|
A SPATIAL EMERGY MODEL
FOR
ALACHUA COUNTY, FLORIDA
BY
JAMES DAVID LAMBERT
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1999
Copyright 1999
By
James David Lambert
Dedicated
To
Marilyn M. Winston
ACKNOWLEDGEMENT
I would like to express my sincerest appreciation to my committee chairman,
Dr. John F. Alexander, and to my other committee members, Dr. H.T. Odum, Dr. Earl M.
Starnes, Dr. Lawrence D. Harris, and Dr. Paul D. Zwick for their support and patience
over the years. Special appreciation is due Dr. Howard T. Odum who has changed my
way of looking at the world forever. This work would not have been completed without
the encouragement and support of my most significant, best friend, and fellow student,
Marilyn Winston.
TABLE OF CONTENTS
page
A C K N O W L E D G E M E N T ............................................................................................... iv
A B S T R A C T ................................................................................................................... v iii
INTRODUCTION ........................................ ......... ........................... 1
T heoretical B asis for the Study ................................. ...................... ................ 1
EM ERGY D defined .................................................... 2
Energy System s D iagram s ............................................... ................. 4
T ransform ity D efined ............................................................. ............. 6
Spatial Distribution of Energy According to Transformation Hierarchy ... 7
E energy Signatures .. ... ................................... ..................... ........... .. 9
Land Area U nit M odel D iagram .......................................... ............. 10
Technological B asis for the Study ................................................... ............. 12
Previous Studies of the Spatial Distribution of EMERGY in the Landscape ..... 13
S tu d y A re a ........................................................................................................... 1 8
Study O objectives ........................................................................................... 2 1
O organization of the Study .................................... ........................ .............. 22
M E T H O D S ...................................................................................................................... 2 5
G general D description of the M odel ................................................... .............. 25
Softw are and H ardw are U sed .................................. ...................... ............. 31
Development of the GIS Database ...................... ............. 32
General Methods for Creating Component Grids ......................................... 37
A nalytical C overages ................ .. ....................... ..................... 37
(Sub)Component Grids from Point-feature Coverages ........................ 38
(Sub)Component Grids from Line-feature Coverages ............................ 42
(Sub)Component Grids from Polygon-feature Coverages ................... 44
Specific Methods for Creating Flow Component Grids ................................ 46
Natural System Flows-Gross Primary Productivity ............................. 46
Natural System Flows-Renewable Sources .................................... 52
Urban System Flows-Water Consumption ..................................... 55
Urban System Flows-Fuels and Electricity .................................... 62
Urban System Flows-Transporation Subcomponent ............................ 62
Urban System Flows-Buildings and Agriculture Subcomponent ...... 64
Urban System Flows-Goods Consumption .................................... 72
Urban System Flows- Human Services .............................. ............. 79
Urban System Flows-Wastes Not Recycled ................................... 90
Urban System Flows- Goods Recycled ................................................. 98
Specific Methods for Creating Storage Component Grids ................................ 104
N natural System Structure ............. ..... ................ ....... ....... ......... 104
Natural System Structure-Biomass Subcomponent ......................... 105
Natural System Structure-Water Storage Subcomponent ............... 106
Natural System Structure-Organic Matter in Soils Subcomponent .... 110
U rban System Structure ................................................... ............. 112
Urban System Structure-Buildings Subcomponent ......................... 112
Urban System Structure-Roads Subcomponent ........................ 116
Urban System Structure-Utility Infrastructure Subcomponent ....... 119
Urban System Structure- People ....... ......... ................................. 122
General Methods for Creating Analytical Grids .......... .. .............. 128
L o g arith m G rid s ............... ................................................................. 12 9
Total EMERGY Consumption and Total EMERGY Storage.............. 130
T ran sform ity G rid s ................................................................................ 132
Other Analytical Grids ............................... 132
Using the Model for Analysis ............................... 134
M ap s and H istogram s ............................................................................ 134
County-wide (Sub)Component EMERGY Signatures ....................... 135
C om p arativ e Stu dies ............................................................................. 136
EMERGY Ratio Analysis ................. .............. 139
Spatial Context Analysis ............................... 140
One-kilometer Cell-size Model ................. .............. 141
R E S U L T S ...................................................................................................................... 14 2
T he G IS D database ..... ... ....................................... ....................... . ........ .. 142
New Primary Data Coverages ........................................ ............... 143
Analytical Coverages ............................... 159
Sum m ary Coverage ........................................................ ............... 181
The Spatial EMERGY Model ............................... 186
EMERGY Flow Components ............... ...................... 192
EMERGY Storage Components ....................................... .............. 271
Total EMERGY Consumption and Total EMERGY Storage......................... 330
Total EMPOWER Density Analytical Grids .......... .............. 330
Total EMSTORAGE Density Analytical Grids ......................... 332
Transform ity A nalytical G rids .................................... ....... ............. 346
County-wide (Sub)Component EMERGY Signatures ....................... 351
County-level Transformities ................. .............. 360
C om parative Studies .................................................. ............................... ..... 361
Land Use and Cover Classification Study .................... ................ 361
Planning Unit Study ............................... 412
Neighborhood Study ................. .............. 426
EM ERGY Ratio A analysis ..................................................... ... .... ............. 440
Resource Use Analytical Grids ................. .............. 440
EM POW ER Ratios ....... ........... .............. .................... 446
EMSTORAGE Ratios ................. .............. 458
Spatial Context A analysis ......................... ............................... .... ............. 470
EMPOWER Density Context Analysis ........................ 472
EMSTORAGE Density Context Analysis ................... .............. 484
One-kilometer-resolution Model .................. .............. 496
D ISCU SSION ............................................................................. ........................ 502
G general Theoretical M ethod ........................................................... ............. 503
Strengths and Weaknesses of the Land Area Unit
Spatial M odeling A approach ..................................... ............... ............. 504
Evidence of Spatial Patterns of Energy Transformation Hierarchy ............... 507
Hierarchical Patterns Associated With Individual Components.......... 507
Total EMERGY Consumption and Total EMERGY Storage.............. 512
Transformities for Total EMERGY Consumption
and Total E M ERGY Storage...... ..... ....................................... 513
Transformities, Mean Densities, and Percentages
for Land U se Classifications ....... .......... .................................... 514
Value Added by Using GIS Technology ....... ......... .................................. 518
Suggestions for Future Research ....... ....... ........ ................... 518
C lo sin g R em ark s ....... .. .................................... .......................... ............. 52 0
A P P E N D IX .................................................................................................................. 5 2 1
REFERENCES ................................................. ......................... .. 558
BIOGRAPH ICAL SKETCH .. .............................................................. ............. 569
Abstract of Dissertation Presented to the Graduate School
Of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
A SPATIAL EMERGY MODEL
FOR
ALACHUA COUNTY, FLORIDA
By
James David Lambert
August 1999
Chairman: John F. Alexander, Jr.
Maj or Department: College of Architecture
A spatial model of the distribution of energy flows and storage in Alachua
County, Florida, was created and used to analyze spatial patterns of energy
transformation hierarchy in relation to spatial patterns of human settlement. Emergy, the
available energy of one kind previously required directly or indirectly to make a product
or service, was used as a measure of the quality of the different forms of energy flows
and storage. Emergy provides a common unit of measure for comparing the productive
contributions of natural processes with those of economic and social processes-it is an
alternative to using money for measuring value.
A geographic information system was used to create a spatial model and make
maps that show the distribution and magnitude of different types of energy and emergy
flows and storage occurring in one-hectare land units. Energy transformities were used
to convert individual energy flows and storage into emergy units.
Maps of transformities were created that reveal a clear spatial pattern of energy
transformation hierarchy. The maps display patterns of widely-dispersed areas with
lower transformity energy flows and storage, and smaller, centrally-located areas with
higher transformities. Energy signature graphs and spatial unit transformities were used
to characterize and compare the types and amounts of energy being consumed and stored
according to land use classification, planning unit, and neighborhood categories. Emergy
ratio maps and spatial unit ratios were created by dividing the values for specific emergy
flows or storage by the values for other emergy flows or storage. Spatial context
analysis was used to analyze the spatial distribution patterns of mean and maximum
values for emergy flows and storage.
The modeling method developed for this study is general and applicable to all
types of landscapes and could be applied at any scale. An advantage of this general
approach is that the results of other studies using this method will be directly comparable
with the results of this study. The results and conclusions of this study reinforce the
hypothesis that an urban landscape will develop a predictable spatial pattern that can be
described in terms of a universal energy transformation hierarchy.
INTRODUCTION
The primary goal of this study is to create a general type of spatial model of the
distribution of energy flows and storage in urban and regional landscapes that can be
used to study spatial patterns of the energy transformation hierarchy and energy
signatures in relation to spatial patterns of human settlement. EMERGY (capitalized
throughout this study to avoid confusion with the word 'energy') is used as a measure of
the quality of the different forms of energy flows and storage that are included in the
model. A secondary goal of the study is to demonstrate how recent advances in computer
and information science technologies can be applied in a way that provides new insights
into the spatial patterns of energy flow and storage.
The primary target audience for this study is urban and regional planners.
However, anyone interested in how man-dominated landscapes are spatially organized to
process and store both local, natural energy sources and imported, nonrenewable energy
sources should be interested in the results. The specific spatial model presented in this
study was implemented at a spatial resolution that makes it particularly useful for
describing, quantifying, and studying patterns that occur within spatial units of urban
systems. Although the focus for this study is on patterns of urban systems, the modeling
methods presented in this study are intended to be general and applicable to all types of
landscapes and most could be applied at any scale.
Theoretical Basis for the Study
This study is based on general and ecological systems theory developed over the
past three decades by H.T. Odum and his associates (Odum, 1971; Odum, et.al., 1976,
Odum, 1983, 1988, 1992; Hall, 1995; Odum,1996; Odum, et.al., 1998). A basic review
of the concepts and principles of this theory that are most important and relevant to this
study are presented in the following sections.
EMERGY Defined
In his latest book, Environmental Accounting EMERGY and Environmental
Decision Making (1996), H.T. Odum explains how the concept of EMERGY can be used
as the basis for an environmental accounting approach that measures and evaluates the
relative contributions to real wealth that are made by both environmental and economic
processes. In this book, Odum (1996, p.6) defines real wealth as anything that ". . is
produced and maintained by work processes from the environment. . ." He explains
how real wealth can be measured in terms of EMERGY in the following paragraph.
To build and maintain the storage of available resources, environmental work has
to be done, requiring energy use and transformation. We can quantitatively
evaluate the storage by the work done in its formation. Work of energy
transformation can be measured by the availability of the energy that is used up.
Thus, real wealth can be measured by the work previously done. EMERGY is a
scientific measure of real wealth in terms of the energy required to do the work of
production. (Odum, 1996, p. 7)
EMERGY is defined as the ". . available energy of one kind previously required
directly or indirectly to make a product or service. ." (Odum, 1996, p.7). This study
will use a variant of EMERGY, called solar EMERGY, as its primary unit of measure.
Solar EMERGY is defined by Odum (1996, p.8) as ". . the available solar energy used
up directly and indirectly to make a service or product. Its unit is the solar emjoule
(abbreviated sej)." These straightforward definitions do not do nearly as much justice to
the elegance of this concept as the following quote from Dr. Odum's book in which he
presents the concept from the perspective of "common folk".
A commonplace idea among people everywhere is that some things take more
effort than others. Long before physical science made a narrow quantitative
definition of energy, the word "energy" was used to refer to the work done. For
example, it was said that a house took more energy to build than a chair. The
universal idea among common folk was, and still is, that putting more energy into
something generates more value. The concept of EMERGY is scientifically
defined to give a quantitative measure to this ancient principle.
Because of the way it is defined, the scientific use of "energy" does not coincide
with the folk concept. The practical, operational scientific measure of energy is
the heat generated when various forms of energy are converted. The scientific
concept makes no allowance for different kinds of energy representing different
levels of effort. The scientific concept rates a calorie of sunlight, electricity,
nuclear fission, and human thinking as equal. In other words, the different levels
of prior effort involved in generating different kinds of energy are ignored ....
EMERGY is a measure that looks back upstream to record what energy went into
the train of transformation processes. The computations recognize that each type
of energy has a different upstream energy input which must be included in order
to summarize all the energy of one type that is required to generate the output.
(Odum, 1996, pp. 13-14)
With these concepts in mind, EMERGY was chosen for this study because it
measures the work contributed by both nature and society to the production and storage
of 'real wealth'. An advantage of using EMERGY for a study such as this is that the
concept provides a method for comparing storage and flows related to natural processes
with those of economic and social processes. EMERGY provides an alternative to using
money for measuring the contributory value of an element of the system. Money does
not work well for valuing the contribution of the 'free' energy sources from nature. The
4
essential contributions from these free resources are often ignored, if not forgotten, in
evaluations of economic systems because of the difficulties of valuing natural resources
(Ahmad et al., 1989; Costanza, 1991; Uno and Bartelmus, 1998).
Energy Systems Diagrams
EMERGY analysis is based on diagramming the flows and storage of energy
using a standard symbol language (Odum, 1983, 1996). These energy system diagrams
are drawn based on an understanding of several energy-related characteristics of the
system being modeled. These characteristics include: what the boundaries of the system
are, what outside energy sources contribute to the system, how energy flows interact
within a system, and what the flow pathways are between interactions and energy storage
elements of a system being modeled. The diagrams are drawn according to a strict set of
rules, defined and described by H.T. Odum (1983), so that the laws of thermodynamics
are observed.
A simple example is shown in Figure 1-1. It has flows from two types of
potential energy sources interacting in a productive process to create a storage or
structure (real wealth). The first law of energy conservation is observed in the
diagramming of the energy flows and storage (all of the energy entering the system is
accounted for-none is created or destroyed). The diagram also illustrates the second
law of thermodynamics by accounting for the fact that some energy loses its ability to do
work in the production interaction process (leaving the system in the form of dispersed
heat), and that all storage will naturally depreciate over time and must be maintained by
productive processes.
Figure 1-1: A simple energy system diagram showing two sources of potential energy
interacting in a productive process that builds a storage or structure. This storage is "real
wealth" as defined by Odum (1996). The diagram illustrates that, because of the second
law of thermodynamics, much of the potential energy used in the productive process is
used up (loses its ability to do work) and leaves the system in a more dispersed form
(usually as heat). The storage also loses potential energy over time. Surviving systems
will self organize to develop structure that has reinforcing feedback to the production
process (Odum, 1983).
Transformity Defined
Transformities are used to calculate the EMERGY in a product or service. H.T.
Odum describes the concept of transformity in the following quotes from his book.
The quotient of a product's EMERGY divided by its energy is defined as its
transformity (Odum, 1976, 1988). The units of transformity are emjoules per
joule .... Solar transformity is the solar EMERGY required to make one joule of
a service or product. Its units are solar emjoules per joule (sej/J). A product's
solar transformity is its solar EMERGY divided by its energy.
The more energy transformations there are contributing to a product, the higher is
the transformity. This is because at each transformation, available energy is used
up to produce a smaller amount of energy of another form. Thus, the EMERGY
increases but the energy decreases, and therefore the EMERGY per unit energy
increases sharply. . Goods and services that have required the most work to
make and have the least energy have the highest transformities. Examples are
human services and information . .most energy transformations are controlled by
inputs of high transformity, whose energy contribution is small but whose
EMERGY contribution may be large. An example is the control of a forest by
people ... (Odum, 1996, p. 11)
Transformities, as defined above, can be used to distribute all forms of energy
along a universal energy hierarchy based on the magnitude of transformity values (Odum,
1996). The concept of universal energy hierarchy is based on the second law of
thermodynamics that states that available energy is degraded in any transformation
process. The second law explains why we see hierarchy in all types of systems.
Odum (1996) points out that a familiar pattern seen in many types of systems can
be described in terms of transformities-many units of lower transformity energy
converge and interact through productive processes with other higher-transformity forms
of energy to create even fewer units of other forms of energy with even higher
transformities.
7
Spatial Distribution of Energy According to Tranformation Hierarchy
Odum and others (Odum and Brown, 1976; Constanza, 1975, Brown, 1980,
Odum, 1983, 1996; Whitfield, 1994; Huang, 1998) have proposed that the spatial
distribution of energy flows and storage tend to follow patterns based on the level of
transformity for the flow or storage (illustrated in Figure 1-2).
For example, low transformity flows tend to be widely dispersed compared to
higher transformity flows. In terms of the spatial arrangement of urban systems, these
previous studies predict that a city center will have a higher energy flow transformity
than the less developed suburbs surrounding the city center, and that rural areas outside
the city will have the lowest transformity for flows.
In the case of storage, there are usually many smaller, lower transformity
storage that are widely dispersed, and there are usually fewer of the larger, higher
transformity storage. Once again, in terms of the urban spatial pattern, this observation
suggests that a city center will have a higher storage transformity than suburbs. Because
of the large inputs of high-transformity nonrenewable resources and human services that
are required to develop and maintain all forms of urban structure, it is likely that the
transformities for urban storage will be larger than those for natural structure in the
surrounding areas.
Brown (1980, 1981) observed a hierarchical phenomenon in the sizes and
distributions of cities in the regional landscape of Florida, and described the phenomenon
in terms of energy and transformity hierarchies. Whitfield (1994) demonstrated this
spatial phenomenon in his study of the energy distribution patterns of Jacksonville,
Florida. Huang (1998) found similar patterns in a study of Taipei, Taiwan.
Increasing Transformity
Figure 1-2: Conceptual diagram for the spatial arrangement of various elements of a
regional landscape arranged in order of predicted transformity (adapted from Huang,
1998, and Brown, 1981).
Energy Signatures
Odum (1983) suggests that all systems self-organize to maximize power flow
according to the energy sources that are available to the system. The term 'energy
signature' has been used by Odum (1983) to describe the type and magnitude of energy
flows to a particular system. The nature and magnitude of these energy sources helps
determine the nature and magnitude of the productive processes and the storage that
occur in the system. This concept can be applied to both ecological and economic
systems. For instance, Twilley (1995) found that energy signatures could be used to
predict the morphological structure and ecological processes of mangrove ecosystems,
and Odum (1983) has used energy signatures to characterize the level of development of
countries and regions.
These energy signatures are typically presented as bar graphs with the
magnitudes of each of the various energy sources for the system accounted for in the
same units of measure (for energy or EMVERGY). Odum (1983) prefers that the sources
be arranged from left to right on the graph in order of increasing transformity to improve
understanding of the energy transformation hierarchy.
Whereas, the concept of energy signatures has been previously used to
characterize total systems, it is proposed here that they could also be used to characterize
distinct spatial units in the landscape. It is also proposed that the concept can logically be
expanded to include the magnitude of storage that are within each spatial unit in the
signature. An energy signature of areal land units that includes both energy flow and
storage magnitudes (measured in units of the same kind-in this case, EMERGY) could
be used as a multivariate land classification scheme that links classes with processes.
Land Area Unit Model Diagram
The unit model system diagram shown in Figure 1-3 is proposed as a general
basis for spatial modeling of the EMERGY flows and storage in equal-size units of land
area in an urban and/or regional system. The diagram shows how the EMERGY sources
from outside the land area unit model system flow in and interact with other flows
through production processes that maintain or build storage of EMERGY. This model
does not attempt to model flows between different land area units-each unit is
considered to be a separate unit model system with flows coming from unspecified points
outside of the unit model system and going to unspecified points outside the unit system.
The spatial model created in this study is based on this diagram. In other words,
the individual storage and flow components of the spatial model correspond to elements
of the unit model system diagram. The diagrammatic elements that have corresponding
components in the spatial model are numbered for cross-referencing in the text.
To create the spatial model, the magnitude of each energy flow into each land
area unit and each energy storage within each unit was estimated based on the
characteristics and features of each land area unit. The EMERGY associated with each
energy flow or storage was calculated using the appropriate transformities for each type
of energy flow or storage. Energy or EMERGY signatures for land area units are based
on the magnitude of each flow and storage included in the diagram.
The horizontal dimension used in this study for the land area unit model was 100-
meter squared (one-hectare). The vertical dimension of each land unit was 250 meters
from the ground to the lower boundary and 100 meters from the ground to the upper
boundary of the land area unit. The time dimension used in this study is one year.
n .........., xr . ,Recycled
1" 8 Solid Waste
Land Area Unit Boundary 7 : Wastes
ndar Not Recycled
= EMERGY Flow ........................ = EMERGY Flow = EMERGY Storage
Evaluated Not Evaluated
Figure 1-3: EVIMERGY flows and storage evaluated for each land area unit in the spatial model. Individual flow and storage elements
have been numbered to provide cross-references in the text (see Figure 2-1). Note: GPP = gross primary production.
Technological Basis for the Study
Throughout much of modern civilization, the paper map has been the preeminent
tool for studying and managing the spatial patterns of natural processes and human
settlement. However, recent advances in computer technologies and in the field of
geographic information sciences have resulted in the development of sophisticated
geographic information systems (GIS) that offer a new paradigm for the study of spatial
processes and patterns-the digital spatial model.
There has been dramatic growth in the use of GIS technology in the last decade.
Most government agencies with land planning or management responsibilities have
adopted the technology, and many have created, or are in the process of creating, GIS
databases that describe the features of the land that they study, regulate, or manage. This
phenomenon has resulted in the widespread public availability of government-developed
GIS databases (Lambert and Zwick, 1997).
Typically these GIS databases have been created to support the very particular
spatial information needs of the agency, but often they can be used for other purposes
because of the general nature of the data. For example, GIS databases developed by
property appraisers contain descriptive attributes and, most significantly, the location of
buildings that can be used to precisely map the location of estimates for the storage of
energy and EMERGY in urban structure.
Detailed municipal GIS databases that describe property, road, and utility
structures form the basis of a new opportunity to examine the spatial distribution patterns
of energy and EMERGY in more detail than has been possible using paper maps and
13
traditional techniques. At the same time that the relevant digital spatial data have become
available, the computer hardware with enough computational power to manipulate these
large, complex databases is becoming less expensive. And finally, the capabilities of the
GIS software have evolved to make working with large databases much more practical
than in the recent past.
As previously stated, a secondary goal of this study is to demonstrate how recent
advances in technology can be applied to the study of the spatial patterns and relations of
energy flows and storage in landscapes. To a large degree, this study is made possible
by the temporal convergence of three necessary ingredients-data, hardware, and
software. It is hoped that by mixing these ingredients together with EMERGY theory
that new insights will be revealed in the results.
Previous Studies of the Spatial Distribution of Energy in the Landscape
About two decades ago there were several studies conducted that used energy
transformation hierarchy theory as a basis for studying urban and regional systems
(Brown, 1973; DeBellevue, 1976; Littlejohn, 1973; Maltby, 1977; H.T. Odum, 1971;
Regan, 1977; Sipe, 1977; Sipe, et.al., 1979; Steller, 1976; Zucchetto, 1975; Kangas,
1983). All of these studies used energy systems diagrams to model aspects of the urban
system. Most of these studies used land use maps to estimate areas of different land uses.
These area estimates were used to make estimates of energy flows and storage in the
systems being studied.
Many textual references were made in these studies regarding the relationships
between energy transformation hierarchy and the spatial patterns observed in the system
14
being studied. However, if any maps were included, they were usually simple hand
drawn diagrammatic maps that were used to document these relations or land use maps
that were used to make the area calculations.
In 1981 an energy analysis workshop was conducted at the University of Florida
to assess the research needs for developing a better understanding of the relationships
between humanity and nature (Brown and Odum, 1981). Most of the primary research
needs that were identified through this process were related to the need for a better
understanding of how ecological and economic processes affect the spatial organization
of landscapes.
There were several studies conducted about two decades ago that were
specifically aimed at examining the spatial patterns of the distribution of energy flows
and storage in regional landscapes as expressed in various embodied energy units. This
body of research was the initial inspiration for this research work. All of these early
studies used similar methods to estimate energy flows and storage per unit of area.
For instance, as part of the study 'Carrying Capacity of Man and Nature in South
Florida' (Odum and Brown, 1976), Costanza(1975) studied the spatial distribution of
land use subsystems in south Florida over the period from 1900 to 1973. He calculated
the metabolism and structure for each subsystem by multiplying the energy flows and
storage occurring in each subsystem by an energy quality factor and expressed the
results in fossil fuel equivalents (FFE).
Costanza's study used methods to measure areas of each land use that would be
considered rather primitive when compared to today's technological methods. First land
use maps were prepared from air photos, then these maps were cut up into pieces and
15
weighed to determine the area of each land use in each county. These rough area
calculations were used to calculate subsystem energy flows and storage by multiplying
the area times a generalized estimate of the flow or storage per unit area for each
subsystem.
He produced coarse resolution regional maps showing the spatial distribution of
incoming energies using isopleths, and spectral distribution graphs showing mean energy
flow per unit area, referred to in this study as energy density. Energy flow per unit area
was later referred to as power density (Odum, Brown, and Costanza, 1976).
Costanza (1979) utilized the data from this earlier research to build a spatial
simulation model for south Florida. This model divided the south Florida region into 88
16 by 16-mile cells. This spatial model was linked to an energy diagram simulation
model. According to Costanza, the model did a "fairly good job" of predicting the
historical development of spatial patterns between 1900 and 1973.
Brown (1980) studied the hierarchical organization of urban and regional
landscapes using embodied energy (expressed as coal equivalents, CE) as a basis for his
calculations. He calculated the power density and volume of structure for 11 urban land
uses. Urban power densities were calculated based on the embodied energy in the
electricity and other primary fuels used, and on the embodied energy in goods and
services. The power densities for natural ecological systems were measured by
estimating the gross primary production of each type of system.
Brown calculated areas of land uses in his study areas in the same manner as
Costanza, by cutting up land use maps and weighing the pieces. Estimates of direct
energy use within urban land use classes were derived from representative samples of
16
yearly energy consumption from utility records. These average figures were then
multiplied by the number of acres of the corresponding land use class. Estimates of
volume of structure per land use were also based on representative samples. Using a ratio
of the embodied energy per dollar of assessed value of 20,000 Cal CE/$, Brown
calculated an average figure for embodied energy in urban structure per acre of each
category of land. Embodied energy in goods and services was assumed to be
proportional to the consumption of direct energy (electricity and fuel). Brown used a
ratio of 21,000 Cal CE/$ to calculate the embodied energy in goods and services.
Using these data, he produced a series of graphs that demonstrated the
hierarchical distribution land use areas within his study areas and the phenomenon of
increasing levels of embodied energy flows and storage with increasing complexity of
land use. His study also demonstrated the hierarchical distribution of cities in Florida
when measured by their total embodied energy flow and storage.
Alexander (1981) used a dynamic landscape simulation model of Flagler County,
Florida to study the effects of regional development policies on settlement patterns. His
model combined energy theory with thematic mapping methods developed by Ian
McHarg (1969). The computer simulation model he used created simple maps using
symbols ( +, x,*,--, etc.) to plot out the spatial results. This is the first example (known to
the author) where a computer-generated map was created that was based on energy
transformation theory. Another early example of using computers to generate energy
maps was a study by Richardson (1988) that used a simple, but general, dynamic
simulation model to demonstrate how systems will self-organize spatially to maximize
power flows from available sources.
17
A more recent study by Whitfield (1994) examined the spatial patterns of
EMERGY flows and storage for Jacksonville, Florida. Whitfield evaluated the resource
flows and storage for 17 different land uses in his study area. He used averaged figures
for each land use to calculate the EMPOWER density, EMERGY storage, and
transformity for each land use.
Whitfield's study includes a collection of graphs he called "resource use
signatures" that show the amount of each resource flow by land use. Most significantly,
he generated maps of the total EMERGY flows and storage for Jacksonville, as well as
spatial profiles that he called "landscape signatures". These landscape signature graphs
show the quantity of EMERGY flow or storage on the y-axis and the x-axis indicates the
distance along a straight transect through the city. From these maps and profiles, the
spatial pattern of resource use and structure in the study area is clearly evident.
In a recent study of energy hierarchies in the landscape, Huang (1998) studied the
effects of the spatial convergence of energy on the evolution of spatial patterns of Taipei,
Taiwan. He used a model (similar to Figure 1-2) to explain how Taipei is spatially
organized to converge lower transformity energy flows from the natural and agricultural
areas into areas of higher concentration, with the area of highest concentration (energy
flow density) being at the urban center of the system. Huang (1998) observed that
"... mutual non-linear interactions between population, area, and assets of each zone
result in a self-organizing spatial system, from which an urban hierarchy evolves."
In this study, Huang estimated the magnitude of 14 different renewable and
nonrenewable flows of energy and EMERGY for 1178 administrative districts in Taipei.
He did not estimate energy stored in urban or natural structure, but implied that the level
18
of urban structure develops in tandem with the level of energy density. Using the flow
values estimated for each district, Huang used statistical analysis to cluster the districts
into six energetic zones to simplify his model and describe the concept of energy
convergence based on these zones.
Huang used the latest GIS technology to make maps and conduct analysis of
spatial patterns. For instance, using the estimated flows, he generated maps of the
individual flows, the combined renewable and nonrenewable flows, and EMERGY
investment ratio maps. Several EMERGY indices were calculated for each of the
energetic zones including: transformity of the zone, total zonal EMPOWER density,
EMERGY investment ratio for the zone, and per capital EMERGY used in the zone.
Study Area
The area chosen for spatial modeling and study is Alachua County, Florida, USA.
The map in Figure 1-4 shows where Alachua County is located in the State of Florida.
The model is based on data from the years 1993 and 1994. According to the 1990 U.S.
Census of Population (U.S. Census Bureau, 1992), the county had a population of
142,081 adults and 39,515 children. The county has a service-oriented economy with
the largest employer being state and local government. There is very little industrial
manufacturing, and only a small agricultural industry (cattle, vegetables, forestry).
The largest city in the county, Gainesville, is located in the center of the county,
and there are several small towns distributed in a radial arrangement around Gainesville.
Most (94%) of the population of the county works in the county, and most of the county's
employers are located in the City of Gainesville. Gainesville is also the location of
19
several regional shopping centers and regional hospitals and health service providers.
The University of Florida is also located in Gainesville. The University exerts a strong
influence on the spatial organization of the county in terms of the location of high-density
housing, service providers, and road infrastructure.
The rural areas of the county are best described according to the eastern and
western halves of the county. The eastern portion of the county is characterized by a
fairly level landscape with many isolated small wetlands and several large lake, stream,
and wetland systems. A large portion of the eastern half of the county is used for forestry
production. There is relatively little population growth in this half of the county. In
contrast, the western half of the county is characterized by upland forest types, pastures,
and agricultural crop fields. There are no large lakes or wetlands in this half of the
county. Compared to the eastern half of the county, the western half is experiencing
dramatic residential development pressure radiating out from the City of Gainesville.
Alachua County was chosen because of the familiarity that the author has with the
area. Familiarity was an important consideration for two reasons: first, because of the
level of spatial detail obtainable by using one-hectare square areas for the land area unit
spatial model; and second, because of the abstraction of this detail resulting from
converting everything into energy and EMERGY terms. Although other potential study
areas already had the spatial data that were required for this type of study, it was decided
that it would be more advantageous to be very familiar with the detailed spatial patterns
of the study area. Unfortunately, this decision meant that considerable effort had to be
expended to develop several of the GIS databases (property parcels, buildings, roads, and
land use classifications) that would be required to conduct the study.
\ Federal Highways
State Highways
M Alachua County
State of Florida
Figure 1-4: Location of the Alachua County study area.
21
Study Objectives
The primary objectives for this study are: 1) to develop a general theoretical
approach for the spatial modeling of energy flows and storage in urban and regional
landscapes; 2) to analyze the spatial patterns of energy transformation hierarchy and
energy signatures in the study area using a specific implementation of the general model;
and, 3) to demonstrate how GIS technologies and methods can be used to add new
insight into spatial patterns of energy hierarchy by conducting analyses that would not be
reasonable to do without GIS technology.
1) Develop a general theoretical method. The first objective is to develop a
general method for the spatial modeling of energy flows and storage in an urban and
regional landscape using energy transformation hierarchy theory and the land area unit
diagram model as a theoretical basis.
The previous studies that were cited suggest that urban and regional systems may
develop similar spatial patterns in terms of energy transformation hierarchies. However,
the non-standard methods used in these studies can make direct quantitative comparisons
of the results difficult. This study will demonstrate that the concept of a land area unit
spatial model is a more general approach to modeling the spatial distribution of energy
and EMERGY flows and storage.
2) Study the spatial patterns of energy hierarchy for the study area. The second
primary objective is to study the spatial patterns of the energy transformation hierarchy in
the Alachua County study area.
Because the target audience for this study is urban planners, the specific spatial
model created for this study should be at a spatial resolution and scale that makes the
22
results most useful from the viewpoint of the planner. The resolution of the spatial model
should make it possible for planners to describe and study the energetic characteristics of
the urban landscape in relation to more commonly used land area terms such as land use
and zoning classifications and neighborhoods. The influence of prominent urban
features, such as major roads or shopping centers, should be recognizable in the results.
3) Demonstrate the value added by using GIS technologies and methods. The
third primary objective of the study is to demonstrate how GIS technologies and methods
increase the potential for studying spatial relations in terms of energy hierarchy.
The availability of very detailed GIS databases of urban structural features and
urban flows such as traffic counts and electricity usage makes very detailed spatial
analysis possible. By using GIS analysis methods, patterns of energy distribution may be
found in this detailed data that were not previously intuited. However, one may want to
ask the question in the end as to whether or not the extra detail adds enough new insight
to be worth the enormous effort required to manipulate the large amounts of data.
Organization of the Study
This study is organized around the creation, description, and use for analysis of a
spatial EMERGY model of Alachua County, Florida. The model was created by using the
values of descriptive attributes in several thematic GIS databases as the input for
algorithms that estimate the flows and storage of energy and EMERGY. Detailed
descriptions of the methods and algorithms used to create the model dominate the
methods chapter of the study. In fact, the methods chapter was designed to be a
'cookbook' for others wishing to create similar models for other cities or regions.
23
A conscious effort was made to avoid as much GIS software-specific
terminology in the methods chapter as possible. In other words, the methods were
designed to not require a specific GIS software package for implementation-in fact, the
procedures used should be available in most basic GIS software. However, some generic
GIS terminology has been used for clarity. A familiarity with basic GIS methods will be
required of someone wishing to create their own model using this approach.
The first section of the results chapter uses maps and basic statistics to describe
the results for each of the individual components of the model. The model components
are associated with a flow or storage that was represented in the land area unit model
system diagram (Figure 1-3). An EMERGY and an energy map was made for each flow
or storage represented in the model diagram. The model component maps represent
energy flow density (joules/hectare/year), EMPOWER density (solar
emjoules/hectare/year), energy storage density (joules/hectare), or EMERGY storage
density, referred to in this study as "EMSTORAGE" (solar emjoules/hectare) for the
component flow or storage.
In the second section of the results, appropriate energy and EMERGY component
flows were added together to create "Total Energy density" and "Total EMPOWER
density" maps. Transformity maps are presented that were created by dividing the "Total
EMPOWER density" map values by the "Total Energy density" map values (solar
emjoules/ joules). Similar maps were created for "Total" energy and EMERGY storage.
County-wide energy and EMERGY signature graphs are presented that were
created by summing the all of the land area unit values for each of the component flows
24
and storage. The county-wide signatures were created to characterize the whole study
area as if the County were a single land area unit.
In the third section of the results, three comparative studies are presented that
should be of particular interest to urban and regional planners. Comparisons of the
energy and EMERGY characteristics (including energy and EMERGY signatures and
transformities) are presented for land use classifications, planning units, and
representative neighborhoods.
The fourth and fifth sections in the results present analyses that would be very
difficult to conduct without using GIS technology. For instance, in the fourth section,
EMERGY ratio maps were produced by dividing the values in one EMERGY component
map by the values in another component map. Examples of the EMERGY ratio maps
include the ratio of nonrenewable to renewable EMERGY-a possible index of
environmental loading, and the ratio of EMERGY in urban structure to EMERGY in
natural structure. In the fifth section, spatial context analyses were conducted that utilize
the special capabilities of GIS technology. In the context analyses, the values of each
land area unit were compared to the values of the contiguous units to generate several
spatial context characterizations for each land area unit.
Finally, the last section of the results presents a version of the spatial model in
which the land area units are one-square kilometer (compared to one-hectare). This
version of the spatial model was created to be able to examine the effects of the size of
the land area unit on the results, and to evaluate the usefulness of a more generalized
model for planning applications.
METHODS
General Description of the Model
The spatial EMERGY model components. A general land area unit system
diagram model describing the EMERGY flows and storage for each land area unit was
presented in Figure 1-3. The spatial model that complements this diagram model
includes both energy and EMERGY components that correspond to each of the diagram's
storage and flow symbols. In the spatial model each component is modeled as a data set
called a 'grid'. A grid is a spatial data set that divides the study area into equal-size,
square land area units called 'cells'. Each cell contains a value corresponding to the
magnitude of the flow or storage that the component grid represents.
The number of grids in the spatial model is directly related to the complexity of
the diagrammatic model. For instance, urban structure can be represented by a single
storage tank in the diagram and as a single component grid in the complementary spatial
model. Or urban structure can be modeled more discretely with separate tanks in the
diagram, and separate grids in the spatial model.
In the spatial model, a 'component grid' may be composed of the sum of two or
more 'subcomponent grids.' For example, in the spatial model the 'urban built structure'
component grid is the sum of the subcomponent grids representing the storage in the
buildings, transportation infrastructure, and utility infrastructure.
A primary purpose of creating each of these subcomponent grids separately was
to facilitate the processing of the original source GIS data layers. However, in this study
subcomponent grids were also considered to be discrete elements of their associated
component flow or storage that were potentially worthy of being used independently in
planning analyses.
The concept of a subcomponent grid and it's related component grid in the spatial
model is implicit. It may help to think of a subcomponent grid representing a storage as a
compartment within a storage tank symbol in the diagram model, and a subcomponent
grid representing a flow as one of the forks along a split flow symbol. Figure 2-1
illustrates these concepts.
Tables 2-1 and 2-2 list the component and subcomponent grids included in the
spatial EMERGY model that was developed for this study. These tables include a cross
reference to the numbers used in Figure 1-3 to identify the individual flow and storage
elements in the land area unit diagram model that are included in the spatial model.
Any given component grid, or even a subcomponent grid, may also be the sum of
one or more 'intermediate component grids'. The split flows shown in Figure 1-3 for
water use (element #3), fuels used in buildings and agriculture (element #4b), and goods
used (element # 5) were calculated using 'intermediate component grids'. Many
intermediate component grids were created as intermediate steps in the creation of
component or subcomponent grids. Unlike subcomponent grids, they exist, in this study,
only to facilitate processing of original GIS data layers, and will not be considered
explicitly in the general analysis of the model or any planning analyses. The intermediate
component grids could, however, be used for more detailed analysis in future studies.
Table 2-1: Component and subcomponent (shown shaded) grids representing flows of
energy and EMERGY that were included in the spatial model.
Description of Figure 1-3 Energy EMERGY
General Individual Flows Item Component Component
Category Calculated Number Grid Name Grid Name
Renewable Transpiration 1 renew_en renew
Resources Used
Natural Systems Gross primary 2 gpp_en gpp
Metabolism productivity
Water Water used 3 wtruse_en wtruse
Used by Man for domestic,
commercial,
and agricultural
purposes
Direct Use of Sum of use in 4 fuel_en fuel
Fuels and buildings, grounds,
Electricity agriculture,
transportation
Use for
transportation 4a trn fulen trn_ful
Use in buildings,
grounds, and 4b bag_ful_en bag_ful
agriculture
Goods/Services Consumable and 5 goods_en goods
Consumption durable goods
(and services) used
Human Services 'In situ' services 6 servicemen service
from local people
Solid and Liquid Solid/liquid wastes 7 waste_en waste
Wastes that are not recycled
Recycled Recycled solid 8 recycle_en recycle
Wastes wastes
Table 2-2: Component and subcomponent (shown shaded) grids representing storage of
energy and EMERGY that were included in the spatial model.
General Description of Figure 1-3 Energy EMERGY
Category Individual Storages Item Component Component
Calculated Number Grid Name Grid Name
Natural Systems Sum of 9 natstr_en natstr
Structure biomass, surface
(includes and ground water,
agriculture, and organic matter
forestry, and in soils
urban forest)
Biomass 9a biostr_en biostr
Surface and 9b wtrstr_en wtrstr
groundwater
Organic matter in 9c soilom_en soilom
soil
Urban System Sum of 10 urbstr_en urbstr
Structure buildings, roads and
utilities infrastructure
Buildings 10a bldg_en bldg
Roads 10b road_en road
Utilities 10c util_en util
infrastructure
People Storage in humans 11 popstr_en popstr
Roads
Buildings
Utilities
Urban Structure
(a)
Figure 2-1: The relationship between a subcomponent grid and it's related component
grid in the spatial model. One can think of a storage subcomponent grid as a
compartment within a storage tank symbol (a), or a flow subcomponent grid as a split
along a flow symbol line (b).
Size of Component Grid Cells. All of the grids in the model have a cell area of
one hectare. Each square cell is 100 meters on each side. Each grid has 583 columns and
596 rows. This results in grids that have a total of 347,468 cells per grid. Only 252,581
cells have data associated with them because the area of the county does not fill the entire
square grid. Each of the cells of each (sub)component grid is spatially coincident with
cells in the other (sub)component grids.
Units of measure. The number associated with each grid cell in a (sub)component
grid represents either the total energy or EMERGY stored or the total energy or
EMERGY flow per year occurring in the geographic area of that cell.
The numbers for all component grids representing EMERGY storage are in units
of solar emjoules per hectare (sej/ha) and are referred to as EMSTORAGE density
values. The energy storage component grids are in units of joules per hectare (j/ha).
EMERGY flows are measured as the total annual flow of solar emjoules per
hectare per year (sej/ha/yr). Energy flows are measured in units of joules per hectare per
year (j/ha/yr).
EMPOWER has been defined as the EMERGY flow per unit of time and
'EMPOWER density' as the EMPOWER per unit of area (EMPOWER/area) (Odum,
1996). In this model, the numbers associated with each cell in each of the EMERGY
flow component grids represent the 'component EMPOWER density' of each cell.
Geographic extent and time period. The spatial model covers the geographic
extent of Alachua County, Florida. All calculations of EMERGY storage and flow were
based on data from either 1993 or 1994. These years were chosen because of the
availability of primary data and statistics. A single reference year for the model was not
possible due to limitations imposed by data availability. All energy or EMERGY flow
numbers only represent the total flow occurring for a period of one year, which occurred
at some time during the two-year reference period. In some cases, calculations represent
an average of data for the two years; in other cases, data were only available for one or
the other of the two years. Materials with a turnover time of less than a year are
calculated in the model as flows. Materials with longer turnover times are calculated as
storage in the model.
Software and Hardware Used
GIS and imagery processing software. The primary GIS software used for this
study was Arc/Info Version 7.1, a product of the Environmental Systems Research
Institute, Inc. (ESRI), Redlands, California. Another ESRI software product called
ArcView GIS Version 3.0 was also used extensively.
GIS software terminology. An effort was made to describe these methods in
terms that do not require the reader to be familiar with GIS software-specific jargon.
However, familiarity with a few terms is essential. In ESRI software terminology, a
vector-type GIS data layer is called a 'coverage.' There are three basic categories of
coverages based on the type of spatial feature that they represent: points, lines, or
polygons. A raster-type GIS data layer is called a 'grid' (ESRI, 1996).
Computer hardware. Because of the intensive computer processing required for
executing some of the steps in creating this model, a workstation computer (SUN Ultral
Model 170 workstation, UNIX operating system) was used to create the model and
perform the analyses. The initial development of the model required the use of
approximately 6 gigabytes of disk storage. This large amount of disk storage was
required primarily to facilitate the collection and processing oOf the raw GIS databases
and for the manipulation of this raw data to create the final model. The model in its final
form can be viewed and analyzed using far less expensive desktop PC computers.
Development of the GIS Database
The first major step in the development of the spatial EMERGY model was the
development of a GIS database that would provide the type of information needed to
estimate energy and EMERGY flows and storage listed in the previous tables. This
required the collection, review, and processing of existing data, and subsequently, the
creation of several new coverages.
Existing coverages. Over the past few years, many GIS databases have been
created by various federal, state, and local government agencies in support of their
specific missions. Many of these GIS databases for Florida have been collected and
processed so that they have a standardized digital format (ESRI coverages) and map
projection through the Florida Geographic Data Library (FGDL) project (Lambert and
Zwick, 1997). In this study the FGDL map projection, described in Figure 2-2, was used
for all existing and new coverages and grids. Approximately 100 GIS databases,
including those available as a result of the FGDL project, were reviewed for their
potential contribution to this study. Based on a review of the available databases, only a
few were eventually used directly in the creation of the final model.
Projection ALBERS EQUAL AREA
Datum NAD27
Units METERS
Spheroid CLARKE1866
Parameters:
1st standard parallel 24 0 0.000
2nd standard parallel 31 30 0.000
central meridian 84 0 0.000
latitude of projection's origin 24 0 0.000
false eating (meters) 400000.000
false nothing (meters) 0.00000
Figure 2-2: Description of the map projection and coordinate system used for all
coverages and grids in the GIS database and model.
Several of the databases that were not used directly still served important roles as
ancillary, supporting data. For instance, municipal limits and boundaries of protected
areas help one understand why a particular observed spatial pattern might have
developed. There is also a potential for some of these databases to be used in the future
for additional analyses.
New coverages. Based on the review of the existing databases, several new
coverages had to be developed including parcels, buildings, and roads. The creation of
these new coverages required a great deal of time and effort. However, they were
indispensable inputs to the development of the model. The following sections briefly
describe the methods used to develop these new coverages.
Property parcel coverage. The property parcel coverage was created by
converting the Alachua County Property Appraiser's GIS database, in MapGrafix
software format, into the Arc/Info GIS polygon coverage format. This effort required the
development of a new software format conversion methodology (Lambert, 1996).
A property parcel coverage is a valuable primary data source for the development
of this type of model because of the attribute data associated with each polygon feature.
For instance, attributes associated with the built structures include dollar value of the
structures, and the year they were built. Other attributes indicate the primary type of
economic activity occurring on the parcel based on Florida Department of Revenue land
use codes (FDOR, 1990) and total square footage of buildings. The property parcel
coverage was also valuable as one of the intermediate data processing inputs for several
of the final components of the model.
Building coverage. The creation of the building coverage was made possible by
the creation of the property parcel coverage described above. The polygon centroids of
those polygons in the parcel coverage that had a dollar value greater than zero for built
structure were converted into a point-feature coverage.
The building coverage includes attribute data for all built structure on a property.
For instance, there are attributes for the dollar value of miscellaneous structure such as
fencing, driveways, etc. in addition to the primary building structures. Consequently, in
some cases, point features may not represent buildings, but rather only other built
structure. The attribution of the coverage allowed for consideration of these cases when
calculating fuel use in buildings, etc., however, it was a complicating factor in the
calculations, and may have introduced a few anomalies in some of the component grids.
All of the point features that were derived from polygons that were 10 acres or
less in area, were assumed to approximate the location of the built structures accurately
enough for the spatial resolution of the model. On the other hand, it was determined that
the accuracy of the location of each of the point features that were derived from parcel
polygons that were greater than 10 acres in size should be reviewed.
One objective of reviewing each of these point features was to identify those
features with obviously anomalous locations, and to manually reposition the feature to a
more likely location within the limits of it's originating polygon. Typical anomalous
conditions included point features in the middle of lakes or wetlands. A second objective
of this review was to identify, and reposition if necessary, those point features with dollar
values greater than one million. A 1995 SPOT panachromatic satellite image
(10-meter resolution) was a used as an aid to reposition each point feature to its most
probable location within its originating parcel polygon. All of this tedious work was
necessary because the spatial accuracy of the building point coverage would be reflected
in several of the final components of the model that would be derived from various
attribute data associated with the building coverage.
Road coverage. A new road coverage had to be created because the existing
coverages for roads were not spatially accurate enough for the model and did not have the
descriptive and quantitative attributes needed to calculate estimated EMERGY flows and
storage.
The initial spatial data for this new coverage was converted from several AutoCad
digital format files that were created as part of the Alachua County Control Densification
and Identification of Land Corners Project (Southern Resource Mapping Co., 1990).
This relatively spatially-accurate digital map of the roads of Alachua County was derived
from 1988-vintage aerial photography. New roads, built since 1988, were added to the
coverage using the parcel coverage as a guide.
Attribute data for functional classification, number of lanes, and estimated
average annual daily traffic was added to each road segment in the coverage (ADOT,
1995, and FDOT, 1995). These attribute data were necessary for the EMERGY storage
and flow calculations as described in later sections.
Land use and cover coverage. A county-wide land use and cover coverage was
created for this study by combining the pre-release version of the 1995 Suwannee River
Water Management District (SRWMD) land use and cover coverage (SRWMD, 1995)
with the 1990 St. John's River Water Management District (SJRWMD) (May, 1993)
land use and cover coverage. Both coverages were derived from aerial photo
interpretation, and are based on the Florida Land Use and Cover Classification System
(FLUCCS) (FDOT, 1985). This resulted in a hybrid land use and cover coverage based
on different dates. Fortunately, this did not affect the analysis appreciably because the
small portion of the county not included in the SRWMD coverage was in the southeast
portion of the County. This part of the County had not changed much since 1990 (based
on a review of the 1990 data using the SPOT panachromatic satellite imagery for 1995
for comparison).
General Methods for Creating Component Grids
The general methods used to process the primary data coverages into the
(sub)component grids of the spatial model are described in this section. Other
(sub)component-specific methods are described in later sections.
Analytical Coverages
Each of the (sub)component grids created was actually derived from an
'analytical' coverage that was created to facilitate the calculations required in this study.
These 'analytical' coverages are significantly modified versions of the existing or new
'raw' primary data coverages.
Every 'analytical coverage' has had several new data items added to its feature
attribute table to facilitate the energy and EMERGY calculations. Each analytical
coverage has also been 'overlaid' with a county-wide polygon coverage, called
'alanetl00cov', which is comprised of 100-meter square polygons corresponding to the
one-hectare cells in the final model. Other unique modifications to specific analytical
coverages are described in later sections.
Each of the model's component grids was created by first making energy and
EMERGY calculations based on attribute data contained in the analytical coverages, and
then converting the analytical coverages into grids using methods developed for each
type of coverage (i.e., point, line, polygon).
Figure 2-3 describes the general conversion steps that were used for all types of
analytical coverages, and the following sections discuss specific methods for point, line,
or polygon-type analytical coverages. The unique energy and EMERGY calculations and
methods used to create each of the intermediate, subcomponent, or component grids in
the model are described in later sections. The actual file names of each grid or coverage
created in this study have been used in this report to facilitate the use of the GIS digital
database (which will be published separately on CD-ROM) by other researchers.
(Sub)Component Grids from Point-feature Coverages
The following steps were used to convert all of the analytical coverages that were
comprised of point features into the grids used in the model. The method was designed
to convert these point coverages into grids that have individual cell values that represent
the sum of all of the energy or EMERGY values (for a particular flow or storage)
calculated for the point features found within the area of each one-hectare cell.
Figure 2-3: General steps used to convert all analytical coverages into the intermediate,
sub-component, or component grids.
Step One
Overlay analytical coverage with 100-meter square polygons
of the 'alanetl 00cov' coverage to transfer the
unique 'id' for each polygon (found in the 'emgrid-id' attribute item)
to the spatially coincident features in the analytical coverage.
Step Two
Calculate the amount of energy and EMERGY
that is associated with each feature in the analytical coverage
using appropriate values in the feature attribute table.
Step Three
Create a 'summary data table'
that contains values for the total energy or EMERGY
of all point, line, or polygon features that is summarized by
the 'unique id' of each 100-meter-square polygon.
Step Four
Merge the 'summary data table' with the feature attribute table of the
'summary coverage' called 'emnetl00cov' using the values in the
'emgrid-id' attribute items (the unique id of the 100-meter-square polygons)
of both tables as the relational values.
Step Five
Create the energy or EMERGY grid
using the POLYGRID Arc/INFO command
and the appropriate item in the 'emnetl00cov' feature attribute table
as the data input source for the command.
Step 1: Each analytical point coverage was overlaid with a polygon coverage
called 'alanetl00cov' using the Arc/Info IDENTITY command (ESRI, 1996). This step
only needs to be executed once for each of the analytical coverages.
The 'alanetl00cov' coverage divides the entire county into 100-meter-square
polygons that are spatially coincident with the grid cells in the final model. Each polygon
in the 'alanetl00cov' coverage has a unique identification number stored in an attribute
item called 'emgrid-id'. As a result of the overlay operation, every point feature that falls
within the area of any one the polygons in the 'alanetl00cov' coverage, as illustrated in
Figure 2-4, is attributed with the same unique identification number from the
'alanetl00cov' coverage.
Step 2: The amount of each energy or EMERGY flow or storage associated with
each point feature was calculated using the appropriate attribute items in the point feature
attribute table. The specific calculations for each type of energy or EMERGY flow or
storage are described later.
Step 3: The Arc/Info STATISTICS command (ESRI, 1996) was used to create a
summary database table of the total energy or EMERGY of all point features found
within each 100 meter-square polygon.
Step 4: This summary database table was then merged (using the unique number
in the 'emgrid-id' attribute item as the relational item value) with the polygon attribute
table of the 'emnetl00cov' summary coverage using the Arc/Info JOINITEM command
(ESRI, 1996).
100 meters
Figure 2-4: Each analytical coverage was overlaid with a polygon coverage that divides
the entire county into 100 meter-square polygons. These polygons are spatially
coincident with the grid cells in the final model. Each polygon in the 'alanetlOOcov'
coverage has a unique identification number stored in an attribute item called 'emgrid-
id'. As a result of the overlay operation, every point, line, or polygon feature that falls
within the area of any one the polygons in the 'alanetlOOcov' coverage is attributed with
the same unique identification number from the 'alanetlOOcov' coverage. As illustrated
above, polygon and line features are split at the points of intersection with each of the
100-meter square polygons.
-------------------------
#2
---------------------------
The 'emnetlOOcov' summary coverage has the same 100 meter-square polygons
(which are spatially coincident with the grid cells in the final model) with the same
unique numbers (in the 'emgrid-id' attribute) as the 'alanetl00cov' coverage. The
'emnetl00cov' summary coverage was used to store all of the final summarized energy
and EMERGY data for each polygon.
Step 5: Each final intermediate, subcomponent, or component grid was created
by executing the Arc/Info GRID function POLYGRID (ESRI, 1996) using the
summarized values for energy or EMERGY flows or storage from the appropriate
attribute items in the 'emnetl00cov' summary coverage as the input source for the
POLYGRID function. The values in the resulting grid represent the total energy or
EMERGY of all of the point features within each 100 meter-square area.
(Sub)Component Grids from Line-feature Coverages
The following steps were used to convert all analytical coverages that were
comprised of linear features into the grids used in the model. Similar to the previous
method employed for conversion of point features, this technique was designed to
convert line feature coverages into grids that have individual cell values that represent the
sum of all of the energy or EMERGY values calculated for the linear features found
within the area of each 100 meter-square cell.
Step 1: Each analytical line coverage was overlaid with the polygon coverage
called 'alanetl00cov' coverage using the Arc/Info IDENTITY command (ESRI, 1996).
In contrast to performing this operation on point coverages, in which the point features do
not get altered in any way other than receiving an additional attribute, this overlay
operation significantly alters the structure and number of linear features. Essentially, all
linear features are 'split' at each of the points where they intersect the border of any one
of the 100-meter-square polygons in the 'alanetl00cov' coverage (see Figure 2-4).
Each line feature in the modified analytical coverage retains the values for each of
its attributes except for the 'length' attribute. The value of the 'length' attribute is
recalculated to reflect the length of the feature that is within each 100 meter-square
polygon. This step needs to be executed only once for each analytical coverage.
Step 2: As a result of step one, each of the split linear features has an 'emgrid-id'
attribute item value indicating which of the 100 meter-square polygons in the
'alanetl00cov' coverage it is located within.
The amount of energy or EMERGY flow or storage associated with each feature
was then calculated using the appropriate attributes multiplied times the length of the
segment of the feature that is contained within each of the polygons. The specific
calculations for each type of energy or EMERGY flow or storage are described later.
Step 3: The Arc/Info STATISTICS command (ESRI, 1996) was used to create a
summary database table of the total energy or EMERGY of all the line features found
within each 100 meter-square polygon.
Step 4: This summary database table was then merged (using the unique number
in the 'emgrid-id' attribute item as the relational item value) with the polygon attribute
table of the 'emnetl00cov' coverage using the Arc/Info JOINITEM command (ESRI,
1996).
Step 5: Each final intermediate, sub-component, or component grid was created
by executing the Arc/Info GRID function POLYGRID (ESRI, 1996) using the
summarized values for energy or EMERGY flows or storage from the appropriate
attribute items in the 'emnetlOOcov' summary coverage as the input source. The values
in the resulting energy or EMERGY grid represent the total energy or EMERGY of all of
the linear features within each 100 meter-square area.
(Sub)Component Grids from Polygon-feature Coverages
The following steps were used to convert all analytical coverages that were
comprised of polygon features into the grids used in the model. This technique was
designed to convert polygon feature coverages into grids that have individual cell values
that represent the sum of all of the energy or EMERGY values calculated for the polygon
features found within the area of each 100 meter-square cell.
Step 1: Each analytical polygon coverage was overlaid with the polygon
coverage called 'alanetl00cov' coverage using the Arc/Info IDENTITY command
(ESRI, 1996). In a manner similar to the 'splitting' of linear features, and in contrast to
point feature coverages, each polygon in the coverage is 'split' along the borders of any
one of the 100 meter-square polygons. This effect is illustrated in Figure 2-4.
Each resulting polygon feature retains the values for each of its attributes except
for the 'area' attribute. The value of the 'area' attribute is recalculated to reflect the area
of the feature that is within each 100 meter-square polygon. This step only needs to be
executed once for each analytical coverage.
Step 2: As a result of step one, each of the split polygon features has an 'emgrid-
id' attribute item value indicating which of the 100 meter-square polygons it is located
within. The amount of energy or EMERGY flow or storage associated with each feature
was then calculated using the appropriate attributes multiplied times the area of the
feature that is contained within each of the 100 meter-square polygons. The specific
calculations for each type of energy or EMERGY flow or storage are described later.
Step 3: The Arc/Info STATISTICS command (ESRI, 1996) was used to create a
summary database table of the total energy or EMERGY of all the features found within
each 100 meter-square polygon.
Step 4: This summary database table was then merged (once again, using the
unique number in the 'emgrid-id' attribute item as the relational item value) with the
polygon attribute table of the 'emnetl00cov' coverage using the Arc/Info JOINITEM
command (ESRI, 1996).
Step 5: Each final component grid was created by using the Arc/Info GRID
function POLYGRID (ESRI, 1996) and the summarized values for energy or EMERGY
flows or storage from the appropriate attribute items in the 'emnetl00cov' summary
coverage as the input source. The values in the resulting intermediate, sub-component, or
component grid represent the total energy or EMERGY of all of the polygon features
within each 100 meter-square area.
Specific Methods for Creating Flow Component Grids
The following sections describe the specific calculations and methods used to
create each of the energy and EMERGY flow component and subcomponent grids for
Alachua County that were listed in Table 2-1 and diagrammed in Figure 1-3.
Natural System Flows-Gross Primary Productivity
The renewable, environmental energies available to a system include sunlight,
deep earth heat, tides, surface wind, physical and chemical energy of rain and rivers.
Through photosynthesis, plants have the ability to efficiently capture and process these
natural energy flows, thus play the role of the primary production component of the
world's natural ecosystems. Science has devised many ways of measuring the value of
the contributions made by the earth's vegetation, but one of the most basic ways of
assessing the amount of work contributed by plants is to measure their gross primary
productivity. Gross primary productivity is a measure of the total energy captured, or
organic matter created, by green plants per unit surface and time. The proportion of this
production that is left after respiration is called net primary production (Whittaker, 1970,
E. P. Odum, 1971).
Gross primary productivity is a more appropriate measure to use for this study
than net primary productivity because it still recognizes the value of production for the
maintenance of structure, as in the case of 'climax' forests, when net production is often
very low (Bernatzky, 1978, Waring and Schlesinger, 1985, Packham et al.., 1992, Tivy,
1993). Several studies similar to this one (Costanza, 1975, Brown, 1980, and Whitfield,
1994) have used gross primary productivity (GPP) for similar reasoning.
Unfortunately, GPP is difficult to measure except in laboratory experiments or by
inference from net production (Whittaker and Marks, 1975). Consequently, data on GPP
for ecosystems are very limited. A major effort to measure the primary productivity of
the world's ecosystems was conducted in the 1970s called the International Biological
Program (IBP) (National Academy of Sciences, 1975). The focus of this coordinated
research effort was to measure net primary productivity (NPP) and biomass of the
world's ecosystems (Lieth, 1975, Lieth, 1978). As a result of these efforts, many
estimates of NPP and biomass have been published from which GPP can be extrapolated
(Art and Marks, 1971, Rodin et al., 1975, Leith, 1975, Olson, 1975, Reichle, 1981).
Similar efforts continue today in recognition of the importance of the role of global forest
resources in the global carbon cycle (Apps and Price, 1996, Archibold, 1995).
Satoo (1982) pointed out that estimating GPP from NPP can, however, be difficult
because it depends on respiration rates that vary in relation to both environmental and
stand conditions. For instance, the proportion of GPP that is measured as NPP varies
with temperature, season, and age of stand. Complicating the issue further, these early
measurements are coming under scrutiny recently because they often only estimated
aboveground biomass (Long et al., 1992). Despite these difficulties in measuring GPP
accurately, it is conceptually an appropriate measure of the work contributed by the
vegetative component of natural systems.
Whereas GPP is a measure of the work contributed by natural systems, it is not a
measure of the amount of natural, renewable energies used in the production. Odum
(1983, 1996) suggests a method of deriving a reasonable measure of the EMERGY from
renewable sources that is used by terrestrial ecosystems is to use the amount of water
transpired by the ecosystem. The EMERGY in transpired water is in effect an integration
of the EMERGY in sun, wind, and rain, and avoids double-counting of these inputs
(Odum, 1996). In cases where additional EMERGY sources (which may be non-
renewable) are significant to production, the EMERGY of these sources must be added as
well to find the total EMERGY contributing to GPP. Transformities for the GPP of
natural systems can be calculated by dividing the total EMERGY contributing to GPP by
the energy in GPP.
Gross primary productivity (GPP) was used in this study as a measure of the work
contributed by natural systems. The amount of water transpired by ecosystems was used
as a measure of the amount of natural, renewable energies used in the work provided by
these ecosystems (Odum 1983, 1996). Additional non-renewable EMERGY sources that
were significant to production, were added to the total EMERGY contributing to GPP.
Transformities for the GPP of natural systems were calculated by dividing the total
EMERGY contributing to GPP by the energy in GPP. Calculations for important
ecosystems in Alachua County (as they were classified by the land use and cover
coverages) were done to determine energy in GPP, natural and cultural EMERGY flows
contributing to GPP, and transformity of GPP for each ecosystem.
The 'em landcov' analytical coverage. The coverage used to create this
component of the spatial model was the analytical coverage called 'em_landcov'. This
coverage was created in an attempt to more precisely model the variation found in
urbanized landscapes than would be possible with the original land use and cover
coverage described earlier. For instance, the areas occupied by roads are not identified in
the land use and cover coverage (except in cases of major four-lane or interstate roads).
Consequently, a method was devised to 'buffer' (using the Arc/Info BUFFER command
(ESRI, 1996)) the linear road coverage features (by a factor considering number of lanes
and average lane width) so that these paved areas, which do contribute any GPP and do
not use environmental flows in the same manner as natural systems, could be treated
differently from the areas with natural systems. Following the same reasoning, the areas
covered by the footprints of buildings were identified and buffered (according to the
square footage of the building) so that these areas could also be treated differently from
the natural systems.
Another characteristic of the original land use and cover coverage is that some
land use classification categories, such as the residential, commercial, and industrial
classes, do not refer to any type of land cover (or natural system). In these cases,
assumptions were made about the type of natural systems found in these categories.
Based on personal observation, the typical Alachua County low- and medium-density
residential yard consists of large trees with an understory that has been cleared in a zone
adjacent to the dwelling and replaced with turfgrass, ornamental shrubs, and flowers.
This maintained zone of residential landscape is supported by a significant amount of
non-renewable energy sources. In an effort to estimate the areal extent of these
maintained landscape zones, the building footprints were buffered by an additional 10
meters.
Other land use classifications, such as high-density residential, commercial, etc.,
are assumed to have proportionately less vegetation in the maintained landscape zone
than low- and medium-density residential land uses because of larger proportions of the
area used for parking lots, etc.. In the cases of more intensive land use types, such as
shopping malls and industrial sites, estimates of proportionately less vegetation were
made and applied to the entire area of the land use.
Data processing steps. The following generalized data processing steps were used
in creating the analytical coverage 'emlandcov' and the gross primary production
components of the spatial EMERGY model (element # 2 in Figure 1-3).
Step 1: The roads coverage was buffered based on an algorithm whereby the
number of lanes of each road segment was multiplied by 2.5 meters to calculate the
buffer distance used for each segment. The resulting buffered road polygon is twice as
wide as the buffer distance. For example, a 4 lane road is assigned a 10 meter buffer
distance resulting in a 20 meter wide polygon representing the paved area of the road.
The buffered road coverage was then unioned with the original land use and cover
coverage.
Step 2: The point features in the 'building' coverage that had a total building and
miscellaneous value of less than $10,000 were deleted from the building coverage. These
features were not considered for buffering as building footprints because most of these
features only represent miscellaneous 'improvements' rather than actual residential or
commercial buildings (52,411 point features remained out of a total of 57,421 features in
the building coverage).
The remaining point features were buffered according to the square footage of the
building to create the building footprint coverage. Estimates of the square footage were
made for buildings where data was not present (2,334 features) by dividing the total
value of the building by 50 (assuming $50/sq.ft.). The buffer distance used to buffer
point features is the radius of a circle. The radius of the building footprint buffer
(resulting in a circle) was calculated based on square footage of the building using the
following equation:
(.305 m/ft)(sqrt(totsqfoot/3.14)) = buffer distance in meters
Finally, the building footprint coverage was unioned with the intermediate land
use and cover coverage that was previously unioned with the buffered roads.
Step 3: The building footprint coverage was buffered by 10 meters to represent
the area of the managed landscape that surrounds the buildings. As noted earlier, this 10
meter buffer distance is based on field observations in Alachua County. In an effort to
introduce a measure of the natural spatial variation that is found in the urban landscape,
the actual size of these buffered areas were purposefully made slightly random by
introducing a coarse 'fuzzy tolerance' into the buffering algorithm. The resulting
coverage was unioned with the last version of the land use and cover coverage to yield
the final analytical coverage, called 'em_landcov', that includes areas of buffered roads,
building footprints, and maintained landscape.
Step 4: Additional attribute items were added to the 'em_landcov' feature
attribute table in order to calculate the EMERGY flows occurring through each classified
polygon in the 'em_landcov' coverage.
Estimates of GPP per hectare were made for representative ecosystems. These
estimates were converted from flows per hectare to flows per meter-squared. Each land
use or cover class was assigned a flow/m2 /yr value according to the most closely
corresponding value calculated for a representative ecosystem, or in the cases of some
commercial and industrial classifications, by assigning a value that has been
proportionately reduced by a factor to estimate the amount of vegetation typically present
on that type of land use. The flow/m2 /yr rate assigned to each polygon was multiplied
times the area of the polygon so that the total flow for each cell could be summed.
Step 5: The coverage was then processed according to the methods described
previously for converting polygonal feature coverages into the EMERGY and energy
component grids, called 'GPP' and 'GPP EN', for gross primary productivity of natural
systems.
Natural System Flows-Renenewable Sources
Renewable environmental energy flows are the basis of all natural processes and
contribute to all economic processes. Each area on the earth's surface has several types
of natural energy sources that contribute renewable energy and EMERGY to the natural
and economic systems of the area. Odum (1983, 1996) has pointed out that the
combination, frequency, and quantity of these environmental energy sources is reflected
in the character of both the area's ecosystems and economic systems.
In Alachua County, sun, wind, and rain are the predominant local sources of
natural energy. Rainwater ends up either being used for transpiration by ecosystems or
flowing into streams, lakes, and aquifers. In this study, the geopotential energy in runoff
water and streams was assumed to be a minor source of natural energy because of the
relatively flat landscape of the county.
A notable exception to this assumption is the Santa Fe River, which flows along
the northern boundary of the county. However, due to technical limitations of the
relevant coverages, this important feature could not be included in the present form of the
model. The storage of water in streams, lakes, and aquifers is however included in the
model and is discussed later.
Odum (1996, 1983) cautions against 'double-counting' in evaluating the total
EMERGY contributed by geobiospheric processes (from sun, wind, rain, waves, land
cycles, etc.) to local areas of natural systems. For instance, he points out that, from a
global perspective, sun, wind, and rain are all coproducts of the same global climatic
process. Hence, the EMERGY content of rainfall contains EMERGY from sun and wind.
In order to avoid double-counting the EMERGY of geobiospheric sources, Odum has
shown that for any local area the largest geobiospheric input of EMERGY should be used
to account for the total contribution of EMERGY from these global processes.
In consideration of this principle, calculations for representative one-hectare areas
within Alachua County were made and it was revealed that the largest geobiospheric
input of EMERGY comes from chemical potential energy of rainfall. In this model,
transpiration by ecosystems was considered to be the primary use of this rainfall within
any given one-hectare cell of the model. The rainfall that is not used in transpiration
either recycles through evaporation, or contributes to long-term groundwater and lake
storage. The chemical potential energy of the rainwater that is used in the process of
transpiration in ecosystems integrates the combined inputs of the energy in sunlight used
and wind absorbed and thus represents the total amount of EMERGY used from
renewable geobiospheric sources without double-counting. This procedure assumes land
use and its erosion is the same EMPOWER or less than the global annual budget.
The coverage used to create this component of the spatial model was the
analytical coverage called 'emlandcov' that was also used for the creation of the gross
primary production component grids.
Calculations made previously to determine the energy and EMIERGY in gross
primary production included estimates of the energy and EMIERGY in transpiration for
each of the ecosystems represented in the county. These calculated values for energy and
EMVIERGY in transpiration were also used to create this component of the model.
The previous section described how building and road footprints were added to
the 'em_landcov' analytical coverage to more precisely model the type and amount of
natural systems present in any one-hectare cell. For these calculations, transpiration flow
values were not assigned to the building and road footprint polygons. Instead, the
amount of annual sunlight energy and EMVIERGY received by each building or road
footprint polygon was used to assign a value to those polygons.
Data Processing Steps. The following data processing steps were used to create
the RENEW_EN and RENEW component grids that represent the amount of renewable
energy and EMERGY used (element # 1 in Figure 1-3).
Step 1: Additional attribute items were added to the 'em_landcov' feature
attribute table to facilitate calculations for the amount of transpiration per unit of area for
each type of ecosystem or land use, and the corresponding energy and EMVIERGY flows.
Energy and EMERGY calculations were made for the building and road footprint
polygons by multiplying the annual solar insolation rate for Gainesville, Florida ((5.95 E9
j(sej)/m2/yr)(Odum 1996)) by the area of the polygon.
Step 2: The coverage was then processed according to the methods described
previously for converting polygon feature coverages into the energy and EMERGY
component grids.
Urban System Flows-Water Consumption
This component grid models the energy and EMERGY in water used by humans
for domestic, commercial, and agricultural purposes (see element # 3 in Figure 1-3 and
note the split flow to urban and agricultural systems). Since water used in agriculture is
transpired, there is a small amount of double counting of the EMERGY calculated for
agricultural renewable resources use.
According to Marella (1992 1993), in Alachua County, an estimated 52.12
million gallons of groundwater per day (Mgd) were withdrawn from the Floridan aquifer
in 1990, while only .36 Mgd were withdrawn from surface waters. Of this total for the
county, an estimated 22.95 Mgd was withdrawn for public-supply use, and 14.75 Mgd of
this was used at residences. Thirteen utility companies provided water to a total of
142,104 customers in 1990. About 88% of these customers were served by Gainesville
Regional Utilities, and the rest were served by minor utilities in the smaller cities. Self-
supply wells supplied the majority of the water needs of those outside of the public-
supply service areas. Agriculture was the largest self-supplier, using an estimated 18.45
Mgd of water for crop irrigation and raising of livestock.
The creation of the final component grid representing the energy and EMERGY
in water used by humans for domestic, commercial, and agricultural purposes required
that several intermediate component grids be created. The 'em resbldgcov' analytical
coverage was used to estimate the quantities of water used for domestic purposes. The
point features in the 'em_resbldgcov' analytical coverage are a subset of the original
'buildings' coverage, which represent residential buildings only. The point features in
the 'em_combldgcov' analytical coverage (which include all other buildings not in the
'em_resbldgcov' coverage, i.e., commercial, institutional, agricultural, and industrial
buildings) were used to estimate the quantities of water used for commercial, industrial,
and institutional purposes. The use of these analytical point coverages was determined
to be the most appropriate, and consistent, method of calculating and locating domestic,
commercial, and industrial water use because most of the water is used inside of, or very
close to, buildings. On the other hand, agricultural water use was estimated using the
analytical polygon coverage called 'em_agwtrcov' because agricultural water use is often
distributed over larger areas.
It was determined that different transformities should be used for public-water
supply and self-supplied well water. The transformity for water supplied by public
utilities has a higher transformity, assumed to be 665714 sej/j (Odum, 1996), than self-
supplied well water, assumed to be 255242 sej/j (Odum, 1996), because of the additional
energies required to process the water and supply it to the end-user. Consequently, it was
necessary to create a coverage of approximate municipal water service boundaries, called
'munwtrbnd', that would facilitate the calculations of EMERGY in water used based on
whether or not the water was from a public water utility or a well.
Four sets of intermediate component grids (described in Table 2-3) were created
and then 'added' together to create the final water use component grids.
Table 2-3: Intermediate grids that were used to calculate the total energy and EMERGY
of water use in buildings and agriculture.
Intermediate Grids Description
RES_WTR EMERGY of domestic water used
RES_WTR_EN energy of domestic water used
COM_WTR EMERGY of Comm/Industrial/Institutional
COM_WTR_EN energy of Comm/Industrial/Institutional use
BLD_WTR Sum of RES_WTR and COM_WTR
BLD_WTR_EN Sum of RES_WTR_EN and COM_WTR EN
AGR_WTR EMERGY of agricultural water used
AGR_WTR_EN energy of agricultural water used
Note: The final EMERGY component grid called WTRUSE is the sum of BLD_WTR
and AGR_WTR. The final energy component grid called WTRUSE EN is the sum of
BLD_WTR_EN and AGR_WTR EN. The intermediate component grid called
AGR_WTR represents the left fork of the split flow in element #3 of Figure 1-3. The
intermediate component grid called BLD_WTR is the right fork of the same element.
These intermediate component grids represent how the land area unit diagram can be
made more or less complex depending on the need for detail. In other words, these
intermediate component grids could be included as subcomponent grids if desired.
Data processing steps. The following generalized data processing steps were used
in creating the intermediate grids used to create this component of the model.
Domestic water use intermediate component grid. The domestic water use grids,
called RES WTR and RES WTR EN were created as follows:
Step 1: The analytical point coverage called 'em_resbldgcov' was overlaid with
the 'munwtrbnd' coverage to identify those buildings which were part of a municipal
water service area.
Step 2: Additional attribute items were added to the 'em_resbldgcov' analytical
coverage feature attribute table to facilitate calculations for estimates of the amount of
water used (gallons) in each building, and the energy and EMERGY.
Step 3: The calculations for the amount of water used in each building were
based on the assumption that within municipal water service areas the rate of water use
was 104 gallons/day/person, and outside the service areas the rate was 162
gallons/day/person (Marella, 1992). The estimated building population data used in these
calculations was calculated as part of the development of the population and service
component grids. The following equations were used:
for public water supply;
gallons of water used = building population 37,960
( note: 104 gal/person/day 365 day/yr = 37,960 gal/per/yr)
energy (joules) = gallons 8.35 Ib/gal 453.6 g/lb 4.9 j/g
EMERGY (sej) = energy (j) 665,714 sej/j
for well water supply;
gallons of water used = building population 59,130
( note: 162 gal/person/day 365 day/yr = 59,130 gal/per/yr)
energy (joules) = gallons 8.35 Ib/gal 453.6 g/Ib 4.9 j/g
EMERGY (sej) = energy (j) 255,242 sej/j
Step 4: The coverage was then processed according to the methods described
previously for converting point feature coverages into the energy and EMERGY
intermediate component grids for residential water use.
Commercial water use intermediate component grid. The
commercial/industrial/institutional water use grids, called COM_WTR and
COM WTR EN were created as follows:
Step 1: The analytical point coverage called 'em_combldgcov' was processed to
identify those buildings in the municipal water service area.
Step 2: Additional attribute items were added to the 'em_combldgcov' coverage
to facilitate calculations for estimates of the amount of water used (gallons) in each
building, and the energy and EMERGY.
Step 3: The calculations for the amount of water used in each building were
based an assumption that the total amount of water used in all commercial, industrial, and
institutional buildings could be distributed proportionally according to the square footage
of each building. The following calculations were used to determine the proportion of
water use to be attributed to each square foot of building depending on whether the
building had public-supplied or well water:
For public water supply;
Total water used = 8.2 Mgd (Marella, 1993)
8.2 Mgd 365 days/yr = 2,993,000,000 gal/yr
2,993,000,000 gal/yr / 46,118,466 sq.ft. = 64.9 gal/sq.ft./yr
For well water supply;
Total water used = 2.29 Mgd (Marella, 1993)
2.29 Mgd 365 day/yr = 835,850,000 gal/yr
835,850,000 g/y / 8,315,778 sq.ft = 100.5 gal/sq.ft./yr
The following equations were used to calculate the gallons of water used, and the
energy and EMERGY of this water:
For municipal supply;
gallons of water used = square footage of building 64.9 gal
energy (joules) = gallons 8.35 Ib/gal 453.6 g/lb 4.9 j/g
EMERGY (sej) = energy (j) 665,714 sej/j
For well water supply;
gallons of water used = square footage of building 100.5 gal
energy (joules) = gallons 8.35 Ib/gal 453.6 g/lb 4.9 j/g
EMERGY (sej) = energy (j) 255,242 sej/j
Step 4: The coverage was then processed according to the methods described
previously for converting point feature coverages into the energy and EMERGY
intermediate component grids for commercial/industrial/ institutional water use.
Agricultural water use intermediate component grid. The agricultural water use
grids, called AGR_WTR and AGR_WTR_EN were created as follows:
Step 1: The analytical polygon coverage called 'em_agwtrcov' was created to
identify those areas of the county being used for various agricultural purposes. Those
types of agriculture using significant amounts of irrigation water (vegetables and
ornamentals) or water for livestock were identified and located.
Step 2: Additional attribute items were added to the 'em_agwtrcov' coverage to
facilitate calculations for estimates of the amount of water used (gallons) by agriculture,
and the energy and EMERGY.
Step 3: The estimates used here for the amount of water used per unit of area by
agriculture are based on the average quantities used by typical crops and animals grown
in Alachua County and predicted aggregated use are consistent with estimates for total
county use made by Marella (1992). For all vegetable and field crops, tree crops, and
intensive livestock operations, the water use rate was estimated to be 45 gallons/m2/year.
For all ornamental nurseries the water use rate was estimated to be 360 gallons/m2/year.
Water used was assumed to be from wells.
The following equations were used:
gallons of water used = area (m2) 45 gal/m2/yr
or,
gallons of water used = area (m2) 360 gal/m2/yr
energy (joules) = gallons 8.35 Ib/gal 453.6 g/Ib 4.9 j/g
EMERGY (sej) = energy (j) 255,242 sej/j
Step 4: The coverage was then processed according to the methods described
previously for converting polygon feature coverages into the energy and EMERGY
intermediate component grids for agricultural water use.
Urban System Flows-Fuels and Electricity
The EMERGY component grid, called FUEL, represents the total energy and
EMERGY of fuel and electricity used within each cell of the model (element # 4 in
Figure 1-3). It was created by adding together the transportation subcomponent grid
(element # 4a in Figure 1-3), called TRN_FUL, and the buildings and agriculture
subcomponent grid called BAG_FUL (element # 4b in Figure 1-3). The energy
component grid called FUEL_EN was created by adding together the TRN_FUL_EN and
BAG_FUL EN subcomponent grids.
Urban System Flows-Transportation Subcomponent
A large proportion of any urban area's total fossil fuel consumption is used for
transportation (Energy Information Administration, 1994). In addition to estimating the
spatial distribution of this use, the amount of fossil fuel used for transportation per unit of
area and time may serve as an indicator of the relative amount of materials (goods) that
are being carried through a given location for ultimate consumption at another
unspecified location.
The primary coverage used to create this component of the spatial model was the
analytical coverage called 'em_roadcov'. To facilitate the estimation of motor fuel use,
each road segment was assigned an estimate of the average annual daily traffic flow
(AADT). Two sources of AADT data were used to create this database. The primary
source was the Alachua County Department of Transportation Traffic Survey Database
(Alachua County, 1995). This tabular database contains sample daily traffic counts for
all of the major urban and rural county roads. These traffic counts were made at specific
known locations along roads, such as intersections or crossroads. Therefore, for the
purposes of this study, it was necessary to extrapolate these point measurements to
estimates for the traffic flow that occurred along the corresponding linear segment of
each road. The other source of AADT data was the Road Characteristics Inventory GIS
Database maintained by the FDOT (FDOT, 1996). This spatial database had AADT
estimates for linear road segments (although this database was also originally derived
from point measurements).
Data processing steps. The following generalized data processing steps were used
in creating this subcomponent of the model.
Step 1: Additional attribute items were added to the analytical coverage, called
'em_roadcov', to enable the calculation of the energy and EMERGY flows of motor fuels
used in each road segment found within each one hectare cell of the model.
Step 2: The energy and EMERGY flows for each road segment were calculated
by using the AADT and an assumed average fuel consumption rate for vehicles in 1993
(Energy Information Administration, 1994). The average fuel consumption rate was
converted from miles per gallon units to liters per meter as shown below.
(1609.34 meters/mile)(20 mpg)/(3.7854 liters/gal) = 8503 meters/liter
(8503 meters/liter)-1 = .000117605 liters/meter
The following equation was used to calculate the liters of fuel used per year along
each road segment.
(Length, meters)(.000117605 liters/meter)(AADT)(365 days) =
liters/yr/segment
Finally, these material flows were converted into energy and EMERGY flows
using the energy factor and transformity of fossil fuels (Odum, 1996) and the equations
shown below.
(liters/yr/segment)(4.52 x 107 J/liter) = energy, j/yr/segment
(energy, j) (6.6 x 104 sej/J) = sej/yr/segment
Step 3: The coverage was then processed according to the methods described
previously for converting linear feature coverages into the EMERGY and energy
subcomponent grids, called TRN_FUL and TRN_FUL_EN, for fossil fuel use in
transportation.
Urban System Flows-Buildings and Agriculture Subcomponent
This subcomponent grid represents the flow of energy and EMERGY in the
electricity and fossil fuels used in all types of buildings and in agriculture (see element #
4b in Figure 1-3 and note the split flow going both to agriculture and building use).
Electricity consumption accounts for approximately 86% of all types of fuel use
in buildings in Florida, fossil fuels (natural gas, fuel oils, and LPG) account for 14% of
the residential site energy consumption (EIA, 1993), and electricity consumption
accounts for approximately 76% of commercial site energy consumption (EIA, 1994).
In Alachua County, the majority of fossil fuel use in buildings occurs within the
Gainesville Regional Utilities (GRU) natural gas service area. Knowing this, ideally this
component grid should be created by first calculating the energy and EMERGY from the
estimated use of electricity, using a transformity of 200,000 sej/j (Odum, 1996), and then
calculating the energy and EMERGY of fossil fuels, using a transformity of 48,000 sej/j
(for natural gas which is the primary fossil fuel used in buildings in Alachua County)
(Odum, 1996), and finally by adding the amounts of energy or EMERGY calculated for
each of the flows together. However, there is a lack of data regarding the location of the
approximately 20,000 GRU customers that received natural gas service in the 1992-93
era (GRU, 1996). Therefore, it was necessary to calculate the energy and EMERGY
representing both electricity and fuel use in buildings by predicting the use of electricity
in each building as if natural gas supplies were not available.
Although this method introduces some error, particularly in the calculations of
EMERGY, the amount of error was deemed to be acceptable for the purposes of this
study. One reason for this is that the primary uses for fossil fuels in residences are for
space and water heating, and in the 'typical' Florida home fossil fuels only account for
about 14% of total energy consumption (EIA, 1993). The majority of residential energy
consumption occurs through the uses of electricity for appliances, lighting, and air
conditioners, and these uses are also found in homes with natural gas service. Therefore,
the effect on the estimates introduced by this assumption only apply to a minor
percentage of the total energy and EMERGY flows, and only in those buildings with gas
service.
In order to estimate the amount of electricity used in each building using the data
available for this study, different usage rates were assigned according to the size and/or
type of building. Residential buildings were calculated using one of two different usage-
rates equations according to the square footage of the building. The reasoning for this
was that a strictly linear equation that calculates the same number of kilowatt-hours per
square foot for all sizes of buildings is not realistic. A more appropriate calculation was
derived from the finding by the Energy Information Administration (EIA) that for
residential buildings about 75% of the electricity consumed is used for appliances and
water heating (EIA, 1993). These basic uses of electricity are found in homes of all sizes
and dollar values. Thus, it is more appropriate to assume a base level of electricity use
for all residential buildings and then calculate additional usage for heating and cooling
based on square footage. Variability due to age of home (new homes are more energy
efficient) and household income (poor households tend to have older, less efficient
appliances) was not included in these calculations.
Various types of commercial and institutional buildings have different typical
energy intensity (kWh/sq.ft.) characteristics that have been established by national and
regional surveys (EIA, 1992, 1991). For this study, each non-residential building was
assigned to an energy intensity class according to the type of commercial or institutional
activity being conducted in the building. The estimated total electricity used in each
building was then calculated by multiplying the appropriate energy intensity, according to
that building's energy intensity class, times the square footage of the building.
To establish the acceptability of these assumptions, the individual building energy
use estimates were summed for all of the buildings within the county and compared with
the known amounts of all types of energy (in kWh units) supplied by all utilities in 1993
(GRU, 1996, BEBR, 1995). The method used predicts the total electricity/heating fuel
energy use to within 4.7% of the actual amount of energy known to be supplied to all
customers.
The creation of the final component grid representing the energy and EMERGY
in electricity and other fuels used by humans for domestic, commercial, and agricultural
purposes required that several intermediate component grids be created in a manner
similar to the method used to create the water use component grid. Four of the
intermediate grids were created using the analytical point coverages called
'em_resbldgcov' and 'em_combldgcov'. The 'em_resbldgcov' coverage was used to
estimate of the electricity used in residential buildings, and the 'em_combldgcov'
coverage was used to estimate the electricity used in commercial, industrial, and
institutional buildings. Agricultural fuel use was estimated using the analytical polygon
coverage called 'emlandcov'.
Four sets (an energy and an EMERGY grid in each set) of intermediate
component grids were created and then 'added' together to create the final electricity and
fuel use subcomponent grids. These intermediate grids are listed in Table 2-4.
The final EMERGY subcomponent grid, called 'BAG_FUL', is the sum of
BLD_ELC and AGR_FUL, and the final energy component grid, called
'BAG FUL EN', is the sum of BLD ELC EN and AGRFUL EN.
Data processing steps. The following three sets of generalized data processing
steps were used in creating the intermediate component grids that were used to create this
subcomponent of the spatial EMERGY Model.
Domestic electricity use intermediate component grid. The residential electricity
use intermediate component grids, were created as follows:
Step 1: Additional attribute items were added to the 'em_resbldgcov' analytical
coverage to facilitate calculations for the amount of electricity used in each building, and
the corresponding energy and EMERGY flows.
Step 2: The calculations for the amount of electricity used in each building were
based on the assumptions about base levels of use that were discussed above. Residential
buildings were divided into two groups for the calculations. The first group included all
buildings with a total square footage of less than 1000 square feet and all multi-family
residential buildings. The multi-family buildings were included based on the assumption
that: although the total square footage of multi-family buildings was usually greater than
1000 square feet, that this total was most often comprised of individual residential units
of less than 1000 square feet with base levels of usage comparable to individual
dwellings.
The following equation was used to calculate the amount of electricity used in
buildings of this group:
Electricity Used (kWh/year) = total square feet 9 kWh/sq.ft./year
The second group includes all other residential buildings (those with more than
1000 square feet). The base level of electricity usage was estimated to be 10,000
kWh/year/building, and additional usage was estimated for heating and cooling uses
based on the square footage of the building (EIA, 1993). The following equation was
used for this group:
Electricity Used, kWh/year =
(10,000 kWh/year) + (total square feet 2.25 kWh/sq.ft./year)
Table 2-4: Intermediate grids that were used to calculate the total energy and EMERGY
of electricity and fuel use in buildings and agriculture.
Intermediate Grids Description
RES_ELC EMERGY of residential electricity used
RES_ELC_EN energy of residential electricity used
COM_ELC EMERGY of Commercial/
Industrial/ Institutional electricity used
COM_ELC_EN energy of Commercial/
Industrial/ Institutional electricity used
BLDELC Sum of RESELC and COMELC
BLDELCEN Sum of RES_ELC_EN and COMELC EN
AGR_FUL EMERGY of agricultural fuels used
AGR_FUL_EN energy of agricultural fuels used
Note: The final EMERGY sub-component grid called BAG_FUL is the sum of
BLD_ELC and AGR_FUL. The final energy sub-component grid called BAG_FUL EN
is the sum of BLD_ELC_EN and AGRFUL EN. The intermediate component grid
called AGR_FUL represents the left fork of the split flow of element #4b in Figure 1-3.
The intermediate component grid called BLD_ELC is the right fork of the split flow of
element #4b.
Finally, the energy and EMERGY of the electricity used was calculated using the
following equations:
Energy, j/year = (kWh/yr) (860 kcal/kWh) (4186 J/kcal)
EMERGY, sej/year = (energy, j/yr) (200,000 sej/j)
Step 3: The coverage was then processed according to the methods described
previously for converting point feature coverages into the EMERGY and energy
intermediate component grids, called RESELC and RESELC EN, for electricity use in
residential buildings.
Commercial electricity use intermediate component grid. The commercial/
industrial/institutional electricity use intermediate component grids were created as
follows:
Step 1: Additional attribute items were added to the 'em_combldgcov' analytical
coverage to facilitate calculations for the amount of electricity used in each building, and
the corresponding energy and EMERGY flows.
Step 2: The calculations for the amount of electricity used in each building were
based on the application of different usage rates for different types of buildings.
Electricity usage rates, in kWh per square foot, were assigned to each building type
(using the Department of Revenue (DOR) use code associated with the building)
according to average usage rates, which were determined by regional surveys conducted
by the Energy Information Administration (EIA, 1991, 1992, 1994). The following
equation was used, applying the appropriate usage rate, to calculate each building's
annual electricity use:
Electricity Used (kWh/year) = total square feet kWh/sq.ft./year
The energy and EMERGY of the electricity used was calculated using the
following equations:
Energy, j/year = (kWh/yr) (860 kcal/kWh) (4186 J/kcal)
EMERGY, sej/year = (energy, j/yr) (200,000 sej/j)
Step 3: The data layer was then processed according to the methods described
previously for converting point feature coverages into the EMERGY and energy
intermediate component grids, called COM_ELC and COM_ELC_EN, for electricity use
in commercial, industrial, and institutional buildings.
Agricultural use intermediate component grid. Agricultural uses of fuel that were
closely associated with a building, such as nursery and greenhouse production, were
calculated in the commercial/industrial/ institutional use grid. The agricultural fuel use
intermediate component grids, called AGRFUL and AGR_FUL_EN, include fuels used
for pasture and row crop production, and maintenance of residential lawns and gardens,
and were created as follows:
Step 1: Additional attribute items were added to the 'emlandcov' analytical
(polygon) coverage to facilitate calculations for the amount of electricity and fossil fuels
used per unit of area for each type of agricultural land use, and the corresponding energy
and EMERGY flows.
Step 2: Estimated fuel energy usage rates for each type of agricultural production
were used that were based on previous calculations done to determine contributions to
gross primary production in natural and agricultural systems and other data (Fluck, 1992,
and 1992b). The following equations were used, applying the appropriate usage rates, to
calculate the energy and EMERGY of the fuels used for the total area of each occurrence
of an agricultural land use:
Energy, j/year = (j/m2/yr) (area, m2)
EMERGY, sej/year = (sej/m2/yr) (area, m2)
Step 3: The coverage was then processed according to the methods described
previously for converting polygon feature coverages into the EMERGY and energy
intermediate component grids, called AGRFUL and AGR_FUL_EN, for fuel use in
agriculture.
Urban System Flows-Goods Consumption
This component grid models the flows of energy and EMERGY of both
consumable and durable goods consumed in residential, commercial, and institutional
buildings and as a result of agricultural production (element #5 in Figure 1-3). The
EMERGY of these goods implicitly includes the EMERGY of services to deliver them to
the point of their consumption (Odum, 1996). This model does not try to track the flow
of goods through wholesale and retail establishments. Durable goods (furniture, cars,
equipment, etc.) are included in these calculations by combining estimates of annual
depreciation of durable goods (in other words, annual consumption) with estimates of the
amounts of consumable goods used annually in each building.
The creation of this component grid required that several intermediate component
grids be created in a manner similar to the methods used to create the water use and fuel
use component grids. Four of the intermediate grids were created using the analytical
point coverages called 'em_resbldgcov' and 'em_combldgcov'. The 'em_resbldgcov'
coverage was used to estimate the goods consumed in residential buildings, and the
'em_combldgcov' coverage was used to estimate the goods consumed in commercial,
industrial, and institutional buildings. Agricultural goods consumption was estimated
using the analytical polygon coverage called 'emlandcov'.
Four sets (an energy and an EMERGY grid in each set) of intermediate
component grids were created and then 'added' together to create the final goods
consumption component grid. These intermediate grids are described in Table 2-5. The
final EMERGY component grid, called 'GOODS', is the sum of 'BLD_GDS' and
'AGR GDS'.
Data processing steps. The following three sets of generalized data processing
steps were used in creating the intermediate grids used to create this component of the
spatial EMERGY Model.
Domestic goods consumption intermediate component grid. The residential
goods consumption grids, called RES_GDS and RES_GDS EN were created as follows:
Step 1: The analytical point coverage called 'em_resbldgcov' was overlaid with
the 'em_census' analytical coverage to add the data required to estimate the income of
each individual single family residential-type building. Additional attribute items were
added to the 'em_resbldgcov' coverage to facilitate calculations of energy and EMERGY
flows.
Table 2-5: Intermediate grids that were used to calculate the total energy and EMERGY
of goods consumption in residential, commercial/public buildings and agriculture.
Intermediate Grids Description
RES_GDS EMERGY of residential-based goods consumed
RES_GDS_EN energy of residential-based goods consumed
COM_GDS EMERGY of Commercial/Industrial/
Institutional-based goods consumption
COM_GDS_EN energy of Commercial/Industrial/
Institutional-based goods consumption
BLD_GDS Sum of RES_GDS and COMGDS
BLDGDSEN Sum of RESGDSEN and COMGDS EN
AGR_GDS EMERGY of goods consumed in
agricultural production
AGR_GDS_EN energy of goods consumed in
agricultural production
Note: The final EMERGY sub-component grid called GOODS is the sum of BLD_GDS
and AGR GDS. The final energy sub-component grid called GOODS_EN is the sum of
BLD_GDS EN and AGR_GDS_EN. The intermediate component grid called
AGR_GDS represents the left fork of the split flow of element #5 in Figure 1-3. The
intermediate component grid called BLD_GDS is the right fork of the split flow of
element #5.
Step 2: The 1990 census data (U.S. Bureau of the Census, 1992) were processed
to find the aggregate income for each census 'blockgroup'. The aggregate income was
apportioned to each building by first determining the total square footage of all
residential buildings in each blockgroup and then assigning each building a proportional
amount of the blockgroup aggregate income according to the square footage of the
building. The aggregate income figures provided by the census data were also adjusted
to 1994 levels (BEBR, 1995). The following equation was used to estimate the income
for each building:
Building Income ($/year) = ((blockgroup aggregate income, $/year)
/ (blockgroup total square feet)) square feet of building
Step 3: The estimated dollar amount of goods consumed in each building was
then calculated based on each building's income estimate using data from the U.S.
Department of Labor (1995). This data represents the sum of national average estimated
expenditures ($) for the following general categories: food at home, household supplies,
apparel, amortized vehicle, entertainment, equipment, appliances and furnishings
purchases, personal care supplies, and miscellaneous supplies. Each building was
assigned a dollar value for goods consumption based on the income range of the building.
Step 4: The EMERGY of the annual goods consumption in each building was
calculated using the value of 1.37 E12 sej/$ (Odum, 1996) in the following equation:
EMERGY, sej/year = (dollars of goods/yr) (1.37 E12 sej/$)
The energy of the annual goods consumption in each building was calculated
based on methods and data used by Brown (1980). In this case the assumption was made
that the average cost of a pound of goods was $5 and each pound of goods contained
1500 kcal of energy. The following equation was used:
Energy, j/year = (dollars of goods/yr) (300 kcal/$) (4186 J/kcal)
Step 5: The coverage was then processed according to the methods described
previously for converting point feature coverages into the EMERGY and energy
intermediate component grids, called RES_GDS and RES_GDS_EN, for goods
consumption in residential buildings.
Commercial goods consumption intermediate component grid. In this model the
commercial/industrial/institutional goods consumption grids, called COM_GDS and
COM_GDS_EN, are intended to estimate the amount of goods consumed annually on site
in the process of conducting the particular type of business associated with each building.
The calculations for estimating the amount of goods consumed in each building were
based on the application of different consumption rates, in dollars of goods consumed per
square foot of building per year, for the different types of businesses associated with each
of the buildings. These consumable and durable goods would typically include office
supplies, and amortized depreciation of office furniture, appliances, and industrial
equipment. These grids are not intended to represent goods in stock, that are intended for
wholesale or retail sale to other consumers. In other words, goods are mapped at their
point of consumption. These intermediate component grids were created as follows:
Step 1: Additional attribute items were added to the 'em_combldgcov' analytical
coverage to facilitate calculations of energy and EMERGY flows in
commercial/industrial/institutional buildings.
Step 2: Goods consumption rates, in dollars per square foot per year, were
assigned to each building type using the Department of Revenue (DOR) use code
associated with the building. These rates were calculated for commercial businesses
according to an estimated percentage of gross annual sales expended on goods (both
consumable and amortized durable goods) during a year for each type of business
(USDOC, 1996a, 1996b, 1996c).
Data on county-wide annual gross sales for each type of business (BEBR, 1993,
and 1995) was multiplied by the estimated percentage of gross sales representing goods
consumed in conducting each type of business. This total expenditure on goods was then
divided by the total square footage of all corresponding buildings to get the estimated rate
of goods consumption per square foot for those buildings associated with each type of
business.
Annual goods expenditures by schools, colleges and local/state/federal
government were based on an estimated percentage of gross annual government
expenditures in each category (BEBR, 1995). Gross annual expenditures in each
category were multiplied by the appropriate percentage rate representing goods consumed
and then divided by the aggregate total square footage of the appropriate type of
buildings.
The following equation was used, applying the appropriate consumption rate, to
calculate the annual goods consumption occurring in each building:
Goods consumed ($/year) = total square feet dollars/sq.ft./year
Step 3: To be consistent with the calculations used for domestic goods
consumption, the energy and EMERGY of the annual goods consumption in each
commercial/institutional building was calculated using the same assumptions and the
following equations:
EMERGY, sej/year = (dollars of goods/yr) (1.37 E12 sej/$)
Energy, j/year = (dollars of goods/yr) (300 kcal/$) (4186 J/kcal)
Step 4: The coverage was then processed according to the methods described
previously for converting point feature coverages into the EMERGY and energy
intermediate component grids called, COM_GDS and COM_GDS_EN, representing
goods consumption in commercial/institutional buildings.
Agricultural goods consumption intermediate component grid. In this model the
intermediate component grids called AGR GDS and AGR_GDS_EN are intended to
estimate the amount of goods consumed annually in the process of pasture and row crop
production, and maintenance of residential lawns and gardens. Consumption of goods
that are closely associated with a building, such as nursery and greenhouse production,
was calculated in the commercial/ institutional intermediate component grid. These grids
were created as follows:
Step 1: Additional attribute items were added to the 'em_landcov' analytical
(polygon) coverage to facilitate calculations for the amount of goods consumed per unit
of area for each type of agricultural land use, and the corresponding energy and
EMERGY flows.
Step 2: Estimated goods consumption rates for each type of agricultural
production were used. These rates were based on previous calculations done to
determine contributions to gross primary production in natural and agricultural systems
and other data (Fluck, 1992, and 1992b).
The following equations were used, applying the appropriate usage rates, to
calculate the energy and EMERGY of the goods consumed for the total area of each
occurrence of an agricultural land use:
Energy, j/year = (j/m2/yr) (area, m2)
EMERGY, sej/year = (sej/m2/yr) (area, m2)
Step 3: The coverage was then processed according to the methods described
previously for converting polygon feature coverages into the EMERGY and energy
intermediate component grids called, AGR_GDS and AGR_GDS EN, for goods
consumption in agricultural production.
Urban System Flows-Human Services
This component grid models the energy and EMERGY of in-situ human services
(those services provided within the area of each grid cell). It does not include the
EMERGY from services that was implicitly included in the previous calculations for the
EMERGY of goods consumed (see element #6 in Figure 1-3).
The creation of this component grid presented a significant challenge since, unlike
buildings, people do not stay in the same location all day. The general approach taken for
this study was to estimate the percentages of time people spend while at home, work or
school, and while 'shopping', and then spatially distribute the corresponding services
according to appropriate locations.
Based on this general approach, several intermediate component grids were
created to facilitate the spatial distribution of these services. Six sets (an energy and an
EMERGY grid in each set) of intermediate component grids were created and then
'added' together to create the final human services component grid. These intermediate
grids are listed in Table 2-6. The final EMERGY component grid, called 'SERVICE', is
the sum of the BLD_SRV and AGR SRV intermediate component grids, and the final
energy component grid, called 'SERVICE_EN', is the sum of the BLD_SRV_EN and
AGR_SRV_EN intermediate component grids.
All of the intermediate grids, except for the agricultural services grid, were
created using the analytical point coverages called 'em_resbldgcov' and
'em_combldgcov'. The 'em_resbldgcov' coverage was used to estimate of the services
provided while people were at their residences. The 'em_combldgcov' coverage was
used to estimate the services provided while people were at workplaces, schools, or were
shopping. The use of these analytical point coverages was determined to be the most
appropriate method of calculating and locating most human services because most people
spend the majority of their time in and around buildings.
The analytical polygon coverage called 'em_landcov' was used to calculate the
agricultural services because these services tend to be spatially dispersed.
Data processing steps. The following generalized data processing steps were used
in creating the intermediate grids used to create this component of the spatial EVIMERGY
model.
Table 2-6: Intermediate grids that were used to calculate the total energy and EMERGY
of human services.
Intermediate Grids Description
EMERGY of services occurring in
RES_SRV residential buildings
RES_SRV_EN energy of services in residential buildings
EMERGY of human services occurring in
WRK_SRV school and work buildings
WRK_SRV_EN energy of services in schoolwork buildings
EMERGY of human services occurring in
SHP_SRV buildings as a result of shopping
SHP_SRV_EN energy of services while shopping
COMSRV WRK SRV + SHPSRV
COMSRVEN WRK SRV EN + SHPSRVEN
BLDSRV RES SRV + COMSRV
BLDSRVEN RES SRV EN + COMSRVEN
EMERGY of human services provided to
AGR_SRV agricultural production
AGR_SRV_EN energy of services in agricultural production
Note: The final EMERGY component grid, called 'SERVICE', is the sum of the
BLD_SRV and AGR_SRV intermediate component grids, and the final energy
component grid, called 'SERVICE_EN', is the sum of the BLD_SRV_EN and
AGR_SRV_EN intermediate component grids.
Services-at-home intermediate component grid. In this study, the assumption was
made that people spend 67% of each 24-hour day at home. The intermediate component
grids, called RES_SRV and RES_SRV EN, representing EMERGY and energy of
services provided at home were created as follows:
Step 1: The analytical point coverage called 'em_resbldgcov' was overlaid with
the 'em_census' data layer to add the data required to estimate the population of each
individual residential-type building.
The census data (U.S. Bureau of the Census, 1992) were processed to identify two
population classes: the total number of adults (over 18 years old) and the total number of
children. Since the census data represent the total population, by age class, for an area
called a 'census block', these population numbers had to be distributed proportionately
across all of the buildings, represented as point features, which were found in each block.
This apportioning of the population was accomplished by first determining the
number of 'residential units' occurring in each building (the number of units in multi-
family buildings is an attribute in the property parcel coverage) and the total number of
units occurring in each census block.
Then, a calculation was performed for each building as follows: the total number
of people (in each age class) of the corresponding census block was divided by the total
number of residential units in that block; this number was then multiplied by the number
of residential units in the building. The result of these calculations was an estimate of
each building's population of both adults and children (people under 18 years old).
Step 2: Additional attribute items were added to the 'em_resbldgcov' coverage to
facilitate calculations for the energy in services. The energy in human services from
adults and children occurring in each residential building were calculated using the
following equations:
Energy in adult services, j/yr/bldg = (# of adults) (2500 kcal/day)
Energy in child services, j/yr/bldg
Total energy in services, j/yr/bldg
* (365 days/year) (4186 j/kcal) (.67)
= (# of children) (2500 kcal/day)
* (365 days/year) (4186 j/kcal) (.67)
= (Adult Services, j/yr/bldg)
+ (Child Services, j/yr/bldg)
Step 3: Additional attribute items were also added to the 'em_resbldgcov'
coverage to facilitate calculations for the EMERGY in services from humans based on
different transformities according to level of education (Odum, 1996). For adults, the
average proportion of high school and college-educated persons (U.S. Census Bureau,
1992) was used to calculate the EMERGY of services using two different transformities.
All services from children use the same transformity. The following equations were
used:
for college-educated adults;
EMERGY, sej/yr/bldg = ((# adults) (.25)) (2500 kcal/day)
(365 days/year) (4186 j/kcal) (.67, %time) (7.33 E7 sej/j)
for high school-educated adults;
EMERGY, sej/yr/bldg = ((# adults) (.75)) (2500 kcal/day)
(365 days/year) (4186 j/kcal) (.67, %time) (2.46 E7 sej/j)
for children;
EMERGY, sej/yr/bldg = (# children) (2500 kcal/day) (365 days/year)
(4186 j/kcal) (.67, %time) (8.90 E6 sej/j)
for the total EMERGY;
Total EMERGY, sej/yr/bldg = EMERGY from college-educated adults
+ EMERGY from high-school educated adults
+ EMERGY from children
Step 4: The 'em_resbldgcov' analytical coverage was then processed according
to the methods described previously for converting point feature coverages into the
EMERGY and energy intermediate component grids, called RES_SRV and
RES_SRV_EN, for residential human services.
Services-at-work/school intermediate component grid. In this study, the
assumption was made that people spend 25% of each day at work or school. The
analytical point coverage called 'em_combldgcov' was used to create this intermediate
component grid. This coverage contains all of the buildings that were not classified as
residential. The intermediate component grids, called WRK_SRV and WRK_SRV_EN,
representing EMERGY and energy of services provided at work or school were created
as follows:
Step 1: Additional attribute items were added to the 'em_combldgcov' coverage
to facilitate an estimate of the average daytime population of each building. These
calculations were based on estimates of the number of employees, or students, per square
foot of each building type according to the Department of Revenue Use Code. These
estimates were calculated by dividing the total number of employees/students for each
type of business/school (BEBR, 1993) by the total square footage of all of the buildings
in each type of business/school. The average daytime building population was calculated
as follows:
Building Population = (employees or students/sq.foot)
(sq. feet of building)
Step 2: Additional attribute items were added to the 'em_combldgcov' coverage
to facilitate calculations for the energy in human services. Only the building population
of primary and secondary schools were used for the calculation of children's services.
The energy in human services from adults and children occurring in each building were
calculated using the following equations:
Energy in adult services, j/yr/bldg = (# of adults) (2500 kcal/day)
(365 days/year) (4186 j/kcal) (.25)
Energy in child services, j/yr/bldg = (# of children) (2500 kcal/day)
(365 days/year) (4186 j/kcal) (.25)
Total energy in services, j/yr/bldg = (Adult Services, j/yr/bldg)
+ (Child Services, j/yr/bldg)
Step 3: Additional attribute items were also added to the 'em_combldgcov'
coverage to facilitate calculations for the EMERGY in services from humans at
work/school. These calculations are also based on different transformities according to
level of education as was done for services occurring at residences. Different types of
buildings were assumed to have different percentages of college and high school
educated adults, and children depending on whether the building was associated with
professional services, manufacturing, etc. The following equations (which are essentially
the same ones used for residential services, except for the percentage of time and the
varying percentages of college and high school-educated adults) were used:
for college-educated adults;
EMERGY, sej/yr/bldg = ((# adults) (%college)) (2500 kcal/day)
(365 days/year) (4186 j/kcal)
(.25, %time) (7.33 E7 sej/j)
for high school-educated adults;
EMERGY, sej/yr/bldg = ((# adults) (%high school) (2500 kcal/day)
(365 days/year) (4186 j/kcal)
(.25, %time) (2.46 E7 sej/j)
for all children;
EMERGY, sej/yr/bldg = (# children) (2500 kcal/day) (365 days/year)
(4186 j/kcal) (.25, %time) (8.90 E6 sej/j)
for the total EMERGY;
Total EMERGY, sej/yr/bldg = EMERGY from college-educated adults
+ EMERGY from high-school educated adults
+ EMERGY from children
Step 4: The 'em_combldgcov' analytical coverage was then processed according
to the methods described previously for converting point feature coverages into the
EMERGY and energy intermediate component grids, called WRK_SRV and
WRK_SRV_EN, for human services provided while at work or school.
Shopping' intermediate component grid. In this study, the assumption was made
that people spend an average of 8% of each day 'shopping' at retail or service
establishments. This works out to about 2 hours per day, and implicitly includes time
spent driving to and from these 'shopping' places. This intermediate grid was created in
an attempt to recognize that people spend time at locations other than work and home.
The energy and EMERGY in human services provided by people while shopping
at these locations was estimated in the intermediate component grids called SHP_SRV
and SHP_SRV_EN. These grids were created as follows:
Step 1: Additional attribute items were added to the analytical point coverage
called 'em_combldgcov' to facilitate the calculations. Only those buildings that were
assumed to be associated with the retail or service industries were qualified for inclusion
in these calculations. The square footage of each shopping-qualified building was
divided by the total square footage of all qualified buildings to determine the percentage
each building represents of the total county-wide qualified building square footage.
Step 2: The primary calculations used to create this grid are based on two
assumptions: first, it was assumed that every person, including children, spends 8% of
their time shopping, and second, it was assumed that the percentage of the total shopping
time of all people in the county that occurs in each qualified building was proportional to
the percentage that building represents of the total county-wide shopping-qualified
building square footage. The total population of the county, 142,081 adults and 39,515
children, was based on U.S. Census Bureau data (1990). Based on these assumptions, the
following equations were used to calculate the energy in shopping services:
Energy in adult services, j/yr/bldg = (% of total shopping sq. footage)
(total number of adults in county) (.08, %time)
(2500 kcal/day)* (365 days/year) (4186 j/kcal)
Energy in child services, j/yr/bldg = (% of total shopping sq. footage)
(total number of children in county) (.08, %time)
(2500 kcal/day)* (365 days/year) (4186 j/kcal)
Total energy in services, j/yr/bldg = (Adult Services, j/yr/bldg)
+ (Child Services, j/yr/bldg)
Step 3: Additional attribute items were also added to the 'em_combldgcov'
coverage to facilitate calculations for the EMERGY in services from humans while
shopping. These calculations are based on the assumptions about education levels used
previously for the residential services intermediate component grid. The following
equations were used:
for college-educated adults;
EMERGY, sej/yr/bldg = (% of total shopping sq. footage)
(# of college educated adults in county)
(.08, %time) (2500 kcal/day) (365 d/year)
(4186 j/kcal) (7.33 E7 sej/j)
for high school-educated adults;
EMERGY, sej/yr/bldg = (% of total shopping sq. footage)
(# of high school educated adults in county)
(.08, %time)* (2500 kcal/day) (365 d/year)
(4186 j/kcal) (2.46 E7 sej/j)
for children;
EMERGY, sej/yr/bldg = (% of total shopping sq. footage)
(# of children in county) (.08, %time)
(2500 kcal/day) (365 days/year)
(4186 j/kcal) (8.90 E6 sej/j)
for the total EMERGY;
Total EMERGY, sej/yr/bldg = EMERGY from college-educated adults
+ EMERGY from high school educated adults
+ EMERGY from children
Step 4: The 'em_daypopcov' analytical coverage was then processed according
to the methods described previously for converting point feature coverages into the
EMERGY and energy intermediate component grids, called SHP_SRV and
SHP_SRV_EN, for human services provided while shopping.
Agricultural services intermediate component grid. Human services related to
agricultural production, which were closely associated with a building, such as nursery
and greenhouse production, were calculated in the intermediate component grid for
work/school services. This grid includes services that are more appropriately modeled as
occurring over areal rather than point locations. These services include those provided to
pasture, forestry, and row crop production, and maintenance of residential lawns and
gardens. The intermediate component grids are called AGR SRV and AGR SRV_EN
and were created as follows:
Step 1: Additional attribute items (were added to the 'em_landcov' polygon
coverage to facilitate calculations for the amount of human services provided per unit of
area for each type of agricultural land use, and the corresponding energy and EMERGY
flows.
Step 2: Estimated service rates for each type of agricultural production were used
based on previous calculations done to determine contributions to gross primary
production in natural and agricultural systems.
The following equations were used, applying the appropriate usage rates, to
calculate the energy and EMERGY of the services provided for the total area of each
occurrence of an agricultural land use:
Energy, j/year = (j/m2/yr) (area, m2)
EMERGY, sej/year = (sej/m2/yr) (area, m2)
Step 3: The coverage was then processed according to the methods described
previously for converting polygon feature coverages into the EMERGY and energy
intermediate component grids, called AGR SRV and AGR SRV EN, for human
services provided in agricultural production.
Urban System Flows-Wastes Not Recycled
This component grid models the flows of energy and EMERGY of the solid and
liquid wastes that were generated in residential, commercial, and institutional buildings
that were not recycled (see element #7 in Figure 1-3).
Data from several sources (FDEP, 1995, USEPA, 1997, and TIA, 1991) on the
municipal solid waste (MSW) stream for Alachua County (for the period of July 1994 to
June 1995) was summarized and used as the basis for estimates of municipal solid waste
(MSW) flows being generated at the sites of various types of buildings. Land-filled
MSW was categorized into: general wastes (urban or rural), yard wastes (urban or rural),
and construction debris. Liquid waste estimates were based on a percentage of water
consumed using data calculated previously for the water use component grid.
The creation of this component grid required that several intermediate component
grids be created in a manner similar to the method used to create previous component
grids. Separate solid and liquid waste intermediate component grids were created using
different methods. The 'em_resbldgcov' coverage was used to estimate the solid and
liquid wastes generated in residential buildings, and the 'em_combldgcov' coverage was
used to estimate the solid and liquid wastes generated in commercial, industrial, and
Table 2-7: Intermediate grids that were used to calculate the total energy and EMERGY
of municipal solid wastes (MSW) and liquid waste flows that were not recycled.
Intermediate Grids Description
RES_MSW EMERGY of MSW originating
in residential buildings
RES_MSW_EN energy of MSW originating
in residential buildings
RES_LWS EMERGY of liquid wastes originating
in residential buildings
RES_LWS_EN energy of liquid wastes originating
in residential buildings
COM_MSW EMERGY of MSW originating
in commercial/institutional buildings
COM_MSW_EN energy of MSW originating
in commercial/institutional buildings
COM_LWS EMERGY of liquid wastes originating
in commercial/institutional buildings
COM_LWS_EN energy of liquid wastes originating
in commercial/institutional buildings
BLDMSW RES MSW + COM MSW
BLDMSWEN RES MSW EN + COMMSWEN
BLDLWS RESLWS + COMLWS
BLD_LWSEN RESLWSEN + COMLWSEN
|
RSS
