation Fields," The Regional Science Association Papers Vol. 11 (1963), pp. 99-109; D. Amedeo and R. G. Golledge, An Introduction to Scientific Reasoning in Geography (New York, John Wiley and Sons, Inc., 1975), pp. 265-271; T. H~gerstrand, Innovation Diffusion as a Spatial Process (Chicago, The University of Chicago Press, 1967); R. Morrill and F. R. Pitts, "Marriage, Migration and Mean Information Field: A Study in Uniqueness and Generality," Annals of the Association of American Geographers, Vol. 57 (1967), pp. 401422; L. Brown, Diffusion Dynamics: A Review and Revision of the Quantitative Theory of Spatial Diffusion of Innovation
(the Royal University of Lund, Sweden, Department of Geography 1968); P. M. Blaikie, "The Spatial Structure of Information Networks and Innovative Behaviour in the Ziz Valley, Southern Morocco," Geografiska Annaler, Ser. 55B; pp. 83-105; K. E. Haynes and W. T. Enders, "Distance, Direction, and Entropy in the Evolution of a Settlement Pattern," Economic Geography, Vol. 51 (1975), pp. 357-365.

12Here are some examples of studies that have adopted these methodologies: L. Brown and K. Cox, "Empirical Regularities in Diffusion of Innovation," Annals of Association of American Geographers, Vol. 61 (1971) pp. 551-560; F. C. Fliegel and J. E. Kirlin, "Attributes of Information Factors in Diffusion," The American Journal of Sociology, Vol. 72 (1966), pp. 235-248; R. J. Colenutt, "Linear Diffusion in an Urban Setting: An Example," Geographical Analysis, Vol. 1 (1969) pp. 106-114; F. R. Pitts, "Problems in Computer Simulation of Diffusion," The Regional Science Association Paer, Vol. 11 (1963) pp. 111-119



Overview of Chapter

There are two components to spatial diffusion. one deals with the distribution of phenomena over time and the other deals with the distribution of the same phenomena over space. These two components are interrelated components of the same process. For the application of diffusion theory to

the analysis of commercial land-use changes to be valid, the results must conform with both the spatial and the temporal elements of the changes.

With this aim, the chapter is divided into three sections. The first section presents a comparison of the simulated landuse changes with the observed changes, over time, without considering location. The second section compares the spatial distributions of the simulated patterns with the observed. The last section analyzes the findings of the first and the second sections.

The Simulation of Change Over Time

Figure 4 shows the simulation of the commercial land-use changes using an exponential function to define the change of




70 _,






60 0



10 /

65 66 67 68 69 70 71 72 73 74 75 76 77



structure, both with and without characteristics.* Both of these cases produce a logistic curve, simulating that, in the initial stage of the process, very few parcels changed to commercial use, but as time progressed many parcels of land changed use before a leveling-off stage was reached. For example, the curve that was obtained by including parcel characteristics (Figure 4) indicates that, in 1965, there was only one parcel of land which changed. By 1966, however, the number had risen to five and by 1967 the number had reached 11. From that time on there has been an upward trend until 1972, when 69 of the 71 parcels of land had changed; from 1972 to 1976, no more changes occurred. This is the type of logistic curve that would be expected in a spatial diffusion process, as can be seen by its similarity to the normative curve (the curve produced without characteristics).

Simulation without characteristics (normative curve) indicates that, during the entire time period (12 years), only 51 of the 71 parcels, of land changed to commercial use, lending support to the fact that "real world" situations cannot be simulated without "real world" facts.

Figure 5 shows the simulation of the commercial land-use

changes using the gravity function to define the change structure, both with and without characteristics. Also, both of these cases produce a logistic curve, simulating that, in the initial stage of the process, there were few parcels which changed use before

Characteristics, as used in this chapter, mean the
attributes of the parcels of land, such as ownership, topography, government regulationsaccessibility, drainage and so on.







Lt) C)

C 2O


1 0 -- i


65 66 67 68 69 70 71 72 73 74 75 76 77



the leveling-off stage was reached. When this function is used, the curve obtained by inputting parcel characteristics indicates

that 71 parcels of land involved changed in 11 years in comparison with the normative curve that yielded 56 parcel changes in 12 years.

The curve obtained by plotting the observed data

does not totally resemble any of the curves discussed earlier. Unlike the simulated results that have been presented (Figures

4 and 5), the observed graph (Figure 6) has a peculiar, although explainable, shape. The observed process has a slow start, common to the curves already presented. In 1968, however, there was a sudden jump in the number of parcels of land that changed (Figure 6). This is accounted for by the opening of the airport which had to attract certain activities before it could function. Most of the airline offices, car rental businesses and freight companies started operation that year.

Since 1969, however, the curve produced by the observed changes resembles a logistic curve. This is evidenced in the fact that, by 1969, only one parcel changed; by 1970 four parcels changed, and by 1971 seven parcels changed. The upward trend continued until 1975 when only eight parcels changed, and by 1976 the number that changed was back to four. The slow pace of 1969 can also be attributed to lack of demand generated by the airport since all those necessary commercial activities started the same year as the airport (1968). J. F. Alvarez, who owns 800 acres of land adjacent to the airport, suggested a reason for the slow trend. He stated that:







30 "
20 S40




I I I 1 1
65 66 67 68 69 70 71 72 73 74 75 76 77


In the 1960s, land value was very high and people
offered good prices. P -,,al estate went down,
however, in the late 1960s. It went way down in the
early 1970s after the airport opened. Before the airport was opened there was a great anticipation
for the airport . everybody was talking that
the big plan . big money was coming. We have
Tenneco coming in, and when the Tenneco deal
failed, the negative effect of the airport was
felt and everything went way down.1

Whether one examines Figure 6 at the initial stage or

from 1969, there are some elements of a logistic curve in the graph. It is not known, however, if the observed curve (Figure 6) conforms more to the curve produced by the simulation with exponential function or the curve produced by the gravity function.

Figure 7 shows the comparison of the observed curve with both the exponential and gravity function simulated curves. For the first three years of the process (1965, 1966, and 1967) both the exponential and gravity functions have the same trend of change. After 1968, however, there was a deviation in the trend of the curves, as that produced by the gravity function is steeper than that produced by the exponential function. Such deviation should be expected since the gravity function implies an inversely proportional distance-decay principle, whereas the exponential function postulates that the increase in activities at any time period is proportional to the size already attained. The law of supply and demand is involved, the number of parcels that changed each year depends on the number of commercial activities that are already present around the airport.







4 04-4~


S 30


65 66 67 68 69 70 71 72 73 74 /b 76 77 Time in Years Figure 7 EXPONENTIAL AND GRAVITY CHANGES WITH CHARACTERISTICS


The observed process of changes in the study area deviates from both the exponential and the gravity functions at the initial stage (1965-1968). The major reason advanced for the deviation is the opening of the airport. Before this occurred in 1968, there were only four commercial activities in the study area; during the same time there were eleven commercial activities simulated with both the exponential and gravity functions, respectively. In 1968, when the airport was opened, there were 21 additional parcels of land that changed in the observed process, as compared with 10 and 11 parcels of land for exponential and gravity functions, respectively.

As can be seen, the exponential graph seems to be closer

to the "real world" situation from 1969 onward. Figure 7 indicates that the spacing of the changes over time from 1969 of the "real world" process may be proportional to the rate of increase and decrease of the exponential function. For example, in 1969, fifteen parcels of land changed, using exponential functions, fifteen additi onal parcels changed in 1970 and twelve, seven, and one parcel changed in 1971, 1972, and 1973, respectively. The "real world", on the other hand, also had one parcel changing in 1969, four parcels in 1970, and seven, five, ten, and nine parcels changed in 1971, 1972, 1973, and 1974, respectively. There is no such comparison, however, with the gravity function. In 1969, 26 parcels of land changed their use to commercial, 19 parcels changed in 1970, four and zero parcels changed in 1971 and 1972, respectively. As noted from Figure 7, the exponential function seems to agree better with "real world" changes.


Figure 8 compares expom.- ntial function simulations, with characteristics and without characteristics, with the observed process. The figure shows that there are some similarities and some dissimilarities among the three curves. The dissimilarities occurred mainly at the beginning of the process. Here the observed distribution deviates completely from the simulation up to 1969. From 1969 onward, the observed process resembles the simulation without characteristics for about four years. During 1969 to 1972, the slopes of the curves produced by both the observed and the simulation without characteristics of exponential function are almost identical. From 1972 to 1974, the slope of the observed process roughly parallels the exponential simulation with characteristics, a further indication that "real world" data will simulate the "real world" better than hypothetical data. The overall evaluation of Figure 8 is that -Ehe observed'proces's of commercial land-use changes around the Jacksonville International Airport tends to conform with the simulation that is based on the exponential function.

Spatial Distribution of the Simulated
Commercial Land-Use Changes

Presented in Figures 9 through 11 are the spatial distributions of the simulated process results, as the concern is now with the spatial component of the diffusion process., In order to better show the distributions on a map, the twelve years covered by the study were divided into six time periods:



70 - --- --/


0 /




1 0


65 66 67 68 69 70 71 72 73 74 75 76 77


Ic At II



A 1965-1966 1971-1972

1967-1968 0 1973-1974

LI 1969-1970 0 1975-1976



k !


Figure 10


A 1965-1966 *1971-1972

@1967-1968 0 1973-1974

O 1969- 19 70 C1975-1976 AA



Slt I


01967-1968 01973-1974

E31969-1970 01975-1976



first period, 1965-1966; second period, 1967-1968; 1969-1970, 1971-1972, 1973-1974, 1975-1976 are the third, fourth, fifth, and sixth time periods respectively.

Figures 9 and 10 demonstrate a well-defined pattern of change to commercial land use. The patterns generated through the use of an exponential distance-decay function appear to be very similar to the classic contagious diffusion pattern. Such similarity would be expected in Figure 10, because "real-world" characteristics were not included, producing a normative simulation. of greater interest is the contagious diffusion type pattern achieved in Figure 9, when characteristics were used in the simulation. Most of the changes occurred in or very close to the airport and extended outward, very similar to the "real world" situation. Recency of commercial development increases with increasing distance.

Another interesting pattern can be noted in Figure 9. Not only does the spatial distribution resemble contagious diffusion, it shows also that a hierarchical diffusion pattern may be developing. In the second time period, seventeen parcels of land changed. The majority cf these parcels were located very close to the airport but two were located further away on a major road that serves the airport. These two parcels then initiated another contagious effect around them. In the third time period (1969-1970) most of the changes occurred along this road.

Figure 11 represents the simulation with the gravity

function. As can be observed, there is no defined pattern that


can be related to any of the L patial diffusion types (i.e. contagious, hierarchical, or contagious-hierarchical). This supports the fact that the gravity function is clearly not appropriate for the simulation of the changes.

Comparison of the Observed Spatial Pattern with
the Simulation Spatial Pattern

By comparing the observed process of commercial land-use

changes (Figure 12) around the Jacksonville International Airport with the simulated pattern shown by Figures 9 and 10, a number of striking similarities can be noted. At first, most of the changes occurred around the terminal building and later spread outward. However, there are some noticeable exceptions in Figure 12 as five of the parcels do not conform with the trend. The main reason is that the land occupied by these establishments had been in use since 1968 (which corresponded with the second time period), but there were changes in ownership in recent years.

Generally, the parcels in the airport and those close

to it changed before those distant from the airport (Figure 12). The commercial activities along the airport road do not indicate any significant change until the third time period (1969 and 1970) of the process. All those along State Road 17 (SR 17) indicate no significant change until the fifth time period (1973 and 1974). Those establishments along the airport road are older than those farther away from the airport as can be noted in Figure 13.

Figure 12


A 1965-1966 1971-1972


These are some of the commercial activities ,that came into being during the third stage of the process (1969-70) on the airport road.

A new establishment that came in the fifth stage of the process (1973-74) at the north end of the study area. On the opposite side of this gas station another tourist stop is being built.

Figure 13