<%BANNER%>

Parcel-Level Methodology for Estimating Commercial, Industrial and Institutional Water Use

Permanent Link: http://ufdc.ufl.edu/UFE0042241/00001

Material Information

Title: Parcel-Level Methodology for Estimating Commercial, Industrial and Institutional Water Use
Physical Description: 1 online resource (151 p.)
Language: english
Creator: Morales, Miguel
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: commercial, conservation, demand, estimating, industrial, institutional, management, parcel, use, water
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, M.E.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Title: Parcel-Level Methodology for Estimating Commercial, Industrial and Institutional Water Use This thesis presents a new methodology to estimate commercial, industrial, and institutional (CII) water use based on evaluation of parcel-level customer attribute and water use billing databases. The Florida Department of Revenue (FDOR) and Florida County Property Appraisers (FCPA) databases provide the heated building area and customer classifications for every CII parcel in the state of Florida. Linking this parcel-level attribute data with parcel-level water use billing data provides a major improvement in our ability to estimate CII use. Existing methods typically use the number of employees as the best single indicator of the size of the activity. Employee data is available periodically through the U.S. Census or private surveys. Census data is not available at the parcel level and survey data are expensive to collect. Evaluation of alternative measures of size that are contained in the FDOR/FCPA databases and customer billing data indicate that heated area is the best single measure of size to use across the 55 CII two-digit FDOR land use categories, or subsectors. It is relatively straightforward to link the FDOR and FCPA databases for utilities because the water management districts provide information on utility boundaries. The more difficult challenge is to link these parcel-level databases with the utility s customer billing database. The level of effort depends on the type of billing system and whether common fields are available to create the necessary relational databases. Only a few Florida water utilities are known to have merged these databases. The base case for this analysis is Hillsborough County Water Resources Services and Gainesville Regional Utilities that provide a relatively large sample of 3,205 CII parcels of which 69% are commercial, 10% are industrial, and 21% are institutional. Property and parcel attributes of this sample dataset including heated area distributions were evaluated at the subsector level and compared to the state-wide totals of CII parcels, to gain a sense of the representativeness of the sample. Then, monthly water use was analyzed to estimate average, base, seasonal, and May peak water use per CII parcel. Finally, water use coefficients expressed in gallons per square foot of heated area per day were developed for each of the available FDOR CII subsectors. Knowing the water use coefficient and the total heated area for each FDOR subsector, it is simple to estimate their total water use. The availability of the FDOR/FCPA databases provides a major improvement in our ability to estimate CII water use. The quality of these estimates will continue to improve as more utilities link their billing data with the FDOR/FCPA databases.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Miguel Morales.
Thesis: Thesis (M.E.)--University of Florida, 2010.
Local: Adviser: Heaney, James.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042241:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042241/00001

Material Information

Title: Parcel-Level Methodology for Estimating Commercial, Industrial and Institutional Water Use
Physical Description: 1 online resource (151 p.)
Language: english
Creator: Morales, Miguel
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: commercial, conservation, demand, estimating, industrial, institutional, management, parcel, use, water
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, M.E.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Title: Parcel-Level Methodology for Estimating Commercial, Industrial and Institutional Water Use This thesis presents a new methodology to estimate commercial, industrial, and institutional (CII) water use based on evaluation of parcel-level customer attribute and water use billing databases. The Florida Department of Revenue (FDOR) and Florida County Property Appraisers (FCPA) databases provide the heated building area and customer classifications for every CII parcel in the state of Florida. Linking this parcel-level attribute data with parcel-level water use billing data provides a major improvement in our ability to estimate CII use. Existing methods typically use the number of employees as the best single indicator of the size of the activity. Employee data is available periodically through the U.S. Census or private surveys. Census data is not available at the parcel level and survey data are expensive to collect. Evaluation of alternative measures of size that are contained in the FDOR/FCPA databases and customer billing data indicate that heated area is the best single measure of size to use across the 55 CII two-digit FDOR land use categories, or subsectors. It is relatively straightforward to link the FDOR and FCPA databases for utilities because the water management districts provide information on utility boundaries. The more difficult challenge is to link these parcel-level databases with the utility s customer billing database. The level of effort depends on the type of billing system and whether common fields are available to create the necessary relational databases. Only a few Florida water utilities are known to have merged these databases. The base case for this analysis is Hillsborough County Water Resources Services and Gainesville Regional Utilities that provide a relatively large sample of 3,205 CII parcels of which 69% are commercial, 10% are industrial, and 21% are institutional. Property and parcel attributes of this sample dataset including heated area distributions were evaluated at the subsector level and compared to the state-wide totals of CII parcels, to gain a sense of the representativeness of the sample. Then, monthly water use was analyzed to estimate average, base, seasonal, and May peak water use per CII parcel. Finally, water use coefficients expressed in gallons per square foot of heated area per day were developed for each of the available FDOR CII subsectors. Knowing the water use coefficient and the total heated area for each FDOR subsector, it is simple to estimate their total water use. The availability of the FDOR/FCPA databases provides a major improvement in our ability to estimate CII water use. The quality of these estimates will continue to improve as more utilities link their billing data with the FDOR/FCPA databases.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Miguel Morales.
Thesis: Thesis (M.E.)--University of Florida, 2010.
Local: Adviser: Heaney, James.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042241:00001


This item has the following downloads:


Full Text






PARCEL-LEVEL METHODOLOGY FOR ESTIMATING COMMERCIAL, INDUSTRIAL
AND INSTITUTIONAL WATER USE




















By

MIGUEL ALFREDO MORALES


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ENGINEERING

UNIVERSITY OF FLORIDA

2010
































2010 Miguel Alfredo Morales



















To my family, for their never-ending support









ACKNOWLEDGEMENTS

I am grateful to Dr. James Heaney, my primary advisor, for his tremendous insight and

assistance throughout the course of my graduate studies. I am also forever indebted to Dr. David

Mazyck for granting me the opportunity to conduct undergraduate research, sparking my interest

in academic research, and always having my best interests at heart. I thank Dr. Ben Koopman for

taking part in my advisory committee; his support was also critical. To Jackie Martin, I am

greatly appreciative of our early collaboration, but even more so of our friendship. I would also

like to thank my other colleagues at the Conserve Florida Water Clearinghouse: John Palenchar,

Ken Friedman, Lukasz Ziemba, Leighton Walker, Camilo Corejo, Randy Switt, and Kristen

Riley. Financial support for this research was provided by the Florida Department of

Environmental Protection, South Florida Water Management District, St. John's River Water

Management District, and the Southwest Florida Water Management District. Critical data for

this research was provided by Hillsborough County Water Resources Services and Gainesville

Regional Utilities. Finally, I would like to thank my family and friends. Their contribution to this

work cannot be easily quantified, but without them none of this would have been possible.









TABLE OF CONTENTS
page

A C K N O W L E D G E M EN T S ...................... .. .. ......... .. ..............................................................4

T A B L E O F C O N TEN T S ......................................................... ..............................................5

L IST O F T A B L E S ...................... ............... ....................................................... 7

LIST OF FIGURES ................... .. .............................. .................. ............... 11

L IST O F A B B R E V IA T IO N S ...................... .. .. .......... ............................................12

A B S T R A C T ............ ................... ............................................................ 13

CHAPTER

1 INTRODUCTION ................................ ... .. .... ..... ................. 15

2 L ITE R A TU R E R E V IE W ............................................................................. .....................18

Previous Commercial, Industrial, and Institutional Water Use Models ..............................18
Employment Data as a Measure of Size ........................................................................ 23

3 BOTTOM-UP APPROACH FOR EVALUATING THE SIZE OF EACH ACTIVITY........28

In tro d u ctio n ................................. ................. ....................... ................ 2 8
Land Use Databases.................. ................... .............................. ............ 28
Florida Department of Revenue ............................ ................ .................... 28
Florida County Property Appraisers...................................................... ...................31
Relationship of Effective Area to Heated Area................................... ...............34
A naly sis of A v ailable D ata ............................................... ......................... ....................... 36
S am p le siz e .................................36.............................
E ffectiv e Y ear B u ilt............ ... .... ....... ............................................... ......... ..... .. 3 7
Heated Area to Effective Area Calculation........... ................... ....... .. ............. 37
Distribution and Representativeness of Sampled Heated Areas ...................................38

4 FLORIDA CASE STUDIES ON EVALUATING WATER USE........................................66

U utility W ater B killing D atabases ................................................................... .....................66
Hillsborough County Water Resources Services .................................... ...............66
G ainesville R regional U tilities............................................................... .....................68
C om bined U tilities.............. .... .................................... .. .. ....... ..............70
Relationship of Heated Area to Water Use.................... ............................... 71









5 COMMERCIAL, INDUSTRIAL, AND INSTITUTIONAL WATER USE
C O E F F IC IE N T S .................................... .................................................... ....................87

Intro du action ................... .......................................................... ................ 87
C om m ercial Sector ............... .... ............ ............................ .. .... ......... .. ..... 89
Total Commercial Coefficient Calculation ........................................ ............... 90
Commercial Subsector Analysis at the Utility Level ............................................... 91
Commercial Subsector Analysis across Utilities.................................. ...............92
Aggregation of Comm ercial Utility Data............................................... .................. 94
In du trial S ecto r......................................................................... .. 9 4
Industrial Subsector Analysis at the Utility Level....................................................95
Industrial Subsector Analysis across Utilities ...................................... ............... 95
Aggregation of Industrial Utility D ata ........................................ ........................ 96
Institutional Sector.............................................................................. 96
Institutional Subsector Analysis at the Utility Level.....................................................97
Institutional Subsector Analysis across Utilities .................................. ............... 97
Aggregation of Institutional Utility Data ............................................. ............... 97
Coefficient Com prison w ith other Studies....................................... ......................... 98
Incorporation of Results into EZ Guide 2....... .... .................................... ............ .....98

6 SUMMARY, CONCLUSIONS, AND NEED FOR ADDITIONAL RESEARCH ..............117

APPENDIX: HEATED AREA SUBSECTOR DISTRIBUTIONS ...........................................120

L IS T O F R E F E R E N C E S ............................................................................................................. 14 9

B IO G R A PH IC A L SK E T C H ............................. ............................................... ..................... 151









LIST OF TABLES


Table page

2-1 Comparison of models in the Institute of Water Resources Municipal and Industrial
N eeds (IW R-M A IN ) m odel tradition........................................... .......................... 27

2-2 Sources of em ploym ent data............................................................................. ............27

3-1 Databases and parcel attributes used to develop water use and area conversion
coefficients based on a sample of 3,205 commercial, industrial, and institutional
(C II) p a rc e ls ............................................................................ 4 1

3-2 Parcel breakdown across commercial subsectors for Hillsborough County Water
Resources Services, Gainesville Regional Utilities, the two utilities combined, and
the entire state of Florida ...................... ...................... ................... .. ......42

3-3 Parcel breakdown across industrial subsectors for Hillsborough County Water
Resources Services, Gainesville Regional Utilities, the two utilities combined, and
the entire state of Florida ...................... ...................... ................... .. ......43

3-4 Parcel breakdown across institutional subsectors for Hillsborough County Water
Resources Services, Gainesville Regional Utilities, the two utilities combined, and
the entire state of Florida ...................... ...................... ................... .. ......44

3-5 Average effective year built across commercial subsectors for Hillsborough County
Water Resources Services, Gainesville Regional Utilities, the two utilities combined,
and the entire state of Florida.................................................................. ............... 45

3-6 Average effective year built across industrial subsectors for Hillsborough County
Water Resources Services, Gainesville Regional Utilities, the two utilities combined,
and the entire state of Florida.................................................................. ............... 46

3-7 Average effective year built across institutional subsectors for Hillsborough County
Water Resources Services, Gainesville Regional Utilities, the two utilities combined,
and the entire state of Florida.................................................................. ............... 46

3-8 Heated area to effective area ratio of the means calculation of commercial subsectors
in Hillsborough County Water Resources Services, Gainesville Regional Utilities,
and the tw o utilities com bined ................................................ ............................... 47

3-9 Heated area to effective area ratio of the means calculation of industrial subsectors in
Hillsborough County Water Resources Services, Gainesville Regional Utilities, and
the tw o utilities com bined ..........................................................................................48

3-10 Heated area to effective area ratio of the means calculation of institutional subsectors
in Hillsborough County Water Resources Services, Gainesville Regional Utilities,
and the tw o utilities com bined ................................................ ............................... 49









3-11 Comparison of heated area to effective area ratio of the means calculation for the
commercial subsectors in Hillsborough County Water Resources Services,
Gainesville Regional Utilities, and the two utilities combined............... ............... 50

3-12 Comparison of heated area to effective area ratio of the means calculation for the
industrial subsectors in Hillsborough County Water Resources Services, Gainesville
Regional Utilities, and the two utilities combined ......................................................51

3-13 Comparison of heated area to effective area ratio of the means calculation for the
institutional subsectors in Hillsborough County Water Resources Services,
Gainesville Regional Utilities, and the two utilities combined............... ............... 51

3-14 Comparison of average heated area across the commercial subsectors of
Hillsborough County Water Resources Services, Gainesville Regional Utilities, the
two utilities combined, and the state of Florida........................... ............................52

3-15 Comparison of average heated area across the industrial subsectors of Hillsborough
County Water Resources Services, Gainesville Regional Utilities, the two utilities
com bined, and the state of Florida..................................................... ............... 54

3-16 Comparison of average heated area across the institutional subsectors of
Hillsborough County Water Resources Services, Gainesville Regional Utilities, the
two utilities combined, and the state of Florida........................... ............................55

3-17 Fitted lognormal probability density equations and associated Anderson-Darling (A-
D) and Kolmogorov-Smirnow (K-S) statistics for State total and combined utility
sam ple of com m ercial parcels........................................................... ...........................56

3-18 Fitted lognormal probability density equations and associated Anderson-Darling (A-
D) and Kolmogorov-Smirnow (K-S) statistics for State total and combined utility
sam ple of indu strial parcels ..................................................................................... ........ 57

3-19 Fitted lognormal probability density equations and associated Anderson-Darling (A-
D) and Kolmogorov-Smirnow (K-S) statistics for State total and combined utility
sam ple of institutional parcels .................................................. ........................................... 58

3-20 Fitted lognormal shift probability density equations and associated Anderson-Darling
(A-D) and Kolmogorov-Smimow (K-S) statistics for State total and combined utility
sam ple of com m ercial parcels........................................................... ...........................59

3-21 Fitted lognormal shift probability density equations and associated Anderson-Darling
(A-D) and Kolmogorov-Smimow (K-S) statistics for State total and combined utility
sam ple of industrial parcels ......... ............. ...................................................61

3-22 Fitted lognormal shift probability density equations and associated Anderson-Darling
(A-D) and Kolmogorov-Smimow (K-S) statistics for State total and combined utility
sample of institutional parcels ...................... ........................................... ..62









4-1 Summary statistics for CII sectors in Hillsborough County Water Resources Services ...74

4-2 Summary statistics for CII sectors in Gainesville Regional Utilities .............................74

4-3 Comparison of water use per commercial account for Hillsborough County Water
Resources Services, Gainesville Regional Utilities, and the two utilities combined.........75

4-4 Comparison of water use per industrial account for Hillsborough County Water
Resources Services, Gainesville Regional Utilities, and the two utilities combined.........77

4-5 Comparison of water use per institutional account for Hillsborough County Water
Resources Services, Gainesville Regional Utilities, and the two utilities combined.........78

4-6 Correlation matrix of Florida Department of Revenue (FDOR) and Florida County
Property Appraiser (FCPA) property attributes and water use for CII parcels in
Hillsborough County Water Resources Services and Gainesville Regional Utilities........79

4-7 Stepwise regression results for CII parcels in Hillsborough County Water Resources
Services and Gainesville Regional Utilities.................................. ......................... 79

4-8 Stepwise regression "goodness-of-fit" measures for CII parcels in Hillsborough
County Water Resources Services and Gainesville Regional Utilities............................79

4-9 ANOVA for stepwise regression of CII parcels in Hillsborough County Water
Resources Services and Gainesville Regional Utilities .......................................... 79

4-10 Step information for stepwise regression of CII parcels in Hillsborough County
Water Resources Services and Gainesville Regional Utilities .......................................80

4-11 Regression statistics between heated square footage and average daily water use for
the FDOR CII subsectors in Hillsborough County Water Resources Services and
G ainesville R regional U utilities ................................................ ............... ............... 81

5-1 Water use coefficients and sector statistics based on sample of 1,177 commercial
parcels and four years of billing records from Hillsborough County Water Resources
Services ........... ..... ...................................................................... .... 100

5-2 Water use coefficients and sector statistics based on sample of 1,037 commercial
parcels and two years of billing records from Gainesville Regional Utilities ...............102

5-3 Percent difference between commercial water use coefficients of Hillsborough
County Water Resources Services and Gainesville Regional Utilities..........................104

5-4 Water use coefficients and sector statistics based on sample of 2,214 commercial
parcels in Hillsborough County Water Resources Services and Gainesville Regional
U tilitie s ........................................................ ................................. 10 5









5-5 Water use coefficients and sector statistics based on sample of 163 industrial parcels
and four years of billing records from Hillsborough County Water Resources
S e rv ic e s ................... .......................... .. ........................ ................ 1 0 7

5-6 Water use coefficients and sector statistics based on sample of 144 industrial parcels
and two years of billing records from Gainesville Regional Utilities ...........................108

5-7 Percent difference between industrial water use coefficients of Hillsborough County
Water Resources Services and Gainesville Regional Utilities .....................................109

5-8 Water use coefficients and sector statistics based on sample of 307 industrial parcels
in Hillsborough County Water Resources Services and Gainesville Regional Utilities .110

5-9 Water use coefficients and sector statistics based on sample of 428 institutional
parcels and four years of billing records from Hillsborough County Water Resources
S erv ic e s ............. ......... .. .. ......... .. .. ......... ......................................... 1 1 1

5-10 Water use coefficients and sector statistics based on sample of 256 institutional
parcels and two years of billing records from Gainesville Regional Utilities ...............12

5-11 Percent difference between institutional water use coefficients of Hillsborough
County Water Resources Services and Gainesville Regional Utilities............................113

5-12 Water use coefficients and sector statistics based on sample of 684 institutional
parcels in Hillsborough County Water Resources Services and Gainesville Regional
U utilities ......... ................. .......... ......................................................114

5-13 Water use coefficient comparison to other studies on CII water use ............. ...............115









LIST OF FIGURES


Figure page

3-1 Levels of Florida Department of Revenue (FDOR) land use disaggregation into 9
residential and 55 commercial, industrial, and institutional (CII) sectors........................63

3-2 Schematic of spatial and attribute database relationships to FDOR..............................63

3-3 Macro to nano-scale evaluation of public water use in Florida..................................64

3-4 Heated and effective area correlation for 3,205 CII parcels in Hillsborough County
Water Resources Services and Gainesville Regional Utilities .......................................64

3-5 Residual plot of heated and effective area simple linear regression for 3,205 CII
parcels in Hillsborough County Water Resources Services and Gainesville Regional
Utilities ....................................................... 65

4-1 Time series plots of monthly water use for 1,177 commercial parcels, 163 industrial
parcels, and 428 institutional parcels in Hillsborough County Water Resources
Services ......... ..... ............... ........... ....................................... ..... 82

4-2 Time series plots of monthly water use for 1,037 commercial parcels, 144 industrial
parcels, and 256 institutional parcels in Gainesville Regional Utilities ..........................83

4-3 Average monthly water use per commercial account in Hillsborough County Water
Resources Services, Gainesville Regional Utilities, and the two utilities combined.........84

4-4 Average monthly water use per industrial account in Hillsborough County Water
Resources Services, Gainesville Regional Utilities, and the two utilities combined.........85

4-5 Average monthly water use per institutional account in Hillsborough County Water
Resources Services, Gainesville Regional Utilities, and the two utilities combined.........86

5-1 EZ Guide 2 water budget summary for a utility in South Florida ........................ 116









LIST OF ABBREVIATIONS


APCA

CFWC

CII

EA

FCPA

FDOR

Gpcd

GRU

HA

HCPA

HCWRS

IWR-MAIN

PMCL

SIC

TAZ


Alachua County Property Appraiser

Conserve Florida Water Clearinghouse

Commercial, industrial, and institutional

effective area

Florida County Property Appraisers

Florida Department of Revenue

gallons per capital per day

Gainesville Regional Utilities

heated area

Hillsborough County Property Appraiser

Hillsborough County Water Resources Services

Institute of Water Resources Municipal and Industrial Needs model

Planning and Management Consulting, Ltd

Standard Industrial Classifications

Traffic Analysis Zone









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Engineering


PARCEL-LEVEL METHODOLOGY FOR ESTIMATING COMMERCIAL, INDUSTRIAL
AND INSTITUTIONAL WATER USE


By

Miguel Alfredo Morales

August 2010

Chair: James P. Heaney
Major: Environmental Engineering Sciences

This thesis presents a new methodology to estimate commercial, industrial, and

institutional (CII) water use based on evaluation of parcel-level customer attribute and water use

billing databases. The Florida Department of Revenue (FDOR) and Florida County Property

Appraisers (FCPA) databases provide the heated building area and customer classifications for

every CII parcel in the state of Florida. Linking this parcel-level attribute data with parcel-level

water use billing data provides a major improvement in our ability to estimate CII use. Existing

methods typically use the number of employees as the best single indicator of the "size" of the

activity. Employee data is available periodically through the U.S. Census or private surveys.

Census data is not available at the parcel level and survey data are expensive to collect.

Evaluation of alternative measures of size that are contained in the FDOR/FCPA databases and

customer billing data indicate that heated area is the best single measure of size to use across the

55 CII two-digit FDOR land use categories, or subsectors.

It is relatively straightforward to link the FDOR and FCPA databases for utilities because

the water management districts provide information on utility boundaries. The more difficult

challenge is to link these parcel-level databases with the utility's customer billing database. The









level of effort depends on the type of billing system and whether common fields are available to

create the necessary relational databases. Only a few Florida water utilities are known to have

merged these databases. The base case for this analysis is Hillsborough County Water Resources

Services and Gainesville Regional Utilities that provide a relatively large sample of 3,205 CII

parcels of which 69% are commercial, 10% are industrial, and 21% are institutional. Property

and parcel attributes of this sample dataset including heated area distributions were evaluated at

the subsector level and compared to the state-wide totals of CII parcels, to gain a sense of the

representativeness of the sample. Then, monthly water use was analyzed to estimate average,

base, seasonal, and May peak water use per CII parcel. Finally, water use coefficients expressed

in gallons per square foot of heated area per day were developed for each of the available FDOR

CII subsectors. Knowing the water use coefficient and the total heated area for each FDOR

subsector, it is simple to estimate their total water use. The availability of the FDOR/FCPA

databases provides a major improvement in our ability to estimate CII water use. The quality of

these estimates will continue to improve as more utilities link their billing data with the

FDOR/FCPA databases.









CHAPTER 1
INTRODUCTION

Municipal water use is commonly measured in gallons per capital per day (gpcd) to

compare use between utilities and across water use sectors. Gpcd can be defined as net gpcd,

which is water use by the residential sector only, or as gross gpcd, water use by all sectors. Gross

gpcd includes total water use and water loss, and should be evaluated by a water budget to better

understand the relative use by different sectors of customers. The Conserve Florida Water

Clearinghouse (CFWC) EZ Guide 2 is a water planning tool to estimate water use and evaluate

conservation best management practices (http://www.conservefloridawater.org/). The major

sectors profiled in EZ Guide 2 are: single family residential, multi-family residential,

commercial, industrial, institutional (CII), and water loss (Friedman and Heaney 2009). Water

use estimates for the CII sectors are addressed in this thesis.

CII water use comprised 23.5% of public water supply withdrawals for the state of Florida

in 2005 (Marella 2009). These estimates of water use were based on county-wide employment

figures from the U.S. Census Bureau multiplied by water use per employee coefficients. These

coefficients come from a nationwide survey of 3,448 commercial and institutional establishments

conducted in the 1980's, as well as from surveys of manufacturers by the U.S. Census Bureau

and the California Department of Water Resources (Dziegielewski and Boland, 1989).

Employment estimates of CII activity can be used for a top-down estimate of water use, but in

order to evaluate the water use patterns of individual sectors, a bottom-up method is needed. This

thesis presents a bottom-up methodology to estimate CII water use based on parcel-level land use

and water billing databases.

For projecting water use of future customers, utilities have historically relied on similar

customers within their service area, or on water use coefficients developed through nationwide









studies. Typically, these water use coefficients use number of employees as the measure of size.

However, it is difficult to get this information, especially for individual parcels. Identifying

similar customers can be a tedious process, and water use coefficients are often difficult to apply,

suffering from lack of standardization of customer classifications and water use normalizations.

To overcome these challenges, this thesis presents a methodology by which to estimate public

supply water use for the CII sectors through publicly available databases from the Florida

Department of Revenue (FDOR) and Florida County Property Appraisers (FCPA) that provide

parcel-level information for 55 CII sectors. The FDOR data are available in a standard format

including geospatial information for every parcel in Florida. The FCPA data vary from county to

county but it is straightforward to link the FDOR and FCPA databases. Parcel-level water use

data was obtained from Hillsborough County Water Resources Services (HCWRS) and

Gainesville Regional Utilities (GRU), and this monthly billing data was linked with the FDOR

and FCPA databases.

The FDOR/FCPA databases provide a standardized classification of CII customers.

Relevant parcel attributes include: building and parcel areas, year built, and spatial location via

GIS. Building area is a good normalization parameter for CII customers. Year built allows for

the incorporation of historical growth patterns and can be used along with other attributes to

further disaggregate customer classifications. By linking FDOR/FCPA databases to water billing

data, water use coefficients for CII customers on public water supply were developed for base,

average and peak flow conditions with standardized FDOR customer classifications. This

methodology greatly improves a utility's ability to understand the nature of how water is used in

their service area when geocoded billing records are not available.









A review of previous and historically prominent CII water use models is presented in

Chapter 2. The land use databases, allowing for a bottom-up approach in estimating measures of

size are described in Chapter 3, along with a rigorous analysis of the sample data and State totals

of CII parcels. Chapter 4 presents the water billing databases employed in this thesis, justifying

the use of point estimates of water use and evaluating variables in predicting CII water use. The

methodology used to develop water use coefficients is then detailed in Chapter 5, which also

describes the application of the developed water use coefficients to EZ Guide 2. Conclusions and

the need for future work are presented in Chapter 6.









CHAPTER 2
LITERATURE REVIEW

Water use models typically forecast water use for supply planning purposes. Estimates to

forecast water use include having a rate of water use for a sector and a measure of its size

throughout the planning period. The rate of water use, or activity water use coefficient, is the

total water use by all customers within that sector normalized by a measure of its size. Some

models may provide default coefficients for sectors of varying levels of customer disaggregation

or require the user to develop their own. Total water use over n sectors is calculated using

Equation 2-1.

n
Q=Z(ak *xk) (2-1)
k-1

Where: Q = water use for n sectors
ak = water use coefficient of sector k
Xk = size of sector k
n = number of sectors

Previous Commercial, Industrial, and Institutional Water Use Models

The historically predominant water use model was the Institute of Water Resources

Municipal and Industrial Needs model (IWR-MAIN). It was the first model to estimate

commercial, industrial, and institutional (CII) water use empirically and disaggregate the general

sector into more distinct categories. IWR-MAIN was created by Hittman Associates, Inc. in

1969, developed under the Institute of Water Resources of the U.S. Army Corps of Engineers

and further refined by Planning and Management Consulting, Ltd (PMCL, now a subsidiary of

CDM). IWR-MAIN was a public domain model and transitioned into a proprietary model after

IWR financial support ended in the 1980s. The original IWR-MAIN is no longer available and

water use estimates are made using spreadsheets that replicate many of the features in the

original model. In IWR-MAIN, the size of each CII sector is estimated by total employment and









CII water use is estimated based on Standard Industrial Classifications (SIC) sectors as

developed by the Department of Commerce (Opitz et al. 1998).

IWR-MAIN Version 5.1 (1988) was used to estimate the demand for water and the

intensity of water use within a sector. The water use coefficients were determined by regression

analysis and the explanatory variables are the number of employees, the price of water and sewer

services, and the presence of conservation programs (Boland 1997). The coefficients in Version

5.1 were developed from the weighted average of water use rates found in a nationwide survey

of 3,448 commercial and institutional establishments, as well as from surveys of manufacturers

by the U.S. Census Bureau and the California Department of Water Resources (Dziegielewski

and Boland 1989). The survey to develop the Version 5.1 activity coefficients was large, random

and parcel-specific to produce stable national averages, but the model could not be adjusted to a

specific region's climate.

The latest release of the IWR-MAIN model was Version 6.1 in 1995. Version 6.1 had a

more sophisticated demand forecast procedure. The employment data used to estimate the water

use coefficients were based on surveys over a decade of over 7,000 CII establishments across the

United States and were developed by regression to account for the elasticity of CII water demand

to economic and climatic independent variables. Users could choose from a library of default

coefficients or input their own estimates (Opitz et al. 1998).

The IWR-MAIN model Version 6.1 had activity coefficients for the eight major industry

groups (n = 8 disaggregated CII sectors), all of the 65 two-digit SIC sectors (n = 65) and all of

the 417 three-digit SIC sectors (n = 417). Because the sample was collected over a period of

time, the forecasts of water use were able to provide daily-demands and summer and winter, or

annual periods (Baumann et al. 1998; Opitz et al. 1998). These coefficients were a major









improvement because they were developed from a larger cross-section of customers over a

longer period of time. The accuracy of the model was still limited by the quality of the input

data. If a user relied on the government Census for employment data, the sectoral and temporal

components of the model were restricted by the resolution of that data. If the user relied on

survey data, their forecast was restricted by the amount of historic data they could obtain.

The traditional IWR-MAIN model with default coefficients is no longer supported and the

coefficients are antiquated. In 1999, PMCL released the IWR-MAIN Water Demand

Management Suite. This spreadsheet model removed the default water use coefficients and

required the user to develop their own coefficients (CDM 2008). This software provided a

different approach to modeling where the user could tailor the model to their region of

implementation.

Following in the tradition of IWR-MAIN, Hazen and Sawyer and PMCL (2004) developed

a utility-wide model for Tampa Bay Water, a wholesale distributor. This model was used to

estimate single family residential, multi-family residential and non-residential water use for

seven different member government planning areas. The model has an equation to estimate the

non-residential water use coefficient based on historical usage, composition of the non-

residential sector, local affluence and climate. A commercial vendor provided the historical

employment and income data for the years 1999 to 2002 by survey. The study included 39,727

non-residential parcels and linked the parcel data to their billing records. The values were

averaged per Traffic Analysis Zone (TAZ) and combined with rainfall data to run a regression

and develop the monthly water use coefficients (Hazen & Sawyer and PMCL 2004). Total

employment is the size of the non-residential sector used to estimate water use.









Tampa Bay Water found that its non-residential model explained only two percent of the

variation in water use (Hazen & Sawyer and PMCL 2004). The modelers attribute this low

explanatory power to the typically heterogeneous nature of non-residential water use. If more

specific customer classifications were developed for the non-residential sector, then each group

of customers could be more homogeneous in their application of water. The coefficients also

came from two years of severe drought (SWFWMD 2006) and likely do not describe average

water use during a normal year.

Similar to the 1999 IWR-MAIN release, the Water Evaluation and Planning model also

requires the user to develop the water use coefficients. This model, developed by the Stockholm

Environmental Institute (2009), is a water forecasting tool for utility supply planning and

demand management. This model utilizes a system of modules that characterize the physical

water demand-supply network and estimate water demands for various water uses as defined in

the study. The activities are standardized per production output for CII and sectors determined by

the quality of the input data. A notable feature of this demand management program is that a

hierarchic branching data infrastructure is used to manage water use data by the sector,

subsector, end-use and device. The model incorporates demand estimates into the distribution

program to simulate supply allocations on a spatial and temporal basis to provide a sophisticated

planning tool (Wurbs 1995). This model allows for greater flexibility and accuracy in water

supply planning but has an extensive requirement of user inputs. Such data requirements also

increase the amount of time and resources required to run such a model.

The models of the IWR-MAIN tradition are tools for forecasting water use. These models

develop their water use projections from a regression analysis of explanatory variables and have

been developed differently over time. Table 2-1 highlights their progression. As forecasting









tools, these models do little to evaluate present water use or indicate conservation potential.

Tampa Bay Water has a rich spatial and temporal dataset for water use by each parcel, and a

more sophisticated CII model could have been developed if the water use coefficients were

standardized by a measure of size that can be related more directly to water use processes.

Maddaus and Maddaus (2004) have pioneered the end-use approach with a Least Cost Planning

Demand Management Decision Support System model. This proprietary forecasting model,

which relies on employment data for the measure of CII water use and can only disaggregate into

C, I and I, calibrates total water use with an estimate of fixtures in a building based on its age,

the frequency of each end-use by an employee, and the fixture's natural replacement rate.

EPA's white paper titled Water Efficiency in the Commercial and Institutional Sector

(2009) summarizes many of the studies conducted on CII water use. Citing studies like the

American Water Works Association Research Foundation Commercial/Institutional End Uses of

Water (Dziegielewski et al. 2000), and the Pacific Institute's Waste Not, Want Not: The Potential

for Urban Water Conservation in California (2003) for documentation on baseline water use by

CII subsectors and end use breakdown. Other studies are cited in addition, such as East Bay

Municipal Utility District's Watersmart Guidebook (2008) and the North Carolina Department of

Environment and Natural Resources' Water Efficiency Manual for CIIFacilities (2009) which

provide water-saving measures and technologies that are applicable to the different CII sectors.

In addition to providing an extensive literature review EPA's Water Efficiency in the

Commercial and Institutional Sector (2009), also documents national and international CII water

efficiency programs. This white paper also outlines the information gaps currently present in

evaluating CII for water conservation. The paper cites a lack of subsector specific data, such as

water usage by facility and end use, and existing benchmarks by which to set targets. Though









other studies (Dziegielewski et al. 2000, Colorado WaterWise 2007), have provided such

subsector data, they have only done so for a limited number of subsectors.

The American Water Works Association Research Foundation Commercial/Institutional

End Uses of Water Study (Dziegielewski et al. 2000) carried out a detailed analysis on five CII

subsectors of interest: schools, hotels/motels, office buildings, restaurants, and food stores. This

study collected survey data from utilities across the southwestern United States as well as direct

monitoring water use for 24 selected commercial and institutional facilities in the five user

categories. The study developed water use coefficient for these subsectors using a variety of

measures of size. Of all measures of size investigated, building area was the sole significant

predictor of water use across all subsectors. This study through submetering, also disaggregated

total subsectoral water use down to the general end uses (e.g., domestic, cooling, landscaping), as

well as presented benchmarks for efficiency based on the 25th percentile of customers sampled.

The need for a standardized classification system for CII water users, along with the expansion

of benchmarking to other subsectors was also addressed. Similarly, Colorado WaterWise (2007)

also developed water use coefficients using various measures of size for four CII subsectors,

including restaurants, schools, hotels/motels, and homes for the aged. Presented in this study was

the need for standardized databases containing relevant measures of size, a classification system

for CII water users, and a national clearinghouse of water utility data.

Employment Data as a Measure of Size

Many of the models reviewed standardize CII water use by the total number of employees

in order to compare the efficiency and variance of different sectors. Other variables have been

investigated. Mercer and Morgan (1974) developed water use coefficients based on number of

employees and this study cites that employee data has historically been readily available

compared to other parameters such as acreage. Kim and McCuen (1979) studied retail stores and









the results showed that the best two predictors of water use were gross area and sales area, the

only two measures of area analyzed, followed by average number of daily man-hours and

employees. As described previously, Dziegielewski et al. (2000) investigated five different

commercial and institutional water users and only building area was found to be a significant

indicator of water use across all customer categories.

Models of the past primarily depended on the user to input local employment data, which

is available from the U.S. Census or from private surveys (Dziegielewski et al. 2000). U.S.

Census employment data are available through three avenues: the Economic Census, County

Business Patterns, and Longitudinal Employer-Household Dynamics. The U.S. Economic

Census (2010) is conducted every five years and the employment data is aggregated to

geographical areas. These areas include state, county, metropolitan area, and city. Statistical data

on employment is provided per North American Industry Classification System code (which

recently replaced the SIC classification system). The size, density and composition of parcels

within these geographical areas vary widely and the precision of total employment estimates for

each sector by U.S. Economic Census is limited because of this aggregation.

County Business Patterns (US Census 2009a) provides annual U.S. Census employment

estimates at the county, zip code and metropolitan level with a two year lag. This data is treated

differently depending on the type of business establishment. Single-unit firm data, where the

firm owns and operates a single establishment, is gathered from a variety of administrative

record and survey sources when available. Multi-unit firm data, where a firm owns and operates

multiple establishments, is obtained through the U.S. Economic Census, and the annual

Company Organization Survey, which only survey companies with 250 employees or more

unless administrative records specify otherwise. This method of collecting employment data is









subject to nonsampling errors such as inability to identify all businesses, definition difficulties,

and estimation of missing or misreported data. Employment data from the U.S. Census County

Business Patterns is also limited in their aggregation of customers to ensure geographic (county,

zip code, or metropolitan area) and classification anonymity. Classification of customers is

provided through the North American Industry Classification System, but employee estimates

are largely presented in bins to protect the identity of individual establishments.

The Longitudinal Employer-Household Dynamics program is a new state/federal

partnership between the U.S. Census Bureau and ten states (CA, FL, IL, MD, MN, NC, NJ, OR,

PA, and TX). Modern statistical and computing techniques are used to combine federal and state

administrative data on employers and employees with core U.S. Census Bureau censuses and

surveys while protecting the confidentiality of people and firms that provide the data. The

Longitudinal Employer-Household Dynamics program provides quarterly employment estimates

at even smaller geographical areas than the two previous sources, including: tract, block group,

zip code, and TAZ. Employment estimates are once again provided per 2-digit North American

Industry Classification System code, and the data is often presented in bins to ensure anonymity.

The program also provides annual employment figures given that the quarterly estimates are still

viewed as experimental (US Census 2009b). Outside of the U.S. Census, employment figures

can be derived from commercial surveys which are more thorough and precise because data is

collected at the customer level. The accuracy of such surveys however depends on the diligence

of the respondent and this data must be purchased. Sources of employment data are profiled in

Table 2-2.

In order to create a bottom-up approach, a model needs a more accurate, frequent and

robust database that disaggregates the CII sector into relatively homogeneous subsectors.









Employment data are primarily averaged per geographical area and available periodically.

Because this metric is so widely used and a better substitute has not been developed, no CII

water forecasting model is appropriate for evaluating conservation options. A new approach is

needed. The parcel-level land-use databases utilized in the proposed methodology to arrive at a

measure of size are described in the following chapter.









Table 2-1. Comparison of models in the Institute of Water Resources Municipal and Industrial
Needs (IWR-MAIN) model tradition (Dziegielewski and Boland 1989; Opitz et al.
1998; Hazen & Sawyer and PMCL 2004)
Item Version 5.1 Version 6.1 Tampa Bay Water


Measure of sector
size
Data used to
develop water
use coefficients
Number of survey
points

Level of data
aggregation
Time span of
historic data


Employment


National survey


3,448


Employment


National survey


7,000


Census block


1 year


Census block


10 years


Employment


Regional survey


39,727


TAZ


3 years


Explanatory
variables






Default CII
coefficients


Number of
employees,
price of water
and sewer
services, and
presence of
conservation
programs

23 commercial
and
institutional,
and 198
industrial


Number of employees,
marginal price of
water, average
productivity of
labor, and number
of cooling degree
days

Aggregated: 8 major
industrial groups,
disaggregated: 65
two-digit SIC, most
disaggregated: 417
three-digit SIC


Total number of
employees,
residential income,
total rainfall, and
fraction of
employment in
commercial,
institutional,
industrial sectors



1 non-residential


Table 2-2. Sources of employment data
Source Available Smallest geographical unit

U.S. Economic Census Every 5 years City

County Business Patterns Annually Zip code

Longitudinal Employer-Household Dynamics Quarterly TAZ

Commercial surveys Varies Customer









CHAPTER 3
BOTTOM-UP APPROACH FOR EVALUATING THE SIZE OF EACH ACTIVITY

Introduction

Given the limitations of past models, including access to reliable data, a new methodology

to estimate commercial, industrial, and institutional (CII) water use based on parcel-level land

use and water billing databases is presented in this thesis. This chapter describes heated building

area, the measure of size proposed for this methodology. The Florida Department of Revenue

(FDOR) database, in conjunction with Florida County Property Appraisers (FCPA), provides the

heated building areas for every parcel in the State along with their land use classification,

allowing for sector specific water use coefficients. The land-use databases used, along with their

attributes of interest, are presented in Table 3-1.

Land Use Databases

Florida Department of Revenue

FDOR maintains a database of legal, physical and economic property-based information

for each of the 9 million parcels of land in the state of Florida. Of this total number, 326,000 are

CII parcels (215,000 commercial, 69,000 industrial, and 42,000 institutional).This database is

publicly available free of charge from the FDOR FTP website (ftp://sdrftp03.dor.state.fl.us/), and

is audited and updated annually. FDOR partitions parcels based on their land use into 100 sectors

using two-digit FDOR codes. These codes are standardized across the State, providing consistent

definitions of terms. The parcel information in this database is provided annually by the State's

67 FCPAs to FDOR for a statewide land-use database. The FDOR has also joined forces with

other state agencies and water management districts to capture and share data for a more

thorough database (FDOR 2009).









The following attributes of interest are provided by the FDOR database:

Parcel ID number

Land use code

Effective year built

Effective building area

Parcel area

The parcel ID number is a unique identifier to a plot of land, and serves as the link between

the various databases presented in this methodology. The coefficients presented in this thesis are

at the parcel level, requiring data to be adjusted to this level of aggregation.

The FDOR land use code is a two-digit classification system that identifies the primary use

of the land by its economic activity. CII sectors in this study are determined using FDOR land

use codes. The FDOR land use classification system allows for various degrees of disaggregation

following the hierarchical structure presented in Figure 3-1. In broadest terms, urban land use

can be broken up into residential and CII sectors based on groups of FDOR codes.

Disaggregation is possible by allocating the residential codes into either single family or multi-

family residential, and CII codes into commercial, industrial, and institutional. The greatest level

of disaggregation available from the FDOR database comes from the two-digit land use code.

Effective year built is defined as the effective or actual year built of major improvements

for a building. The year built provides valuable time series information to estimate trends, and is

an essential tool in forecasting number of accounts, building and parcel characteristics, and water

use rates.

The effective or adjusted building area field, defined as the total effective area of all floors

of all buildings on a given parcel, is not a true area, but rather a calculated field. Effective area

incorporates economic factors to weight the various building area types found within a parcel









differently. To calculate effective area, a parcel is first divided into primary and secondary

subareas. Then, subareas within a parcel are adjusted by a reference standard cost per square foot

of construction. Primary areas are the interior finished living areas, and by definition have an

Effective Area Factor of 1. Secondary areas are defined as an area that has a cost per square foot

of construction different from the living area reference standard. All secondary areas have an

Effective Area Factor greater or less than 1 (Oliver 2009). The effective area calculation is

presented in Equation 3-1. By definition, the effective area of a parcel will always be greater

than or equal to its heated building area, since other non-heated areas are associated with

effective area. Through the use of FDOR alone, the relative importance of sectors can be

quantified by simply summing total effective areas of parcels within sectors.


EA = Pa, + (Sai Eaf) (3-1)
i=1 i=1

Where:

Pai = primary areas
Sai = secondary areas

Sa, Cost ,2
Eafi = Effective Area Factor (of secondary area "i") -
Pa, 1Cost f


Parcel area is a derived field from the FDOR database. Though FDOR provides a parcel

area field, [LNDSQFOOT], this field is seldom populated. The FDOR database however, also

provides polygon shapefiles delineating every parcel in the State. Using standard GIS tools, the

area of each parcel can thus calculated, and joined to the other parcel information provided in the

FDOR attribute data. Besides, parcel dimensions, these polygon shapefiles also offer the spatial

location of every parcel in the State. This allows simple spatial queries to determine which

parcels are within the service boundaries of a given utility. South Florida Water Management









District (WMD), St. Johns River WMD, and Southwest Florida WMD provide the water service

area boundaries of utilities in their districts as polygon shapefiles available in their respective

websites to be viewed in GIS. The parcels are identified by a unique parcel identification number

which can be related to the FDOR database to find the attributes for the parcels in the utility

being analyzed.

FDOR serves as the foundation for a Florida urban water database allowing for both spatial

and attribute joins, and providing consistent definition of terms. Other public water supply

modeling parameters such as: population data from U.S. Census, utility service boundary

information, utility flow data from the Florida Department of Environmental Protection, and

water billing records from select utilities can be joined as appropriate. Figure 3-2 presents a

schematic of how databases, along with their attributes of interest, can be related to FDOR. The

parcel-level data from FDOR is powerful, allowing analysis across spatial scales: macro (state,

water management district, or county), meso (city or utility), micro (parcel), and nano (end use

such as toilets) as shown in Figure 3-3.

Florida County Property Appraisers

Each of the 67 counties in Florida maintains a FCPA database that contains the same

information as the FDOR database, along with additional attributes that vary from county to

county. Attributes of interest in all FCPAs are:

Parcel ID number

Heated building area

Parcel ID number is a unique identifier to a parcel, and serves as the link between FCPA

and FDOR. FCPA provides the heated areas of buildings in a parcel, defined as all building area

under climate control. Unlike effective building area, provided by FDOR, heated area is a

physical building area. The relationship between heated area and effective area is described later









in this chapter. The two FCPA databases analyzed in our study are: Hillsborough County

Property Appraiser (HCPA) and Alachua County Property Appraiser (ACPA). These FCPA

databases encompass the two utilities with available water billing data for this thesis: HCWRS

(Hillsborough County) and GRU (Alachua County). These water-billing databases are described

in following chapter.

Hillsborough County Property Appraiser

Hillsborough County is located in west-central Florida, and is the fourth most populous

county in the State (Census 2000). HCPA data includes all the FDOR parcel-attributes for the

county, along with other attributes of interest including:

Parcel ID number

Heated building area

HCPA land use code

Each local FCPA may have the same two-digit land use code as FDOR, or subdivide the land

use further with a four-digit code. In the case of the four-digit code, the first two digits are

consistent with the original FDOR classification system and the second two break the sector

down into further land use detail. Four-digit FCPA land use codes allow for greater

disaggregation of customers, but since these codes are neither required nor standardized across

the State, the methodology presented in this thesis solely addresses two-digit FDOR codes.









Alachua County Property Appraiser

Alachua County is located in north-central Florida, and is the 20th most populous county in

the State. ACPA provides considerably more property attributes than FDOR, including:

Parcel ID number

Heated building area

ACPA land use code

Information on impervious areas

Presence of in-ground pool

Presence of in-ground irrigation system

Presence of well

Number of bedrooms

Number of baths

Number of stories

As described in the previous section, four-digit FCPA land use codes vary from county to

county. The ACPA and HCPA four-digit land use classifications are dissimilar. Additional

attributes from ACPA include those labeled as, "miscellaneous areas." These areas present the

square footage of impervious areas on a parcel, such as parking lots, patios, sheds, and pools.

Summing a parcel's associated impervious areas, labeled as miscellaneous areas, along with the

footprint of all buildings on a parcel provides an estimate of the impervious area for that parcel.

By subtracting the impervious area estimate from the total parcel area, one arrives at an estimate

of the pervious or irrigable area on a parcel (Equation 3-2).











PA = TA FS AIA


Where:

PA = parcel pervious area
TA = total parcel area
FS = footprint of all buildings on parcel
AIA = associated impervious areas on parcel

Within miscellaneous areas, ACPA also tags other property attributes of interest in public

water supply evaluations, including presence of in-ground sprinkler systems and pools, and

wells. Additional attributes of interest provided for all parcels in Alachua County include the

number of bedrooms, baths, and stories. For CII parcels, number of baths denotes the count of

flushable toilets and urinals in all structures on a parcel, the reported accuracy of which has not

been investigated. Number of stories is critical in calculating the footprint of buildings on a

parcel. A simple estimate of footprint is calculated by dividing HA of all buildings on a parcel by

their respective number of stories (Equation 3-3).

n HAi
FS=i (3-3)
i=1 Ni

Where:

FS = footprint of all buildings on parcel
HAi = total heated area in building i
Ni = number of stories in building i

Relationship of Effective Area to Heated Area

Effective building area, provided by FDOR, is not a physical area, but rather a calculated

value incorporating market values of the structures within a parcel. For this reason, water use

coefficients were developed using heated building area, a physical area not prone to

misinterpretation and available from FCPA. Application of water use coefficients directly with


(3-2)









FDOR requires coefficients to convert the effective building area to heated building area. A

simple linear regression of Heated Area (HA) as a function of Effective Area (EA) based on

sample of 3,205 CII parcels in Hillsborough and Alachua County, Florida is shown in Figure 3-4.

The resulting Equation 3-4 was forced through the origin:

HA = 0.9526*EA R2 = 0.9921 (3-4)

The very high R2 indicates that the fit is excellent. According to the regression equation,

HA is about 95.26% of EA. A slightly different result (Equation 3-5) is obtained if the HA/EA

ratio is defined as follows:

HA/EA= HA/ EA =0.9392 (3-5)

Both methods of estimating the relationship between HA and EA assume linearity between

the two variables through the origin. For all 3,205 CII parcels analyzed in Florida, as shown in

Figure 3-4, linearity through the origin is a safe assumption to make. The regression method

presented (Equation 3-4), works best when the variance is homogeneous. The method described

in Equation 3-5, also known as the ratio of the means, is most efficient if the variance of the

dependent variable (HA) is proportional to each value of the independent variable (EA) (Swank

and Schreude 1974). The residual plot for the simple linear regression of heated and effective

area for 3,205 CII parcels in Hillsborough and Alachua County (Figure 3-5) indicates that

variance across EAs is not homogeneous, hence the ratio of the means method appears to be

most appropriate for estimating the relationship between EA and HA.

Given the strong correlation between EA and HA across all CII sectors (Figure 3-4), it is

likely that similarly strong relationships are presented at the individual 2-digit FDOR subsector

level. Hence, EA to HA conversion coefficients using the ratio of the means approach were

developed at the 2-digit FDOR subsector level. These area conversion coefficients allow for the









application of water use coefficients, normalized by HA, to the EA measures available from

FDOR.

Analysis of Available Data

HCWRS, located in Hillsborough County, provided utility billing data for 1,768 CII

accounts (67% commercial, 9% industrial, and 24% institutional) for 48 months beginning in

January 2003. GRU, located in Alachua County, supplied two complete years of monthly water

billing from January 2008 to December 2009, for 1,437 CII parcels (72% commercial, 10%

industrial, and 18% institutional). The subsector heated area statistics of these two utilities, along

with their sample size and other property attributes from FDOR and FCPA are addressed in this

section. The subsector characteristics are compared across utilities, with the combined utility

data set, and with subsector data for the entire state of Florida from FDOR.

Sample size

The combined utility dataset from HCWRS and GRU, provide samples for 24 of the 28

FDOR commercial subsectors (Table 3-2). The available subsector samples from the combined

utility datasets account for between 0.2% and 4.8% of the State total of commercial parcels.

Overall, the combined utility dataset provides a 1% sample of commercial parcels in the state of

Florida. The available sampled subsectors account for over 97% of commercial parcels in the

State.

The available utility data for the industrial (Table 3-3) and institutional (Table 3-4)

subsectors provides samples between 0.2% and 3.7% of total State parcels at the subsector level.

Data is available for eight of the 11 industrial FDOR subsectors, and 14 of the 16 institutional

subsectors. The sample data accounts for 0.4% and 1.6% of all industrial and institutional parcels

in the State. The sampled subsectors account for nearly 99% of industrial and 97% of

institutional parcels in the State.









Effective Year Built

The average effective year built of major improvements on a parcel for the available

commercial subsectors in HCWRS, GRU, and combined utilities, as well as in the entire state of

Florida, are presented in Table 3-5. Similar tables are also available for the industrial (Table 3-

6) and institutional (Table 3-7) sectors. These tables show that the average effective year built

across the majority of subsectors varies little between the sample dataset and the State totals of

CII parcels.

Heated Area to Effective Area Calculation

The heated to effective area ratio of the means calculation at the subsector level is shown

for the commercial, industrial, and institutional sectors in Table 3-8, Table 3-9, and Table 3-10,

respectively. The calculation, presented for the sample of HCWRS and GRU CII subsectors, as

well as for the combined dataset, involves dividing the average heated area of each subsector

provided by FCPA by its average effective heated area from FDOR. The difference between the

heated area to effective area ratios from HCWRS and GRU is presented for the commercial,

industrial, and institutional sectors in Table 3-11, Table 3-12, Table 3-13, respectively. This

percent difference varies from -16% to 20% depending on the subsector being analyzed, and is

due in large part to the available sample sizes.

Even though the ratio of the means calculation was chosen over a linear regression

approach, also shown in these tables is the associated R2 for each of the available subsectors,

presented to show the generally strong relationship between heated area and effective area across

CII subsectors. For the most part, R2 values presented approach 1. Comparisons of the subsector-

heated area to effective area ratios indicate negligible differences across the two utilities. This

fact solidifies our confidence in using these area conversion factors to apply the water use

coefficients developed in this thesis with the statewide data available from FDOR. For the CII









sectors, heated area to effective area relationships likely do not significantly vary across the

State.

Distribution and Representativeness of Sampled Heated Areas

HCWRS, GRU, combined utilities and State comparisons of average heated area at the

subsector level are presented for the commercial, industrial, and institutional sectors in Table 3-

14, Table 3-15, and Table 3-16, respectively. The state of Florida subsector information is

available from FDOR, which only provides the effective area of each parcel. However, using the

heated to effective area (HA/EA) conversion ratios calculated for each subsector, estimates of

heated area can be derived for the entire State. The use of these coefficients is reasonable

following the strong relationship between heated and effective area across CII subsectors

described in the previous section. For the subsectors where no sample data was available to

calculate the area conversion ratio, the average HA/EA of the available subsectors from the

given sector was utilized.

The percent difference of subsectoral average heated area between the combined utility

sample from HCWRS and GRU and the entire state of Florida is also presented in Table 3-14,

Table 3-15, and Table 3-16. Following the heterogeneous nature of CII customers, these tables

demonstrate the relatively large variability (from -80% to 533%) in average heated area within

subsectors. The limited number of 2-digit CII FDOR codes ensures that multiple facility types

with differing size characteristics and drivers of water use are grouped within each code.

Disaggregated groupings within 2-digit FDOR codes can be achieved by developing size

categories based on heated building area, or age built of a facility which might also affect water

use, given the requirement or availability of certain end-use devices at the time of construction.

For example, the residential sector is broken up into three age groups (pre-1983, 1983-1994,

1995-present) corresponding with State regulations requiring minimum plumbing fixture water









efficiencies. Even facilities within the same subsector might offer new services requiring

different end-use water devices as they respond to changing conditions. Predicting what fixture

types are prevalent in certain customer groups greatly improves estimates of water use, as well as

facilitates the weighing of water conservation options. Unfortunately, sample size limitations do

not allow this analysis be carried out.

A preliminary assessment of the representativeness of the available combined utility

sampled dataset as it pertains to the State totals of CII parcels is provided by Table 3-14, Table

3-15, and Table 3-16. These tables utilize the average heated area statistic to make this

inference. For greater insight into both the combined utility dataset and the State totals of CII

parcels, it is best to look at the distribution of heated areas across all parcels in a given subsector.

Histograms detailing both the combined utility and State heated area distributions of each CII

subsector are presented in the appendix. By comparing the histogram of the sample combined

utility dataset, to that of the State, one can gain a measure of how representative the sample

dataset is to the State totals of CII parcels at the subsector level. Though the sample dataset

might not be representative of the entire subsector State distribution, it might be representative of

a segment within that State distribution.

The fitted lognormal probability density equations for heated area subsector distributions

of both the combined utility sample and State total are presented in Table 3-17, Table 3-18, and

Table 3-19, for the commercial, industrial, and institutional sectors, respectively. The parameters

are estimated using @Risk (Palisade Corporation 2010). The log normal probability density

function is shown in Equation 3-6.









I1 1 In(x) )2
f(x)= xe 2 (3-6)



Where:

t = the location parameter or log mean
c = the scale parameter or log standard deviation


The log-normal distribution was chosen because it has a defined lower bound (typically

0) and fits the data well. The corresponding shifted lognormal equations are shown in Table 3-

20, Table 3-21, and Table 3-22. In these tables, along with sample size, the Anderson-Darling

and Kolmogorov-Smirnov statistics are also presented to provide a measure of the goodness-of-

fit of the data to the lognormal distribution. From these statistics, and the histograms presented in

the appendix, it is clear that the representativeness of the combined utility sample varies across

subsectors. This discrepancy is largely attributable to sample size, and the homogeneity of a

given subsector throughout the State. A general trend is apparent in that as sample size increases,

the State and sample distributions converge. Thus, the parcel-level heated area statistics

presented in this thesis would improve with increased sample sizes.

The following chapter will present the sampled water use data from HCWRS and GRU.

The chapter will explain their database structure, preparation required for analysis, as well as

present their time series information.









Table 3-1. Databases and parcel attributes used to develop water use and area conversion
coefficients based on a sample of 3,205 commercial, industrial, and institutional (CII)
parcels
Database Attributes of interest Period of record
Parcel ID number
Land use code
FDOR Effective year built 1920 2008
Effective building area
Parcel area
Parcel ID number
HCPA Heated building area 1920 2008
HCPA land use code
Parcel ID number
Heated building area
ACPA land use code
Information on impervious areas
ACPA Presence of in-ground pool 19202008
Presence of in-ground irrigation system
Presence of well
Number of bedrooms
Number of baths
Number of stories











Table 3-2. Parcel breakdown across commercial subsectors for Hillsborough County Water
Resources Services, Gainesville Regional Utilities, the two utilities combined, and the
entire state of Florida
SP c Parcel subsector Percent of
FDOR Parcel count Sa
code Description breakdown State
codeHCWRS GRU Combined State Combined State Combined
HCWRS GRU Combined State Combined State Combined


11 Stores, one-story
12 Mixed use
13 Department stores

14 Supermarkets /
conv. stores
15 Regional malls
Community
16 shopping
centers
17 Office, one-story
18 Office, multi-story
19 Medical office
20 Transit terminals
21 Restaurants
Fast-food
22
restaurants
Financial
23
institutions
24 Insurance company
offices
25 Service shops
26 Service stations
27 Auto sales / repair
29 Wholesale outlets
30 Florist / greenhouses
Drive-in theaters /
31
open stadiums
Enclosed theaters /
32
auditoriums
33 Nightclubs / bars
3 Bowling alleys /
skating rinks
35 Tourist attractions
36 Camps
37 Race tracks
Golf courses /
38
driving ranges
39 Hotels / motels


39,002
18,931
829


117 6

2 1


123 2,571

3 449


161 78 239 7,726


55 50

63 35

11

18 31
5
116 58
5
2



1 2


384 37,629
73 16,033
264 20,953
6 2,825
120 7,629

105 4,377

98 4,732

11 283


4,309
3,434
15,129
729
352


3 242


1,914

492

742
330
129


20 6


26 1,392


13.1%
6.5%
0.9%


18.1%
8.8%
0.4%


5.6% 1.2%

0.1% 0.2%

10.8% 3.6%


17.3%
3.3%
11.9%
0.3%
5.4%


0.5%

2.2%
0.2%
7.9%
0.2%
0.1%



0.1%

0.9%

0.1%


17.5%
7.5%
9.8%
1.3%
3.6%


1.2% 0.6%


10 40 50 21.702 2.3% 10.1% 0.2%


Total commercial 1,177 1,037 2,214 214,893 100.0% 100.0% 1.0%


4.7% 2.0%

4.4% 2.2%


0.7%
0.8%
2.3%

4.8%

0.7%

3.1%

1.0%
0.5%
1.3%
0.2%
1.6%

2.4%

2.1%



1.1%


1.2%


1.2%

1.0%

0.6%





1.9%


0.1%

2.0%
1.6%
7.0%
0.3%
0.2%

0.0%

0.1%

0.9%

0.2%

0.3%
0.2%
0.1%










Table 3-3. Parcel breakdown across industrial subsectors for Hillsborough County Water
Resources Services, Gainesville Regional Utilities, the two utilities combined, and the
entire state of Florida
FDOR Parcel count Parcel subsector Percent of
FDOR Parcel count
code Description breakdown State
code
HCWRS GRU Combined State Combined State Combined
41 Light manufacturing 10 23 33 18,398 10.7% 26.7% 0.2%
42 Heavy industrial 2 1 3 677 1.0% 1.0% 0.4%
43 Lumberyards 1 1 471 0.3% 0.7% 0.2%
44 Packing plants 526 0.8%
45 Bottler / canneries 1 1 90 0.3% 0.1% 1.1%
46 Food processing 295 0.4%
47 Mineral processing 8 2 10 598 3.3% 0.9% 1.7%
Warehousing /
48 W treusiong 118 110 228 42,406 74.3% 61.5% 0.5%
distribution
49 Open storage 17 2 19 2,190 6.2% 3.2% 0.9%
91 Utility, gas & elec. 8 4 12 3,185 3.9% 4.6% 0.4%
92 Mining / petroleum 112 0.2%
Total industrial 163 144 307 68,948 100.0% 100.0% 0.4%










Table 3-4. Parcel breakdown across institutional subsectors for Hillsborough County Water
Resources Services, Gainesville Regional Utilities, the two utilities combined, and the
entire state of Florida
SP c Parcel subsector Percent of
FDOR Parcel count Sa
code Description breakdown State
code
HCWRS GRU Combined State Combined State Combined
71 Churches 221 116 337 18,888 49.3% 45.1% 1.8%
Private Schools &
72 e oos 65 46 111 3,243 16.2% 7.7% 3.4%
Colleges
73 Private Hospitals 2 4 6 570 0.9% 1.4% 1.1%
74 Homes for the Aged 12 12 4,447 1.8% 10.6% 0.3%
75 Orphanages/Non- 67 15 82 2,214 12.0% 5.3% 3.7%
profits
Mortuaries /
76 rtues7 6 13 887 1.9% 2.1% 1.5%
Cemeteries
77 Clubs / Union Halls 15 40 55 3,244 8.0% 7.8% 1.7%
Sanitariums /
78 aitar1 1 457 0.1% 1.1% 0.2%
Convalescents
Cultural
79 cultural 1 1 234 0.1% 0.6% 0.4%
Organizations
81 Military 43 0.1%
Parks and
82 arksand 1 1 850 0.1% 2.0% 0.1%
Recreation
83 Public County 51 1 52 2,953 7.6% 7.1% 1.8%
Schools
84 Colleges 9 9 288 1.3% 0.7% 3.1%
85 Hospitals 3 3 173 0.4% 0.4% 1.7%
Gov. Leased
90 erests 1,276 3.0%
Interests
Outdoor
97 Recreati 1 1 2,090 0.1% 5.0% 0.0%
ToaRecreational 428 256 684
Total Institutional 428 256 684 41,857 100.0% 100.0% 1.6%









Table 3-5. Average effective year built across commercial subsectors for Hillsborough County
Water Resources Services, Gainesville Regional Utilities, the two utilities combined,
and the entire state of Florida

FDOR es Average effective year built
cde Description
code r--------------


HCH


Stores, one-story
Mixed use
Department stores
Supermarkets / conv. stores
Regional malls
Community shopping centers
Office, one-story
Office, multi-story
Medical office
Transit terminals
Restaurants
Fast-food restaurants
Financial institutions
Insurance company offices
Service shops
Service stations
Auto sales / repair
Wholesale outlets
Florist / greenhouses
Drive-in theaters / open stadiums
Enclosed theaters / auditoriums
Nightclubs / bars
Bowling alleys / skating rinks
Tourist attractions
Camps
Race tracks
Golf courses / driving ranges
Hotels / motels
Total commercial


WVRS GRU
1989 1982
1975 1978
1995 1982
1991 1984
1997 1995
1989 1985
1985 1984
1988 1986
1989 1991
1979 2000
1993 1981
1998 1989
1991 1992
1988
1984 1979
1986
1985 1983
1972
1966


1999
1981
1983




1991
1986
1988


2001
1966
1988




1986
1981
1984


Combined State


1985
1976
1994
1991
1996
1988
1984
1987
1990
1982
1988
1994
1992
1988
1981
1986
1984
1972
1966

2000
1972
1986




1990
1982
1986


1981
1971
1992
1986
1992
1990
1988
1990
1989
1995
1985
1991
1992
1984
1980
1983
1980
1977
1980
1975
1984
1974
1985
1988
1979
1985
1990
1989
1985









Table 3-6. Average effective year built across industrial subsectors for Hillsborough County
Water Resources Services, Gainesville Regional Utilities, the two utilities combined,
and the entire state of Florida
FDOR Dn Average effective year built
code Description
codeHCWRS GRU Combined State
41 Light manufacturing 1982 1979 1979 1986
42 Heavy industrial 1995 1998 1996 1979
43 Lumber yards 1990 1990 1980
44 Packing plants 1974
45 Bottler/ canneries 1969 1969 1974
46 Food processing 1976
47 Mineral processing 1972 1984 1974 1982
48 Warehousing/distribution 1990 1985 1988 1987
49 Open storage 1979 1999 1981 1989
91 Utility, gas & elec. 1974 1983 1977 1984
92 Mining / petroleum 1982
Total industrial 1987 1984 1986 1986


Table 3-7. Average effective year built across institutional subsectors for Hillsborough County
Water Resources Services, Gainesville Regional Utilities, the two utilities combined,
and the entire state of Florida
FDOR D Average effective year built
code DesctonHCWRS GRU Combined State
71 Churches 1957 1979 1965 1975
72 Private schools & colleges 1986 1981 1984 1980
73 Private hospitals 1997 1988 1991 1984
74 Homes for the Aged 1993 1993 1986
75 Orphanages/non-profits 1987 1985 1986 1985
76 Mortuaries / cemeteries 1989 1974 1982 1979
77 Clubs / union halls 1981 1935 1947 1978
78 Sanitariums / convalescents 1975 1975 1985
79 Cultural organizations 1989 1989 1980
81 Military 1972
82 Parks and Recreation 1998 1998 1982
83 Public county schools 1987 1980 1987 1980
84 Colleges 1970 1970 1980
85 Hospitals 1988 1988 1986
90 Gov. leased interests 1989
97 Outdoor recreational 1950 1950 1989
Total institutional 1971 1973 1972 1979











Table 3-8. Heated area to effective area ratio of the means calculation of commercial subsectors in Hillsborough County Water
Resources Services, Gainesville Regional Utilities, and the two utilities combined

FDOR Average heated area (sf) Average effective area (sf) HA/EA
code Description
codeHCWRS GRU Combined HCWRS GRU Combined HCWRS GRU Combined
HCWRS GRU Combined HCWRS GRU Combined HCWRS GRU Combined


Stores, one-story
Mixed use
Department stores
Supermarkets / conv. stores
Regional malls
Community shopping centers
Office, one-story
Office, multi-story
Medical office
Transit terminals
Restaurants
Fast-food restaurants
Financial institutions
Insurance company offices
Service shops
Service stations
Auto sales / repair
Wholesale outlets
Florist / greenhouses
Enclosed theaters / auditoriums
Nightclubs / bars
Bowling alleys / skating rinks
Golf courses / driving ranges
Hotels / motels


Total commercial


1


8


9,375 6,532
13,601 6,390
30,076 94,109
5,576 10,065
120,551 928,070
39,269 39,269
5,446 6,334
38,283 20,700
6,770 8,070
10,670 2,193
5,084 4,664
3,105 2,697
4,424 6,338
10,736
2,787 6,906
1,829
9,095 5,836
23,700
3,376
97,632 27,989
3,686 5,336
30,784 34,410
18,091 9,633
36,875 31,626
16,444 11,457


7,644
11,483
128,183
5,795
856,391
39,269
5,983
30,576
7,543
9,257
4,902
2,910
5,108
10,736
5,393
1,829
8,009
23,700
3,376
51,203
4,676
33,201
16,139
32,676
14,108


9,808 7,258
14,225 7,618
135,104 117,614
5,746 11,143
939,712 936,397
41,224 41,350
5,674 6,567
39,230 21,743
7,046 8,268
11,057 2,235
5,311 4,816
3,205 2,809
5,080 6,791
11,741
3,563 8,489
2,591
10,353 7,053
31,323
3,679
105,622 27,987
3,684 5,770
31,278 36,864
20,444 10,744
38,761 33,574
17,443 12,224


8,255
12,284
134,183
6,009
938,607
41,265
6,214
31,565
7,773
9,586
5,097
3,016
5,691
11,741
6,679
2,591
9,253
31,323
3,679
53,865
4,936
35,002
18,206
34,611
14,999


0.96 0.90
0.96 0.84
0.96 0.80
0.97 0.90
0.87 0.99
0.95 0.95
0.96 0.96
0.98 0.95
0.96 0.98
0.96 0.98
0.96 0.97
0.97 0.96
0.87 0.93
0.91
0.78 0.81
0.71
0.88 0.83
0.76
0.92
0.92 1.00
1.00 0.92
0.98 0.93
0.88 0.90
0.95 0.94
0.94 0.94


0.93
0.93
0.96
0.96
0.91
0.95
0.96
0.97
0.97
0.97
0.96
0.96
0.90
0.91
0.81
0.71
0.87
0.76
0.92
0.95
0.95
0.95
0.89
0.94
0.94










Table 3-9. Heated area to effective area ratio of the means calculation of industrial subsectors in Hillsborough County Water
Resources Services, Gainesville Regional Utilities, and the two utilities combined

FDOR Average heated area (sf) Average effective area (sf) HA/EA
code Description
code
HCWRS GRU Combined HCWRS GRU Combined HCWRS GRU Combined
41 Light manufacturing 80,659 20,851 38,975 89,552 23,130 43,258 0.90 0.90 0.90
42 Heavy industrial 44,432 41,893 43,586 45,382 58,506 49,756 0.98 0.72 0.88
43 Lumber yards 18,488 18,488 19,276 19,276 0.96 0.96
44 Packing plants
45 Bottler/canneries 34,541 34,541 43,783 43,783 0.79 0.79
46 Food processing
47 Mineral processing 102,201 23,001 86,361 104,177 24,106 88,163 0.98 0.95 0.98
48 Warehousing/distribution 41,608 16,966 29,719 43,010 18,929 31,392 0.97 0.90 0.95
49 Open storage 1,860 7,589 2,463 2,036 7,799 2,642 0.91 0.97 0.93
91 Utility, gas & elec. 15,714 50,505 27,311 16,425 56,926 29,926 0.96 0.89 0.91
92 Mining / petroleum
Total industrial 41,596 18,777 30,893 43,318 21,023 32,860 0.96 0.89 0.94










Heated area to effective area ratio of the means calculation of institutional subsectors in Hillsborough County Water


Resources Services, Gainesville Regional Utilities, and the two utilities combined

FDOR Average heated area (sf) Average effective area (sf) HA/EA
code Description
code
HCWRS GRU Combined HCWRS GRU Combined HCWRS GRU Combined
71 Churches 12,838 13,555 13,085 13,678 14,143 13,838 0.94 0.96 0.95
72 Private schools & colleges 7,309 9,149 8,071 7,720 9,658 8,524 0.95 0.95 0.95
73 Private hospitals 424,916 141,224 235,788 434,468 147,756 243,326 0.98 0.96 0.97
74 Homes forthe aged 116,675 116,675 126,513 126,513 0.92 0.92
75 Orphanages/non-profits 9,741 10,894 9,952 10,277 11,349 10,473 0.95 0.96 0.95
76 Mortuaries/ cemeteries 7,560 5,286 6,511 8,593 5,933 7,365 0.88 0.89 0.88
77 Clubs / union halls 10,654 13,308 12,584 12,846 13,460 13,292 0.83 0.99 0.95
78 Sanitariums / convalescents 43,505 43,505 43,644 43,644 1.00 1.00
79 Cultural organizations 2,302 2,302 3,125 3,125 0.74 0.74
81 Military
82 Parks and recreation 5,288 5,288 5,771 5,771 0.92 0.92
83 Public county schools 127,905 59,448 126,588 130,491 60,959 129,153 0.98 0.98 0.98
84 Colleges 175,786 175,786 175,937 175,937 1.00 1.00
85 Hospitals 123,175 123,175 128,633 128,633 0.96 0.96
90 Gov. leased interests
97 Outdoor recreational 9,233 9,233 10,021 10,021 0.92 0.92
Total institutional 26,988 26,395 26,766 28,014 27,465 27,809 0.96 0.96 0.96


Table 3-10.










Table 3-11. Comparison of heated area to effective area ratio of the means calculation for the
commercial subsectors in Hillsborough County Water Resources Services,
Gainesville Regional Utilities, and the two utilities combined
FDOR HA/EA Percent HA/EA 2
Code DescriHCWRS GRU difference Combined
11 Stores, one-story 0.96 0.90 6% 0.93 1.00
12 Mixed use 0.96 0.84 14% 0.93 0.99
13 Department stores 0.96 0.80 20% 0.96 0.98
14 Supermarkets/ conv. stores 0.97 0.90 7% 0.96 0.99
15 Regional malls 0.87 0.99 -12% 0.91 0.84


16 Community shopping
centers
17 Office, one-story
18 Office, multi-story
19 Medical office
20 Transit terminals
21 Restaurants
22 Fast-food restaurants
23 Financial institutions
24 Insurance company offices
25 Service shops
26 Service stations
27 Auto sales / repair
29 Wholesale outlets
30 Florist/ greenhouses
31 Drive-in theaters / open
stadiums
Enclosed theaters /
32
auditoriums
33 Nightclubs / bars
SBowling alleys / skating
34 .
rinks
35 Tourist attractions
36 Camps
37 Race tracks
38 Golf courses / driving ranges
39 Hotels / motels
Total commercial


0.95 0.95

0.96 0.96
0.98 0.95
0.96 0.98
0.96 0.98
0.96 0.97
0.97 0.96
0.87 0.93
0.91
0.78 0.81
0.71
0.88 0.83
0.76
0.92



0.92 1.00

1.00 0.92

0.98 0.93


0.88
0.95
0.94


0.90
0.94
0.94


0%

0%
3%
-2%
-2%
-1%
1%
-7%

-4%

6%






-8%

8%

5%





-1%
1%
1%


0.95

0.96
0.97
0.97
0.97
0.96
0.96
0.90
0.91
0.81
0.71
0.87
0.76
0.92



0.95

0.95

0.95





0.89
0.94
0.94


1.00

1.00
1.00
1.00
1.00
0.99
0.99
0.99
0.98
0.92
0.93
0.99
0.95
0.98



1.00

0.97

0.87





0.98
0.99
0.99










Table 3-12. Comparison of heated area to effective area ratio of the means calculation for the
industrial subsectors in Hillsborough County Water Resources Services,
Gainesville Regional Utilities, and the two utilities combined
FDOR HA/EA Percent HA/EA 2
code escponHCWRS GRU difference Combined
41 Light manufacturing 0.90 0.90 0% 0.90 1.00
42 Heavy industrial 0.98 0.72 37% 0.88 0.98
43 Lumber vards 0.96 0.96


44
45
46
47
48
49
91
92


Packing plants
Bottler / canneries
Food processing
Mineral processing
Warehousing / distribution
Open storage
Utility, gas & elec.
Mining / petroleum


0.98
0.97
0.91
0.96


0.79

0.95
0.90
0.97
0.89


3%
8%
-6%
8%


0.79

0.98
0.95
0.93
0.91


1.00
1.00
1.00
1.00


Total industrial 0.96 0.89 8% 0.94 1.00


Table 3-13. Comparison of heated area to effective area ratio of the means calculation for the
institutional subsectors in Hillsborough County Water Resources Services,
Gainesville Regional Utilities, and the two utilities combined
FDOR D n HA/EA Percent HA/EA 2
Description R
code HCWRS GRU difference Combined
71 Churches 0.94 0.96 -2% 0.95 0.99
72 Private schools & colleges 0.95 0.95 0% 0.95 0.99
73 Private hospitals 0.98 0.96 2% 0.97 1.00
74 Homes for the aged 0.92 0.92 1.00
75 Orphanages/non-profits 0.95 0.96 -1% 0.95 1.00
76 Mortuaries / cemeteries 0.88 0.89 -1% 0.88 0.99
77 Clubs / union halls 0.83 0.99 -16% 0.95 0.98
78 Sanitariums / convalescents 1.00 1.00
79 Cultural organizations 0.74 0.74
81 Military
82 Parks and recreation 0.92 0.92
83 Public county schools 0.98 0.98 1% 0.98 0.99
84 Colleges 1.00 1.00 1.00
85 Hospitals 0.96 0.96 1.00
90 Gov. leased interests
97 Outdoor recreational 0.92 0.92
Total institutional 0.96 0.96 0% 0.96 0.99











Table 3-14. Comparison of average heated area across the commercial subsectors of Hillsborough County Water Resources
Services, Gainesville Regional Utilities, the two utilities combined, and the state of Florida
Average Average Percent Total heated Percent of
Average Percent
heated area total
FDOR Average heated area (sf) effective HA/EA difference (million h
code Des n area (sf) area from State
(sf) sf) area
HCWRS GRU Combined State Combined State Combined State State


Stores, one-story
Mixed use
Department
stores
Supermarkets /
conv. stores
Regional malls
Community
shopping
centers
Office, one-story
Office, multi-
story
Medical office
Transit terminals
Restaurants
Fast-food
restaurants
Financial
institutions
Insurance
company
offices
Service shops
Service stations
Auto sales /
repair


9,375
13,601

130,076

5,576

820,551

39,269

5,446

38,283

6,770
10,670
5,084

3,105

4,424


6,532
6,390

94,109

10,065

928,070

39,269

6,334

20,700

8,070
2,193
4,664

2,697

6,338


10,736

2,787 6,906
1,829

9,095 5,836


0.93
0.93

0.96

0.96

0.91

0.95

0.96

0.97

0.97
0.97
0.96

0.96

0.90


7,644
11,483

128,183

5,795

856,391

39,269

5,983

30,576

7,543
9,257
4,902

2,910

5,108


10,736

5,393
1,829

8,009


7,063
4,658

106,078

12,456

183,859

38,295

4,749

20,363

5,740
5,378
4,526

3,209

7,139


7,627
4,983

111,043

12,915

201,510

40,241

4,933

21,022

5,915
5,569
4,705

3,326

7,954


10,547

6,865
3,617

7,955


0.2%
130.4%

15.4%

-55.1%

325.0%

-2.4%

21.3%

45.4%

27.5%
66.2%
4.2%

-12.5%

-35.8%


1.8%

-21.4%
-49.4%

0.7%


275.5
88.2

87.9

32.0

82.6

295.9

178.7

326.5

120.3
15.2
34.5

14.0

33.8


2.7

23.9
8.8

104.2


13.3%
4.2%

4.2%

1.5%

4.0%

14.2%

8.6%

15.7%

5.8%
0.7%
1.7%

0.7%

1.6%


0.1%

1.2%
0.4%

5.0%


9,644

5,542
2,554

6,885











Table 3-14. Continued
Average Average Percent Total heated Percent of
Average Percent
heated area total
FDOR Average heated area (sf) effective HA/EA difference (million h
code Desc o area (sf)area from State
(sf) sf) area
HCWRS GRU Combined State Combined State Combined State State
Wholesale
29 outlets 23,700 23,700 19,928 0.76 15,078 18.9% 11.0 0.5%
outlets
30 greenhse 3,376 3,376 3,945 0.92 3,620 -14.4% 1.3 0.1%
greenhouse
Drive-in
theaters /
31 theaters77,719 0.92* 71,501 2.0 0.1%
open
Stadiums
Enclosed
32 theaters/ 97,632 27,989 51,203 29,303 0.95 27,855 74.7% 6.7 0.3%
auditorium
33 Nightclubs 3,686 5,336 4,676 4,881 0.95 4,624 -4.2% 8.9 0.4%
bars
Bowling alleys /
34 skating 30,784 34,410 33,201 22,931 0.95 21,751 44.8% 10.7 0.5%
rinks
Tourist
35Tourst 36,730 0.92* 33,791 25.1 1.2%
attractions
36 Camps 9,184 0.92* 8,450 2.8 0.1%
37 Race tracks 44,563 0.92* 40,998 5.3 0.3%
Golf courses /
38 driving 18,091 9,633 16,139 18,908 0.89 16,762 -14.6% 23.3 1.1%
ranges
39 Hotels/motels 36,875 31,626 32,676 12,211 0.94 11,528 167.6% 250.2 12.0%
Total
Tommercia 16,444 11,457 14,108 10,272 0.94 9,662 37.3% 2,076.3 100.0%
*HAEA no commercial sector, aEA of available
*HA/EA not available for subsector, average HA/EA of available subsectors used










Comparison of average heated area across the industrial subsectors of Hillsborough County Water Resources Services,


Gainesville Regional Utilities, the two utilities combined, and the state of Florida
Average Average Percent Total heated Percent of
Average Percent
heated area total
FDOR Average heated area (sf) effective HA/EA eaedifference (million h
code Description area (sf) area from State
(sf) sf) area
HCWRS GRU Combined State Combined State Combined State State
Light
41 Light 80,659 20,851 38,975 16,282 0.90 14,670 165.7% 269.9 21.9%
manufacturing
42 Heavy industrial 44,432 41,893 43,586 79,078 0.88 69,271 -37.1% 46.9 3.8%
43 Lumberyards 18,488 18,488 30,446 0.96 29,201 -36.7% 13.8 1.1%
44 Packing plants 38,997 0.91* 35,488 18.7 1.5%
45 Bottler/canneries 34,541 34,541 121,454 0.79 95,817 -64.0% 8.6 0.7%
46 Food processing 41,368 0.91* 37,645 11.1 0.9%
47 Mineralprocessing 102,201 23,001 86,361 17,919 0.98 17,553 392.0% 10.5 0.9%
Warehousing /
48 warehousing/ 41,608 16,966 29,719 19,855 0.95 18,797 58.1% 797.1 64.7%
distribution
49 Open storage 1,860 7,589 2,463 7,576 0.93 7,063 -65.1% 15.5 1.3%
91 Utility, gas & elec. 15,714 50,505 27,311 8,577 0.91 7,828 248.9% 24.9 2.0%
Mining /
92 Minng/ 20,465 0.91* 18,623 2.1 0.2%
petroleum
Total industrial 41,596 18,777 30,893 19,000 0.94 17,862 73.0% 1,231.5 100.0%
*HA/EA not available for subsector, average HA/EA of available subsectors used


Table 3-15.











Table 3-16. Comparison of average heated area across the institutional subsectors of Hillsborough County Water Resources
Services, Gainesville Regional Utilities, the two utilities combined, and the state of Florida
e Average Percent Total heated Percent of
Aveag heated difference area total
FDOR Average heated area (sf) effective HA/EA h d a
Description area from (million heated
(sf) State sf) area
HCWRS GRU Combined State Combined State Combined State State
71 Churches 12,838 13,555 13,085 10,643 0.95 10,064 30.0% 190.1 20.6%
Private schools &
72 vateschools 7,309 9,149 8,071 18,864 0.95 17,863 -54.8% 57.9 6.3%
colleges
73 Private hospitals 424,916 141,224 235,788 132,165 0.97 128,071 84.1% 73.0 7.9%
Homes for the
74 omes f e 116,675 116,675 19,986 0.92 18,432 533.0% 82.0 8.9%
aged
75 Orphanages/ 9,741 10,894 9,952 11,948 0.95 11,353 -12.3% 25.1 2.7%
non-profits
Mortuaries /
76 ceeteries 7,560 5,286 6,511 6,745 0.88 5,963 9.2% 5.3 0.6%
cemeteries
77 hlls 10,654 13,308 12,584 7,699 0.95 7,289 72.7% 23.6 2.6%
hallsni s
78 onvalescnt 43,505 43,505 41,039 1.00 40,908 6.3% 18.7 2.0%
convalescent
Cultural
79 organization 2,302 2,302 15,616 0.74 11,503 -80.0% 2.7 0.3%
organizations
81 Military 85,651 0.93 79,655 3.4 0.4%
Parks and
82 Parksand 5,288 5,288 7,244 0.92 6,638 -20.3% 5.6 0.6%
recreation
83 Pulic co 127,905 59,448 126,588 106,648 0.98 104,529 21.1% 308.7 33.5%
schools
84 Colleges 175,786 175,786 195,205 1.00 195,037 -9.9% 56.2 6.1%
85 Hospitals 123,175 123,175 154,688 0.96 148,126 -16.8% 25.6 2.8%
Gov. leased
90 eased 28,526 0.93 26,530 33.9 3.7%
interests
Outdoor
97 recreation 9,233 9,233 3,488 0.92 3,214 187.3% 6.7 0.7%
recreational
Total institutional 26,988 26,395 26,766 22,880 0.96 22,022 21.5% 921.8 100.0%
*HA/EA not available for subsector, average HA/EA of available subsectors used











Table 3-17. Fitted lognormal probability density equations and associated Anderson-Darling (A-D) and Kolmogorov-Smimow (K-
S) statistics for State total and combined utility sample of commercial parcels
FDOR State total Combined utility sample


code N


11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
29
30
31
32
33
34
35
36
37
38
39


39,002
18,931
829
2,571
449
7,726
37,629
16,033
20,953
2,825
7,629
4,377
4,732
283
4,309
3,434
15,129
729
352
28
242
1,914
492
742
330
129
1,392
21,702


N Fitted lognormal equation A-D statistic K-S statistic


Fitted lognormal equation
Lognorm(6205,9267.4)
Lognorm(3979.6,4892.6)
Lognorm(130417.5,226144.1)
Lognorm(10466.9,18876)
Lognorm(504479.3,5563273.4)
Lognorm(42572.1,101645)
Lognorm(4189.8,6972.1)
Lognorm(13586.8,56193.4)
Lognorm(5223.1,9767.9)
Lognorm(5701,102409.2)
Lognorm(4487.7,3618.4)
Lognorm(3224.7,2156.5)
Lognorm(6571.5,6560.8)
Lognorm(4570.4,6424.4)
Lognorm(4647.3,5794.8)
Lognorm(2487.9,1936.4)
Lognorm(6238.3,9484.1)
Lognorm(14316,23222.4)
Lognorm(3075.9,3956.8)
Lognorm(39865,187722.3)
Lognorm(28010.4,54268.4)
Lognorm(4346.6,3742.9)
Lognorm(30297.7,81922.8)
Lognorm(18060.9,73944.1)
Lognorm(8234,14388.4)
Lognorm(28736.7,139346.2)
Lognorm(25060.1,84420.3)
Lognorm(4718,18708.6)


A-D statistic
72.11
129.69
42.49
107.52
15.40
30.59
357.76
322.42
196.25
94.64
7.27
59.65
110.42
13.10
35.40
13.02
45.83
1.69
0.69
0.85
2.03
8.95
11.72
1.59
1.15
3.12
26.99
1399.08


2.01
1.84
0.44
8.91


0.07
0.09
0.16
0.21


K-S statistic
0.03
0.05
0.21
0.15
0.15
0.06
0.08
0.11
0.07
0.17
0.03
0.09
0.11
0.14
0.06
0.05
0.04
0.04
0.04
0.17
0.08
0.05
0.13
0.04
0.05
0.13
0.10
0.21


Lognorm(6915.4,9009.3)
Lognorm(7664.1,8870.4)
Lognorm(128058.9,44261.3)
Lognorm(4533.9,4488)


Lognorm(41189.7,72650)
Lognorm(5592.5,6137.9)
Lognorm(31701.2,48051.2)
Lognorm(6619.7,6010.2)
Lognorm(8147.7,16823.5)
Lognorm(5078.8,3785.5)
Lognorm(2992.3,1775.5)
Lognorm(5030.5,2929.5)
Lognorm(10121.5,17828.6)
Lognorm(5224.4,5678.8)
Lognorm(1829.1,256.26)
Lognorm(6043.8,7416.4)
Lognorm(24295.6,17153.8)




Lognorm(4556,3802.7)


0.45


0.17


1.11
0.86


0.19
0.11


2.58
3.17
0.45
4.03
0.38
1.91
4.36
3.60
0.29
0.35
0.32
1.48
0.23


26 Lognorm(20768.6,37549)
50 Lognorm(37478.7,54315.1)


0.08
0.08
0.07
0.12
0.21
0.10
0.16
0.15
0.17
0.08
0.23
0.09
0.18










Table 3-18. Fitted lognormal probability density equations and associated Anderson-Darling (A-D) and Kolmogorov-Smimow (K-
S) statistics for State total and combined utility sample of industrial parcels
FDOR State total Combined utility sample
code N Fitted lognormal equation A-D statistic K-S statistic N Fitted lognormal equation A-D statistic K-S statistic
41 18,398 Lognorm(13375.8,32016.2) 137.52 0.08 33 Lognorm(35589.2,82016.2) 1.05 0.22
42 677 Lognorm(70712.4,196744.3) 0.35 0.02 3
43 471 Lognorm(33928.7,86991.8) 0.98 0.04 1
44 526 Lognorm(39030,99836.6) 0.87 0.03
45 90 Lognorm(121878.7,494517.2) 0.25 0.05 1
46 295 Lognorm(37283.9,113069.3) 1.81 0.06
47 598 Lognorm(16233,50857.9) 0.59 0.03 10 Lognorm(128958.6,1274954) 0.16 0.12
48 42,406 Lognorm(17258.9,39654.1) 99.08 0.03 228 Lognorm(31655.2,54270.5) 1.04 0.06
49 2,190 Lognorm(12588.5,66705) 38.64 0.08 19 Lognorm(2249.2,2004.3) 0.66 0.15
91 3,185 Lognorm(6195.7,24310.6) 37.32 0.09 12 Lognorm(23727.1,22734.2) 0.91 0.21
92 112 Lognorm(14980.6,58563.9) 1.60 0.11










Table 3-19. Fitted lognormal probability density equations and associated Anderson-Darling (A-D) and Kolmogorov-Smimow (K-
S) statistics for State total and combined utility sample of institutional parcels
FDOR State total Combined utility sample
code N Fitted lognormal equation A-D statistic K-S statistic n Fitted lognormal equation A-D statistic K-S statistic
71 18,888 Lognorm(9755.6,15478.2) 58.40 0.03 337 Lognorm(12941.2,18379.4) 1.53 0.06
72 3,243 Lognorm(12686.5,25968.3) 53.08 0.09 111 Lognorm(7458.2,7491) 0.92 0.09
73 570 Lognorm(149847.5,612156.5) 2.21 0.05 6 Lognorm(301959.4,981345.9) 0.25 0.19
74 4,447 Lognorm(6225.8,24872.7) 245.89 0.21 12 Lognorm(127980.6,448598.9) 0.52 0.19
75 2,214 Lognorm(9998.3,20236.8) 9.98 0.05 82 Lognorm(9547.1,28900.4) 0.61 0.09
76 887 Lognorm(6442.2,7980.9) 13.90 0.10 13 Lognorm(6581.3,3300.4) 0.32 0.21
77 3,244 Lognorm(6931.3,8460.6) 3.36 0.03 55 Lognorm(13045.8,14582.2) 0.43 0.10
78 457 Lognorm(48378,98509.5) 12.84 0.16 1
79 234 Lognorm(9926.2,20234.2) 1.25 0.07 1
81 43 Lognorm(68159.2,204065.6) 1.01 0.15
82 850 Lognorm(6238.2,15500.5) 0.68 0.02 1
83 2,953 Lognorm(465926.8,5919503.5) 225.66 0.21 52 Lognorm(130157.5,94366.6) 2.62 0.18
84 288 Lognorm(423964.5,5902238.9) 1.69 0.04 9 Lognorm(199478.1,1154125.5) 0.61 0.23
85 173 Lognorm(231446.7,1827373.5) 1.81 0.07 3
90 1,276 Lognorm(27445.3,149544.1) 7.07 0.07
97 2,090 Lognorm(2856.7,6198.6) 7.71 0.06 1










Table 3-20. Fitted lognormal shift probability density equations and associated Anderson-Darling (A-D) and Kolmogorov-Smirnow
(K-S) statistics for State total and combined utility sample of commercial parcels
FDOR State total Combined utility sample
code N Fitted lognormal equation A-D statistic K-S statistic n Fitted lognormal equation A-D statistic K-S statistic
11 39,002 Lognorm(6191.4,9148.7,Shift(- 73.11 0.03 289 Lognorm(6879.1,10162.8,Shift(
9.8419)) 219.34))
12 18,931 Lognorm(3971.2,4809.7,Shift(-2.14 0.05 143
1 718,931 132.14 0.05 143
10.617))
13 829 19 Lognorm(105097.2,46236,Shift 0.41 0.14
(23164))
14 2,571 Lognorm(10491,19976.4,Shift( 99.82 0.14 123 Lognorm(3851.9,6063.4,Shift(8
112.83)) 35.36))
15 449 Lognorm(677996.5,11046462.1, 1647 16 3
15 449 16.47 0.16 3
16 7,726 Lognorm(43500.7,109776.5,Shi 27.80 0.05 239 Lognorm(42574.9,88032.7,Shif
ft(162.45)) t(871.69))
17 37,629 384 Lognorm(5363.2,9389.2,Shift(7 0.28 0.03
02.9))
18 16,033 Lognorm(13597.5,56316,Shift(321.85 0.11 73 Lognorm(33856.1,73791.5,Shif 0.41 0.07
0.55288)) t(1683.9))
19 20,953 Lognorm(4878.2,7364.5,Shift(- 114.84 0.05 264 Lognorm(5952.2,7472.9,Shift(8
108.5)) 68.41))
20 2,825 Lognorm(6270.7,127055.2,Shift 6
(0.71497))
21 7,629 120 Lognorm(8645,2775.3,Shift(- 0.53 0.06
3737.2))
22 4,377 105 Lognorm(22534,1247.4,Shift(-0 0.06
22 4,377 105 0.90 0.06
19623))
23 4,732 98
24 283 Lognorm(4425.7,7098.1,Shift(2 10.89 0.13 11 Lognorm(15285.7,77251.8,Shif 0.27 0.16
08.47)) t(831.87))
25 4,309 Lognorm(4629,6006.4,Shift(53. 32.32 0.06 49 Lognorm(5224.9,5675.9,Shift(-
395)) 1.0112))
26 3,434 5 Lognorm(2263.2,254.85,Shift(- 32 023
434.2))
27 15,129 Lognorm(6230.3,9407.6,Shift(- 46.40 0.04 174 Lognorm(6048.8,7296.1,Shift(-
6.5661)) 27.336))










Table 3-20. Continued

FDOR State total Combined utility sample
code N Fitted lognormal equation A-D statistic K-S statistic n Fitted lognormal equation A-D statistic K-S statistic
9 7 Lognorm(14417.7,24479.5,Shif1.57 0.04 5 Lognorm(41463.5,13411.9,Shif 0
29 729 1.57 0.04 5 0.19 0.18
t(103.52)) t(-17713))
30 352 Lognorm(3071.4,4261.6,Shift(5 0.81 0.04 2
6.464))
31 28 Lognorm(66864.5,776190.2,Sh 0.37 0.12
ift(702.06))
32 242 Lognorm(29357.2,68186.4,Shif 1.28 0.06 3
32 242 1.28 0.06 3
t(573.64))
33 1,914 20 Lognorm(3892.9,6069.7,Shift(9 0.20 0.10
85.81))
34 492 Lognorm(24285.3,38140.6,Shif 6.90 0.09 3
t(-1017.9))
35 742 Lognorm(19303.9,88847.3,Shif 1.65 0.04
t(49.97))
36 330 Lognorm(8076.2,12907.7,Shift(
36 330 0.81 0.05
-96.975))
37 129 Lognorm(33492.9,215636.3,Sh 2.16 0.11
ift(195.85))
38 1,392 Lognonn(20406.4,45509.6,Shif
t(-373.38))
39 21,702 Lognorm(4693.3,18461.7,Shift( 1403.58 0.21 50 Lognorm(38324.6,33788.4,Shif 0.47 0.09
-0.81817)) t(-4828.5))










Table 3-21. Fitted lognormal shift probability density equations and associated Anderson-Darling (A-D) and Kolmogorov-Smirnow
(K-S) statistics for State total and combined utility sample of industrial parcels
FDOR State total Combined utility sample
code N Fitted lognormal equation A-D statistic K-S statistic n Fitted lognormal equation A-D statistic K-S statistic

41 18,398 Lognorm(13469,33184,Shift(41 0.07 33
18,398 126.86 0.07 33
.897))
42 677 Lognorm(69618.8,187617.9,Sh 0.32 0.02 3
ift(-100.43))
43 471 Lognorm(33597.4,84239.6,Shif 0.92 0.04 1
t(-41.52))
44 526 Lognorm(37471.6,87439.9,Shif 0.50 0.03
44 526 0.50 0.03
t(-195.3))
45 90 Lognorm(115994.8,434806.5,S 0.24 0.05 1
hift(-164.33))
46 295 Lognorm(39449,135557,Shift(2 1.19 0.05
29))
47 598 Lognorm(16955,58041.8,Shift( 0.39 0.02 10 Lognorm(15291500000,181412 2.26 0.38
58.741)) .3,Shift(-15291500000))
48 42,406 Lognorm(17209.7,39262.3,Shif 102.42 0.03 228 Lognorm(31697.6,54708.6,Shif .
t(-8.7696)) t(29.426))
49 2,190 Lognorm(7474.9,18983.8,Shift( 2.16 0.03 19 Lognorm(66065071.7,3037.2,S
-72.531)) hift(-66062608))
91 3,185 Lognorm(6754.5,31730.1,Shift( 23.92 0.07 12 Lognorm(30900.2,243072.7,Sh
40.436)) ift(8314.8))
92 112 Lognorm(1843610000,49323.8, 24.11 0.35
Shift(-1843590000))










Table 3-22. Fitted lognormal shift probability density equations and associated Anderson-Darling (A-D) and Kolmogorov-Smirnow
(K-S) statistics for State total and combined utility sample of institutional parcels
FDOR State total Combined utility sample
code N Fitted lognormal equation A-D statistic K-S statistic n Fitted lognormal equation A-D statistic K-S statistic


71 18,888 Lognorm(9492.5,14051.4,
Shift(-52.418))
72 3,243 Lognorm(12106.6,22765,
Shift(-41.379))
73 570

74 4,447 Lognorm(8023.4,54233.7,
Shift(117.32))
75 2,214 Lognorm(10167.6,21827.5,
Shift(62.384))
76 887

77 3,244

78 457 Lognorm(48081.6,95982.2,
Shift(-88.963))
79 234 Lognorm(10208.5,23085.5,
Shift(113.43))
81 43 Lognorm(85496900000,180027
.1, Shift(-85496900000))
82 850 Lognorm(6588.5,18670.1,
Shift(50.818))
83 2,953 Lognorm(110758.3,142790.4,
Shift(-10436))
84 288 Lognorm(271500.7,2238345.1,
Shift(-96.53))
85 173 Lognorm(293003.6,3293062.8,
Shift(523.62))
90 1,276 Lognorm(27953.8,156847,
Shift(11.64))
97 2,090 Lognorm(2841,6084.1, Shift(-
2.6213))


68.23

53.10


164.38

8.36


12.71

0.90

9.93

0.74

47.42

1.08

1.25

6.54

8.01


0.04 337


09 111 Lognorm(7117.8,8751.9,Shift(5
37.89))
6

0.17 12

0.05 82 Lognorm(9852.3,31747.8,Shift(
23.7))
1 Lognorm(18004.1,2742.8,Shift(
-11492))
55 Lognorm(13149.6,13893.8,Shif
t(-228.41))
0.16 1

0.06 1

0.38

0.03 1

0.09 52 Lognorm(150120.7,73721.4,Sh
ift(-24686))
0.06 9

0.08 3

0.07

0.06 1


0.59


0.60


0.42


2.57









-
Residentia
L- -


I
E


Im
Urban\Wate


I


-

U


Levels of Florida Department of Revenue (FDOR) land use disaggregation into 9
residential and 55 commercial, industrial, and institutional (CII) sectors


Census Block



" PWSID
" n = 2,623'III'I II i
^f "DE P aO^^

otl vraeadpa


*


*


Schematic of spatial and attribute database relationships to FDOR


ml
1 O
01020


Figure 3-1.


Wel


SpatialJoin
AttributeJoin


I


Figure 3-2.


M
1
Isttn
ON


" Prcl" ID
: No oflrd bilding
", Yelal bui ltr
", Total pacl area
", Totl bulding area
:, n =8.8milio










oa Parcels GRU
55,551

STotal Parcels
8,807,768







Parcels Alachua
99,305




Figure 3-3. Macro to nano-scale evaluation of public water



1,200,000

1,000,000

S800,000

600,000

400,000

200,000

o1


N,o


Effective Area (sf)


Heated and effective area correlation for 3,205 CII parcels in Hillsborough
County Water Resources Services and Gainesville Regional Utilities


Figure 3-4.


o oo


ooo oo0
I? %PZ










50,000


0

-
-50,000



-100,000



-150,000



-200,000


Figure 3-5.


~ A


. .


NO,
bc0O
11


Effective Area (sf)


Residual plot of heated and effective area simple linear regression for 3,205 CII
parcels in Hillsborough County Water Resources Services and Gainesville
Regional Utilities


000
~~LdU'


d^z









CHAPTER 4
FLORIDA CASE STUDIES ON EVALUATING WATER USE

Utility Water Billing Databases

Water billing data vary widely in both content and availability depending on the policies of

the individual utilities. Without standardized databases, these differences make water use

comparisons across utilities very difficult. In our study, parcel-level land use characteristics from

the Florida Department of Revenue (FDOR) and Florida County Property Appraiser (FCPA)

database were linked with historic water billing data for 3,205 commercial, industrial, and

institutional (CII) parcels (1,768 in Hillsborough County Water Resources Services (HCWRS)

and 1,437 in Gainesville Regional Utilities (GRU)) to develop water use coefficients normalized

by heated building area. HCWRS provided four complete years of monthly water billing from

January 2003 through December 2006, while GRU supplied two complete years of monthly

water billing from January 2008 to December 2009.

Hillsborough County Water Resources Services

HCWRS data provided utility billing data for 1,768 CII accounts (67% commercial, 9%

industrial, and 24% institutional) for 48 months beginning in January 2003. The FDOR and

Hillsborough County Property Appraiser (HCPA) property attribute data are updated annually.

These databases include the year built so it is possible to show how the size of any sector has

changed over time. Thus, it is possible to adjust the monthly water use statistics to the number of

accounts that existed in each of the years from 2003 to 2006. By using FDOR land use codes, CII

water customers were grouped into the commercial, industrial, and institutional sectors.

HCWRS provided the crucial link to FDOR via a parcel ID. Parcel ID is the common

identifier which allows parcel attributes from FDOR to be related to water use. HCWRS data

also provided valuable time series information about the nature of water use in its CII sectors. By









using FDOR land use codes, CII water customers can be aggregated into either the commercial,

industrial, or institutional sector (Figure 3-1). The time series plots of these sectors, presented in

Figure 4-1, shows minimal trending and monthly variability among the sectors. By obtaining the

average water use for each month within the four year time series, the minimum and maximum

average monthly water use is taken to be the base and peak usage, respectively. The seasonal

component of the time series is extracted by subtracting the base usage from the overall monthly

average water use, as shown in Equation 4-1.

QSeasonal =Avg Base (4-1)

Where:
Qseasonal = average monthly seasonal water use
QAvg = average monthly water use
QBase = minimum month water use


The seasonal component divided by the average usage (Equation 4-2) results in an

estimate of percent seasonality, a measure of the significance of seasonal water use for the given

system or sector.


% Seasonality = Qseasonal 100% (4-2)
QAvg

Where:

% Seasonality = estimate of percent of average monthly water use that is seasonal
Qseasonal = average monthly seasonal water use
QAvg = average monthly water use

These water use statistics per sector are presented in Table 4-1 for HCWRS. The percent

seasonality estimates are 3.2%, 7.8%, and 11.0% for the commercial, industrial and institutional

sectors, respectively. Such seasonal estimates indicate that point estimates of water use

coefficients are reasonable for the CII sectors. The lack of seasonality within the CII sectors of









HCWRS is also apparent in the time series information presented in Figure 4-1. This figure also

indicates minimal trending within the sectors following a least square regression trend line. The

trend lines in Figure 4-1 indicate a 0.004%, -0.007%, and 0.005% monthly change in total water

use amongst the commercial, industrial, and institutional sectors, respectively. Such minimal

trending, given that fact that only customers with water billing throughout the entirety of the

billing period were included in the time series, once again indicated that point estimates of water

use are appropriate for the CII sectors.

To develop the water use coefficients, the HCWRS water billing records were adjusted so

that all customer records on a given parcel in a given month were summed. This procedure

aggregated the water billing to the parcel level, and accounted for customers with multiple water

use entries for a given month. Multiple entries are common in billing records, where a utility

seeks to correct for a billing over/under-charge with a separate billing entry. This method of

aggregating the billing records maintains the water use in a time series for each parcel, and

allows for parcel land use classification via FDOR. Only parcels reporting monthly water use

through the entire study period were included. The adjusted billing records for 1,177 commercial

parcels, 163 industrial parcels, and 428 institutional parcels were then linked to FDOR and

HCPA via the unique parcel ID.

Gainesville Regional Utilities

GRU supplied two complete years of monthly water billing from January 2008 to

December 2009. Unlike HCWRS, the GRU billing records did not include a Parcel ID field. To

arrive at this field, a geocoded GIS point shapefile which included the utility specific,

[Premise Key], was provided by GRU. This file was spatially joined to the FDOR parcel

polygon shapefile, so that every [Premise_ Key] was tagged with their corresponding Parcel ID.

This spatial join resulted in a look-up table linking [Premise Key] to [Parcel_ID] which allowed









billing records provided by GRU to be joined with the parcel attributes from FDOR and Alachua

County Property Appraiser.

The GRU water billing records were also not rectified, meaning that each customer had

billing periods which differed depending on the date in which each meter was read. In order to

compare and aggregate the water billing within sectors, the water billing records required

rectification. The water billing records included several key fields that made rectification

possible, these included:

Usage quantity

Usage units

Bill period length

Bill period end date

[USAGE QUANTITY] is the total water billed to a customer for a given billing period.

[USAGE_UNITS] is the units in which this water billing is reported.

[BILLPERIODLENGTH] is the number of days which encompass the billing period. This

field allows water use per day to be calculated for each billing period, which is key to rectifying

billing records. [BILLPERIODENDDATE] is the date in which a given billing period ends.

This field allows for calculation of how many billing period days are within the current "end

period month." By knowing how many days are in the month to be rectified, these fields provide

all the required information to rectify billing records following Equation 4-3.












Qrecd = Bx + (M B)y


Where:
Qrectified = rectified month water use
B = number of billing days in current "end of billing period" month
M = number of days in rectification month
x = water use per day for current "end of billing" period
y = water use per day for following billing period
(M B) is less than the following billing period


With the billing rectified, the water billing records were then grouped so that all customer

records on a given parcel in a given month were summed. The time series plots for the

aggregated 1,037 commercial parcels, 144 industrial parcels and 256 institutional parcels in GRU

are presented in Figure 4-2. This figure shows the lack of monthly variability among the CII

sectors, as well as minimal trending. The trend lines in Figure 4-2 indicate a -0.02%, -0.04%, and

-0.01% monthly change in total water use amongst the commercial, industrial, and institutional

sectors, respectively. Only parcels reporting monthly water use through the entirety of the study

period were included in the analysis. These water use statistics per sector for GRU are shown in

Table 4-2. The percent seasonality estimates are 7.2%, 17.8%, and 10.1% for the commercial,

industrial and institutional sectors, respectively. Such small percent seasonal estimates once

again demonstrate that point estimates of water use are reasonable for the CII sectors.

Combined Utilities

In order to increase the overall sample size and include as many different customers with

dissimilar building and water use characteristics, the billing records from HCWRS and GRU

were combined. The average monthly water use per account time series for the commercial

sector in HCWRS, GRU and the two utilities combined is presented in Figure 4-3. Similar plots

are available for the industrial, and institutional sectors (Figure 4-4, Figure 4-5). The presentation


(4-3)









of this time series information once again shows the lack of seasonality within the CII sectors of

water use. These figures also demonstrate that the average water use per account across the

utilities is relatively similar, despite such sectors being composed of varied subsectors.

In order to ease the bias of different land use mixes within the two utilities Table 4-3,

Table 4-4, and Table 4-5 present the average gallons per account per day (gpad) for HCWRS,

GRU, and the two utilities combined, for the commercial, industrial, and institutional subsectors,

respectively. The percent difference in gpad between HCWRS and GRU is calculated to provide

a measure of the water use similarity of water users across utilities. Subsector percent differences

in gpad across utilities ranged from -92% to 274%. Differences in the type of water users and

measures of size, as shown in the previous chapter, are evident across utilities. Normalization

through a measure of size is required to compensate for the heterogeneous nature of CII water

users. The measure of size proposed in this methodology is heated building area as was discussed

in the previous chapter. The relationship between water use and heated building area is evaluated

in the following section.

Relationship of Heated Area to Water Use

The ability for heated area to be a good estimator of water use is critical for its utilization

as a measure of size to normalize water use. By linking the water billing databases of HCWRS

and GRU with the land use databases of FDOR and FCPA, the relationship between property

attributes and CII water use can be evaluated. The strong correlation between heated area and

water use for all 3,205 CII parcels in HCWRS and GRU, shown in Table 4-6, indicates that

heated area, with a correlation coefficient of 0.63 is good predictor of water use within the CII

sector.

Other property attributes such as parcel area and effective year built, can be evaluated

alongside heated area through stepwise multivariate regression (Neter et al. 1996). The stepwise









regression was carried out using StatTools (Palisade Tools 2010), which employs a method that

models the choice of entering predictive variables based on their p-value, whereby if the p-value

is less than 0.05, the variable is entered in the regression, and if the p-value is greater than 0.1,

then the variable leaves the regression. The result for this stepwise regression is presented in

Table 4-7. The table shows that all three predictive variables are entered in the regression, and

provides their coefficients in the regression equation, which passed through the origin. The

adjusted R2 value of the stepwise regression equation (Equation 4-4) is shown in Table 4-8 to

equal 0.41. The ANOVA (Table 4-9) shows that the p-value for the overall regression equation

is less than 0.0001, which indicates that the relationship between these variables in predicting

water use is highly statistically significant.

Q, = (0.064)HA, + (0.373)EYB, (0.646)TA, (4-4)

Where:
Qi = average gallons per day water use for parcel i
HAi = heated square footage of all buildings on parcel i
EYBi = effective year built of major improvements on parcel i (e.g., 1984)
TAi = parcel area of parcel i in acres

The order that the predictive variables enter the regression model is dependent on both

their correlation to water use and to each other, and is critical in determining the best-fit

regression model. From Table 4-10, it is shown that heated square footage is the first predictive

variable entered into the regression, given that it is the most highly correlated variable to water

use. The Adjusted R2 value for the regression model of water use solely using heated square

footage as the predictive variable is 0.39. Hence, by adding effective year built, and parcel

acreage, little predictive power is gained, since the overall regression equation produces an

adjusted R2 of 0.41.









Since heated area is the best predictor of water use available from the property attributes

evaluated, and little is gained from the other variables, this methodology proposes water use

relationships be based solely on heated square footage of buildings on a parcel. The regression

statistics through the origin between heated square footage and average daily water use for the

individual FDOR CII subsectors in the combined utilities is presented in Table 4-11. Though p-

values across subsectors vary from <0.0001 to 0.6352, for the most part the relationship between

these two variables is highly significant.

Heated square footage is also a good measure of size for CII parcels since it is available

free for every parcel in the State through the FDOR database, and relationships between effective

and heated areas presented in this thesis. This database is of high quality since it is used for

setting property taxes and is updated annually. Heated area is also a standardized area across

most fields and outside the state of Florida, whereby heated area is all building areas under

climate control. Such a standardized metric as a measure of size allows water use coefficients

normalized by heated area to be readily applied to other property databases outside of the State.

Property databases such as FDOR and FCPA, also provide an added benefit in that they provide

heated area at the parcel level, which is a finer spatial resolution than Traffic Analysis Zones

(TAZ). TAZ is the finest geographical area by which the U.S. Census aggregates employment

figures. In the state of Florida, there are 12,747 TAZs and nearly 9 million parcels. Parcel-level

data allows for greater precision in estimating water use, and identifying sectors and drivers of

demand.

The following chapter will address the development of CII water use coefficients, carried

out by using heated area to normalize water use.









Table 4-1. Summary statistics for CII sectors in Hillsborough County Water Resources Services
Commercial Industrial Institutional
Sample size 1,177 163 428
Total (MGD) 2.64 0.29 0.86
Base (MGD) 2.56 0.27 0.77
Base month September January January
Peak (MGD) 2.70 0.30 0.97
Peak month May August August
Seasonal (MGD) 0.09 0.02 0.10
% Seasonal 3.2% 7.8% 11.0%


Table 4-2. Summary statistics for CII sectors in Gainesville Regional Utilities
Commercial Industrial Institutional
Sample size 1,037 144 256
Total (MGD) 1.50 0.18 0.57
Base (MGD) 1.39 0.15 0.51
Base month December December December
Peak (MGD) 1.62 0.21 0.63
Peak month May July September
Seasonal (MGD) 0.11 0.03 0.06
% Seasonal 7.2% 17.8% 10.1%











Table 4-3. Comparison of water use per commercial account for Hillsborough County Water Resources Services, Gainesville
Regional Utilities, and the two utilities combined
HCWRS GRU Combined utility
Water use Water use Utility % Water use
Total water per Total water per diff. Total water per
FDOR use account use account in use account
code Description n (gal/d) (gpad) n (gal/d) (gpad) gpad n (gal/d) (gpad)


78,275
131,528


11 Stores, one-story
12 Mixed use
Department
13 stores
Supermarkets /
14 conv. stores
15 Regional malls
Community
shopping
16 centers
Office, one-
17 story
Office, multi-
18 story
19 Medical office
Transit
20 terminals
21 Restaurants
Fast-food
22 restaurants
Financial
23 institutions
Insurance
company
24 offices
25 Service shops
26 Service stations
Auto sales /
27 repair


693
1,302


8,230

1,577
71,750


176 137,392
42 19,957

1 2,344


7,833
43,354


4,555 78 193,460

978 232 147,634


3,034 32
885 157


3,724
4,195


30,168
219,951

217
150,690


18 148,133

117 184,474
2 143,501


161 733,342

152 148,692

41 124,393
107 94,648

5 18,622
68 285,242

55 116,724

63 68,994




18 21,753


116 138,436


781
475


2,344

1,305
43,354


2,480

636

943
1,401

217
2,898

1,684

3,366


782
798
311

587


13%
-64%


289 215,668
143 151,485


-72% 19 150,477


-17%
-40%


123 192,306
3 186,855


-46% 239 926,802

-35% 384 296,327

-69% 73 154,561
58% 264 314,599


-94%
-31%


6 18,840
120 435,932


-21% 105 200,911

207% 98 186,801


11
-34% 49
5


8,604
46,496
1,555


-51% 174 172,493


746
1,059

7,920

1,563
62,285


3,878


2,117
1,192

3,140
3,633

1,913

1,906


2,122 50 84,187

1,095 35 117,807


11 8,604
1,208 31 24,744
5 1,555

1,193 58 34,057










Table 4-3. Continued
HCWRS GRU Combined utility
Water use Water use Utility % Water use
Total water per Total water per diff. Total water per
FDOR use account use account in use account
code Description n (gal/d) (gpad) n (gal/d) (gpad) gpad n (gal/d) (gpad)
Wholesale
29 outlets 5 2,991 598 5 2,991 598
Florist /
30 greenhouses 2 1,459 730 2 1,459 730
Enclosed
theaters /
32 auditoriums 1 11,802 11,802 2 6,653 3,326 -72% 3 18,455 6,152
Nightclubs /
33 bars 8 7,848 981 12 10,668 889 -9% 20 18,516 926
Bowling alleys /
skating
34 rinks 1 1,228 1,228 2 2,532 1,266 3% 3 3,759 1,253
Golf courses /
driving
38 ranges 20 54,602 2,730 6 23,430 3,905 43% 26 78,032 3,001
39 Hotels/motels 10 99,301 9,930 40 278,572 6,964 -30% 50 377,873 7,557
Total
commercial 1,177 2,611,538 2,219 1,037 1,550,259 1,495 -33% 2,214 4,161,797 1,880










Table 4-4. Comparison of water use per industrial account for Hillsborough County Water Resources Services, Gainesville Regional
Utilities, and the two utilities combined
HCWRS GRU Combined utility
Water use Water use Utility % Water use
Total water per Total water per diff. Total water per
FDOR use account use account in use account
code Description n (gal/d) (gpad) n (gal/d) (gpad) gpad n (gal/d) (gpad)
41 Light
41 manufa g 10 14,866 1,487 23 55,852 2,428 63% 33 70,718 2,143
manufacturing
42 Heavy industrial 2 3,515 1,758 1 1,397 1,397 -21% 3 4,912 1,637
43 Lumber yards 1 10,765 10,765 1 10,765 10,765
45 Bottler/ canneries 1 339 339 1 339 339
Mineral
47 process 8 140,064 17,508 2 280 140 -99% 10 140,344 14,034
processing
48 istriuion 118 120,308 1,020 110 113,492 1,032 1% 228 233,800 1,025
distribution
49 Open storage 17 6,410 377 2 703 352 -7% 19 7,113 374
91 Utility, gas & elec. 8 2,836 354 4 5,303 1,326 274% 12 8,139 678
Total industrial 163 287,998 1,767 144 188,132 1,306 -26% 307 405,412 1,321











Table 4-5. Comparison of water use per institutional account for Hillsborough County Water Resources Services, Gainesville
Regional Utilities, and the two utilities combined
HCWRS GRU Combined utility
Water use Water use Utility % Water use
Total water per Total water per diff. Total water per
FDOR use account use account in use account
code Description n (gal/d) (gpad) n (gal/d) (gpad) gpad n (gal/d) (gpad)
71 Churches 221 118,775 537 116 98,282 847 58% 337 217,057 644
Private schools &
72 vateschools 65 70,238 1,081 46 41,872 910 -16% 111 112,110 1,010
colleges
73 Private hospitals 2 73,145 36,572 4 77,174 19,294 -47% 6 150,319 25,053
Homes for the
74 age 12 140,993 11,749 12 140,993 11,749
aged
75 Orphanages / non- 67 122,526 1,829 15 20,027 1,335 -27% 82 142,552 1,738
profits
Mortuaries /
76 ceeteries 7 3,344 478 6 9,387 1,565 228% 13 12,731 979
cemeteries
77 Clubs/union halls 15 25,422 1,695 40 105,102 2,628 55% 55 130,524 2,373
Sanitariums /
78 convaleents 1 16,464 16,464 1 16,464 16,464
convalescents
Cultural
79 ltural 1 375 375 1 375 375
organizations
Parks and
82 recreatn 1 1,209 1,209 1 1,209 1,209
recreation
Public county
83 Pulic co 51 448,492 8,794 1 1,474 1,474 -83% 52 449,966 8,653
schools
84 Colleges 9 54,621 6,069 9 54,621 6,069
85 Hospitals 3 2,208 736 3 2,208 736
Outdoor
97 1 239 239 1 239 239
recreational
Total institutional 428 861,941 2,014 256 569,429 2,224 10% 684 1,431,370 2,093









Table 4-6. Correlation matrix of Florida Department of Revenue (FDOR) and Florida County
Property Appraiser (FCPA) property attributes and water use for CII parcels in
Hillsborough County Water Resources Services and Gainesville Regional Utilities
Average
Heated Effective Parcel Effective Avera
monthly
area area area year w
Sc\water
(sf) (sf) (acres) built
use
Heated area (sf) 1.000
Effective area (sf) 0.996 1.000
Parcel area (acres) 0.347 0.356 1.000
Effective year built 0.028 0.030 0.003 1.000
Average monthly water use 0.631 0.639 0.096 0.021 1.000


Table 4-7. Stepwise regression results for CII parcels in Hillsborough County Water Resources
Services and Gainesville Regional Utilities
Confidence
Coefficient rd t-value p-value interval 95%
error
errorLower Upper
Constant 0 N/A N/A N/A N/A N/A
Heated area (sf) 0.0639 0.0014 46.9384 < 0.0001 0.0612 0.0665
Effectiveyear 0.3727 0.0355 10.4866 < 0.0001 0.3030 0.4424
built
Parcel area
Parcel area -0.6461 0.0652 -9.9052 < 0.0001 -0.7740 -0.5182
(acres)


Table 4-8. Stepwise regression "goodness-of-fit" measures for CII parcels in Hillsborough
County Water Resources Services and Gainesville Regional Utilities
Multiple R R2 Adjusted R2 StErr of estimate
0.6439 0.4146 0.4142 3738.1719


Table 4-9. ANOVA table for stepwise regression of CII parcels in Hillsborough County Water
Resources Services and Gainesville Regional Utilities
Degrees of Mean of
Degrees of Sum of squares F-ratio p-value
freedom Su sq s squares
Explained 3 31519189783 10506396594 751.8570 < 0.0001
Unexplained 3185 44506963976 13973929









Table 4-10.



Heated area (
Effective year
Parcel area (a


Step information for stepwise regression of CII parcels in Hillsborough County
Water Resources Services and Gainesville Regional Utilities
Multiple R2 AdjustedR2 StErrof Enter or
R R Adjusted R
R estimate exit
sf) 0.6275 0.3938 0.3938 3802.7815 Enter
r built 0.6297 0.3966 0.3964 3794.7158 Enter
cres) 0.6439 0.4146 0.4142 3738.1719 Enter










Table 4-11.


Regression statistics between heated square footage and average daily water use for the FDOR CII subsectors in
Hillsborough County Water Resources Services and Gainesville Regional Utilities


FDOR FDOR
e Description p-value F R Description p-value
code code
11 Stores, one-story 0.0126 41 Light manufacturing 0.6252
12 Mixed use 0.0103 42 Heavy industrial 0.0302
13 Department stores 0.3288 43 Lumber yards
14 Supermarkets / conv. stores <0.0001 44 Packing plants
15 Regional malls <0.0001 45 Bottler / canneries
16 Community shopping centers <0.0001 46 Food processing
17 Office, one-story 0.0042 47 Mineral processing 0.0006
18 Office, multi-story <0.0001 48 Warehousing / distribution 0.3862
19 Medical office <0.0001 49 Open storage <0.0001
20 Transit terminals <0.0001 91 Utility, gas & elec. <0.0001
21 Restaurants <0.0001 92 Mining / petroleum
22 Fast-food restaurants <0.0001 Total industrial <0.0001
23 Financial institutions <0.0001 71 Churches <0.0001
24 Insurance company offices 0.063 72 Private schools & colleges <0.0001
25 Service shops <0.0001 73 Private hospitals 0.0004
26 Service stations 0.6087 74 Homes for the aged 0.424
27 Auto sales / repair <0.0001 75 Orphanages / non-profits 0.0005
29 Wholesale outlets <0.0001 76 Mortuaries / cemeteries <0.0001
30 Florist / greenhouses 77 Clubs / union halls <0.0001
31 Drive-in theaters / open stadiums 78 Sanitariums / convalescents
32 Enclosed theaters / auditoriums 0.0086 79 Cultural organizations
33 Nightclubs/ bars 0.0007 81 Military
34 Bowling alleys / skating rinks 0.1708 82 Parks and recreation
35 Tourist attractions 83 Public county schools <0.0001
36 Camps 84 Colleges 0.0009
37 Race tracks 85 Hospitals <0.0001
38 Golf courses / driving ranges 0.0223 90 Gov. leased interests
39 Hotels / motels <0.0001 97 Outdoor recreational
Total commercial <0.0001 Total institutional <0.0001










3.00


2.40


1.80


1.20


0.60


0.00


y = 0.0001x- 1.9135
R2 = 0.2506


v = 4E-05x 0.7445
R _R2= 0.0422 A


S= -2E-05x + 0.9238
'R2= 0.0562

C D C D C C\ CO C
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
,2 iU O 4 5 i O O
^S ^ ^s ^ ^s ^ ^s ^


Commercial Industrial Institutional


Figure 4-1. Time series plots of monthly water use for 1,177 commercial parcels, 163
industrial parcels, and 428 institutional parcels in Hillsborough County Water
Resources Services










2.00


1.60


1.20


" 0.80


0.40


0.00


Commercial Industrial Institutional


Figure 4-2. Time series plots of monthly water use for 1,037 commercial parcels, 144
industrial parcels, and 256 institutional parcels in Gainesville Regional Utilities


y =-0.0003x + 14.594
R2 = 0.5446





y= -4E-05x + 1.9954
R2 = 0.0236

y -8E-05x 3.425
R2 =059










3,000


2,400


1,800


1,200


o 600
Q?


S I I I I I I I I I I I I





-GRU -HCWRS -Combined

Figure 4-3. Average monthly water use per commercial account in Hillsborough County
Water Resources Services, Gainesville Regional Utilities, and the two utilities
combined










2,500


S2,000
a

1,500


, 1,000


: 500


-GRU -HCWRS -Combined


Figure 4-4. Average monthly water use per industrial account in Hillsborough County Water
Resources Services, Gainesville Regional Utilities, and the two utilities combined












3,000
o
" 2,400


: 1,800

S1,200


o 600
Q%


-GRU -HCWRS -Combined


Figure 4-5. Average monthly water use per institutional account in Hillsborough County
Water Resources Services, Gainesville Regional Utilities, and the two utilities
combined


_.. _ZMN .....................









CHAPTER 5
COMMERCIAL, INDUSTRIAL, AND INSTITUTIONAL WATER USE COEFFICIENTS

Introduction

Commercial, industrial and institutional (CII) public water use activity coefficients were

developed by linking the parcel-level property attribute data with the water billing data from

Hillsborough County Water Resources Services (HCWRS) and Gainesville Regional Utilities

(GRU). Because the Florida Department of Revenue (FDOR) database identifies the

classification of land use for each parcel, this database provides the means to aggregate or

disaggregate the water use activity coefficients for sectors by different levels of specification.

The most aggregated coefficients are for the general CII sectors. These coefficients can be

disaggregated into 55 subsectors based on the FDOR two-digit land use categories. By this

method, every utility in the state of Florida can determine the relative water use by different

subsectors of customers in their service area. The relative use of each sector can be calibrated

with known total water use in order to identify how the water is used at present and which

subsectors are the more significant water users.

Four water use coefficients are presented in this chapter: average, base, seasonal, and

May peak. The average water use coefficients were developed by summing the average monthly

water use of all parcels within a given subsector and dividing by their total heated area, and the

average number of days in the months billed (Equation 5-1). This method of calculating the

coefficients provides a weighted average which compensates for the skewness often found in the

distribution of CII water users.









AWU =[YQ, /Y HA, ]AD (5-1)
-i1 1=1

Where:
AWU = average weighted water use coefficient (gallons/heated sq. ft./day)
Qi = average monthly water use of parcel i (gallons/month)
HAi = heated square footage of all buildings on parcel i (square feet)
AD = average number of days in months billed (days)


Peak (Equation 5-2) and base (Equation 5-3) water use coefficients were also developed

by correspondingly summing the average May and average minimum monthly water use of all

parcels in a subsector, and dividing by the total heated area of the subsector. The average May

usage is the peak month use for most water utilities in Florida. Thus, it is appropriate to use May

as the peak water use of interest. Unlike the peak coefficient, where the overall system peak is of

concern, the base coefficient provides a measure of the seasonality of a given subsector, and is

dependent on that given subsector's own time series. Only parcels reporting monthly water use

through the entirety of the study period were included in the analysis.



PWU = [ Q,, / HA, ]/ ADM (5-2)
1=1 1=1


Where:
PWU = May peak weighted water use coefficient (gallons/heated sq. ft./day)
QMay,i = average May monthly water use of parcel i (gallons/month)
HAi = heated square footage of all buildings on parcel i (square feet)
ADMay = average number of days in May months billed (days)


BWU = [ ,, / HA, I/ADm, (5-3)
1=1 1=1

Where:
BWU = base weighted water use coefficient (gallons/heated sq. ft./day)
QMin,i = average minimum monthly water use of parcel i (gallons/month)
HAi = heated square footage of all buildings on parcel i (square feet)
ADMin = average number of days in minimum month months billed (days)









The measure of size used to normalize the water use data and develop the activity

coefficients is heated area that is available from the FCPA. As documented in Chapter 3, in order

to use these water use coefficients directly with the FDOR statewide database, effective building

area must be converted to heated building area. The coefficients to convert from effective area

(EA) to heated area (HA) are presented under the subheading of HA/EA.

Commercial Sector

The developed water use coefficients for the available commercial FDOR subsector

categories in HCWRS and GRU are shown in Table 5-1 and Table 5-2, respectively. These

tables include the sample sizes from which the coefficients were derived, the average effective

year built and heated building areas, and percent seasonal, a measure of the significance of

seasonal water use. This measure is obtained by subtracting the base water use coefficient from

the average water use coefficient to arrive at the seasonal water use coefficient (Equation 5-4).

SWU =AWU BWU (5-4)

Where:
SWU= seasonal weighted water use coefficient (gallons/heated sq. ft./day)
AWU= average weighted water use coefficient (gallons/heated sq. ft./day)
BWU= base weighted water use coefficient (gallons/heated sq. ft./day)


The seasonal weighted water use coefficient is then divided by the average water use

coefficient to arrive at an estimate of the percentage of total water use that is seasonal

(Equation 5-5).


% Seasonal = SWU 100%(55)
AWU

Where:
% Seasonal = percentage of total water use that is seasonal
SWU= seasonal weighted water use coefficient (gallons/heated sq. ft./day)
AWU= average weighted water use coefficient (gallons/heated sq. ft./day)









Also included in these tables is the percent of total heated area and water use in each

subsector. These columns provide a measure of the relative importance of each subsector when it

comes to CII land use and water use for the given utility. Not shown in Table 5-1 are insurance

company offices (FDOR 24), service stations (FDOR 26), wholesale outlets (FDOR 29),

florists/greenhouses (FDOR 30), open stadiums (FDOR 31), enclosed theaters (FDOR 32),

tourist attractions (FDOR 35), camps (FDOR 36), and race tracks (FDOR 37), since data on

these commercial subsectors was unavailable from HCWRS. The absent subsectors for GRU are

open stadiums (FDOR 31), tourist attractions (FDOR 35), camps (FDOR 36), and race tracks

(FDOR 37).

Total Commercial Coefficient Calculation

The total average commercial water use coefficient, located at the bottom of Table 5-1 and

Table 5-2, is area weighted, and is thus heavily dependent on the commercial land use mix for a

given utility. As shown in Equation 5-6, for the total average water use coefficient, a weighted

average based on the total heated area of the two-digit FDOR subsectors is used. For the area

conversion coefficients, a similar operation follows, though the total effective area is used in the

weighting (Equation 5-7).

S(AWU HAotl,j ) (5-6)
AWSector = J HAol,
J-1

Where:
AWUsector = total sector weighted average water use coefficient (gallons/heated sq. ft./day)
AWUj = weighted average water use coefficient of subsector j (gallons/heated sq. ft./day)
HATotalj = total heated square footage of all parcels in subsector j (sq. ft.)









(HA/EA EAro. ,)
HA / EAsctor = J
C (EAotal,,) (5-7)
J-1

Where:
HA/EAsector = total sector effective area to heated area conversion coefficient
HA/EAj = effective area to heated area conversion coefficient of subsector j
EATotalj = total effective square footage of all parcels in subsectorj (sq. ft.)

If the same subsector water use coefficients are applied to another utility where only the

heated area is known, then the area weighted average for the commercial sector would reflect the

relative importance of the commercial subsectors. Indeed, the area weighting provides an

important improvement in the accuracy of CII estimates since the sizes of the various activities

are included directly in the calculations. In the case of HCWRS, the point estimate for total

commercial average water use is the sum of the water use, divided by the summed heated

building values over the 1,177 commercial parcels in HCWRS. Following this calculation, the

HCWRS commercial sector uses an area-weighted average of 0.135 gallons of water per square

foot of heated building area per day (gal/hsf/d), which is close to the GRU total commercial

average water use coefficient of 0.130 gal/hsf/d. Total base and seasonal water use coefficients

are dependent on the overall sectoral time series.

Commercial Subsector Analysis at the Utility Level

The two-digit breakdown in Table 5-1 and Table 5-2, allows for the evaluation of which

subsectors are the most important as judged by the combination of their water use rate and their

size as measured by heated area. If HCWRS was to pursue water conservation in the commercial

sector, for example, the level of disaggregation in Table 5-1 can justify targeting a specific class

of customers. Restaurants (FDOR 21) have the highest rate of water use per square foot of heated

area. Though their heated area only accounts for 1.8% of the heated area for the commercial

sector, their overall water use totals 10.9% of the commercial water use. Supermarkets/









convenience stores (FDOR 14) have many customers in that subsector and a relatively high

water use rate. For GRU (Table 5-2), restaurants (FDOR 21) and fast-food restaurants (FDOR

22) once again have a high water use rate, and make up a significant fraction (15.1%) of

commercial water use. The largest commercial water use subsector for GRU is hotels/motels

(FDOR 39), which utilizes close to 18% of total commercial water use for the utility, at an

average rate of 0.220 gal/hsf/d. Such sectors could be analyzed through further disaggregation

for water conservation potential.

The percent seasonal column in Table 5-1 and Table 5-2 allows for the evaluation of

seasonality at the subsector level. In Chapter 3 it was determined that the aggregate CII sectors

of water use lack significant variability, and thus the argument was made that point estimates of

water use were reasonable. At the commercial, industrial, and institutional level of aggregation

however, it is impossible to determine whether the individual subsectors within the aggregated

sectors actually lack seasonal variability. Each subsector might be highly variable, but at

different points in the time series. Thus, the aggregation of the individual subsector time series

might mute the overall sector time series. The percent seasonal estimates for the commercial

subsectors in Table 5-1 and Table 5-2 indicate small seasonal components of water use

throughout the majority of subsectors, further justifying the use of point estimates of water use.

The May peak provides a good indication of the extent to which a sector impacts the utility

wide peak. The May peak is caused primarily by irrigation needs during the spring dry season.

CII use may be lower in May since a significant number of winter residents have left Florida. In

this case, the CII users may not be significant contributors to the May peak.

Commercial Subsector Analysis across Utilities

Through the availability of water use and property attribute data from multiple utilities,

evaluation of the inter-utility water use coefficient differences between subsectors is possible.









Heated area and water use comparisons across utilities were made in Chapters 3 and 4,

respectively. Inter-utility differences between commercial water use coefficients at the subsector

level show that for subsectors where both utilities had a significant sample size, water use

coefficients were largely within an order of magnitude (Table 5-3).

A significant sample size is important in analyzing CII water use given its heterogeneous

nature of customers, and skewed measures of size. These water use coefficients and heated area

statistics would greatly benefit from increased sample sizes provided by other utilities,

particularly for the insignificantly sampled subsectors. Heated areas, as shown in Chapter 3, may

vary widely within an FDOR subsector. The limited number of 2-digit CII FDOR subsectors

ensures grouping of multiple facility types with differing drivers of water use.

Disaggregated groupings within subsectors are possible by developing size categories

based on heated building area. Year built of a facility might also affect water use, given the

requirement or availability of certain end-use devices at the time of construction. For example,

the residential sector is broken up into three age groups (pre-1983, 1983-1994, 1995-present)

corresponding with State regulations requiring minimum plumbing fixture water efficiencies.

Even facilities within the same subsector might offer new services requiring different end-use

water devices as they respond to changing conditions. Predicting what fixture types are prevalent

in certain customer groups greatly improves estimates of water use, as well as facilitates the

weighing of water conservation options.

Two subsectors of note in Table 5-1, Table 5-2, and Table 5-3 are financial institutions

(FDOR 23) and service shops (FDOR 25). The average water use coefficient for financial

institutions in GRU is over two times larger than that of HCWRS. Though their average year

built is nearly identical, GRU financial institutions are close to 2,000 square feet larger than









those in HCWRS, and thus likely represent a different customer subset of financial institutions.

Service shops in HCWRS use nearly four times as much water per square foot on the average as

those in GRU. Service shops in HCWRS however, are slightly older than and nearly two and a

half times as large as those in GRU. Thus, once again, service stations across these two utilities

are dissimilar in both their property attributes and in the way they use water. As described in

Chapter 3, the appendix presents the subsector distributions of heated areas for the State and

combined utilities, which can be used to gauge the representativeness of the subsector samples

from HCWRS and GRU to the entire State's total of CII parcels.

Aggregation of Commercial Utility Data

Given that the water use and property attributes of the subsectors are similar across

utilities, in order to increase our overall sample size and include as many different customer

types as possible, the combined commercial coefficients including all sampled parcels in

HCWRS and GRU are presented in Table 5-4. The top commercial water users for the combined

utilities are community shopping centers (FDOR 16), restaurants (FDOR 21), and hotels/motels.

Data for the following commercial categories was unavailable from HCWRS or GRU: open

stadiums (FDOR 31), tourist attractions (FDOR 35), camps (FDOR 36), and race tracks (FDOR

37). The combined parcel-level water use and property attribute statistics presented in Table 5-4

provide a more comprehensive view of the commercial water use sector in Florida.

Industrial Sector

The available industrial subsectors, along with their water use and area conversion

coefficients are presented in Table 5-5 and Table 5-6 for HCWRS and GRU, respectively. Like

the commercial sector, the total industrial water use coefficients are a calculated heated area

weighted average of the eleven FDOR codes that make up that sector. The industrial sector

accounts for the least amount of CII water users in both HCWRS and GRU. With a total









weighted average water use coefficient of 0.042 gal/hsf/d (HCWRS) and 0.070 gal/hsf/d (GRU),

this sector also presents the smallest rate of water use amongst the CII sectors. Not shown in

Table 5-5 are lumber yards (FDOR 43), packing plants (FDOR 44), bottlers/canneries (FDOR

45), food processing (FDOR 46), and mining/petroleum (FDOR 92). Packing plants (FDOR 44),

food processing (FDOR 46), and mining/petroleum (FDOR 92) were the unavailable industrial

subsectors from GRU. Industrial water use statistics are only for those industries that are served

by the public water supply system. Many larger industries are self-supplied and need to be

evaluated separately.

Industrial Subsector Analysis at the Utility Level

The largest industrial subsector in terms of parcel count and heated building area for both

HCWRS and GRU is warehousing/distribution (FDOR 48), despite its relatively small average

water use coefficient. Given that the water use calculation is a product of a sectors size (total

heated building area) and water use coefficient, it is not surprising that warehousing/distribution

is by far the largest and second largest industrial water user in HCWRS and GRU at 60% and

42% of total industrial water use, respectively. Throughout the industrial subsectors, large

average heated areas and small water use coefficients are prevalent. Given this fact, it seems fair

to infer that these customers likely do not utilize their potable water connections for industrial

processes.

Industrial Subsector Analysis across Utilities

The industrial subsectors lacked significant sample sizes for inter-utility comparisons of

water use and property attributes. For the two subsectors with significant sample sizes across

utilities, light manufacturing (FDOR 23) and warehousing/distribution (FDOR 48), water use

and property attributes vary considerably, evidence of the even greater heterogeneity of

industrial water users (Table 5-7). The industrial sector has the fewest number of FDOR









categories across CII with only 11 subsectors. Despite having such few subsectors, the industrial

sector comprises the most diverse number of customer types and drivers of demand in CII. Light

manufacturing (FDOR 23) facilities in GRU use on the average 6.3 times more water per square

foot than those in HCWRS. In addition, GRU light manufacturing customers are nearly 60,000

square feet smaller than those in HCWRS, a clear indication that these facilities across the two

utilities vary significantly. Warehousing/distribution facilities in HCWRS are over twice the size

as those in GRU, while utilizing nearly half as much water use per square foot on the average.

The lack of disaggregation inherent in the FDOR classification of CII ensure that multiple

customer types with drastically different property attributes and drivers of demand are grouped

within single 2-digit FDOR subsectors.

Aggregation of Industrial Utility Data

The combined industrial coefficients including all sampled parcels in HCWRS and GRU

are shown in Table 5-8. The top industrial water users for the combined utilities are

warehousing/distribution (FDOR 48), mineral processing (FDOR 46), and light manufacturing

(FDOR 41). Data for the following industrial categories was unavailable from HCWRS or GRU:

packing plants (FDOR 44), food processing (FDOR 46), and mining/petroleum (FDOR 92). In

this more comprehensive view of industrial public water use in the state of Florida, it is once

again apparent that small water use coefficients and large heated areas characterize industrial

customers.

Institutional Sector

The institutional sector is disaggregated into 18 FDOR subsectors. The water use and area

conversion coefficients for these institutional subsectors are presented in Table 5-9 for HCWRS,

and Table 5-10 for GRU. Not shown in Table 5-9 are homes for the aged (FDOR 74),

sanitariums/convalescents (FDOR 78), cultural organizations (FDOR 79), military (FDOR 81),









parks and recreation (FDOR 82), colleges (FDOR 84), hospitals (FDOR 85), government leased

interested (FDOR 90), and outdoor recreational (FDOR 97). Military (FDOR 81), and

government leased interested (FDOR 90) were the unavailable institutional subsectors from

GRU.

Institutional Subsector Analysis at the Utility Level

The largest institutional water user for HCWRS is public county schools (FDOR 83).

Though this subsector has a relatively small average water use coefficient of 0.069 gal/hsf/d,

public county schools in the utility have a large average heated building area of 127,905 square

feet and there are many of them. Another subsector of note is churches (FDOR 71), which make

up the largest institutional subsector for both HCWRS and GRU in terms of parcel count, while

also being a significant institutional water user for both utilities.

Institutional Subsector Analysis across Utilities

The institutional subsectors largely share similar water use coefficients, largely within a

measure of magnitude (Table 5-11). Throughout the subsectors, average water use coefficients

are stable towards the small side, and seasonality appears to be minimal. The relatively small

water use coefficients indicate several possibilities in terms of end uses, such as utilization of

private wells for irrigation. The small seasonal measures also justify the use of point estimates of

water use.

Aggregation of Institutional Utility Data

Given that the water use and property attributes of the subsectors are similar across

utilities, in order to increase our overall sample size and include as many different customer

types as possible, the combined institutional coefficients including all sampled parcels in

HCWRS and GRU are presented in Table 5-12. The top institutional water users for the

combined utilities are public county schools (FDOR 83), churches (FDOR 71), and private









hospitals (FDOR 73). Data for the following institutional categories was unavailable from

HCWRS or GRU: military (FDOR 81), and government leased interested (FDOR 90).

Coefficient Comparison with other Studies

Other studies have also developed CII water use coefficients. The American Water Works

Association Research Foundation's CIEnd Uses of Water (2000) analyzed the water use patterns

of five major CII subsectors, including supermarkets, office buildings, restaurants, hotels, and

schools. The Colorado WaterWise Council's Industrial, Commercial & Institutional Water

Conservation (2007) report, presented water use coefficients or benchmarks for restaurants,

hotels, schools and homes for the aged. Comparison across the coefficients presented in this

thesis and others developed in previous studies can be carried out by mapping the FDOR

subsectors used in this thesis accordingly: supermarkets (FDOR 14), office buildings (FDOR 17

& 18), restaurants (FDOR 21), hotels (FDOR 39), schools (FDOR 83), and homes for the aged

(FDOR 74). The comparison of CII coefficients from this and previous studies is shown in Table

5-13. For the most part, the coefficients are comparable. The percent difference between

coefficients presented in this thesis and other studies ranged from

-38% to 348% across the available subsectors. Large discrepancies in coefficients can be

attributed to the other studies being specific to the southwestern region of the United States

where other factors such as varied climatic conditions can affect water use.

Incorporation of Results into EZ Guide 2

By employing a measure of size that is standard and reliable across the CII subsectors,

along with default water use coefficients, any utility within the State can estimate the subsectoral

breakdown of CII water use within their service boundary. The FDOR database is accompanied

by polygon shapefiles that delineate every parcel in the State. This database can be queried to

determine which parcels are within the service boundaries of a given utility. South Florida Water









Management District (WMD), St. Johns River WMD, and Southwest Florida WMD provide the

water service area boundaries of utilities in their districts as polygon shapefiles available in their

respective websites to be viewed in GIS. The parcels are identified by a unique parcel

identification number which can be related to the FDOR database to find the attributes for the

parcels in the utility being analyzed. These shapefiles need to be checked to verify their

accuracy.

EZ Guide 2 utilizes the coefficients shown in Table 5-4, Table 5-8, and Table 5-12 to

estimate CII water use for any utility in the State. These coefficients are applied within the water

budget section of EZ Guide 2 (Figure 5-1). By estimating the individual water use for each CII

subsector, a utility or planning agency can develop a conservation strategy according to the

relative importance and water use intensity of its subsectors. To estimate the amount of water use

for the single and multi-family residential sectors, a similar data-driven measure of size approach

is also taken. EZ Guide 2 is available free online, and the Conserve Florida Water Clearinghouse

can assist water utilities and water management districts in generating the necessary information

(http://www.conservefloridawater.org/).











Table 5-1. Water use coefficients and sector statistics based on sample of 1,177 commercial parcels and four years of billing records
from Hillsborough County Water Resources Services
Average Weighted water use coef. % Heated % Avg.
DAverage
S Sample Avere heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
0 size area EA seasonal in use in
s Ayear built Base Seasonal sector sector
(sf) Avg. Base Seasonal Maypeak sector sector


S Stores, one-
story
12 Mixed use
13 Department
stores
Supermarkets /
14 conv.
stores
15 Regional malls
Community
16 shopping
centers
17 Office, one-
story
18 Office, multi-
18
story
19 Medical office
S Transit
20
terminals
21 Restaurants
S Fast-food
22
restaurants
S Financial
23
institutions
25 Service shops
Auto sales /
27
repair
Enclosed
32 theaters/
auditorium


1989

1975

1995


1991


9,375

13,601

130,076


0.074

0.096

0.063


0.067

0.092

0.056


0.007

0.003

0.008


0.078

0.099

0.065


5,576 0.97 0.283 0.248 0.035 0.306


1997 820,551 0.87 0.087 0.080 0.007 0.081


1989


1985

1988

1989

1979

1993

1998

1991

1984

1985


1999


39,269 0.95 0.116 0.114 0.002 0.116


5,446 0.96 0.180 0.170 0.010 0.184


38,283

6,770

10,670

5,084

3,105

4,424

2,787

9,095


0.079

0.131

0.349

0.825

0.684

0.248

0.434

0.131


0.075

0.120

0.262

0.778

0.647

0.221

0.393

0.116


0.004

0.011

0.087

0.047

0.037

0.027

0.041

0.015


0.084

0.143

0.358

0.846

0.719

0.259

0.438

0.133


97,632 0.92 0.121 0.080 0.041 0.131


9.0%

3.5%

11.9%


5.47%

7.10%

12.10%


12.2% 3.37%

8.4% 8.48%


1.5% 32.67% 28.08%


5.4% 4.28%


5.1%

8.5%

25.1%

5.7%

5.4%

10.9%

9.4%

11.5%


8.11%

3.74%

0.28%

1.79%

0.88%

1.44%

0.26%

5.45%


33.6% 0.50%


3.00%

5.04%

5.67%


7.06%

5.49%


5.69%

4.76%

3.62%

0.71%

10.92%

4.47%

2.64%

0.83%

5.30%


0.45%










Table 5-1. Continued

Average Weighted water use coef. % Heated % Avg.
Q Average
Sample Average heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
size area EA My seasonal in use in
year blt (sf) Avg. Base Seasonal peak sector sector

Nightclubs /
33 8Nightclubs 1981 3,686 1.00 0.266 0.226 0.040 0.328 15.0% 0.15% 0.30%
bars
Bowling alleys /
34 skating 1 1983 30,784 0.98 0.040 0.036 0.004 0.036 8.9% 0.16% 0.05%
rinks
Golf courses /
38 driving 20 1991 18,091 0.88 0.151 0.142 0.009 0.150 6.2% 1.87% 2.09%
ranges
39 Hotels/motels 10 1986 36,875 0.95 0.269 0.234 0.035 0.276 12.9% 1.91% 3.80%
Total
Total 1,177 1988 16,444 0.90 0.135 0.131 0.004 0.138 3.0% 100.00% 100.00%
commercial











Table 5-2. Water use coefficients and sector statistics based on sample of 1,037 commercial parcels and two years of billing records
from Gainesville Regional Utilities
Average Weighted water use coef. % Heated % Avg.
DAverage
SzSample Average heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
Size area EA seasonal in use in
year built Seasonal sector sector
(sf) Avg. Base Seasonal M peak sector sector


SStores, one-
story
12 Mixed use
13 Department
stores
Supermarkets /
14 conv.
Stores
15 Regional malls
Community
16 shopping
centers
17 Office, one-
story
18 Office, multi-
18
story
19 Medical office
STransit
20
terminals
21 Restaurants
SFast-food
22
restaurants
SFinancial
23
institutions
Insurance
24 company
offices
25 Service shops


1982

1978

1982


1984


6,532

6,390

94,109


0.120

0.074

0.025


0.109

0.069

0.014


0.010

0.005

0.011


0.127

0.072

0.027


10,065 0.90 0.130 0.106 0.024 0.132


1995 928,070 0.99 0.047 0.033 0.014 0.058


1985


1984

1986

1991

2000

1981

1989

1992


1988

1979


39,269 0.95 0.063 0.058 0.005 0.069


6,334 0.96 0.100 0.085 0.016 0.112


20,700

8,070

2,193

4,664

2,697


0.046

0.174

0.099

0.621

0.624


0.040

0.158

0.077

0.582

0.608


0.006

0.016

0.023

0.039

0.016


0.059

0.183

0.121

0.631

0.631


6,338 0.93 0.531 0.464 0.067 0.570


10,736 0.94 0.073 0.060 0.012 0.086

6,906 0.81 0.116 0.100 0.016 0.129


8.6%

7.0%

43.8%


9.68%

2.26%

0.79%


18.2% 0.51%

29.9% 7.81%


8.0% 25.78% 12.48%


15.6% 12.37%


13.1%

9.2%

22.8%

6.3%

2.6%


5.58%

10.66%

0.02%

2.04%

1.13%


12.7% 1.87%


17.0% 0.99%

13.8% 1.80%


8.86%

1.29%

0.15%


0.51%

2.80%


9.52%

1.95%

14.19%

0.01%

9.72%

5.43%

7.60%


0.55%

1.60%











Table 5-2. Continued

Average Weighted water use coef. % Heated % Avg.
Q Average
Sample Average heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
size area EA Ma seasonal in use in
yer b (sf) Avg. Base Seasonal peak sector sector

26 Service stations 5 1986 1,829 0.71 0.170 0.145 0.025 0.213 14.7% 0.08% 0.10%
Auto sales /
27 Autosales 58 1983 5,836 0.83 0.101 0.085 0.015 0.106 15.1% 2.85% 2.20%
repair
Wholesale
29 outlets 5 1972 23,700 0.76 0.025 0.021 0.004 0.030 17.7% 1.00% 0.19%
outlets
30 greenhou 2 1966 3,376 0.92 0.216 0.144 0.072 0.250 33.5% 0.06% 0.09%
greenhouses
Enclosed
32 theaters/ 2 2001 27,989 1.00 0.119 0.097 0.022 0.115 18.7% 0.47% 0.43%
auditoriums
33 Nightclubs/bars 12 1966 5,336 0.92 0.167 0.134 0.033 0.209 19.6% 0.54% 0.69%
Bowling alleys /
34 B ling ys/ 2 1988 34,410 0.93 0.037 0.032 0.005 0.033 14.2% 0.58% 0.16%
skating rinks
Golf courses /
38 driving 6 1986 9,633 0.90 0.405 0.219 0.186 0.452 46.0% 0.49% 1.51%
ranges
39 Hotels/motels 40 1981 31,626 0.94 0.220 0.186 0.034 0.236 15.7% 10.65% 17.97%
Total
commercial 1,037 1984 11,457 0.96 0.130 0.121 0.010 0.141 7.6% 100.00 100.00
commercial









Table 5-3. Percent difference between commercial water use coefficients of Hillsborough
County Water Resources Services and Gainesville Regional Utilities
FDOR Percent difference
FDOR
Description May
code Avg. Base Seasonal pea

11 Stores, one-story 62% 63% 43% 63%
12 Mixed use -23% -25% 67% -27%
13 Department stores -60% -75% 38% -58%
14 Supermarkets / conv. stores -54% -57% -31% -57%
15 Regional malls -46% -59% 100% -28%
16 Community shopping centers -46% -49% 150% -41%
17 Office, one-story -44% -50% 60% -39%
18 Office, multi-story -42% -47% 50% -30%
19 Medical office 33% 32% 45% 28%
20 Transit terminals -72% -71% -74% -66%
21 Restaurants -25% -25% -17% -25%
22 Fast-food restaurants -9% -6% -57% -12%
23 Financial institutions 114% 110% 148% 120%
24 Insurance company offices
25 Service shops -73% -75% -61% -71%
26 Service stations
27 Auto sales / repair -23% -27% 0% -20%
29 Wholesale outlets
30 Florist / greenhouses
31 Drive-in theaters / open stadiums
32 Enclosed theaters / auditoriums -2% 21% -46% -12%
33 Nightclubs/bars -37% -41% -18% -36%
34 Bowling alleys / skating rinks -8% -11% 25% -8%
35 Tourist attractions
36 Camps
37 Racetracks
38 Golf courses / driving ranges 168% 54% 1967% 201%
39 Hotels / motels -18% -21% -3% -14%
Total commercial -4% -8% 150% 2%











Table 5-4. Water use coefficients and sector statistics based on sample of 2,214 commercial parcels in Hillsborough County Water
Resources Services and Gainesville Regional Utilities
Average Weighted water use coef. % Heated % Avg.
DAverage
S Sample Avere heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
0 size area EA seasonal in use in
s Ayear built Base Seasonal sector sector
(sf) Avg. Base Seasonal peak sector sector


SStores, One-
Story
12 Mixed Use
13 Department
Stores
Supermarkets /
14 Conv.
Stores
15 Regional Malls
Community
16 Shopping
Centers
17 Office, One-
Story
SOffice, Multi-
Story
19 Medical Office
STransit
20
Terminals
21 Restaurants
SFast-Food
22
Restaurants
SFinancial
23
Institutions
Insurance
24 Company
Offices
25 Service Shops


1985

1976


7,644 0.93 0.098 0.093 0.004 0.104

11,483 0.92 0.092 0.089 0.003 0.095


1994 128,183 0.89 0.062 0.054 0.008 0.063


1991


5,795 0.93 0.270 0.238 0.032 0.291


1996 856,391 0.93 0.073 0.065 0.008 0.073


1988


1984

1987

1990

1982

1988

1994

1992


1988

1981


39,269 0.95 0.099 0.098 0.001 0.101


5,983 0.96 0.129 0.117 0.012 0.138

30,576 0.97 0.069 0.065 0.005 0.077

7,543 0.97 0.158 0.144 0.014 0.168

9,257 0.97 0.339 0.254 0.085 0.349

4,902 0.96 0.741 0.711 0.030 0.757

2,910 0.96 0.657 0.636 0.021 0.680

5,108 0.91 0.373 0.349 0.024 0.397


10,736 0.94 0.073 0.060 0.012 0.086

5,393 0.80 0.176 0.159 0.017 0.187


4.2% 7.07%

3.4% 5.26%

12.2% 7.80%


11.8% 2.28%

10.6% 8.23%


0.9% 30.05% 22.27%


9.0% 7.36%

6.7% 7.15%

8.7% 6.38%

25.0% 0.18%


4.0% 1.88% 10.47%


3.3% 0.98%

6.6% 1.60%


17.0% 0.38%

9.9% 0.85%


5.18%

3.64%

3.62%


4.62%

4.49%


7.12%

3.71%

7.56%

0.45%


4.83%

4.49%


0.21%

1.12%











Table 5-4. Continued

Average Weighted water use coef. % Heated % Avg.
Q Average
Sample Average heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
size area EA Ma seasonal in use in
yer b (sf) Avg. Base Seasonal peak sector sector

26 Service stations 5 1986 1,829 0.71 0.170 0.145 0.025 0.213 14.7% 0.03% 0.04%
Auto sales /
27 Autosales 174 1984 8,009 0.86 0.124 0.110 0.014 0.126 11.3% 4.46% 4.14%
repair
Wholesale
29 outlets 5 1972 23,700 0.76 0.025 0.021 0.004 0.030 17.7% 0.38% 0.07%
outlets
30 greenhou 2 1966 3,376 0.92 0.216 0.144 0.072 0.250 33.5% 0.02% 0.04%
greenhouses
Enclosed
32 theaters/ 3 2000 51,203 0.94 0.120 0.095 0.025 0.125 20.9% 0.49% 0.44%
auditoriums
33 Nightclubs/bars 20 1972 4,676 0.95 0.198 0.164 0.034 0.247 17.3% 0.30% 0.44%
Bowling alleys /
34 B ling ys/ 3 1986 33,201 0.96 0.038 0.033 0.004 0.034 11.3% 0.32% 0.09%
skating rinks
Golf courses /
38 driving 26 1990 16,139 0.89 0.186 0.161 0.025 0.191 13.4% 1.34% 1.87%
ranges
39 Hotels/motels 50 1982 32,676 0.95 0.231 0.206 0.025 0.245 10.9% 5.23% 9.08%
Total
Tommercia 2,214 1986 14,108 0.93 0.133 0.129 0.004 0.139 2.8% 100.00% 100.00%
commercial










Table 5-5. Water use coefficients and sector statistics based on sample of 163 industrial parcels and four years of billing records from
Hillsborough County Water Resources Services
Average Weighted water use coef. % Heated % Avg.
DAverage
z Sample ve ce heated HA (gallons/heated square foot/day) Percent area Water
w Description effective
O size area EA seasonal in use in
year built May
year built (s) Avg. Base Seasonal peak sector sector

41 Light 10 1982 80,659 0.90 0.018 0.014 0.004 0.024 23.7% 11.90% 5.16%
manufacturing
42 Heavy industrial 2 1995 44,432 0.98 0.040 0.028 0.012 0.050 30.2% 1.31% 1.22%
47 Mineralprocessing 8 1972 102,201 0.98 0.171 0.162 0.009 0.162 5.5% 12.06% 48.63%
Warehousing /
48 istribuion 118 1990 41,608 0.97 0.025 0.022 0.003 0.025 10.9% 72.41% 41.77%
distribution
49 Open storage 17 1979 1,860 0.97 0.203 0.175 0.027 0.213 13.5% 0.47% 2.23%
91 Utility, gas & elec. 8 1974 15,714 0.96 0.023 0.021 0.002 0.025 7.0% 1.85% 0.98%
Total industrial 163 1987 41,596 0.96 0.042 0.039 0.003 0.043 7.8% 100.00% 100.00%










Table 5-6. Water use coefficients and sector statistics based on sample of 144 industrial parcels and two years of billing records from
Gainesville Regional Utilities
Average Weighted water use coef. % Heated % Avg.
DAverage
z Sample ve ce heated HA (gallons/heated square foot/day) Percent area Water
w Description effective
O size area EA seasonal in use in
year built May
year built (s) Avg. Base Seasonal peak sector sector

41 Light 23 1979 20,851 0.90 0.116 0.095 0.021 0.112 18.3% 17.74% 29.69%
manufacturing
42 Heavy industrial 1 1998 41,893 0.72 0.033 0.024 0.009 0.031 26.7% 1.55% 0.74%
43 Lumberyards 1 1990 18,488 0.96 0.582 0.361 0.222 0.639 38.1% 0.68% 5.72%
45 Bottler/canneries 1 1969 34,541 0.79 0.010 0.007 0.003 0.010 31.9% 1.28% 0.18%
47 Mineralprocessing 2 1984 23,001 0.95 0.006 0.003 0.003 0.007 42.9% 1.70% 0.15%
Warehousing /
48 areiousibion 110 1985 16,966 0.90 0.061 0.050 0.011 0.069 17.6% 69.02% 60.33%
distribution
49 Open storage 2 1999 7,589 0.97 0.046 0.022 0.025 0.078 53.0% 0.56% 0.37%
91 Utility, gas & elec. 4 1983 50,505 0.89 0.026 0.015 0.011 0.020 42.1% 7.47% 2.82%
Total industrial 144 1984 18,777 0.85 0.070 0.058 0.012 0.075 17.2% 100.00% 100.00%









Table 5-7. Percent difference between industrial water use coefficients of Hillsborough County
Water Resources Services and Gainesville Regional Utilities
FDOR Percent difference
FDOR
Description May
code Avg. Base Seasonal May
peak
41 Light manufacturing 544% 579% 425% 367%
42 Heavy industrial -18% -14% -25% -38%
43 Lumber yards
44 Packing plants
45 Bottler / canneries
46 Food processing
47 Mineral processing -96% -98% -67% -96%
48 Warehousing/ distribution 144% 127% 267% 176%
49 Open storage -77% -87% -7% -63%
91 Utility, gas & elec. 13% -29% 450% -20%
92 Mining / petroleum
Total industrial 67% 49% 300% 74%










Table 5-8. Water use coefficients and sector statistics based on sample of 307 industrial parcels in Hillsborough County Water
Resources Services and Gainesville Regional Utilities
Average Weighted water use coef. % Heated % Avg.
DAverage
S Sample Avere heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
O size area EA seasonal in use in
year built May
yearblt (sf) Avg. Base Seasonal peak sector sector

41 Light 33 1979 38,975 0.90 0.055 0.044 0.011 0.057 19.4% 13.56% 14.85%
manufacturing
42 Heavy industrial 3 1996 43,586 0.83 0.038 0.031 0.006 0.044 16.6% 1.38% 1.03%
43 Lumberyards 1 1990 18,488 0.96 0.582 0.361 0.222 0.639 38.1% 0.19% 2.26%
45 Bottler/canneries 1 1969 34,541 0.79 0.010 0.007 0.003 0.010 31.9% 0.36% 0.07%
47 Mineral processing 10 1974 86,361 0.98 0.163 0.154 0.009 0.154 5.5% 9.11% 29.48%
Warehousing /
48 dareiousibin 228 1988 29,719 0.95 0.035 0.031 0.003 0.037 9.1% 71.45% 49.10%
distribution
49 Open storage 19 1981 2,463 0.97 0.152 0.132 0.020 0.169 13.2% 0.49% 1.49%
91 Utility, gas & elec. 12 1977 27,311 0.90 0.025 0.018 0.007 0.022 26.3% 3.46% 1.71%
Total industrial 307 1986 30,893 0.91 0.050 0.045 0.005 0.052 10.6% 100.00% 100.00%










Table 5-9. Water use coefficients and sector statistics based on sample of 428 institutional parcels and four years of billing records
from Hillsborough County Water Resources Services
Average Weighted water use coef. % Heated % Avg.
DAverage
z Sample Avere heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
0 size area EA seasonal in use in
year built May
yea lt (sf) Avg. Base Seasonal peak sector sector

71 Churches 221 1957 12,838 0.94 0.042 0.038 0.004 0.047 9.2% 24.56% 13.78%
Private schools
72 & colleges 65 1986 7,309 0.95 0.148 0.131 0.016 0.166 11.1% 4.11% 8.15%
& colleges
73 Private hospitals 2 1997 424,916 0.98 0.086 0.078 0.008 0.083 9.2% 7.36% 8.49%

75 Orphanages 67 1987 9,741 0.95 0.188 0.163 0.025 0.204 13.1% 5.65% 14.22%
non-profits
Mortuaries /
76 Mortuares 7 1989 7,560 0.88 0.063 0.048 0.015 0.073 23.3% 0.46% 0.39%
cemeteries
77 ubsunion 15 1981 10,654 0.83 0.159 0.141 0.018 0.166 11.5% 1.38% 2.95%
halls
83 Publiccounty 51 1987 127,905 0.98 0.069 0.059 0.010 0.075 14.8% 56.47% 52.03%
schools
Total
otal 428 1971 26,988 0.94 0.075 0.066 0.008 0.081 11.1% 100.00% 100.00%
institutional










Table 5-10. Water use coefficients and sector statistics based on sample of 256 institutional parcels and two years of billing records
from Gainesville Regional Utilities
Average Weighted water use coef. % Heated % Avg.
DAverage
S Sample efece heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
size area EA seasonal in use in
year built May
yearbuilt (sf) Avg. Base Seasonal M peak sector sector

71 Churches 116 1979 13,555 0.96 0.063 0.054 0.009 0.070 14.0% 23.27% 17.26%
Private schools &
72 vateschools 46 1981 9,149 0.95 0.099 0.085 0.014 0.118 14.1% 6.23% 7.35%
colleges
73 Private hospitals 4 1988 141,224 0.96 0.137 0.112 0.025 0.141 18.0% 8.36% 13.55%
Homes for the
74 mesforthe 12 1993 116,675 0.92 0.101 0.087 0.013 0.108 13.4% 20.72% 24.76%
aged
75 Orphanages/non- 15 1985 10,894 0.96 0.123 0.104 0.019 0.137 15.5% 2.42% 3.52%
profits
Mortuaries /
76 ceeteries 6 1974 5,286 0.89 0.296 0.265 0.031 0.294 10.5% 0.47% 1.65%
cemeteries
Clubs / union
77 ubsuon 40 1935 13,308 0.99 0.197 0.154 0.044 0.178 22.1% 7.88% 18.46%
halls
Sanitariums /
78 convalescs 1 1975 43,505 1.00 0.378 0.328 0.050 0.360 13.3% 0.64% 2.89%
convalescents
Cultural
79 cultural 1 1989 2,302 0.74 0.163 0.154 0.009 0.159 5.7% 0.03% 0.07%
organizations
Parks and
82 recreatn 1 1998 5,288 0.92 0.229 0.174 0.055 0.258 24.0% 0.08% 0.21%
recreation
83 Public county 1 1980 59,448 0.98 0.025 0.015 0.010 0.026 38.9% 0.88% 0.26%
schools
84 Colleges 9 1970 175,786 1.00 0.035 0.026 0.008 0.027 23.5% 23.41% 9.59%
85 Hospitals 3 1988 123,175 0.96 0.006 0.002 0.004 0.009 66.9% 5.47% 0.39%
Outdoor
97 recreaio 1 1950 9,233 0.92 0.026 0.015 0.011 0.037 42.6% 0.14% 0.04%
T recreational 256 1973 2
Total institutional 256 1973 26,395 0.96 0.084 0.076 0.009 0.086 10.1% 100.00% 100.00%









Table 5-11.


Percent difference between institutional water use coefficients of Hillsborough
County Water Resources Services and Gainesville Regional Utilities


FDOR Percent difference
FDOR
Description May
code Avg. Base Seasonal pea

71 Churches 50% 42% 125% 49%
72 Private schools & colleges -33% -35% -13% -29%
73 Private hospitals 59% 44% 213% 70%
74 Homes for the aged
75 Orphanages/ non-profits -35% -36% -24% -33%
76 Mortuaries/ cemeteries 370% 452% 107% 303%
77 Clubs / union halls 24% 9% 144% 7%
78 Sanitariums / convalescents
79 Cultural organizations
81 Military
82 Parks and recreation
83 Public county schools -64% -75% 0% -65%
84 Colleges
85 Hospitals
90 Gov. leased interests
97 Outdoor recreational
Total institutional 12% 15% 13% 6%










Table 5-12. Water use coefficients and sector statistics based on sample of 684 institutional parcels in Hillsborough County Water
Resources Services and Gainesville Regional Utilities
Average Weighted water use coef. % Heated % Avg.
DAverage
S Sample efece heated HA (gallons/heated square foot/day) Percent area Water
a Description effective
size area EA seasonal in use in
year built May
yearbuilt (sf) Avg. Base Seasonal M peak sector sector

71 Churches 337 1965 13,085 0.95 0.049 0.044 0.005 0.055 10.0% 24.09% 15.16%
Private schools &
72 vateschools 111 1984 8,071 0.95 0.125 0.110 0.015 0.143 12.2% 4.89% 7.83%
colleges
73 Private hospitals 6 1991 235,788 0.97 0.106 0.096 0.010 0.106 9.6% 7.73% 10.50%
Homes for the
74 mesforthe 12 1993 116,675 0.92 0.101 0.087 0.013 0.108 13.4% 7.65% 9.85%
aged
75 Orphanages/non- 82 1986 9,952 0.91 0.175 0.151 0.023 0.191 13.4% 4.46% 9.96%
profits
Mortuaries /
76 ceeteries 13 1982 6,511 0.88 0.150 0.130 0.021 0.156 13.9% 0.46% 0.89%
cemeteries
Clubs / union
77 lubs on 55 1947 12,584 0.91 0.189 0.152 0.037 0.175 19.5% 3.78% 9.12%
halls
Sanitariums /
78 convalescs 1 1975 43,505 1.00 0.378 0.328 0.050 0.360 13.3% 0.24% 1.15%
convalescents
Cultural
79 cultural 1 1989 2,302 0.74 0.163 0.154 0.009 0.159 5.7% 0.01% 0.03%
organizations
Parks and
82 recreatn 1 1998 5,288 0.92 0.229 0.174 0.055 0.258 24.0% 0.03% 0.08%
recreation
83 Public county 52 1987 126,588 0.98 0.068 0.058 0.010 0.074 14.7% 35.96% 31.44%
schools
84 Colleges 9 1970 175,786 1.00 0.035 0.026 0.008 0.027 23.5% 8.64% 3.82%
85 Hospitals 3 1988 123,175 0.96 0.006 0.002 0.004 0.009 66.9% 2.02% 0.15%
Outdoor
97 recreaio 1 1950 9,233 0.92 0.026 0.015 0.011 0.037 42.6% 0.05% 0.02%
T recreational 684 1972 2
Total institutional 684 1972 26,766 0.95 0.078 0.070 0.008 0.083 9.8% 100.00% 100.00%









Table 5-13. Water use coefficient comparison to other studies on CII water use
Florida study AWWARF CI end 2007 Colorado
coefficients uses study* Waterwise**
(gal/hsf/d) (gal/hsf/d) % Difference (gal/hsf/d) % Difference
Supermarkets (FDOR 14) 0.270 0.223 -17%
Office buildings (FDOR 17& 18) 0.100 0.103 3%
Restaurants (FDOR 21) 0.741 0.845 14% 0.526 -29%
Hotels (FDOR 39) 0.231 0.248 7% 0.329 42%
Schools (FDOR 83) 0.068 0.306 348% 0.042 -38%
Homes for the aged (FDOR 74) 0.101 0.219 118%
Dziegielewski et al. 2000
** Colorado Waterwise 2007















Conv owd wa
Cn..c. Flodd Wctmr


W Multi-Family
Unaccountd 2%
For Institutional. Industrial
8% 1%
Figure 3.2.1 Calibrated Water Budget by Sector


Table 3.2.1 Percentage and gpcd Summary by Sector

% or Total Breakdown ofr eak n
Secor Galmtd Sq.
edo Water Use Gross gpcd RMonth

Single Family
--------- 71.3% 95 .... 5.01........
Single Family Indoor 38.2% 51 268

Single Family Outdoor 33.1% 44 232
Multi-Fam!ly 1.6% 2 0.96
Commerdal 14.7% 20 4.72
industrial 1.2% 2 0.78
insLttuLonal 3.2% 4 3.38
Unaccounted For 8.0% 11 0.37
TOTAL 100.0% 134 465


EZ Guide 2 water budget summary for a utility in South Florida


Figure 5-1.









CHAPTER 6
SUMMARY, CONCLUSIONS, AND NEED FOR ADDITIONAL RESEARCH

This commercial, industrial, and institutional (CII) water use estimating method should

offer a significant improvement over traditional methods of estimating CII water use by

combining water billing records with parcel-level land use databases, principally Florida

Department of Revenue (FDOR). These databases allow for the size of subsectors and their

activity coefficients to be developed by parcel-level data, which is a finer resolution than Traffic

Analysis Zones (TAZ) or Census block data. They also provide a standardized classification

system to categorize land uses across the State. The 55 CII FDOR land use subsectors allow

water users to be classified within various degrees of disaggregation based on the level of

homogeneity desired in a sector. The FDOR database is public information and is capable of

being linked to any utility billing records through the parcel identification number.

Existing utility linkages between the FDOR and water billing records are available for only

a few utilities in the State. This thesis develops activity water use coefficients and presents a

methodology to carry out water budgets regardless of this link. By employing a measure of size

that is standard and reliable across the CII sectors, along with default water use coefficients, any

utility within the State is able to develop a water budget. The water budget is an essential tool to

investigate the gross gpcd of a utility by sectors. It also allows a utility to estimate how the water

in their service boundary is being used. This information about water users is crucial to develop

an accurate forecast for water use or in planning and evaluating conservation efforts.

Water use coefficients presented in this thesis were calculated from historical billing

records from Hillsborough County Water Resources Services (HCWRS) and Gainesville

Regional Utilities (GRU), and heated areas from Florida County Property Appraisers (FCPA).

HCWRS and GRU provided for a relatively large cross-section of customers and long time









series. More utilities from across the State should be incorporated into the analysis to account for

regional differences in water use, as well as increase the sample size for the various water use

subsectors.

The extensive amount of data available from FDOR/FCPA allow for substantial amounts

of future refinements. The available data can be used to improve the accuracy of water estimates,

as well as further disaggregate water use to the end use or process level. For example, FDOR

incorporates year built in its database. This information can be used to carry out time series

analysis and find trends for both heated areas and activity water use coefficients over time. This

analysis improves the accuracy of water use estimates and forecasts, and could provide insight

into end uses. Also, by estimating irrigated area and analyzing the billing records for seasonal

components, it should be possible to break down water use further into the seasonally dependent

irrigation and cooling end uses. For this reason, seasonal water use might best be normalized by

a measure of size such as irrigable area, which is available through the FCPA as presented in

Chapter 3.

Previous studies (Dziegielewski et al. 2000, U.S. EPA 2009, 1997) present valuable

information on the breakdown of end uses within CII water use. Future work should include

estimates on the number, efficiency, and frequency of use of water using devices in the CII

subsectors. Such estimates should include both indoor domestic uses such as toilets, urinals, and

faucets, as well as outdoor uses such as irrigation application rates based on estimates of irrigable

area and cooling water use for CII subsectors where applicable. End use estimates should be

linked with available best management practices, and incorporated with cost/benefit data to

optimize for the best blend of water conservation controls. Future work should also include a

study to analyze the accuracy of the water use estimates described in this thesis and the reliability









of savings with conservation efforts. Such a study would allow for a measure of uncertainty to be

associated with these estimates.

The availability of the FDOR database provides a major improvement in our ability to

estimate CII water use. The Conserve Florida Water Clearinghouse

(http://www.conservefloridawater.org/) is developing these water use coefficients and heated

area statistics and will make them available to interested utilities. Though seasonal components

of water use across CII subsectors were found to be minimal, base water use coefficients

presented in this thesis can be used to arrive at an estimate of non-seasonal water use for CII

customers. These non-seasonal estimates of water use should be relatively stable outside of

Florida, and thus these coefficients should provide good estimates for CII users elsewhere.

Average water use coefficients should also be applicable outside of Florida except where

landscape irrigation is a significant component of water use. Utilities should link their billing

data with the FDOR/CPA databases and share this information so that these estimates can be

improved over time.









Appendix
HEATED AREA SUB SECTOR DISTRIBUTIONS

This appendix presents the histograms for each of the 55 commercial, industrial, and

institutional (CII) subsectors in the Florida Department of Revenue (FDOR) database. Within

each figure, histograms of both the state of Florida total of heated areas per parcel and the

combined sample parcel heated areas from Hillsborough County Water Resources Services and

Gainesville Regional Utilities are presented. These histograms along with the cumulative

frequency curve, and fitted lognormal distribution equations, can be used to gauge to what extent

the sample data presented in this thesis is applicable to the rest of the State. For example, Figure

A-i presents the heated area histograms and cumulative frequency curves for the State and

combined utility FDOR 11 (store, one-story) parcels. Figure A-i, with its matching histograms,

cumulative frequency curve, and similar fitted lognormal equations, presents a CII subsector

were the combined utility sample is highly representative of the State total of FDOR 11 parcels.

A differing circumstance is presented in Figure A-3, where the sample data only appears to be

representative of a segment of the total distribution of FDOR 13 parcels in the State (i.e.,

department stores larger than 60,000 square feet).










1*


100%


80%
State (n= 39,002)
Lognorm(6205,9267.4) 60%
Lognorm(6191.4,9148.7,Shift(-9.8419))
Combined Utilities (n = 289)
Lognorm(6915.4,9009.3)
| Lognorm(6879.1,10162.8,Shift(219.34)) "
S,__- 20%

I M EP .0%

r' 1b, k, o (1 11 00,9N ob


Figure A-1.


35%


S28%

" 21%

P 14%

S7%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 11
(stores, one-story) parcels

S100%


80% a

60%

40% "

20% I

0%


Heated Area (sf)
Figure A-2. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 12 (mixed
use) parcels


25%


>. 20%

& 15%

. 10%

S5%


i State (n = 18,931)
S/ Lognorm(3979.6,4892.6)
Lognorm(3971.2,4809.7,Shift(-10.617))
m Combined Utilities (n = 143)
Lognorm(7664.1,8870.4) -
Lognorm Shift N/A
/ iiln l~~ a._ __ __..









20%


S16%

. 12%

S8%

| 4%


State (n = 829)
Lognorm(130417.5,226144.1)
Lognorm Shift N/A
0 Combined Utilities (n = 19)
Lognorm(128058.9,44261.3)
Lognorm( 105097.2,46236,Shift(23164))






- -(^*^(^ ^O^ ^^ ^(^5^(^^o00 ^O0 5^^ ^0^ ^O^ ^(^^O^ ^0^ ^O
- \


Figure A-3.


65%

S52%

39%

S26%

. 13%


Figure A-4.


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 13
(department stores) parcels


100%

80%
0 State (n = 2,571)
Lognorm(10466.9,18876) 60%
Lognorm(10491,19976.4,Shift(l 12.83))
0 Combined Utilities (n = 123)
Lognorm(4533.9,4488) 40
Lognorm(3851.9,6063.4,Shift(835.36))

20%
O%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 14
(supermarkets / conv. stores) parcels


80%

60%

40% "

20%

0%


100%









100%


80%

60%
r,

40% -

20% 2
Q


13 CO A, A A o <^ o<^o00 < 00 ^ O O0


Figure A-5.


25%

20%

15%

10%

5%

0%


Figure A-6.


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 15
(regional malls) parcels


100%

80% |

60%

40% "

20%

0%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 16
(community shopping centers) parcels


45%


36%

S27%

18%

S9%
OS


I I I


* State (n = 449)
Lognorm(504479.3,5563273.4)
Lognorm(677996.5,11046462.1 ,Shift(320.63))
* Combined Utilities (n = 3)
Lognorm N/A
Loanorm Shift N/A


State (n = 7,726)
Lognorm(42572.1,101645)
Lognorm(43500.7,109776.5,Shift(162.45))
Combined Utilities (n = 239)
Lognorm(41189.7,72650)
Lognorm(42574.9,88032.7,Shift(871.69))



(A dOO ,,, ,- ... .O .(^o^ do d A A o^ < A o d d dO
L,^ ^ 3 Z'3' i 3 R^ ^ ^^ 3'411C) 0 8^ 0 9^^^^^~Z;- ZiC~jOj~~b~RL~jO


-










1*


100%


I _-_- 80% |
State (n = 37,629)
Lognorm(4189.8,6972.1) 60
60%
Lognorm Shift N/A
| Combined Utilities (n = 384) 4
Lognorm(5592.5,6137.9)
Lognorm(5363.2,9389.2,Shift(702.9)) "
20%O

I Nip.I. ..0%


Figure A-7.


65% -

S52% -

I 39%

S26% -

S13%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 17
(office, one-story) parcels


100%


80%

60% ,
Q,
40% -

20% S


V /oU I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I V /oU



Heated Area (sf)
Figure A-8. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 18
(office, multi-story) parcels


35%


>, 28%

S21%

S14%

S7%


* State (n= 16,033)
Lognorm(13586.8,56193.4)
Lognorm(13597.5,56316,Shift(0.55288))
* Combined Utilities (n = 73)
Lognorm(31701.2,48051.2)
Lognorm(33856.1,73791.5,Shift(1683.9))










1*


100%


80%
State (n = 20,953)
Lognorm(5223.1,9767.9)
60%
Lognorm(4878.2,7364.5,Shift(-108.5)) .o
SCombined Utilities (n = 264) 40
Lognorm(6619.7,6010.2) 40
Lognorm(5952.2,7472.9,Shift(868.41)) "
20%



r b 1 1 1 0 9 5 ( 9 Z


Figure A-9.


80%

S64%

- 48%

S32%

t 16%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 19
(medical office) parcels


100%

80% ,
SState (n = 2,825) "
Lognorm(5701,102409.2) 600
Lognorm(6270.7,127055.2,Shift(0.71497))
U Combined Utilities (n = 6)
Lognorm(8147.7,16823.5) 40%
Lognorm Shift N/A

TM ------- 2-0%


Heated Area (sf)
Figure A-10. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 20 (transit
terminals) parcels


30%


> 24%

" 18%

. 12%

O 6%










15%


Figure A-11.


15%

12%

9%

6%

3%

0%


Figure A-12.


Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 21
(restaurants) parcels


E State (n = 4,377)
Lognorm(3224.7,2156.5)
Lognorm Shift N/A
I Combined Utilities (n = 105)
Lognorm(2992.3,1775.5)
Lognorm(22534,1247.4,Shift(-19623))




I- I I ILI r, rL ee H atII e d bI1 Are (f. .
Heated Area (sf)


100%

80% a

60%

40% I

20%

0%


Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 22 (fast-
food restaurants) parcels


E State (n = 7,629)
Lognorm(4487.7,3618.4)
Lognorm Shift N/A
Combined Utilities (n = 120)
Lognorm(5078.8,3785.5)
Lognorm(8645,2775.3, Shift(-3737.2))






Heated Area (sf)


. 12%

- 9%

S6%

3%


100%


80% a

60% ,

40% "

20% |
Q


0%









35%


S28%

" 21%

P 14%

0 7%


Figure A-13.


45%

> 36%

| 27%

S18%

0 9%


100%


0 State (n = 4,732)
Lognorm(6571.5,6560.8)
Lognorm Shift N/A
Combined Utilities (n = 98)
Lognorm(5030.5,2929.5)
Lognorm Shift N/A

ill-M ,-.... -,-. .


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 23
(financial institutions) parcels


100%

80%
I State (n = 283)
Lognorm(4570.4,6424.4)
60%
Lognorm(4425.7,7098.1,Shift(208.47))
S __Combined Utilities (n = 11)
40%
Lognorm(10121.5,17828.6) 0
J Lognorm(15285.7,77251.8,Shift(831.87))

20%
i l .....i. I 1 1 1 1 i I 10


Heated Area (sf)
Figure A-14. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 24
(insurance company offices) parcels


80% a

60%

40% "

20%

0%










- 1*


100%


30%

24%

18%

12%

6%

0%


Figure A-15.


65%

^ 52%

g 39%

. 26%

< 13%

0%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 25
(service shops) parcels


100%

80%

60%

40% -^

20% |

0%


Heated Area (sf)
Figure A-16. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 26
(service stations) parcels


80%
E State (n = 4,309)
Lognorm(4647.3,5794.8) -- 60%
Lognorm(4629,6006.4,Shift(53.395))
m- w -- Combined Utilities (n = 49) -- 40% '
Lognorm(5224.4,5678.8)
Lognorm(5224.9,5675.9,Shift(-1.0112)) 20%

._ oII l d, ,.. ] ... _


" State (n = 3,434)
Lognorm(2487.9,1936.4)
Lognorm Shift N/A
" Combined Utilities (n = 5)
Lognorm(1829.1,256.26)
Lognorm(2263.2,254.85,Shift(-434.2))










25%


>, 20%

- 15%

S10%

5%


Figure A-17.


25%

20% -

S15% -
10%
-
10% -

5%


100%


a State (n = 15,129)
Lognorm(6238.3,9484.1)
Lognorm(6230.3,9407.6,Shift(-6.5661))
i Combined Utilities (n = 174)
Lognorm(6043.8,7416.4)
Lognorm(6048.8,7296.1,Shift(-27.336))


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 27 (auto
sales / repair) parcels

S100%

80%
State (n = 729)
S_____Lognorm(14316,23222.4) 60%
Lognorm(14417.7,24479.5,
i i_ Shift(103.52))
Combined Utilities (n = 5) 40% '-
Lognorm(24295.6,17153.8) 2
S-- Lognorm(41463.5,13411.9, 20%
e I Shift(-17713))
A M .n 0/


z? ^y '\ ?-^y^?^y^\ ^y^^y^ ? ^y^?'^y SI, &Z, &Iy, &yo\ &I, Zy
Heated Area (sf)
Figure A-18. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 29
(wholesale outlets) parcels


80% t

60% ,

40% "

20% =

0%










55%


> 44%

I 33%

S22%

0 11%

0%


Figure A-19.


60%

S48%

" 36%

P 24%

0 12%


100%


| State (n = 352)
Lognorm(3075.9,3956.8)
Lognorm(3071.4,4261.6,Shift(56.464))
SCombined Utilities (n = 2)
Lognorm N/A
Lognorm Shift N/A





4R A,,i Z sd4dd4,d 9,lAm,. A.. .. Z.R0^dd ^s


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 30 (florist
/ greenhouses) parcels


100%

80% =

60%
t_

40% "

20% |

0%


Heated Area (sf)
Figure A-20. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 31 (drive-
in theaters / open stadiums) parcels


80% =

60%
t_

40% "

20% I

0%


I State (n = 28)
Lognorm(39865,187722.3)
Lognorm(66864.5,776190.2,Shift(702.06))
Combined Utilities (n = 0)
Lognorm N/A
Lognorm Shift N/A


I I









35%


28%

g 21%

S14%

7%


Figure A-21.


100%


80%

60% ,

40% >

20% |
U


0 State (n = 242)
Lognorm(28010.4,54268.4)
-- Lognorm(29357.2,68186.4,Shift(573.64))
0 Combined Utilities (n = 3)
-- Lognorm N/A
Lognorm Shift N/A


0^cooo II 1 nn nn Onn II m^ nn = =^ m = nn^ = _^ =^ =^ = = 0 =
3z? 3^h(~ '~ y .D ^ ?^^~LL~~ ~~6 5'8 ^yr^^ ^y^ ^^ yc^^ y

Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 32
(enclosed theaters / auditoriums) parcels


30%

> 24%

" 18%

. 12%

O 6%


100%

80% =

60% ,

40% "

20% =
U
0%


State (n = 1,914)
Lognorm(4346.6,3742.9)
Lognorm Shift N/A
Combined Utilities (n = 20)
Lognorm(4556,3802.7)
I Lognorm(3892.9,6069.7,Shift(985.81))
11 I I I I,, I j. M M.i.. I


Heated Area (sf)
Figure A-22. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 33
(nightclubs / bars) parcels


0%









70%


>, 56%

" 42%

. 28%

0 14%

0%


Figure A-23.


40%

S32% -

S24% -
-

S16% -

0 8% -


100%


* State (n = 492)
Lognorm(30297.7,81922.8)
Lognorm(24285.3,38140.6,Shift(-1017.9))
J Combined Utilities (n = 3)
Lognorm N/A
Lognorm Shift N/A


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 34
(bowling alleys / skating rinks) parcels

100%


80%

60%

40%

20% |
2o


z? y '\ p^^?^ ^ ~ P gL~j5 31 RO RL R ?^^, S, &Z, i &,0 ^ 'b&65 ?&A, ?Z
Heated Area (sf)
Figure A-24. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 35 (tourist
attractions) parcels


80%

60%

40% "

20%

0%


p r1


2


* State (n = 742)
Lognorm(18060.9,73944.1)
Lognorm(19303.9,88847.3,Shift(49.97))
* Combined Utilities (n = 0)
Lognorm N/A
Losnorm Shift N/A


------------------- --










15%

12%

9%

6%

3%

0%


Figure A-25.


75% -r


. 60%

S45%

. 30%

S15%


100%


State (n = 330)
Lognorm(8234,14388.4)
Lognorm(8076.2,12907.7,Shift(-96.975))
m Combined Utilities (n = 0)
r Lognorm N/A -
Lognorm Shift N/A

S. .


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 36
(camps) parcels

100%


80% =
* State (n = 129)
Lognorm(28736.7,139346.2) 6
Lognorm(33492.9,215636.3,Shift(195.85))
* Combined Utilities (n = 0)
Lognorm N/A -- 40%
Lognorm Shift N/A"
20%
0%


Heated Area (sf)
Figure A-26. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 37 (race
tracks) parcels


80%

60%

40% "

20% |

0%









25%

20%

15%

10%

5%

0%


Heated Area (sf)
Figure A-27. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 38 (golf
courses / driving ranges) parcels


100%

80%

I State (n= 21,702)
Lognorm(4718,18708.6) 60% .
Lognorm(4693.3,18461.7,Shift(-0.81817))
" Combined Utilities (n = 50) 40% "
Lognorm(37478.7,54315.1)
Lognorm(38324.6,33788.4,Shift(-4828.5))- 20%
0%


Heated Area (sf)
Figure A-28. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 39 (hotels
/ motels) parcels


100%

80%
E State (n = 1,392)
Lognorm(25060.1,84420.3) 6
--60%
Lognorm(20406.4,45509.6,Shift(-373.38))
* Combined Utilities (n = 26)
Lognorm(20768.6,37549) 40%
Lognorm Shift N/A
-,20%

MI I 0%


85%

^ 68%

| 51%

. 34%

S17%










55%


Figure A-29.


40%

^ 32%

" 24%

. 16%

S8%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 41 (light
manufacturing) parcels


100%



Lognorm(70712.4,196744.3)
Lognorm(69618.8,187617.9,Shift(-100.43)) 6
SCombined Utilities (n = 3)
SLognorm N/A -- 40%
Lognorm Shift N/A
20% I

i m. ... .. 0%


Heated Area (sf)
Figure A-30. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 42 (heavy
industrial) parcels


SState (n = 18,398)
Lognorm(13375.8,32016.2)
Lognorm(13469,33184,Shift(41.897))
Combined Utilities (n = 33)
-1--Lognorm(35589.2,82016.2) --
Lognorm Shift N/A




_-_] -II I I
C~u ~7~ Z j~ja i ~~~ 3 5 1~?~ ~,8'q0 C 0~0~~~U ~~~~~3~C p'R~~~


> 44%

| 33%

S22%

0 11%

0%


100%


80%

60%

40% "

20% |

0%










100%

80%

60%

40%

20%

0%


Figure A-31.


25% -

S20% -

S15% -
*_
.-0
10% -

S50


100%


I State (n = 471)
Lognorm(33928.7,86991.8)
Lognorm(33597.4,84239.6,Shift(-41.52))
Combined Utilities (n = 1)
Lognorm N/A
Lognorm Shift N/A



/I I I I ^ .. .._______________________ocococooooooo~coc~oooocooo


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 43
(lumber yards) parcels

100%

80% *
State (n = 526)
Lognorm(39030,99836.6) 60%
60%
Lognorm(37471.6,87439.9,Shift(-195.3))
Combined Utilities (n = 0)
40%
Lognorm N/A
Lognorm Shift N/A
20%
M I M M 0%


Heated Area (sf)
Figure A-32. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 44
(packing plants) parcels


80%

60%

40% "

20% |

0%













0 State (n = 90)
Lognorm(121878.7,494517.2)
Lognorm(1 15994.8,434806.5,Shift(-164.33))
_ Combined Utilities (n = 1)
Lognorm N/A
Lognorm Shift N/A


S800%

S60%

S40%

( 20%


80% =

60%

40%

20% I
20o


Heated Area (sf)
Figure A-33. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 45
(bottler / canneries) parcels

55% 1 1 100


%


80%
| State (n = 295)
Lognorm(37283.9,113069.3) 60
Lognorm(39449,135557,Shift(229))
Combined Utilities (n = 0) 400 "
Lognorm N/A
Lognorm Shift N/A
20%

I 0%


Heated Area (sf)
Figure A-34. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 46 (food
processing) parcels


S44%

33%

S22%

11%0


.Lognorm( 121.7..7 .2)


100%


100%










85%


>1 68%

. 51%

S34%

S17%


Figure A



400/


> 32%

24%

. 16%

0 8%


100%


80% =

60%

40% /

20% I


Heated Area (sf)
-35. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 47
(mineral processing) parcels


-/ 1
S-r-


100%


80% |

60%
t_

40%

20%
U
0%


Heated Area (sf)
Figure A-36. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 48
(warehousing / distribution ) parcels


* State (n = 598)
Lognorm(16233,50857.9)
Lognorm(16955,58041.8,Shift(58.741))


* Combined Utilities (n = 10)
Lognorm(128958.6,1274954)
Lognorm(15291500000,181412.3,Shift(-
15291500000)) I


0 State (n = 42,406)
Lognorm(17258.9,39654. 1)
Lognorm(17209.7,39262.3,Shift(-8.7696))
0 Combined Utilities (n = 228)
Lognorm(31655.2,54270.5)
Lognorm(31697.6,54708.6,Shift(29.426))


.........................I


I










60%


S48%

| 36%

S24%

0 12%


Figure A-37.


85%

>, 68%

1 51%

. 34%

0 17%


100%


80%

60%
t_

40% -

20% |

0%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 49 (open
storage) parcels


100%

80%
E State (n = 3,185)
Lognorm(6195.7,24310.6) 60
Lognorm(6754.5,31730.1,Shift(40.436))
* Combined Utilities (n = 12) 40%
Lognorm(23727.1,22734.2)
Lognorm(30900.2,243072.7,Shift(8314.8)) 2
20% .

0%


Heated Area (sf)
Figure A-38. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 91
(utility, gas & elec.) parcels


/ State (n = 2,190)
Lognorm(12588.5,66705)
Lognorm(7474.9,18983.8,Shift(-72.531))
S 5* Combined Utilities (n = 19)
Lognorm(2249.2,2004.3)
Lognorm(66065071.7,3037.2,Shift(-66062608))




O ^ dd d d 0^ d ^ ^ ^ gP ^ ^ 00 ^ 0^ ^^ 0^ c^ ^ 0^ ^ 0^
\?~~ 6 z^u~L \? wC~ '\? ~8 ^y^^ y^ yrzxyrri,?^^ I~i33'2'3' 1R 1611,19 RiaZ aa


_









60%


48%

" 36%

. 24%

12%


80% =

60%

40% "

20% I


Figure A-39.


30% 1


S24%

18%

S12%

6%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 92
(mining / petroleum) parcels

100%


80%
" State (n = 18,886)
Lognorm(9755.6,15478.2)
Lognorm(9492.5,14051.4,Shift(-52.418)) 60% r6
| Combined Utilities (n = 337)
Lognorm(12941.2,18379.4) -- 40%
Lognorm Shift N/A
20%

0%


Heated Area (sf)
Figure A-40. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 71
(churches) parcels


SState (n = 112)
Lognorm(14980.6,58563.9)
Lognorm(1843610000,49323.8,Shift(-1843590000))
| Combined Utilities (n = 0)
Lognorm N/A
Lognorm Shift N/A


R m A .


100%










100%


_I__


* State (n = 3,243)
Lognorm(12686.5,25968.3)
Lognorm( 12106.6,22765,Shift(-41.379))
* Combined Utilities (n = 111)
Lognorm(7458.2,7491)
Lognorm(7117.8,8751.9,Shift(537.89))


r-.I.


2
---

- -


80%

60%
.1
40%

20% |
20o


Figure A-41.


45%

S36%

" 27%

. 18%

( 9%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 72
(private schools & colleges) parcels


100%

80% M

60%
>
40% "

20% |

0%


Heated Area (sf)
Figure A-42. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 73
(private hospitals) parcels


40%


S32%

2 24%

o 16%

8%


- -


I


SA' i5 V W S, SI, Nz, oboe, &o, RA, sz, C51 ;1 (oO W (A ^
T-? 5 2'i 2\?^ 20..L3^ ZLV')1) ^t^-'t^-' ^^^ ^^^y ^^ ^*PLrC~~~ ~' O 2i













* State (n = 4,447)
Lognorm(6225.8,24872.7)
Lognorm(8023.4,54233.7,Shift(1 17.32))
* Combined Utilities (n = 12)
Lognorm(127980.6,448598.9)
Lognorm Shift N/A


80%

60%

. 40%

S20%


80% =

60%
t_

40%

20% |
U
0%


Heated Area (sf)
Figure A-43. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 74
(homes for the aged) parcels

60% 1 100'


S48%

g 36%

S24%

, 12%


%


80%
State (n = 2,214)
Lognorm(9998.3,20236.8) 60%
Lognorm(10167.6,21827.5,Shift(62.384))
U Combined Utilities (n = 82) 40%
Lognorm(9547.1,28900.4)
| ___Lognorm(9852.3,31747.8,Shift(23.7))
20%


Heated Area (sf)
Figure A-44. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 75
(orphanages / non-profits) parcels


100%


100%









30%


, 24%

" 18%

P 12%

S6%


Figure A-45.


25%

, 20%

" 15%

S10%

5%


Figure A-46.


100%


* State (n = 887)
Lognorm(6442.2,7980.9)
Lognorm Shift N/A

- Combined Utilities (n = 13)
Lognorm(6581.3,3300.4)
SLognorm(18004.1,2742.8,Shift(-
11492))





IL l l b, b, Q0 ,NY,b. N' i


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 76
(mortuaries / cemeteries) parcels


100%

80%
m State (n = 3,244)
Lognorm(6931.3,8460.6)
I60%
Lognorm Shift N/A
I Combined Utilities (n = 55) 40
Lognorm(13045.8,14582.2)
Lognorm(13149.6,13893.8,Shift(-228.41))
200%

oz'oo~ Io c I,;:oo~ I I 0%

0 AOza 0" d" 0 dd 0 \d. d4d Atzl dd 1 O( k ,d,0 I


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 77 (clubs
/ union halls) parcels


80% >

60% "

40% .|

20% 2

0%










100%


Figure A-47.


100%

S80%

" 60%

. 40%

( 20%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 78
sanitariumss / convalescents) parcels

100%

80%
State (n = 234)
Lognorm(9926.2,20234.2)
I-- 60%
Lognorm(10208.5,23085.5,Shift(1 13.43))
Combined Utilities (n = 1)
Lognorm N/A 40%
Lognorm Shift N/A
20%


Heated Area (sf)
Figure A-48. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 79
(cultural organizations) parcels


o 80%

S60%

. 40%

0 20%


] State (n = 457)
Lognorm(48378,98509.5)
Lognorm(48081.6,95982.2,Shift(-88.963))
Combined Utilities (n = 1)
Lognorm N/A
Lognorm Shift N/A




N,^ J (^~~~0 (0 O CO C O 00 & I00 1 O O (I O 600 1COO CiA(^ ( Z ( 0 pO g(OK 6N 0 (0 .
\? %? ?\foK K6 K?^;N


-i -i i .


80%
u

60%

40% "

20% |

0%


100%












0 State (n = 43)
Lognorm(68159.2,204065.6)
Lognorm(85496900000,180027.1,Shift(-
S85496900000))
Combined Utilities (n = 0)
Lognorm N/A
Lognorm Shift N/A

I .


1 80%

6 60%

. 40%

( 20%


80% |

60%

40%

20%
U
0%


Heated Area (sf)
Figure A-49. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 81
(military) parcels


100%

> 80%

" 60%

S40%

M 20%


100%

80%

60%

40% "

20% 1

0%


Heated Area (sf)
Figure A-50. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 82 (parks
and recreation) parcels


SState (n = 850)
Lognorm(6238.2,15500.5)
Lognorm(6588.5,18670.1,Shift(50.818))
SCombined Utilities (n = 1)
Lognorm N/A
Lognorm Shift N/A
:/ h -.7 m m


100%


100%









30%


24%
State (n= 2,953)
18% Lognorm(465926.8,5919503.5)
Lognorm(110758.3,142790.4,Shift(-10436))
12% Combined Utilities (n = 52)
Lognorm(130157.5,94366.6)
% Lognorm(150120.7,73721.4,Shift(-24686))
6% -

6 /o -^ d----*-*-u~dd-h d -d--1----------------------------------------A----d-
U o ~ ^^^^^^r k ^ ^f ^ ^ f i i ^f i i ^ ^^ ^^ ^
3 ^Q^^P 6 ^ ^^b^JI^,"^\\^<^^^^f^^ijil^j


Figure A-51.


70%

> 56%

S42%

. 28%

0 14%

0%


Heated Area (sf)
Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 83 (public
county schools) parcels


100%

80% =

60%
>
40% :

20% 1

0%


Heated Area (sf)
Figure A-52. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 84
(colleges) parcels


100%


80%

60%

40% '"

20% I

0%


I 1


State (n = 288)
Lognorm(423964.5,5902238.9)
Lognorm(271500.7,2238345.1,Shift(-96.53))
_--* Combined Utilities (n = 9)
Lognorm(199478.1,1154125.5)
Lognorm Shift N/A


[


I I I I I I I I I I I I I I I I I I I


I I I I










100%

80% =

60%
t_

40%

20%

0%


Heated Area (sf)
-53. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 85
(hospitals) parcels


1.-i.


* State (n = 1,276)
Lognorm(27445.3,149544.1)
Lognorm(27953.8,156847,Shift(11.64))
* Combined Utilities (n = 0)
Lognorm N/A
Lognorm Shift N/A


-i 1


100%

80%

60%

40% "

20% |
Q9


Heated Area (sf)
Figure A-54. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 90 (gov.
leased interests) parcels


2
- -

- -


70%

56%

42%

S28%

S14%


Figure A


550

> 440/


State (n = 173)
Lognorm(231446.7,1827373.5)
Lognorm(293003.6,3293062.8,Shift(523.62))
_____ Combined Utilities (n = 3)
Lognorm N/A
Lognorm Shift N/A


0o

0 "


33%

S22%

S11%









100%


I I


100%


S80% 80%
SState (n = 2,090)
Lognorm(2856.7,6198.6)
60% Lognorm(2841,6084.1,Shift(- 60%
S2.6213))
40% Combined Utilities (n = 1) -40% '"
Lognorm N/A
( 20% Lognorm Shift N/A 20%

0% -I I I 0%


Heated Area (sf)
Figure A-55. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 97
(outdoor recreational) parcels









LIST OF REFERENCES

Baumann, D., Boland, J., and Hanemann, D. (1998). Urban Water Demand Management and
Planning, McGraw-Hill, New York.

Boland, J. (1997). "Assessing Urban Water Use and the Role of Water Conservation Measures
Under Climate Uncertainty." Climatic Change, 37(1), 157-176.

CDM. (2008). "IWR-MAIN." (June 9, 2009).

Colorado WaterWise Council Benchmarking Task Force. (2007). "Industrial, Commercial &
Institutional Water Conservation."

Dziegielewski, B., and Boland, J. (1989). "Forecasting Urban Water-Use The IWR-MAIN
Model." Water Resources Bulletin, 25(1), 101-109.

Dziegielewski, B., Kiefer, J., Opitz, E., Porter, G., Lantz, G., DeOreo, W., Mayer, P., and
Nelson, J. (2000). Commercial and Institutional End Uses of Water, AWWARF, Denver,
CO.

East Bay Municipal Utility District. (2008). Watersmart Guidebook: A Water-Use Efficiency
Plan-Review Guide, EBMUD, Oakland.

Florida Department of Renevue. (2009). "Property Tax Oversight."
(June 10, 2009).

Friedman, K. (2009). "Evaluation of Indoor Urban Water Use and Water Loss Management as
Conservation Options in Florida." Master of Engineering Thesis, Dept. of Environmental
Engineering Sciences, University of Florida, Gainesville.

Hazen & Sawyer and PMCL. (2004). "The Tampa Bay Water Long-Term Demand Forecasting
Model." 2004.pdf> Clearwater, FL (June 9, 2009).

Kim, J., and McCuen, R. (1979). "Factors for Predicting Commercial Water-Use." Water
Resources Bulletin, 15(4), 1073-1080.

Maddaus, W., and Maddaus, M. (2004). "Evaluating Water Conservation Cost-Effectiveness
with an End Use Model." Water Sources Conference, Austin, TX, 13.

Marella, R. (2009). "Water Withdrawals, Use, and Trends in Florida, 2005." Dept. of the Interior,
ed., U.S. Geological Survey, Reston, 49.

Mayer, P. at al. (2005). "Water and Energy Savings from High Efficiency Fixtures and
Appliances in Single Family Homes." US EPA.

Mayer, P. W., and DeOreo, W. B. (1999). Residential End Uses of Water, American Water
Works Association Research Foundation, Denver, Colorado.


149









Mercer, L., and Morgan, D. (1974). "Estimation of Commercial, Institutional and Governmental
Water Use for Local Areas." Water Resources Bulletin, 10(4), 794-801.

Neter, J., Kutner, M., Nachtsheim, C., and Wasserman, W. (1996). Applied Linear Statistical
Models, McGraw-Hill.

Opitz, E., Langowski, J., Dziegielewski, B., Hanna-Somers, N., Willet, J., and Hauer, R. (1998).
"Forecasting Urban Water Use: Models and Applications." Urban Water Demand
Management and Planning, D. Baumann, J. Boland, and W. M. Haneman, eds., McGraw-
Hill, Inc., New York, 95-135.

Palisade Corporation. (2010). "Palisade DecisionTools."

Southwest Florida Water Management District (2006). "Regional Water Supply Plan."
(June 10, 2009).

Southwest Florida Water Management District (1997). "ICI Conservation in the Tri-County Area
of the SWFWMD." SWFWMD, Brooksville.

Stockholm Environmental Institute (2009) "Water Evaluation and Planning System (WEAP)."
(September 17, 2009).

Swank, W. T., and Schreude.H.T. (1974). "Comparison of Three Methods of Estimating Surface-
Area and Biomass for a Forest of Young Eastern White-Pine." Forest Science, 20(1), 91-
100.

U.S. Census. (2010). "2007 Economic Census." . (June
21,2010).

U.S. Census. (2009a). "County Business Patterns."
. (June 21, 2010).

U.S. Census. (2009b). "Longitudinal Employer-Household Dynamics."
. (June 21, 2010).

U.S. Environmental Protection Agency (1997) Study of Potential Water Efficiency
Improvements in Commercial Businesses. Final Report to State of California Department
of Water Resources, Sacramento.

U.S. Environmental Protection Agency (2009) Water Efficiency in the Commercial and
Institutional Sector: Considerations for a WaterSense Program. US EPA WaterSense
Program, Washington, D.C.

Wurbs, R. (1995). Water Management Models: A Guide to Software, Prentice Hall, Englewood
Cliffs.









BIOGRAPHICAL SKETCH

Miguel Morales was born in Caracas, Venezuela in 1986. At the age of seven, his family

relocated to the United States seeking greater educational and professional opportunities. When

it came time for college, Miguel's passion for the environment drove him to environmental

engineering. His environmental studies began in the Environmental Engineering Sciences (EES)

department at the University of Florida. EES provided him with a grueling and comprehensive

curriculum that quickly sparked his interest in water engineering. Miguel fueled this interest by

conducting undergraduate research on novel water treatment technologies under the advisement

of Dr. David W. Mazyck. His undergraduate research included involvement in the University

Scholars Program, and culminated with his senior thesis: Development of Silica-Titania Coated

Packing Material for use in Photocatalytic Reactors. Miguel graduated summa cum laude with a

Bachelor of Science in environmental engineering in the fall of 2008.

Having developed a passion for research, the semester following graduation Miguel joined

the Conserve Florida Water Clearinghouse, under the advisement of Dr. James P. Heaney. His

studies towards a Master of Engineering degree began in evaluating the commercial, industrial

and institutional sectors of water use for water conservation potential. Miguel received his

Master of Engineering from the University of Florida in the summer of 2010 and will continue

his doctoral work at the University of Florida.





PAGE 1

1 PARCEL LEVEL METHODOLOGY FOR ESTIMATING COMMERCIAL, INDUSTRIAL AND INSTITUTIONAL WATER USE By MIGUEL ALFREDO MORALES A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQ UIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Miguel Alfredo Morales

PAGE 3

3 To my family, for their never ending support

PAGE 4

4 ACKNOWLEDGEM ENTS I am grateful to Dr. James Heaney, my primary advisor, for his tremendous insight and assistance throughout the course of my graduate studies. I am also forever indebted to Dr. David Mazyck for granting me the opportunity to conduct undergraduate rese arch, sparking my interest in academic research, and always having my best interests at heart. I thank Dr. Ben Koopman for taking part in my advisory committee; his support was also critical. To Jackie Martin, I am greatly appreciative of our early collabo ration, but even more so of our friendship. I would also like to thank my other colleagues at the Conserve Florida Water Clearinghouse: John Palenchar, Ken Friedman, Lukasz Ziemba, Leighton Walker, Camilo Cornejo, Randy Switt, and Kristen Riley. Financial support for this research was provided by the Florida Department of Management District, and the Southwest Florida Water Management District. Critical data for this r esearch was provided by Hillsborough County Water Resources Services and Gainesville Regional Utilities. Finally, I would like to thank my family and friends. Their contribution to this work cannot be easily quantified, but without them none of this would have been possible.

PAGE 5

5 TABLE OF CONTENTS p age ACKNOWLEDGEMENTS ................................ ................................ ................................ ............. 4 TABLE OF CONTENTS ................................ ................................ ................................ ................. 5 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ....................... 11 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 12 ABSTRACT ................................ ................................ ................................ ................................ ... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ................... 15 2 LITERATURE REVIEW ................................ ................................ ................................ ........ 18 Previou s Commercial, Industrial, and Institutional Water Use Models ................................ 18 Employment Data as a Measure of Size ................................ ................................ ................. 23 3 BOTTOM UP APPROACH FOR EVA LUATING THE SIZE OF EACH ACTIVITY ........ 28 Introduction ................................ ................................ ................................ ............................. 28 Land Use Databases ................................ ................................ ................................ ................ 28 Florida Department of Revenue ................................ ................................ ...................... 28 Florida County Property Appraisers ................................ ................................ ................ 31 Relationship of Effective Area to Heated Area ................................ ................................ ...... 34 Analysis of Available Data ................................ ................................ ................................ ..... 36 Sample size ................................ ................................ ................................ ...................... 36 Effective Year Built ................................ ................................ ................................ ......... 37 Heated Area to Effective Area Calculation ................................ ................................ ..... 37 Distribution and Representativeness of Sampled Heated Areas ................................ ..... 38 4 FLORIDA CASE STUDIES ON EVALUATING WATER USE ................................ .......... 66 Utility Water Billing Databases ................................ ................................ .............................. 66 Hillsborough Coun ty Water Resources Services ................................ ............................ 66 Gainesville Regional Utilities ................................ ................................ .......................... 68 Combined Utilities ................................ ................................ ................................ ........... 70 Relationship of Heated Area to Water Use ................................ ................................ ............. 71

PAGE 6

6 5 COMMERCIAL, INDUSTRIAL, AND INSTITUTIONAL WATER USE COEFFICIENTS ................................ ................................ ................................ ..................... 87 Introduction ................................ ................................ ................................ ............................. 87 Commercial Sector ................................ ................................ ................................ ................. 89 Total Commercial Coefficient Calculation ................................ ................................ ..... 90 Commercial Subsector Analysis at the Utility Level ................................ ...................... 91 Commercial Subsector Analysis across Utilities ................................ ............................. 92 Aggregation of C ommercial Utility Data ................................ ................................ ........ 94 Industrial Sector ................................ ................................ ................................ ...................... 94 Industrial Subsector Analysis at the Utility Level ................................ ........................... 95 Industrial Subsector Analysis across Utilities ................................ ................................ 95 Aggregation of Industrial Utility Data ................................ ................................ ............ 96 Instituti onal Sector ................................ ................................ ................................ .................. 96 Institutional Subsector Analysis at the Utility Level ................................ ....................... 97 Institutional Subsector Analysis across Utilities ................................ ............................. 97 Aggregation of Institutional Utility Data ................................ ................................ ........ 97 Coefficient Comparison with other Studies ................................ ................................ ............ 98 Incorporation of Results into EZ Guide 2 ................................ ................................ ............... 98 6 SUMMARY, CONCLUSIONS, AND NEED FOR ADDITIONAL RESEARCH .............. 117 APPENDIX : HEAT ED AREA SUBSECTOR DISTRIBUTIONS ................................ ............ 120 LIST OF REFERENCES ................................ ................................ ................................ ............. 149 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 151

PAGE 7

7 LIST OF TABLES Table p age 2 1 Comparison of models in the Institute of Water Resources Municipal and Industrial Needs (IWR MAIN) model tradition ................................ ................................ ................. 27 2 2 Sources of employment data ................................ ................................ ............................. 27 3 1 Databases and parcel attributes used to develop water use and area conversion coefficients based on a sample of 3,205 commercial, industrial and institutional (CII) parcels ................................ ................................ ................................ ....................... 41 3 2 Parcel breakdown across commercial subsectors for Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilities combi ned, and the entire state of Florida ................................ ................................ ................................ ... 42 3 3 Parcel breakdown across industrial subsectors for Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilities co mbined, and the entire state of Florida ................................ ................................ ................................ ... 43 3 4 Parcel breakdown across institutional subsectors for Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilit ies combined, and the entire state of Florida ................................ ................................ ................................ ... 44 3 5 Average effective year built across commercial subsectors for Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilities combined, and the entire state of Florida ................................ ................................ ............................. 45 3 6 Average effective year built across industrial subsectors for Hillsborough County Water Resources Services, Gainesville Regi onal Utilities, the two utilities combined, and the entire state of Florida ................................ ................................ ............................. 46 3 7 Average effective year built across institutional subsectors for Hillsborough County Water Resources Services Gainesville Regional Utilities, the two utilities combined, and the entire state of Florida ................................ ................................ ............................. 46 3 8 Heated area to effective area ratio of the means calculation of commercial subsectors in Hi llsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ................................ ................................ ........................... 47 3 9 Heated area to effective area ratio of the means calculation of industrial sub sectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ................................ ................................ ................................ .. 48 3 10 Heated area to effective area ratio of the means calculation of institutional subsectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ................................ ................................ ........................... 49

PAGE 8

8 3 11 Comparison of heated area to effective area rat io of the means calculation for the commercial subsectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ................................ .......... 50 3 12 Comparison of heated area to effective area ratio of the means calculation for the industrial subsectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ................................ ............................. 51 3 13 Comparison of heated area to effective area ratio of the means calculation for the institutional subsectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ................................ .......... 51 3 14 Comparison of average heated area across the commercial subsectors of Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilities combined, and the state of Florida ................................ ................................ 52 3 15 Comparison of average heated area across the industrial subsectors of Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilities combined, and the state of Florida ................................ ................................ ..................... 54 3 16 Comparison of average heated area across the institutional subsectors of Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilitie s combined, and the state of Florida ................................ ................................ 55 3 17 Fitted lognormal probability density equations and associated Anderson Darling (A D) and Kolmogorov Smirnow (K S) statistics for State total and co mbined utility sample of commercial parcels ................................ ................................ ............................ 56 3 18 Fitted lognormal probability density equations and associated Anderson Darling (A D) and Kolmogorov Smirnow (K S) statistics for State total and combined utility sample of industrial parcels ................................ ................................ ............................... 57 3 19 Fitted lognormal probability density equations and associated Anderson Darling (A D) and Kolmogorov Smirnow (K S) statistics for Stat e total and combined utility sample of institutional parcels ................................ ................................ ........................... 58 3 20 Fitted lognormal shift probability density equations and associated Anderson Darling (A D) and Kolmogorov Smirnow (K S) sta tistics for State total and combined utility sample of commercial parcels ................................ ................................ ............................ 59 3 21 Fitted lognormal shift probability density equations and associated Anderson Darling (A D) and Kolmogorov Smir now (K S) statistics for State total and combined utility sample of industrial parcels ................................ ................................ ............................... 61 3 22 Fitted lognormal shift probability density equations and associated Anderson Darling (A D) and Ko lmogorov Smirnow (K S) statistics for State total and combined utility sample of institutional parcels ................................ ................................ ........................... 62

PAGE 9

9 4 1 Summary statistics for CII sectors in Hillsborough County Water Resources Services ... 74 4 2 Summary statistics for CII sectors in Gainesville Regional Utilities ................................ 74 4 3 Comparison of water use per commercial accou nt for Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ......... 75 4 4 Comparison of water use per industrial account for Hillsborough Count y Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ......... 77 4 5 Comparison of water use per institutional account for Hillsborough County Water Resources Serv ices, Gainesville Regional Utilities, and the two utilities combined ......... 78 4 6 Correlation matrix of Florida Department of Revenue (FDOR) and Florida County Property Appraiser ( FCPA ) property att ributes and water use for CII parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities ........ 79 4 7 Stepwise regression results for CII parcels in Hillsborough Coun ty Water Resources Services and Gainesville Regional Utilities ................................ ................................ ....... 79 4 8 of County Water Resources Services and Gainesv ille Regional Utilities .............................. 79 4 9 ANOVA for stepwise regression of CII parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities ................................ ..................... 79 4 10 Step information for stepwise regression of CII parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities ................................ .......... 80 4 11 Regres sion statistics between heated square footage and average daily water use for the FDOR CII subsectors in Hillsborough County Water Resources Services and Gainesville Regional Utilities ................................ ................................ ............................ 81 5 1 Water use coefficients and sector statistics based on sample of 1,177 commercial parcels and four years of billing records from Hillsborough County Water Resources Services ................................ ................................ ................................ ............................ 100 5 2 Water use c oefficients and sector statistics based on sample of 1,037 commercial parcels and two years of billing records from Gainesville Regional Utilities ................. 102 5 3 Percent difference between commercial water use coefficients of Hillsborough County Water Resources Services and Gainesville Regional Utilities ............................ 104 5 4 Water use coefficients and sector statistics based on sample of 2,214 commerc ial parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities ................................ ................................ ................................ ............................ 105

PAGE 10

10 5 5 Water use coefficients and sector statistics based on sample of 163 industrial parcels a nd four years of billing records from Hillsborough County Water Resources Services ................................ ................................ ................................ ............................ 107 5 6 Water use coefficients and sector statistics based on sample of 144 industrial parcels and two years of billing records from Gainesville Regional Utilities ............................. 108 5 7 Percent difference between industrial water use coefficients of Hillsborough County Water Resources Services and Gainesville Regi onal Utilities ................................ ........ 109 5 8 Water use coefficients and sector statistics based on sample of 307 industrial parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities 110 5 9 Water use coefficients and sector statistics based on sample of 428 institutional parcels and four years of billing records from Hillsborough County Water Resources Services ................................ ................................ ................................ ............................ 111 5 10 Water use coefficients and sector statistics based on sample of 256 institutional parcels and two years of billing records from Gainesville Regional Utilities ................. 112 5 11 Percent difference between institutional water use coefficients of Hillsborough County Water Resources Services and Gainesville Regional Utilities ............................ 113 5 12 Water use coefficients and sector statistics based on sample of 684 institutional parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities ................................ ................................ ................................ ............................ 114 5 13 Water use coefficient comparison to other studies on CII water use ............................... 115

PAGE 11

11 LIST OF FIGURES Figure p age 3 1 Levels of Florida Department of Revenue ( FDOR ) lan d use disaggregation into 9 residential and 55 commercial, insdustrial, and institutional ( CII ) sectors ........................ 63 3 2 Schematic of spatial and attribute database relationships to FDOR ................................ .. 63 3 3 Macro to nano scale evaluation of public water use in Florida ................................ ......... 64 3 4 Heated and effective area correlation for 3,205 CII parcels in H illsborough County Water Resources Services and Gainesville Regional Utilities ................................ .......... 64 3 5 Residual plot of heated and effective area simple linear regression for 3,205 CII parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities ................................ ................................ ................................ .............................. 65 4 1 Time series plots of monthly water use for 1,177 commercial parcels, 163 industrial parcels, and 428 institutional parcels in Hillsborough County Water Resources Services ................................ ................................ ................................ .............................. 82 4 2 Time series plots of monthly water use for 1,037 commercial parcels, 144 industrial parcels, and 256 institutional parcels in Ga inesville Regional Utilities ............................ 83 4 3 Average monthly water use per commercial account in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities com bined ......... 84 4 4 Average monthly water use per industrial account in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ......... 85 4 5 Average monthly water use per institutional account in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined ......... 86 5 1 EZ Guide 2 water budget summary for a utility in South Florida ................................ ... 116

PAGE 12

12 LIST OF ABBREVIATION S APCA Alachua County Property Appraiser CFWC Conserve Florida Water Clearinghouse CII Commercial, industrial, and institutional EA effective area FC PA Florida County Property Appraisers FDOR Florida Department of Revenue Gpcd g allons per capita per day GRU Gainesville Regional Utilities HA heated area HCPA Hillsborough County Property Appraiser HCWR S Hillsborough County Water Resources Services IWR MAIN Institute of Water Resources Municipal and Industrial Needs model PMCL Planning and Management Consulting, Ltd SIC Standard Industrial Classifications TAZ Traffic Analysis Zone

PAGE 13

13 Abstract of Thesis Presented to the Graduate School of the University of Florida i n Partial Fulfillment of the Requirements for the Degree of Master of Engineering PARCEL LEVEL METHODOLOGY FOR ESTIMATING COMMERCIAL, INDUSTRIAL AND INSTITUTIONAL WATER USE By Miguel Alfredo Morales August 2010 Chair: James P. Heaney Major: Environmental Engineering Sciences This thesis presents a new methodology to estimate commercial, industrial, and institutional (CII) water use based on evaluation of parcel level customer attribute and water use billing databases. The Florida Department of Revenue ( F DOR) and Florida County Property Appraisers ( FC PA) databases provide the heated building area and customer classifications for every CII parcel in the s tate of Florida. Linking this parcel level attribute data with parcel level water use billing data provides a major improvement in our ability to estimate CII use. Existing methods typically use the number of employees activity. E mployee data is available periodically through the U.S. Census or private surveys. Census data is not available at the parcel level and survey data are expensive to collect. Evaluation of alternative measures of size that are contained in the FDOR/FCPA dat abases and customer billing data indicate that heated area is the best single measure of size to use across the 55 CII two digit FDOR land use categories or subsectors It is relatively straightforward to link the FDOR and FCPA databases for utilities bec ause the water management districts provide information on utility boundaries. The more difficult ch allenge is to link these parcel

PAGE 14

14 level of effort depends on the type of billing system and whether common fields are available to create the necessary relational databases. Only a few Florida water utilities are known to have merged these databases. The base case for this analysis is Hillsborough County Water Resources Services and Gainesville R egional Utilities that provide a relatively large sample of 3,205 CII parcels of which 6 9 % are commercial, 10 % are industrial, and 21 % are instit utional. Property and parcel attributes of this sample dataset including heated area distributions were evaluat ed at the subsector level and compared to the state wide total s of CII parcels to gain a sense of the representativeness of the sample Then, m onthly water use was analyzed to estimate average base, seasonal, and M ay peak water use per CII parcel. Finall y, water use coefficients expressed in gallons per square foot of heated area per day were developed for each of the available FDOR CII subsectors Knowing the water use coefficient and the total heated area for each FDOR subsector it is simple to estimat e the ir total water use. The availability of the FDOR/ F CPA databases provides a major improvement in our ability to estimate CII water use. The quality of these estimates will continue to improve as more utilities link their billing data with the FDOR/ F CPA databases.

PAGE 15

15 CHAPTER 1 INTRODUCTION Municipal water use is commonly measured in gallons per capita per day (gpcd) to compare use between utilities and across water use sectors. Gpcd can be defined as net gpcd, which is water use by the residential sector only, or as gross gpcd, water use by all sectors. Gross gpcd includes total water use and water loss, and should be evaluated by a water budget to better understand the relative use by different sectors of customers. The Conserve Florida Water Clearinghouse (CFWC) EZ Guide 2 is a water planning tool to estimate water use and evaluate conservation best management practices ( http://www.conservefloridawater.org/ ) The major sectors profiled in EZ Guide 2 are: singl e family residential multi family residential, commercial, industrial, instit utional (CII), and water loss (Friedman and Heaney 2009) W ater use estimates for the CII sectors are addressed in this thesis CII water use comprised 23.5% of public water sup ply withdrawals for the state of Florida in 2005 (Marella 2009). These estimates of water use were based on county wide employment figures from the U.S. Census Bureau multiplied by water use per employee coefficients. These coefficients come from a nationw ide survey of 3,448 commercial and institutional establishments and the California Department of Water Resources (Dziegielewski and Boland, 1989). Employment estima tes of CII activity can be used for a top down estimate of water use, but in order to evaluate the water use patterns of individual sectors, a bottom up method is needed. This thesis pr esents a bottom up methodology to estimate CII water use based on parce l level land use and water billing databases. For projecting water use of future customers, utilities have historically relied on similar customers within their service area, or on water use coefficients developed through nationwide

PAGE 16

16 studies. Typically, th ese water use coefficients use number of employees as the measure of size. However, it is difficult to get this information, especially for individual parcels. Identifying similar customers can be a tedious process, and water use coefficients are often dif ficult to apply, suffering from lack of standardization of customer classifications and water use normalizations. To overcome these challenges, this thesis presents a methodology by which to estimate public supply water use for the CII sectors through publ icly available databases from the Florida Department of Revenue (FDOR) and Florida County Property Appraisers (FCPA) that provide parcel level information for 55 CII sectors. The FDOR data are available in a standard format including geo spatial information for every parcel in Florida. The FCPA data vary from county to county but it is straightforward to link the FDOR and FCPA databases. Parcel level water use data w as obtained from Hillsborough County Water Resources Services (HCWRS) and Gainesville Regiona l Utilities (GRU), and this monthly billing data was linked with the FDOR and FCPA databases. The FDOR/FCPA databases provide a standard ized classification of CII customers. Relevant parcel attributes include: building and parcel areas, year built, and spa tial location via GIS. Building area is a good normalization parameter for CII customers. Year built allows for the incorporation of historical growth patterns and can be used along with other attributes to further disaggregate customer classifications. By linking FDOR/FCPA databases to water billing data, water use coefficients for CII customers on public water supply were developed for base, average and peak flow conditions with standardized FDOR customer classifications. This methodology greatly improves their service area when geocoded billing records are not available.

PAGE 17

17 A review of previous and historically prominent CII water use models is presented in Chapter 2. The land use database s, allowing for a bottom up approach in estimating measures of size are described in Chapter 3, along with a rigorous analysis of the sample data and State total s of CII parcels. Chapter 4 presents the water billing databases employed in this thesis, justi fying the use of point estimates of water use and evaluating variables in predicting CII water use. The methodology used to develop water use coefficients is then detailed in Chapter 5, which also describes the application of the developed water use coeffi cients to EZ Guide 2. Conclusions and the need for future work are present ed in Chapter 6.

PAGE 18

18 CHAPTER 2 LITERATURE REVIEW Water use models typically forecast water use for supply planning purposes. Estimates to forecast water use include having a rate of water use f or a sector and a measure of its size throughout the planning period. The rate of water use, or activity water use coefficient, is the total water use by all customers within that sector normalized by a measure of its size. Some models may provide default coefficients for sectors of varying levels of customer disaggregation or require the user to develop their own. Total water use over n sectors is calculated using Equation 2 1. ( 2 1 ) Where: Q = water use for n sectors k = water use coefficient of sector k x k = size of sector k n = number of sectors Previous C ommercial, Industrial, and Institutional Water Use Models The historically predominant water use model was the Institute of Water Resources Municipal and I ndustrial Needs model (IWR MAIN). It was the first model to estimate commercial, industrial, and institutional ( CII ) water use empirically and disaggregate the general sector into more distinct categories. IWR MAIN was created by Hittman Associates, Inc. i n 1969, developed under the Institute of Water Resources of the U.S. Army Corps of Engineers and further refined by Planning and Management Consulting, Ltd (PMCL, now a subsidiary of CDM). IWR MAIN was a public domain model and transitioned into a propriet ary model after IWR financial support ended in the 1980s. The original IWR MAIN is no longer available and water use estimates are made using spreadsheets that replicate many of the features in the original model. In IWR MAIN, the size of each CII sector i s estimated by total employment and

PAGE 19

19 CII water use is estimated based on Standard Industrial Classifications (SIC) sectors as developed by the Department of Commerce (Opitz et al. 1998). IWR MAIN V ersion 5.1 (1988) was used to estimate the demand for water and the intensity of water use within a sector. The water use coefficients were determined by regression analysis and the explanatory variables are the number of employees, the price of water and sewer services, and the presence of conservation programs ( Boland 1997). The coefficients in Version 5.1 were developed from the weighted average of water use rates found in a nationwide survey of 3,448 commercial and institutional establishments, as well as from surveys of manufacturers by the U.S. Ce nsus Bureau and the California D epartment of Water Resources (Dziegielewski and Boland 1989) The survey to develop the Version 5.1 activity coefficients was large, random and parcel specific to produce stable national averages, but the model could not be adjusted to a The latest release of the IWR MAIN model was Version 6.1 in 1995. Version 6.1 had a more sophisticated demand forecast procedure The e mployment data used to estimate the water use coefficients were based on surveys over a dec ade of over 7,000 CII establishments across the United States and were developed by regression to account for the elasticity of CII water demand to economic and climatic independent variables. Users could choose from a library of default coefficients or in put their own estimates (Opitz et al. 1998). Th e IWR MAIN model Version 6.1 had activity coefficients for the eight major industry groups (n = 8 disaggregated CII sectors), all of the 65 two digit SIC sectors (n = 65) and all of the 417 three digit SIC se ctors (n = 417). Because the sample was collected over a period of time, the forecasts of water use were able to provide daily demands and summer and winter, or annual periods (Baumann et al. 1998; Opitz et al. 1998) These coefficients were a major

PAGE 20

20 improv ement because they were developed from a larger cross section of customers over a longer period of time. The accuracy of the model was still limited by the quality of the input data. If a user relied on the government Census for employment data, the sector al and temporal components of the model were restricted by the resolution of that data. If the user relied on survey data, their forecast was restricted by the amount of historic data they could obtain The traditional IWR MAIN model with default coefficie nts is no longer supported and the coefficients are antiquated. In 1999, PMCL released the IWR MAIN Water Demand Management Suite. This spreadsheet model removed the default water use coefficients and required the user to develop their own coefficients (CD M 2008) This software provided a different approach to modeling where the user could tailor the model to their region of implementation. Following in the tradition of IWR MAIN, Hazen and Sawyer and PMCL (2004) developed a utility wide model for Tampa Bay Water, a wholesale distributor. This model was used to estimate single family residential multi family residential and non residential water use for seven different member government planning areas. The model has an equation to estimate the non residenti al water use coefficient based on historical usage, composition of the non residential sector, local affluence and climate. A commercial vendor provided the historical employment and income data for the years 1999 to 2002 by survey. The study included 39,7 27 non residential parcels and linked the parcel data to their billing records. The values were averaged per Traffic Analysis Zone (TAZ) and combined with rainfall data to run a regression and develop the monthly water use coefficients (Hazen & Sawyer and PMCL 2004) Total employment is the size of the non residential sector used to estimate water use.

PAGE 21

21 T ampa B ay W ater found that its non residential model explained only two percent of the variation in water use (Hazen & Sawyer and PMCL 2004). The modelers at tribute this low explanatory power to the typically heterogeneous nature of non residential water use. If more specific customer classifications were developed for the non residential sector then each group of customers could be more homogeneous in their application of water. The coefficients also c a me from two years of severe drought (SWFWMD 2006) and likely do not describe average water use during a normal year. Similar to the 1999 IWR MAIN release, the Water Evaluation and Planning model also require s the user to develop the water use coeffi cients. This model developed by the Stockholm Environmental Institute (2009) is a water forecasting tool for utility supply planning and demand management. This model utilize s a system of modules that characterize the physical water demand supply network and estimate water demands for various water uses as defined in the study. The activities are standardized per production output for CII and sectors determined by the quality of the input data. A notable feature of this demand management program is that a hierarchic branching data infrastructure is used to manage water use data by the sector, subsector, end use and device. The model incorporate s demand estimates into the distribution program to simulate supply alloca tions on a spatial and temporal basis to provide a sophisticated planning tool (Wurbs 1995). This model allows for greater flexibility and accuracy in water supply planning but has an extensive requirement of user inputs. Such data requirements also increa se the amount of time and resources required to run such a model. The models of the IWR MAIN tradition are tools for forecasting water use. These models develop their water use projections from a regression analysis of explanatory variables and have been developed differently over time. Table 2 1 highlights their progression. As forecasting

PAGE 22

22 tools, these models do little to evaluate present water use or indicate conservation potential. T ampa B ay W ater has a rich spatial and temporal dataset for water use by each parcel, and a more sophisticated CII model could have been developed if the water use coefficients were standardized by a measure of size that can be related more dire ctly to water use processes. Maddaus and Maddaus (2004) have pioneered the end use approach with a Least Cost Planning Demand Management Decision Support System model. This proprietary forecasting model which rel ies on employment data for the measure of CII water use and can on ly disaggregate into C, I and I, calibrates total water use with an estimate of fixtures in a building based on its age, the frequency of each end white paper titled Water Efficiency in the Commercial and Institutional Sector (2009 ) summarizes m any of the studies conducted on CII water use. Citing studies like the American Water Works Association Research Foundation Commercial/Institutional End Uses of Water (Dziegielewski et al. 2000) and Waste Not, Want Not: The Potenti al for Urban Water Conservation in California (2003) for documentation o n baseline water use by CII subsectors and end use breakdown. Other studies are cited in addition such as East Bay s Watersmart Guidebook (2008) and the Nor th Carolina Department of Water Efficiency Manual for CII Facilities (2009) which provide water saving measures and technologies that are applicabl e to the different CII sectors. In addition to providing an extensive lit Water Efficiency in the Commercial and Institutional Sector (2009), also documents national and international CII water efficiency programs. This white paper also outlines the information gaps currently present in evaluating CII for wa ter conservation. The paper cites a lack of subsector specific data, such as water usage by facility and end use, and existing benchmarks by which to set targets. Though

PAGE 23

23 other studies ( Dziegielewski et al. 2000 Colorado WaterWise 2007 ) have provided such subsector data, they have only done so for a limited number of subsectors. T he American Water Works Association Research Foundation Commercial/Institutional End Uses of Water Study (Dziegielewski et al. 2000) carried out a detailed analysis on five CII s ubsectors of interest: schools, hotels/motels, office buildings, restaurants, and food stores. This study collected survey data from utilities across the southwestern United States as well as direct monitoring water use for 24 selected commercial and insti tutional facilities in the five user categories. The study developed water use coefficient for these subsectors using a variety of measures of size. Of all measures of size investi gated, building area was the sole significant predictor of water use across all subsectors. This study through submetering, also disaggregated total subsectoral water use down to the general end uses (e.g., domestic, cooling, landscaping), as well as presented benchmarks for efficiency based on the 25 th percentile of customers sam pled The need for a standardized classification system for CII water users, along with the expansion of benchmarking to other subsectors was also addressed. Similarly, Colorado WaterWise (2007) also developed water use coefficients using various measures of size for four CII subsectors, including restaurants, schools, hotels/motels, and homes for the aged. Presented in this study was the need for standardized databases containing relevant measures of size, a classification system for CII water users, and a national clearinghouse of water utility data. Employment Data as a Measure of Size Many of the models reviewed standardize CII water use by the total number of employees in order to compare the efficiency and variance of different sectors. Other variable s have been investigated. Mercer and Morgan (1974) developed water use coefficients based on number of employees and this study cites that employee data has historically been readily available compared to other parameters such as acreage. Kim and McCuen (1 979) studied retail stores and

PAGE 24

24 the results showed that the best two predictors of water use were gross area and sales area, the only two measures of area analyzed, followed by average number of daily man hours and employees. As described previously, Dziegi elewski et al. (2000) investigated five different commercial and institutional water users and only building area was found to be a significant indicator of water use across all customer categories. Models of the past primarily depended on the user to inp ut local employment data, which is available from the U.S. Census or from private surveys (Dziegielewski et al. 2000). U.S. Census employment data are available through three avenues: the Economic Census, County Business Patterns, and Longitudinal Employer Household Dynamics. The U.S. Economic Census ( 2010) is conducted every five years and the emp loyment data is aggregated to geographical areas These areas include state, county, metropolitan area, and city. Statistical data on employment is provided per N orth American Industry Classification System code (which recently replaced the SIC classification system). The size, density and composition of parcels within these geographical areas vary widely and the precision of total employment estimates for each sec tor by U.S. Economic Census is limited because of this aggregation. County Business Patterns ( US Census 20 09a ) provides annual U.S. Census employment estimates at the county, zip code and metropolitan level with a two year lag. This data is treated differ ently depending on the type of business establishment. Single unit firm data, where the firm owns and operates a single establishment, is gathered from a variety of administrative record and survey sources when available. Multi unit firm data, where a firm owns and operates multiple establishments, is obtained through the U.S. Economic Census, and the annual Company Organization Survey, which only survey companies with 250 employees or more unless administrative records specify otherwise This method of col lecting employment data is

PAGE 25

25 subject to nonsampling errors such as inability to identify all businesses, definition difficulties, and estimation of missing or misreported data. Employment data from the U.S. Census County Business Patterns is also limited in their aggregation of customers to ensure geographic (county, zip code, or metropolitan area) and classification anonymity. Classification of customers is provided through the North American Industry Classification System but employee estimates are largely presented in bins to protect the identity of individual establishments. The Longitudinal Employer Household Dynamics program is a new state/federal partnership between the U.S. Census Bureau and ten states (CA, FL, IL, MD, MN, NC, NJ, OR, PA, and TX). M o dern statistical and computing techniques are used to combine federal and state administrative data on employers and employees with core U.S. Census Bureau censuses and surveys while protecting the confidentiality of people and firms that provide the data. The Longitudinal Employer Household Dynamics program provides quarterly employment estimates at even smaller geographical areas than the two previous sources, including: tract, block group, zip code, and TAZ. Employment estimates are o nce again provided p er 2 digit North American Industry Classification System code and the data is often presented in bins to ensure anonymity. The program also provides annual employment figures given that the quarterly estimates are still viewed as experimental (US Census 2 009b) Outside of the U.S. Census, employment figures can be derived from c ommercial surveys which are more thorough and precise because data is collected at the customer level. T he accuracy of such surveys however depends on the diligence of the responden t and this data must be purchased. Sources of employment data are profiled in Table 2 2 In order to create a bottom up approach, a model needs a more accurate, frequent and robust data base that disaggregates the CII sector into relatively homogeneous subs ectors

PAGE 26

26 Employment data are primarily averaged per geographical area and available periodically. Because this metric is so widely used and a better substitute has not been developed, no CII water forecasting model is appropriate for evaluating conservation options. A new approach is needed. The parcel level land use databases utilized in the proposed methodology to arrive at a measure of size are described in the following chapter.

PAGE 27

27 Table 2 1 Comparison of models in the Institute of Water Resources Munic ipal and Industrial Needs (IWR MAIN) model tradition (Dziegielewski and Boland 1989; Opitz et al. 1998; Hazen & Sawyer and PMCL 2004) Item Version 5.1 Version 6.1 Tampa Bay Water Measure of sector size Employment Employment Employment Data used to develo p water use coefficients National survey National survey Regional survey Number of survey points 3,448 7,000 39,727 Level of data aggregation Census block Census block TAZ Time span of historic data 1 year 10 years 3 years Explanatory variables N umber of employees price of water and sewer services and presence of conservation programs N umber of employees marginal price of water average productivity of labor and number of cooling degree days T otal number of employees residential income total rai nfall and fraction of employment in commercial, institutional, industrial sectors Default CII coefficients 23 commercial and institutional and 198 industrial Aggregated : 8 major industrial groups d isaggregated : 65 two digit SIC m ost disaggregated : 41 7 three digit SIC 1 non residential Table 2 2 Sources of employment data Source Available Smallest g eographical u nit U.S. Economic Census Every 5 years City County Business Patterns Annually Zip code Longitudinal Employer Household Dynamics Quarter ly TAZ Commercial s urveys Varies Customer

PAGE 28

28 CHAPTER 3 BOTTOM UP APPROACH FOR EVAL UATING THE SIZE OF E ACH ACTIVITY In t roduction Given the limitations of past models, including access to reliable data, a new methodology to estimate commercial, industrial, and instit utional ( CII ) water use based on parcel level land use and water billing databases is presented in this thesis. This chapter describes heated building area, the measure of size proposed for this methodology. The Florida Department of Revenue ( FDOR ) databas e, in conjunction with Florida County Property Appraisers ( FCPA ) provides the heated building areas for every parcel in the State along with their land use classification, allowing for sector specific water use coefficients. The land use databases used, a long with their attributes of interest, are presented in Table 3 1 Land Use Databases Florida Department of Revenue FDOR maintains a database of legal, physical and economic property based information for e ach of the 9 million parcel s of land in the stat e of Florida. Of this total number, 326,000 are CII parcels (215,000 commercial, 69,000 industrial, and 42,000 institutional). This database is publicly available free of charge from the FDOR FTP website ( ftp: //sdrftp03.dor.state.fl.us/ ) and is audited and updated annually. FDOR partitions p arcels based on their land use into 100 sect ors using two digit FDOR codes. T hese codes are standardized across the S tate, providing consistent definitions of terms. The p arcel information in this database is provided annually by the S 67 F CPAs to FDOR for a statewide land use database. The FDOR has also joined forces with other state agencies and water management districts to capture and share data for a more thoroug h database (FDOR 2009)

PAGE 29

29 The following attributes of interest are provided by the FDOR database: Parcel ID number Land use code Effective year built Effective building area Parcel area The parcel ID number is a unique identifier to a plot of land, and se rves as the link between the various databases presented in this methodology. The coefficients presented in this thesis are at the parcel level, requiring data to be adjusted to this level of aggregation. The FDOR land use code is a two digit classificati on system that identifies the primary use of the land by its economic activity. CII sectors in this study are determined using FDOR land use codes. The FDOR land use classification system allows for various degrees of disaggregation following the hierarchi cal structure presented in Figure 3 1 In broadest terms, urban land use can be broken up into residential and CII sectors based on groups of FDOR codes. Disaggregation is possible by allocating the residential codes into either single family or multi fami ly residential and CII codes into commercial, industrial, and institutional. The greatest level of disaggregation available from the FDOR database comes from the two digit land use code. Effective year built is defined as the effective or actual year buil t of major improvements for a building. The year built provides valuable time series information to estimate trends, and is an essential tool in forecasting number of accounts, building and parcel characteristics, and water use rates. The effective or adj usted building area field, defined as the total effective area of all floors of all buildings on a given parcel is not a true area, but rather a calculated field Effective area incorporates economic factors to weight the various building area types found within a parcel

PAGE 30

30 differently. To calculate effective area, a parcel is first divided into primary and secondary s ubareas. Then, s ubareas within a parcel are adjusted by a reference standard cost per square foot of construction Primary areas are the interi or finished living areas, and by definition have an Effective Area Factor of 1. Secondary areas are defined as an area that has a cost per square foot of construction different from the li ving area reference standard. All secondary areas have an Effective Area Factor greater or less than 1 (Oliver 2009). The effective area calculation is presented i n E quation 3 1 By definition, the effective area of a parcel will always be greater than or equal to its heated building area, since other non heated areas are associated with effective area. Through the use of FDOR alone, the relative importance of sectors can be quantified by simply summing total effective areas of parcels within sectors ( 3 1 ) Where: Pa i = p rimary areas Sa i = s econdary a reas Eaf i = Eff Parcel area is a derived field from the FDOR database. Though FDOR provide s a parcel area field, [ LNDSQFOOT], this field is seldom populated. The FDOR database however, also provides polygon shapefiles deline ating every parcel in the State. Using standard GIS tools the area of each parcel can thus calculated, and joined to the other parcel information provided in the FDOR attribute data. Besides, parcel dimensions, these polygon shapefiles also offer the spat ial location of every parcel in the State. This allows simple spatial queries to determine which parcels are within the service boundaries of a given utility South Florida Water Management

PAGE 31

31 District ( WMD ), St. Johns River WMD, and Southwest Florida WMD pro vide the water service area boundaries of utilities in their d istricts as polygon shapefiles available in their respective websites to be viewed in GIS The parc els are identified by a unique p arcel identification number which can be related to the FDOR da tabase to find the attributes for the parcels in the utility being analyzed. FDOR serves as the foundation for a Florida urban water database allowing for b oth spatial and attribute joins, and providing consistent definition of terms. Other public water s upply modeling parameters such as: population data from U.S. Census, utility service boundary information, utility flow data from the Florida Department of Environmental Protection, and water billing records from select utilities can be joined as appropria te. Figure 3 2 presents a schematic of how databases, along with their attributes of interest, can be related to FDOR The parcel level data from FDOR is powerful, allowing analysis across spatial scales: macro (state, water management district, or county) meso (city or utility), micro (parcel), and nano (end u se such as toilets) as shown in Figure 3 3 Florida County Property Appraisers Each of the 67 counties in Florida maintains a F CPA database that contain s the same information as the FDOR database, al ong with additional attributes that vary from county to county. Attributes of interest in all FCPAs are : Parcel ID number Heated building area Parcel ID number is a unique identifier to a parcel, and serves as the link between FCPA and FDOR. F CPA provide s the heated areas of buildings in a parcel, defined as all building area under climate control. Unlike effective building area, provided by FDOR, heated area is a physical building area. The relationship between heated area and effective area is described l ater

PAGE 32

32 in this chapter. The two FCPA databases analyzed in our study are: Hillsborough County Property Appraiser (HCPA) and Alachua County Property Appraiser (ACPA). These FCPA databases encompass the two utilit ies with available water billing data for this thesis: HCWRS (Hillsborough County) and GRU (Alachua County). These water billing databases are described in following chapter. Hillsborough County Property Appraiser Hillsborough County is located in west central Florida, and is the fourth most populous county in the State (Census 2000). HCPA data includes all the FDOR parcel attributes for the county, along with other attributes of interest including: Parcel ID number Heated building area H CPA land use code Each local F CPA may have the same two digit la nd use code as FDOR, or subdivide the land use further with a four digit code. In the case of the four digit code, the first two digits are consistent with the original FDOR classification system and the second two break the sector down into further land u se detail. Four digit F CPA land use codes allow for greater disaggregation of customers, but since these codes are neither required nor stan dardized across the State, the methodology presented in this thesis solely addresses two digit FDOR codes.

PAGE 33

33 Alachu a County Property Appraiser Alachua County is located in north central Florida, and is the 20 th most populous county in the State. ACPA provides considerably more property attributes than FDOR, including: Parcel ID number Heated building area A CPA land use code Information on impervious areas Presence of in ground pool Presence of in ground irrigation system Presence of well Number of bedrooms Number of baths Number of stories As described in the previous section, four digit FCPA land use codes vary from co unty to county. The ACPA and HCPA four digit land use classifications are dissimilar. Additional square footage of impervious areas on a parcel, such as parking l ots, patios, sheds, and pools. footprint of all buildings on a parcel provides an estimate of the impervious area for that parcel. By subtracting the impervious area estimate from the total parcel area, one arrives at an estimate of the pervious or irrigable area on a parcel ( Equation 3 2 )

PAGE 34

34 PA = TA FS AIA ( 3 2 ) Where: PA = parcel pervious area TA = total parcel area FS = footprint of all buildings on p arcel AIA = associated impervious areas on parcel Within miscellaneous areas, ACPA also tags other property attributes of interest in public water supply evaluations, including presence of in ground sprinkler systems and pools, and wells. Additional attr ibutes of interest provided for all parcels in Alachua County include the number of bedrooms, baths, and stories. For CII parcels, number of baths denotes the count of flushable toilets and urinals in all structures on a parcel, the reported accuracy of wh ich has not been investigated. Number of stories is critical in calculating the footprint of buildings on a parcel. A simple estimate of footprint is calculated by dividing HA of all buildings on a parcel by their respective number of stories ( Equation 3 3 ) ( 3 3 ) Where: FS = footprint of all buildings on parcel HA i = total heated area in building i N i = number of stories in building i Relationship of Effective Area to Heated Area Effective building area provided by FDOR, is no t a physical area, but rather a calculated value incorporating market values of the structures within a parcel. For this reason, water use coefficients were developed using heated building area, a physical area not prone to misinterpretation and avai lable from F CPA. A pplication of water use coefficients directly with

PAGE 35

35 FDOR requires coefficients to convert the effective building area to heated building area. A simple linear regression of Heated Area (HA) as a function of Effective Area (EA) based on sam ple of 3,205 CII parcels in Hillsborough and Alachua County, Florida is shown in Figure 3 4 The resulting E quation 3 4 was forced through the origin: HA = 0.9526*EA R 2 = 0.9921 ( 3 4 ) The very high R 2 indicates that the fit is excellent. According to th e regression equation, HA is about 95.26% of EA. A slightly different result ( E quation 3 5 ) is obtained if the HA/EA ratio is defined as follows: = 0.9392 ( 3 5 ) Both methods of estimating the relationship between HA and EA assume linearity between the two variables through the origin. For all 3,205 CII parcels analyzed in Florida, as shown in Figure 3 4 linearity through the origin is a safe assumption to make. The regression method presented ( Equation 3 4 ) works best when the variance is homogeneous The method described in Equation 3 5 also known as the ratio of the means, is most efficient if the variance of the dependent variable (HA) is proportional to each value of the independent variable (EA) (Swank and Schreude 1974) The residual plot for t he simple linear regression of heated and effective area for 3,205 CII parcels in Hillsborough and Alachua County ( Figure 3 5 ) indicate s that variance across EAs is not homogeneous, hence the ratio of the means method appears to be most appropriate for est imating the relationship between EA and HA. Given the strong correlation between EA and HA across all CII sectors ( Figure 3 4 ), it is likely that similarly strong relationships are presented at the individual 2 digit FDOR subsector level. Hence, EA to HA conversion coefficients using the ratio of the means approach were developed at the 2 digit FDOR subsector level. These area conversion coefficients allow for the

PAGE 36

36 application of water use coefficients, normalized by HA, to the EA measures available from F DOR. Analysis of Available Data HCW RS, located in Hillsborough County, provided utility billing data for 1,768 CII accounts (67% commercial, 9% industrial, and 24% institutional) for 48 months beginning in January 2003 GRU, located in Alachua County, su pplied two complete years of monthly water billing from January 2008 to December 2009, for 1,437 CII parcels (72% commercial, 10% industrial, and 18% institutional). The subsector heated area statistics of these two utilities, along with their sample size and other property attributes from FDOR and FCPA are addressed in this section. The subsector characteristics are compared across utilities, with the combined utility data set, and with subsector data for the entire state of Florida from FDOR. Sample s ize The combined utility dataset from HCWRS and GRU, provide samples for 24 of the 28 FDOR commercial subsectors ( Table 3 2 ) The available subsector samples from the combined utility datasets account for between 0.2% and 4.8% of the State total of commercial parcels. Overall, the combined utility dataset provides a 1% sample of commercial parcels in the state of Florida. The available sampled subsectors account for over 97% of commercial parcels in the State. The available utility data for the industrial ( T able 3 3 ) and institutional ( Table 3 4 ) subsectors provide s samples between 0.2% and 3.7% of total State parcels at the subsector level. Data is available for eight of the 11 industrial FDOR subsectors, and 14 of the 16 institutional subsectors. The sample data accounts for 0.4% and 1.6% of all industrial and institutional parcels in the State. The sampled subsectors account for nearly 99% of industrial and 97% of institutional parcels in the State.

PAGE 37

37 Effective Year Built The average effective year built of major improvements on a parcel for the available commercial subsectors in HCWRS, GRU, and combined utilities, as well as in the entire state of Florida are presented in Table 3 5 Similar tables are also available for the industrial ( Table 3 6 ) and instit utional ( Table 3 7 ) sectors. These tables show that the average effective year built across the majority of subsectors varies little between the sample dataset and the State total s of CII parcels. Heated Area to Effective Area C alculation The heated to e ffective area ratio of the means calculation at the subsector level is shown for the commercial, industrial, and institutional sectors in Table 3 8 Table 3 9 and Table 3 10 respectively. The calculation, presented for the sample of HCWRS and GRU CII sub sectors, as well as for the combined dataset, involves dividing the average heated area of each subsector provided by FCPA by its average effective heated area from FDOR. The difference between the heated area to effective area ratios from HCWRS and GRU is presented for the commercial, industrial, and institutional sectors in Table 3 11 Table 3 12 Table 3 13 respectively. This percent difference varies from 16% to 20% depending on the subsector being analyzed, and is due in large part to the available s ample sizes. Even though the ratio of the means calculation was chosen over a linear regression approach, also shown in these tables is the associated R 2 for each of the available subsectors, presented to show the generally strong relationship between hea ted area and effective area across CII subsectors. For the most part, R 2 values presented approach 1. Comparisons of the subsector heated area to effective area ratios indicate negligible differences across the two utilities. This fact solidifies our confi dence in using these area conversion factors to apply the water use coefficients developed in this thesis with the statewide data available from FDOR. For the CII

PAGE 38

38 sectors, heated area to effective area relationships likely do not significantly vary across the State. Distrib ution and Representativeness of Sampled Heated Areas HCWRS, GRU, combined utilities and State comparisons of average heated area at the subsector level are presented for the commercial, industrial, and institutional sectors in Table 3 14 Table 3 15 and Table 3 16 respectively. The state of Florida subsector information is available from FDOR, which only provides the effective area of each parcel. However, using the heated to effective area (HA/EA) conversion ratios calculated for each subsector, estimates of heated area can be derived for the entire State. The use of these coefficients is reasonable following the strong relationship between heated and effective area across CII subsectors described in the previous section. For the subsec tors where no sample data was available to calculate the area conversion ratio, the average HA/EA of the available subsectors from the given sector was utilized. The percent difference of subsectoral average heated area between the combined utility sampl e from HCWRS and GRU and the entire state of Florida is also presented in Table 3 14 Table 3 15 and Table 3 16 Following the heterogeneous nature of CII customers, these tables demonstrate the relatively large variability (from 80% to 533%) in average heated area within subsectors. The limited number of 2 digit CII FDOR codes ensures that multiple facility types with differing size characteristics and drivers of water use are grouped within each code Disaggregated groupings within 2 digit FDOR codes ca n be achieved by developing size categories based on heated building area, or age built of a facility which might also affect water use, given the requirement or availability of certain end use devices at the time of construction. For example, the resident ial sector is broken up into three age groups (pre 1983, 1983 1994, 1995 present) corresponding with State regulations requiring minimum plumbing fixture water

PAGE 39

39 efficiencies. Even facilities within the same sub sector might offer new services requiring diffe rent end use water devices as they respond to changing conditions. Predicting what fixture types are prevalent in certain customer groups greatly improves estimates of water use, as well as facilitates the weighing of water conservation options. Unfortunat ely, sample size limitations do not allow this analysis be carried out. A preliminary assessment of the representativeness of the available combined utility sampled dataset as it pertains to the State total s of CII parcels is provided by Table 3 14 Tabl e 3 15 and Table 3 16 These tables utilize the average heated area stati sti c to make this inference. For greater insight into both the combined utility dataset and the State total s of CII parcels, it is best to look at the distribution of heated areas ac ross all parcels in a given subsector. Histograms detailing both the combined utility and State heated area distributions of each CII subsector are presented in the appendix By comparing the histogram of the sample combined utility dataset, to that of the State, one can gain a measure of how representative the sample dataset is to the State total s of CII parcels at the subsector level. Though the sample dataset might not be representative of the entire subsector State distribution it might be representati ve of a se gment within that State distribution The fitted lognormal probability density equations for heated area subsector distributions of both the combined utility sample and State total are presented in Table 3 17 Table 3 18 and Table 3 19 for the commercial, industrial, and institutional sectors, respectively. The parameters are estimated using @Risk (Palisade Corporation 2010). The log normal probability density function is shown in Equation 3 6

PAGE 40

40 ( 3 6 ) W here : = t he location parameter or log mean = the scale parameter or log standard deviation The log normal distribution was chosen because it has a defined lower bound (typically 0) and fits the data well. The corresponding shifted lognormal equations are shown in Table 3 20 Table 3 21 and Table 3 22 In these tables, along with sample size, the Anderson Darling and Kolmogorov Smirnov statistics are also presented to provide a measure of the goodness of fit of the data to the lognormal distribution. From these statistics, and the histograms presented in the appendix it is clear that the representativeness of the combined utility sample varies across subsectors. This discrepancy is largely attributable to sample size, and the homogeneity of a given subsector thr oughout the State. A general trend is apparent in that as sample size increases, the State and sample distributions converge. Thus, the parcel level heated area statistics presented in this thesis would improve with increased sample sizes. The following chapter will present the sampled water use data from HCWRS and GRU. The chapter will explain their database structure, preparation required for analysis, as well as present their time series information.

PAGE 41

41 Table 3 1 Databases and parcel attributes used to develop water use and area conversion coefficients based on a sample of 3,205 commercial, industrial, and institutional ( CII ) parcels Database Attributes of i nterest Period of r ecord FDOR Parcel ID n umber Land u se c ode Effective y ear b uilt Effective b uil ding a rea Parcel a rea 1920 2008 HCPA Parcel ID n umber Heated b uilding a rea HCPA l and u se c ode 1920 2008 ACPA Parcel ID number Heated building area ACPA land use code Information on impervious areas Presence of in ground pool Presence of in ground irr igation system Presence of well Number of bedrooms Number of baths Number of stories 1920 2008

PAGE 42

42 Table 3 2 Parcel breakdown across commercial subsectors for Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilitie s combined, and the entire state of Florida FDOR c ode Description Parcel c ount Parcel s ubsector b reakdown Percent of State HCWRS GRU Combined State Combined State Combined 11 Stores, one s tory 113 176 289 39,002 13.1% 18.1% 0.7% 12 Mixed u se 101 42 14 3 18,931 6.5% 8.8% 0.8% 13 Department s tores 18 1 19 829 0.9% 0.4% 2.3% 14 Supermarkets / conv. s tores 117 6 123 2,571 5.6% 1.2% 4.8% 15 Regional m alls 2 1 3 449 0.1% 0.2% 0.7% 16 Community shopping c enters 161 78 239 7,726 10.8% 3.6% 3.1% 17 Office, o ne s tory 152 232 384 37,629 17.3% 17.5% 1.0% 18 Office, multi s tory 41 32 73 16,033 3.3% 7.5% 0.5% 19 Medical o ffice 107 157 264 20,953 11.9% 9.8% 1.3% 20 Transit t erminals 5 1 6 2,825 0.3% 1.3% 0.2% 21 Restaurants 68 52 120 7,629 5.4% 3.6% 1.6% 22 F ast f oo d r estaurants 55 50 105 4,377 4.7% 2.0% 2.4% 23 Financial i nstitutions 63 35 98 4,732 4.4% 2.2% 2.1% 24 Insurance company o ffices 11 11 283 0.5% 0.1% 25 Service s hops 18 31 49 4,309 2.2% 2.0% 1.1% 26 Service s tations 5 5 3,434 0.2% 1.6% 27 Auto s a les / r epair 116 58 174 15,129 7.9% 7.0% 1.2% 29 Wholesale o utlets 5 5 729 0.2% 0.3% 30 Florist / greenhouses 2 2 352 0.1% 0.2% 31 Drive in theaters / open s tadiums 28 0.0% 32 Enclosed theaters / auditoriums 1 2 3 242 0.1% 0.1% 1.2% 33 Nightc lubs / bars 8 12 20 1,914 0.9% 0.9% 1.0% 34 Bowling alleys / skating rinks 1 2 3 492 0.1% 0.2% 0.6% 35 Tourist attractions 742 0.3% 36 Camps 330 0.2% 37 Race tracks 129 0.1% 38 Golf courses / driving ranges 20 6 26 1,392 1.2% 0.6 % 1.9% 39 Hotels / motels 10 40 50 21,702 2.3% 10.1% 0.2% Total c ommercial 1,177 1,037 2,214 214,893 100.0% 100.0% 1.0%

PAGE 43

43 Table 3 3 Parcel breakdown across industrial subsectors for Hillsborough County Water Resources Services, Gainesville Regional Util ities, the two utilities combined, and the entire state of Florida F DOR code Description Parcel count Parcel subsector breakdown Percent of S tate HCWRS GRU Combined State Combined State Combined 41 Light manufact uring 10 23 33 18,398 10.7% 26.7% 0.2% 42 Heavy industrial 2 1 3 677 1.0% 1.0% 0.4% 43 Lumber yards 1 1 471 0.3% 0.7% 0.2% 44 Packing plants 526 0.8% 45 Bottler / canneries 1 1 90 0.3% 0.1% 1.1% 46 Food processing 295 0.4% 47 Minera l processing 8 2 10 598 3.3% 0.9% 1.7% 48 Warehousing / distribution 118 110 228 42,406 74.3% 61.5% 0.5% 49 Open storage 17 2 19 2,190 6.2% 3.2% 0.9% 91 Utility, gas & elec. 8 4 12 3,185 3.9% 4.6% 0.4% 92 Mining / petroleum 112 0.2% Total industrial 163 144 307 68,948 100.0% 10 0.0% 0.4%

PAGE 44

44 Table 3 4 Parcel breakdown across institutional subsectors for Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilities combined, and the entire state of Florida FDOR c ode Description Parcel count Parce l subsector breakdown Percent of S tate HCWRS GRU Combined State Combined State Combined 71 Churches 221 116 337 18,888 49.3% 45.1% 1.8% 72 Private Schools & Colleges 65 46 111 3,243 16.2% 7.7% 3.4% 73 Private Hospitals 2 4 6 570 0.9% 1.4% 1.1% 74 Ho mes for the Aged 12 12 4,447 1.8% 10.6% 0.3% 75 Orphanages / Non profits 67 15 82 2,214 12.0% 5.3% 3.7% 76 Mortuaries / Cemeteries 7 6 13 887 1.9% 2.1% 1.5% 77 Clubs / Union Halls 15 40 55 3,244 8.0% 7.8% 1.7% 78 Sanitariums / Convalescents 1 1 457 0 .1% 1.1% 0.2% 79 Cultural Organizations 1 1 234 0.1% 0.6% 0.4% 81 Military 43 0.1% 82 Parks and Recreation 1 1 850 0.1% 2.0% 0.1% 83 Public County Schools 51 1 52 2,953 7.6% 7.1% 1.8% 84 Colleges 9 9 288 1.3% 0.7% 3.1% 85 Hospitals 3 3 173 0 .4% 0.4% 1.7% 90 Gov. Leased Interests 1,276 3.0% 97 Outdoor Recreational 1 1 2,090 0.1% 5.0% 0.0% Total Institutional 428 256 684 41,857 100.0% 100.0% 1.6%

PAGE 45

45 Table 3 5 Average effective year built across commercial subsectors for Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilities combined, and the entire state of Florida FDOR c ode Description Average effective year built HCWRS GRU Combined State 11 Stores, one story 1989 1982 1985 1981 12 Mixed use 1975 1978 1976 1971 13 Department stores 1995 1982 1994 1992 14 Supermarkets / conv. s tores 1991 1984 1991 1986 15 Regional malls 1997 1995 1996 1992 16 Community shopping centers 1989 1985 1988 1990 17 Office, one story 1985 1984 1984 1988 18 Of fice, multi story 1988 1986 1987 1990 19 Medical office 1989 1991 1990 1989 20 Transit terminals 1979 2000 1982 1995 21 Restaurants 1993 1981 1988 1985 22 Fast food restaurants 1998 1989 1994 1991 23 Financial institutions 1991 1992 1992 1992 24 Insu rance company offices 1988 1988 1984 25 Service shops 1984 1979 1981 1980 26 Service stations 1986 1986 1983 27 Auto sales / repair 1985 1983 1984 1980 29 Wholesale outlets 1972 1972 1977 30 Florist / greenhouses 1966 1966 1980 31 Drive in theate rs / open s tadiums 1975 32 Enclosed theaters / auditoriums 1999 2001 2000 1984 33 Nightclubs / bars 1981 1966 1972 1974 34 Bowling alleys / skating rinks 1983 1988 1986 1985 35 Tourist attractions 1988 36 Camps 1979 37 Race tracks 1985 3 8 Golf courses / driving ranges 1991 1986 1990 1990 39 Hotels / motels 1986 1981 1982 1989 Total commercial 1988 1984 1986 1985

PAGE 46

46 Table 3 6 Average effective year built across industrial subsectors for Hillsborough County Water Resources Services, Gai nesville Regional Utilities, the two utilities combined, and the entire state of Florida FDOR c ode Description Average effective year built HCWRS GRU Combined State 41 Light manufacturing 1982 1979 1979 1986 42 Heavy industrial 1995 1998 1996 1979 43 Lumber yards 1990 1990 1980 44 Packing plants 1974 45 Bottler / cann eries 1969 1969 1974 46 Food processing 1976 47 Mineral processing 1972 1984 1974 1982 48 Warehousing / distribution 1990 1985 1988 1987 49 Open storage 1979 1999 1981 1989 91 Utility, gas & elec. 1974 1983 1977 1984 92 Mining / petroleum 1982 Total industrial 1987 1984 1986 1986 Table 3 7 Average effective year built across institutional subsectors for Hillsborough County Water Resources Services, Gainesville Regio nal Utilities, the two utilities combined, and the entire state of Florida F DOR c ode Description Average effective year built HCWRS GRU Combined State 71 Churches 1957 1979 1965 1975 72 Private schools & colleges 1986 1981 1984 1980 73 Private hospit als 1997 1988 1991 1984 74 Homes for the Aged 1993 1993 1986 75 Orphanages / n on profits 1987 1985 1986 1985 76 Mortuaries / cemeteries 1989 1974 1982 1979 77 Clubs / union halls 1981 1935 1947 1978 78 Sanitariums / convalescents 1975 1975 1985 79 Cultural organizations 1989 1989 1980 81 Military 1972 82 Parks and Recreation 1998 1998 1982 83 Public county schools 1987 1980 1987 1980 84 Colleges 1970 1970 1980 85 Hospitals 1988 1988 1986 90 Gov. l eased interests 1989 97 Outdoor recr eational 1950 1950 1989 T otal institutional 1971 1973 1972 1979

PAGE 47

47 Table 3 8 Heated area to effective area ratio of the means calculation of commercial subsectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the t wo utilities combined FDOR c ode Description Average h eated a rea (sf) Average e ffective a rea (sf) HA/EA HCWRS GRU Combined HCWRS GRU Combined HCWRS GRU Combined 11 Stores, one story 9,375 6,532 7,644 9,808 7,258 8,255 0.96 0.90 0.93 12 Mixed use 13,601 6,390 11,483 14,225 7,618 12,284 0.96 0.84 0.93 13 Department stores 130,076 94,109 128,183 135,104 117,614 134,183 0.96 0.80 0.96 14 Supermarkets / conv. s tores 5,576 10,065 5,795 5,746 11,143 6,009 0.97 0.90 0.96 15 Regional malls 820,551 928,070 856 ,391 939,712 936,397 938,607 0.87 0.99 0.91 16 Community shopping centers 39,269 39,269 39,269 41,224 41,350 41,265 0.95 0.95 0.95 17 Office, one story 5,446 6,334 5,983 5,674 6,567 6,214 0.96 0.96 0.96 18 Office, multi story 38,283 20,700 30,576 39,230 21,743 31,565 0.98 0.95 0.97 19 Medical office 6,770 8,070 7,543 7,046 8,268 7,773 0.96 0.98 0.97 20 Transit terminals 10,670 2,193 9,257 11,057 2,235 9,586 0.96 0.98 0.97 21 Restaurants 5,084 4,664 4,902 5,311 4,816 5,097 0.96 0.97 0.96 22 Fast food restaurants 3,105 2,697 2,910 3,205 2,809 3,016 0.97 0.96 0.96 23 Financial institutions 4,424 6,338 5,108 5,080 6,791 5,691 0.87 0.93 0.90 24 Insurance company offices 10,736 10,736 11,741 11,741 0.91 0.91 25 Service shops 2,787 6,906 5,393 3,563 8, 489 6,679 0.78 0.81 0.81 26 Service stations 1,829 1,829 2,591 2,591 0.71 0.71 27 Auto sales / repair 9,095 5,836 8,009 10,353 7,053 9,253 0.88 0.83 0.87 29 Wholesale outlets 23,700 23,700 31,323 31,323 0.76 0.76 30 Florist / g reenhouses 3,376 3 ,376 3,679 3,679 0.92 0.92 32 Enclosed theaters / auditoriums 97,632 27,989 51,203 105,622 27,987 53,865 0.92 1.00 0.95 33 Nightclubs / bars 3,686 5,336 4,676 3,684 5,770 4,936 1.00 0.92 0.95 34 Bowling alleys / skating rinks 30,784 34,410 33,201 31,2 78 36,864 35,002 0.98 0.93 0.95 38 Golf courses / driving ranges 18,091 9,633 16,139 20,444 10,744 18,206 0.88 0.90 0.89 39 Hotels / motels 36,875 31,626 32,676 38,761 33,574 34,611 0.95 0.94 0.94 Total commercial 16,444 11,457 14,108 17,443 12,224 14, 999 0.94 0.94 0.94

PAGE 48

48 Table 3 9 Heated area to effective area ratio of the means calculation of industrial subsectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined FDOR c ode Description Avera ge h eated a rea (sf) Average e ffective a rea (sf) HA/EA HCWRS GRU Combined HCWRS GRU Combined HCWRS GRU Combined 41 Light manufacturing 80,659 20,851 38,975 89,552 23,130 43,258 0.90 0.90 0.90 42 Heavy industrial 44,432 41,893 43,586 45,382 58,506 49,75 6 0.98 0.72 0.88 43 Lumber yards 18,488 18,488 19,276 19,276 0.96 0.96 44 Packing plants 45 Bottler / canneries 34,541 34,541 43,783 43,783 0.79 0.79 46 Food processing 47 Mineral processing 102,201 23,001 86,361 104,177 24,106 88,163 0.98 0.95 0.98 48 Warehousing / distribution 41,608 16,966 29,719 43,010 18,929 31,392 0.97 0.90 0.95 49 Open storage 1,860 7,589 2,463 2,036 7,799 2,642 0.91 0.97 0.93 91 Utility gas & elec. 15,714 50,505 27,311 16,425 56,926 29,926 0.96 0.89 0.91 92 Mining / petroleum Total industrial 41,596 18,777 30,893 43,318 21,023 32,860 0.96 0.89 0.94

PAGE 49

49 Table 3 10 Heated area to effective area ratio of the means calculation of in stitutional subsectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined FDOR c ode Description Average h eated a rea (sf) Average e ffective a rea (sf) HA/EA HCWRS GRU Combined HCWRS GRU Combined HCWRS GRU Combined 71 Churches 12,838 13,555 13,085 13,678 14,143 13,838 0.94 0.96 0.95 72 Private schools & colleges 7,309 9,149 8,071 7,720 9,658 8,524 0.95 0.95 0.95 73 Private hospitals 424,916 141,224 235,788 434,468 147,756 243,326 0.98 0.96 0.97 74 Homes for the a ged 116,675 116,675 126,513 126,513 0. 92 0.92 75 Orphanages / n on profits 9,741 10,894 9,952 10,277 11,349 10,473 0.95 0.96 0.95 76 Mortuaries / cemeteries 7,560 5,286 6,511 8,593 5,933 7,365 0.88 0.89 0.88 77 Clubs / union halls 10,654 13,308 12,584 12,846 13,460 13,292 0.83 0.99 0.95 78 Sanitariums / convalescents 43,505 43,505 43,644 43,644 1.00 1.00 79 Cultural organizations 2,302 2,302 3,125 3,125 0.74 0.74 81 Military 82 Parks and r ecreation 5,288 5,288 5,771 5,771 0.92 0.92 83 Public county schools 127,905 59,44 8 126,588 130,491 60,959 129,153 0.98 0.98 0.98 84 Colleges 175,786 175,786 175,937 175,937 1.00 1.00 85 Hospitals 123,175 123,175 128,633 128,633 0.96 0.96 90 Gov. l eased interests 97 Outdoor recreational 9,233 9,233 10,021 10,021 0. 92 0.92 Total institutional 26,988 26,395 26,766 28,014 27,465 27,809 0.96 0.96 0.96

PAGE 50

50 Table 3 11 Comparison of heated area to effective area ratio of the means calculation for the commercial subsectors in Hillsborough County Water Resources Services, G ainesville Regional Utilities, and the two utilities com bined FDOR Code Description HA/EA Percent d ifference HA/EA R 2 HCWRS GRU Combined 11 Stores, one story 0.96 0.90 6% 0.93 1.00 12 Mixed use 0.96 0.84 14% 0.93 0.99 13 Department stores 0.96 0.80 20% 0.96 0.98 14 Supermarkets / conv. s tores 0.97 0.90 7% 0.96 0.99 15 Regional malls 0.87 0.99 12% 0.91 0.84 16 Community shopping centers 0.95 0.95 0% 0.95 1.00 17 Office, one story 0.96 0.96 0% 0.96 1.00 18 Office, multi story 0.98 0.95 3% 0.97 1 .00 19 Medical office 0.96 0.98 2% 0.97 1.00 20 Tran sit terminals 0.96 0.98 2% 0.97 1.00 21 Restaurants 0.96 0.97 1% 0.96 0.99 22 Fast food restaurants 0.97 0.96 1% 0.96 0.99 23 Financial institutions 0.87 0.93 7% 0.90 0.99 24 Insurance company o ffices 0.91 0.91 0.98 25 Service shops 0.78 0.81 4% 0.81 0.92 26 Service stations 0.71 0.71 0. 93 27 Auto sales / repair 0.88 0.83 6% 0.87 0.99 29 Wholesale outlets 0.76 0.76 0.95 30 Florist / greenhouses 0.92 0.92 0.98 31 Drive in t heaters / open s tadiums 32 Enclosed theaters / auditoriums 0.92 1.00 8% 0.95 1.00 33 Nightclubs / bars 1.00 0.92 8% 0.95 0.97 34 Bowling alleys / skating rinks 0.98 0.93 5% 0.95 0.87 35 Tourist attractions 36 Camps 37 Race tracks 38 Golf courses / driving ranges 0.88 0.90 1% 0.89 0.98 39 H otels / motels 0.95 0.94 1% 0.94 0.99 Total commercial 0.94 0.94 1% 0.94 0.99

PAGE 51

51 Table 3 12 Comparison of heated area to effective area ratio of the means calculation for the industrial subsectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined FDOR c ode Description HA/EA Percent d ifference HA/EA R 2 HCWRS GRU Combined 41 Light manufacturing 0.90 0.90 0% 0.90 1.00 42 Heavy industr ial 0.98 0.72 37% 0.88 0.98 43 Lumber yards 0.96 0.96 44 Packing plants 45 Bottler / canneries 0.79 0.79 46 Food processing 47 Mineral processing 0.98 0.95 3% 0.98 1.00 48 Warehousing / distribution 0.97 0.90 8% 0.95 1.00 49 Open storage 0.91 0.97 6% 0.93 1.00 91 Utility, gas & elec. 0.96 0.89 8% 0.91 1.00 92 Mining / petroleum Total industrial 0.96 0.89 8% 0.94 1.00 Table 3 13 Comparison of heated area to effective area ratio of the means calculati on for the institutional subsectors in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined FDOR c ode Description HA/EA Percent d ifference HA/EA R 2 HCWRS GRU Combined 71 Churches 0.94 0.96 2% 0 .95 0.99 72 Private schools & colleges 0.95 0.95 0% 0.95 0.99 73 Private hospitals 0.98 0.96 2% 0.97 1.00 74 Homes for the a ged 0.92 0.92 1.00 75 Orphanages / n on profits 0.95 0.96 1% 0.95 1.00 76 Mortuari es / cemeteries 0.88 0.89 1% 0.88 0.99 77 Clubs / union halls 0.83 0.99 16% 0.95 0.98 78 Sanitariums / convalescents 1.00 1.00 79 Cultural organizations 0.74 0.74 81 Military 82 Parks and r ecreation 0.92 0.92 83 Public county schools 0.98 0.98 1% 0.98 0.99 84 Colleges 1.00 1.00 1.00 85 Hospitals 0.96 0.96 1.00 90 Gov. l eased interests 97 Outdoor recreational 0.92 0.92 Total institutional 0.96 0.96 0% 0.96 0.99

PAGE 52

52 Table 3 14 Comparison of average heated area across the commercial su bsectors of Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilities com bined, and the state of Florida FDOR c ode Description Av erage h eated a rea (sf) Av erage effective a rea (sf) HA/EA Average heated a rea (sf) Percent d ifference from State Total heated a rea ( m illion sf) Percent of total heated a rea HCWRS GRU Combined State Combined State Combined State State 11 Stores, one story 9,375 6,532 7,644 7,627 0.93 7,063 0.2% 275.5 13.3% 12 Mixed use 13,601 6,390 11,483 4 ,983 0.93 4,658 130.4% 88.2 4.2% 13 Department stores 130,076 94,109 128,183 111,043 0.96 106,078 15.4% 87.9 4.2% 14 Supermarkets / conv. s tores 5,576 10,065 5,795 12,915 0.96 12,456 55.1% 32.0 1.5% 15 Regi onal malls 820,551 928,070 856,391 201,510 0.9 1 183,859 325.0% 82.6 4.0% 16 Community shopping centers 39,269 39,269 39,269 40,241 0.95 38,295 2.4% 295.9 14.2% 17 Office, one story 5,446 6,334 5,983 4,933 0.96 4,749 21.3% 178.7 8.6% 18 Office, multi st ory 38,283 20,700 30,576 21,022 0.97 20,363 45 .4% 326.5 15.7% 19 Medical office 6,770 8,070 7,543 5,915 0.97 5,740 27.5% 120.3 5.8% 20 Transit terminals 10,670 2,193 9,257 5,569 0.97 5,378 66.2% 15.2 0.7% 21 Restaurants 5,084 4,664 4,902 4,705 0.96 4,526 4.2% 34.5 1.7% 22 Fast food restaurants 3,1 05 2,697 2,910 3,326 0.96 3,209 12.5% 14.0 0.7% 23 Financial institutions 4,424 6,338 5,108 7,954 0.90 7,139 35.8% 33.8 1.6% 24 Insurance company offices 10,736 10,736 10,547 0.91 9,644 1.8% 2.7 0.1% 25 S ervice shops 2,787 6,906 5,393 6,865 0.81 5,54 2 21.4% 23.9 1.2% 26 Service stations 1,829 1,829 3,617 0.71 2,554 49.4% 8.8 0.4% 27 Auto sales / repair 9,095 5,836 8,009 7,955 0.87 6,885 0.7% 104.2 5.0%

PAGE 53

53 Table 3 14 Continued FDOR code Description Av erage h eated a rea (sf) Average effective a rea (sf) HA/EA Average heated a rea (sf) Percent d ifference from State Total heated a rea ( m illion sf) Percent of total heated a rea HCWRS GRU Combined State Combined State Combined State State 29 Wholesale outlets 23,700 23,700 19,928 0.76 15,078 18.9% 11.0 0.5% 30 Florist / greenhouse 3,376 3,376 3,945 0.92 3,620 14.4% 1.3 0.1% 31 Drive in theaters / o pen Stadiums 77,719 0.92* 71,501 2.0 0.1% 32 Enclosed theaters / auditorium 97,632 27,989 51,203 29,303 0.95 27,855 74.7% 6.7 0.3% 33 Nightclubs / b ars 3,686 5,336 4,676 4,881 0.95 4,624 4.2% 8.9 0.4% 34 Bowling alleys / skating rinks 30,784 34,410 33,201 22,931 0.95 21,751 44.8% 10.7 0.5% 35 Tourist attractions 36,730 0.92* 33,791 25.1 1.2% 36 Camps 9,184 0.92* 8,450 2.8 0.1% 37 Race tracks 44,563 0.92* 40,998 5.3 0.3% 38 Golf courses / driving ranges 18,091 9,633 16,139 18,908 0.89 16,762 14.6% 23.3 1.1% 39 Hotels / motels 36,875 31,626 32,676 12,211 0.94 11,528 167.6% 250.2 12.0% Total commercial 16,444 11,457 14,108 10, 272 0.94 9,662 37.3% 2,076.3 100.0% *HA/EA not available for subsector, average HA/EA of available subsectors used

PAGE 54

54 Table 3 15 Comparison of average heated area across the industrial subsectors of Hillsborough County Water Resources Services, Gainesvill e Regional Utilities, the two utilities com bined, and the state of Florida FDOR code Description Average heated area (sf) Average effective area (sf) HA/EA Average heated area (sf) Percent difference from State Total heated area (million sf) Percent of tot al heated area HCWRS GRU Combined State Combined State Combined State State 41 Light manufacturing 80,659 20,851 38,975 16,282 0.90 14,670 165.7% 269.9 21.9% 42 Heavy industrial 44,432 41,893 43,586 79,078 0.88 69,271 37.1% 46.9 3.8% 43 Lumber yards 18,488 18,488 30,446 0.96 29,201 36.7% 13.8 1.1% 44 Packing plants 38,997 0.91 35,488 18.7 1.5% 45 Bottler / canneries 34,541 34,541 121,454 0.79 95,817 64.0% 8.6 0.7% 46 Food processing 41,368 0.91 37,645 11.1 0.9% 47 Mineral processing 102,201 23,001 86,361 17,919 0.98 17,553 392.0% 10.5 0.9% 48 Warehousing / distribution 41,608 16,966 29,719 19,855 0.95 18,797 58.1% 797.1 64.7% 49 Open storage 1,860 7,589 2,463 7,576 0.93 7,063 65.1% 15.5 1.3% 91 Utility, gas & elec. 15, 714 50,505 27,311 8,577 0.91 7,828 248.9% 24.9 2.0% 92 Mining / petroleum 20,465 0.91 18,623 2.1 0.2% Total industrial 41,596 18,777 30,893 19,000 0.94 17,862 73.0% 1,231.5 100.0% *HA/EA not available for subsector, average HA/EA of available subsectors used

PAGE 55

55 Table 3 16 Comparison of average heated area across the institutional subsectors of Hillsborough County Water Resources Services, Gainesville Regional Utilities, the two utilities com bined, and the state of Florida FDOR code Descriptio n Average heated area (sf) Average effective area (sf) HA/EA Average heated area (sf) Percent difference from State Total heated area (million sf) Percent of total heated area HCWRS GRU Combined State Combined State Combined State State 71 Churches 12, 838 13,555 13,085 10,643 0.95 10,064 30.0% 190.1 20.6% 72 Private schools & colleges 7,309 9,149 8,071 18,864 0.95 17,863 54.8% 57.9 6.3% 73 Private hospitals 424,916 141,224 235,788 132,165 0.97 128,071 84.1% 73.0 7.9% 74 Homes for the a ged 116,675 1 16,675 19,986 0.92 18,432 533.0% 82.0 8.9% 75 Orphanag es / n on profits 9,741 10,894 9,952 11,948 0.95 11,353 12.3% 25.1 2.7% 76 Mortuaries / cemeteries 7,560 5,286 6,511 6,745 0.88 5,963 9.2% 5.3 0.6% 77 Clubs / union halls 10,654 13,308 12,584 7,699 0 .95 7,289 72.7% 23.6 2.6% 78 Sanitariums / convalescen t 43,505 43,505 41,039 1.00 40,908 6.3% 18.7 2.0% 79 Cultural organizations 2,302 2,302 15,616 0.74 11,503 80.0% 2.7 0.3% 81 Military 85,651 0.93 79,655 3.4 0.4% 82 Parks and r ecreation 5,28 8 5,288 7,244 0.92 6,638 20.3% 5.6 0.6% 83 Pu blic county schools 127,905 59,448 126,588 106,648 0.98 104,529 21.1% 308.7 33.5% 84 Colleges 175,786 175,786 195,205 1.00 195,037 9.9% 56.2 6.1% 85 Hospitals 123,175 123,175 154,688 0.96 148,126 16.8% 2 5.6 2.8% 90 Gov. l eased interests 28,526 0.93 26,530 33.9 3.7% 97 Outdoor recreational 9,233 9,233 3,488 0.92 3,214 187.3% 6.7 0.7% Total institutional 26,988 26,395 26,766 22,880 0.96 22,022 21.5% 921.8 100.0% *HA/EA not available for subsector, average HA/EA of available subsectors used

PAGE 56

56 Table 3 17 Fitted lognormal probability density equations and associated Anderson Darling (A D) and Kolmogorov Smirnow (K S) statistics for State total and combined utili ty sample of commercial parcels FDOR c ode State total Combined utility sample N Fitted lognormal equation A D statis tic K S statistic N Fitted lognormal equation A D statistic K S statistic 11 39,002 Lognorm(6205,9267.4) 72.11 0.03 289 Lognorm(6915.4,9009.3) 2.01 0.07 12 18,931 Lognorm(3979.6 ,4892.6) 129.69 0.05 143 Lognorm(7664.1,8870.4) 1.84 0.09 13 829 Lognorm(130417.5,226144.1) 42.49 0.21 19 Lognorm(128058.9,44261.3) 0.44 0.16 14 2,571 Lognorm(10466.9,18876) 107.52 0.15 123 Lognorm(4533.9,4488) 8.91 0.21 15 449 Lognorm(504479.3,5563273. 4) 15.40 0.15 3 16 7,726 Lognorm(42572.1,101645) 30.59 0.06 239 Lognorm(41189.7,72650) 2.58 0.08 17 37,629 Lognorm(4189.8,6972.1) 357.76 0.08 384 Lognorm(5592.5,6137.9) 3.17 0.08 18 16,033 Lognorm(13586.8,56193.4) 322.42 0.11 73 Lognorm(31701.2,48051 .2) 0.45 0.07 19 20,953 Lognorm(5223.1,9767.9) 196.25 0.07 264 Lognorm(6619.7,6010.2) 4.03 0.12 20 2,825 Lognorm(5701,102409.2) 94.64 0.17 6 Lognorm(8147.7,16823.5) 0.38 0.21 21 7,629 Lognorm(4487.7,3618.4) 7.27 0.03 120 Lognorm(5078.8,3785.5) 1.91 0.10 22 4,377 Lognorm(3224.7,2156.5) 59.65 0.09 105 Lognorm(2992.3,1775.5) 4.36 0.16 23 4,732 Lognorm(6571.5,6560.8) 110.42 0.11 98 Lognorm(5030.5,2929.5) 3.60 0.15 24 283 Lognorm(4570.4,6424.4) 13.10 0.14 11 Lognorm(10121.5,17828.6) 0.29 0.17 25 4,309 Log norm(4647.3,5794.8) 35.40 0.06 49 Lognorm(5224.4,5678.8) 0.35 0.08 26 3,434 Lognorm(2487.9,1936.4) 13.02 0.05 5 Lognorm(1829.1,256.26) 0.32 0.23 27 15,129 Lognorm(6238.3,9484.1) 45.83 0.04 174 Lognorm(6043.8,7416.4) 1.48 0.09 29 729 Lognorm(14316,23222. 4) 1.69 0.04 5 Lognorm(24295.6,17153.8) 0.23 0.18 30 352 Lognorm(3075.9,3956.8) 0.69 0.04 2 31 28 Lognorm(39865,187722.3) 0.85 0.17 32 242 Lognorm(28010.4,54268.4) 2.03 0.08 3 33 1,914 Lognorm(4346.6,3742.9) 8.95 0.05 20 Lognorm(4556,3802.7) 0.45 0.17 34 492 Lognorm(30297.7,81922.8) 11.72 0.13 3 35 742 Lognorm(18060.9,73944.1) 1.59 0.04 36 330 Lognorm(8234,14388.4) 1.15 0.05 37 129 Lognorm(28736.7,139346.2) 3.12 0.13 38 1,392 Lognorm(25060.1,84420.3) 26.99 0.10 26 Lognorm(2 0768.6,37549) 1.11 0.19 39 21,702 Lognorm(4718,18708.6) 1399.08 0.21 50 Lognorm(37478.7,54315.1) 0.86 0.11

PAGE 57

57 Table 3 18 Fitted lognormal probability density equations and associated Anderson Darling (A D) and Kolmogorov Smirnow (K S) statistics for State total and combined utility sample of industrial parcels FDOR code State total Combined utility sample N Fitted lognormal e quation A D s tatistic K S s tatistic N Fitted lognormal e quation A D s tatistic K S s tatistic 41 18,398 Lognorm(13375.8,32016.2) 137. 52 0.08 33 Lognorm(35589.2,82016.2) 1.05 0.22 42 677 Lognorm(70712.4,196744.3) 0.35 0.02 3 43 471 Lognorm(33928.7,86991.8) 0.98 0.04 1 44 526 Lognorm(39030,99836.6) 0.87 0.03 45 90 Lognorm(121878.7,494517.2) 0.25 0.05 1 46 295 Lognorm(372 83.9,113069.3) 1.81 0.06 47 598 Lognorm(16233,50857.9) 0.59 0.03 10 Lognorm(128958.6,1274954) 0.16 0.12 48 42,406 Lognorm(17258.9,39654.1) 99.08 0.03 228 Lognorm(31655.2,54270.5) 1.04 0.06 49 2,190 Lognorm(12588.5,66705) 38.64 0.08 19 Lognorm(2249.2 ,2004.3) 0.66 0.15 91 3,185 Lognorm(6195.7,24310.6) 37.32 0.09 12 Lognorm(23727.1,22734.2) 0.91 0.21 92 112 Lognorm(14980.6,58563.9) 1.60 0.11

PAGE 58

58 Table 3 19 Fitted lognormal probability density equations and associated Anderson Darling (A D) and Kol mogorov Smirnow (K S) statistics for State total and combined utility sample of institutional parcels FDOR c ode State total Combined utility sample N Fitted lognormal equation A D s tatistic K S s tatistic n Fitted lognormal equation A D s tatistic K S s tat istic 71 18,888 Lognorm(9755.6,15478.2) 58.40 0.03 337 Lognorm(12941.2,18379.4) 1.53 0.06 72 3,243 Lognorm(12686.5,25968.3) 53.08 0.09 111 Lognorm(7458.2,7491) 0.92 0.09 73 570 Lognorm(149847.5,612156.5) 2.21 0.05 6 Lognorm(301959.4,981345.9) 0.25 0.19 74 4,447 Lognorm(6225.8,24872.7) 245.89 0.21 12 Lognorm(127980.6,448598.9) 0.52 0.19 75 2,214 Lognorm(9998.3,20236.8) 9.98 0.05 82 Lognorm(9547.1,28900.4) 0.61 0.09 76 887 Lognorm(6442.2,7980.9) 13.90 0.10 13 Lognorm(6581.3,3300.4) 0.32 0.21 77 3,244 L ognorm(6931.3,8460.6) 3.36 0.03 55 Lognorm(13045.8,14582.2) 0.43 0.10 78 457 Lognorm(48378,98509.5) 12.84 0.16 1 79 234 Lognorm(9926.2,20234.2) 1.25 0.07 1 81 43 Lognorm(68159.2,204065.6) 1.01 0.15 82 850 Lognorm(6238.2,15500.5) 0.68 0.02 1 83 2,953 Lognorm(465926.8,5919503.5) 225.66 0.21 52 Lognorm(130157.5,94366.6) 2.62 0.18 84 288 Lognorm(423964.5,5902238.9) 1.69 0.04 9 Lognorm(199478.1,1154125.5) 0.61 0.23 85 173 Lognorm(231446.7,1827373.5) 1.81 0.07 3 90 1,276 Lognorm(27445.3,14 9544.1) 7.07 0.07 97 2,090 Lognorm(2856.7,6198.6) 7.71 0.06 1

PAGE 59

59 Table 3 20 Fitted lognormal shift probability density equations and associated Anderson Darling (A D) and Kolmogorov Smirnow (K S) statistics for State total and combined utility sam ple of commercial parcels FDOR c ode State total Combined utility sample N Fitted lognormal equation A D s tatistic K S s tatistic n Fitted lognormal equation A D s tatistic K S s tatistic 11 39,002 Lognorm(6191.4,9148.7,Shift( 9.8419)) 73.11 0.03 289 Lognor m(6879.1,10162.8,Shift( 219.34)) 1.37 0.06 12 18,931 Lognorm(3971.2,4809.7,Shift( 10.617)) 132.14 0.05 143 13 829 19 Lognorm(105097.2,46236,Shift (23164)) 0.41 0.14 14 2,571 Lognorm(10491,19976.4,Shift( 112.83)) 99.82 0.14 123 Lognorm(3851.9,6063.4,S hift(8 35.36)) 5.72 0.15 15 449 Lognorm(677996.5,11046462.1, Shift(320.63)) 16.47 0.16 3 16 7,726 Lognorm(43500.7,109776.5,Shi ft(162.45)) 27.80 0.05 239 Lognorm(42574.9,88032.7,Shif t(871.69)) 1.96 0.08 17 37,629 384 Lognorm(5363.2,9389.2,Shift(7 02.9 )) 0.28 0.03 18 16,033 Lognorm(13597.5,56316,Shift( 0.55288)) 321.85 0.11 73 Lognorm(33856.1,73791.5,Shif t(1683.9)) 0.41 0.07 19 20,953 Lognorm(4878.2,7364.5,Shift( 108.5)) 114.84 0.05 264 Lognorm(5952.2,7472.9,Shift(8 68.41)) 2.02 0.08 20 2,825 Lognorm(6 270.7,127055.2,Shift (0.71497)) 90.27 0.17 6 21 7,629 120 Lognorm(8645,2775.3,Shift( 3737.2)) 0.53 0.06 22 4,377 105 Lognorm(22534,1247.4,Shift( 19623)) 0.90 0.06 23 4,732 98 24 283 Lognorm(4425.7,7098.1,Shift(2 08.47)) 10.89 0.13 11 Logno rm(15285.7,77251.8,Shif t(831.87)) 0.27 0.16 25 4,309 Lognorm(4629,6006.4,Shift(53. 395)) 32.32 0.06 49 Lognorm(5224.9,5675.9,Shift( 1.0112)) 0.35 0.08 26 3,434 5 Lognorm(2263.2,254.85,Shift( 434.2)) 0.32 0.23 27 15,129 Lognorm(6230.3,9407.6,Shift( 6.5 661)) 46.40 0.04 174 Lognorm(6048.8,7296.1,Shift( 27.336)) 1.44 0.09

PAGE 60

60 Table 3 20. Continued FDOR c ode State total Combined utility sample N Fitted lognormal equation A D s tatistic K S s tatistic n Fitted lognormal equation A D s tatistic K S s tatistic 29 729 Lognorm(14417.7,24479.5,Shif t(103.52)) 1.57 0.04 5 Lognorm(41463.5,13411.9,Shif t( 17713)) 0.19 0.18 30 352 Lognorm(3071.4,4261.6,Shift(5 6.464)) 0.81 0.04 2 31 28 Lognorm(66864.5,776190.2,Sh ift(702.06)) 0.37 0.12 32 242 Lognorm(29357.2,68186.4 ,Shif t(573.64)) 1.28 0.06 3 33 1,914 20 Lognorm(3892.9,6069.7,Shift(9 85.81)) 0.20 0.10 34 492 Lognorm(24285.3,38140.6,Shif t( 1017.9)) 6.90 0.09 3 35 742 Lognorm(19303.9,88847.3,Shif t(49.97)) 1.65 0.04 36 330 Lognorm(8076.2,12907.7,Shift( 9 6.975)) 0.81 0.05 37 129 Lognorm(33492.9,215636.3,Sh ift(195.85)) 2.16 0.11 38 1,392 Lognorm(20406.4,45509.6,Shif t( 373.38)) 14.02 0.07 26 39 21,702 Lognorm(4693.3,18461.7,Shift( 0.81817)) 1403.58 0.21 50 Lognorm(38324.6,33788.4,Shif t( 4828.5) ) 0.47 0.09

PAGE 61

61 Table 3 21 Fitted lognormal shift probability density equations and associated Anderson Darling (A D) and Kolmogorov Smirnow (K S) statistics for State total and combined utility sample of industrial parcels FDOR c ode State total Combined u tility sample N Fitted lognormal equation A D s tatistic K S s tatistic n Fitted lognormal equation A D s tatistic K S s tatistic 41 18,398 Lognorm(13469,33184,Shift(41 .897)) 126.86 0.07 33 42 677 Lognorm(69618.8,187617.9,Sh ift( 100.43)) 0.32 0.02 3 43 471 Lognorm(33597.4,84239.6,Shif t( 41.52)) 0.92 0.04 1 44 526 Lognorm(37471.6,87439.9,Shif t( 195.3)) 0.50 0.03 45 90 Lognorm(115994.8,434806.5,S hift( 164.33)) 0.24 0.05 1 46 295 Lognorm(39449,135557,Shift(2 29)) 1.19 0.05 47 598 Lognor m(16955,58041.8,Shift( 58.741)) 0.39 0.02 10 Lognorm(15291500000,181412 .3,Shift( 15291500000)) 2.26 0.38 48 42,406 Lognorm(17209.7,39262.3,Shif t( 8.7696)) 102.42 0.03 228 Lognorm(31697.6,54708.6,Shif t(29.426)) 1.02 0.06 49 2,190 Lognorm(7474.9,18983.8,Shi ft( 72.531)) 2.16 0.03 19 Lognorm(66065071.7,3037.2,S hift( 66062608)) 2.91 0.32 91 3,185 Lognorm(6754.5,31730.1,Shift( 40.436)) 23.92 0.07 12 Lognorm(30900.2,243072.7,Sh ift(8314.8)) 0.21 0.11 92 112 Lognorm(1843610000,49323.8, Shift( 1843590000)) 24.11 0.3 5

PAGE 62

62 Table 3 22 Fitted lognormal shift probability density equations and associated Anderson Darling (A D) and Kolmogorov Smirnow (K S) statistics for State total and combined utility sample of institutional parcels FDOR code State total Combined util ity sample N Fitted lognormal equation A D s tatistic K S s tatistic n Fitted lognormal equation A D s tatistic K S s tatistic 71 18,888 Lognorm(9492.5,14051.4, Shift( 52.418)) 68.23 0.04 337 72 3,243 Lognorm(12106.6,22765, Shift( 41.379)) 53.10 0.09 11 1 Lognorm(7117.8,8751.9,Shift(5 37.89)) 0.59 0.08 73 570 6 74 4,447 Lognorm(8023.4,54233.7, Shift(117.32)) 164.38 0.17 12 75 2,214 Lognorm(10167.6,21827.5, Shift(62.384)) 8.36 0.05 82 Lognorm(9852.3,31747.8,Shift( 23.7)) 0.60 0.09 76 887 13 L ognorm(18004.1,2742.8,Shift( 11492)) 0.22 0.15 77 3,244 55 Lognorm(13149.6,13893.8,Shif t( 228.41)) 0.42 0.10 78 457 Lognorm(48081.6,95982.2, Shift( 88.963)) 12.71 0.16 1 79 234 Lognorm(10208.5,23085.5, Shift(113.43)) 0.90 0.06 1 81 43 Lognorm( 85496900000,180027 .1, Shift( 85496900000)) 9.93 0.38 82 850 Lognorm(6588.5,18670.1, Shift(50.818)) 0.74 0.03 1 83 2,953 Lognorm(110758.3,142790.4, Shift( 10436)) 47.42 0.09 52 Lognorm(150120.7,73721.4,Sh ift( 24686)) 2.57 0.19 84 288 Lognorm(27150 0.7,2238345.1, Shift( 96.53)) 1.08 0.06 9 85 173 Lognorm(293003.6,3293062.8, Shift(523.62)) 1.25 0.08 3 90 1,276 Lognorm(27953.8,156847, Shift(11.64)) 6.54 0.07 97 2,090 Lognorm(2841,6084.1, Shift( 2.6213)) 8.01 0.06 1

PAGE 63

63 Figure 3 1 Lev els of Florida Department of Revenue (FDOR) land use disaggregation into 9 residential and 55 commercial, insdustrial, and institutional (CII) sectors Figure 3 2 Schematic of spatial and attribute database relations hips to FDOR Urban Water Use Residential Single Family FDOR 01, 02, 04 Multi family FDOR 03, 05 08, 28 CII Commercial FDOR 11 27, 29 39 Industrial FDOR 41 49, 91 92 Institutional FDOR 71 79, 81 85, 90, 97

PAGE 64

64 Figure 3 3 Macro to nano scale evaluation of public water use in Florida Figure 3 4 Heated and effective area correlation for 3,205 CII parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities y = 0.9526x R = 0.9921 0 200,000 400,000 600,000 800,000 1,000,000 1,200,000 Heated Area (sf) Effective Area (sf)

PAGE 65

65 Figure 3 5 Residual plot of heated and e ffective area simple linear regression for 3,205 CII parcels in Hillsborough County Water Resources Services and Gainesville Reg ional Utilities 200,000 150,000 100,000 50,000 0 50,000 Heated Area Residuals (sf) Effective Area (sf)

PAGE 66

66 CHAPTER 4 FLORIDA CASE STUDIES ON EVALUATING WATER USE Utility Water Billing Databases Water billing data var y widely in both content and availability depending on the policies of the individual utilities. Without standardized databases, these differences make water use comparisons across utilities very difficult. In our study p arcel level land use characteristics from t he Florida Department of Revenue ( FDOR ) and Florida County Property Appraiser ( FCPA ) database were linked with historic water billing data for 3,205 commercial, industrial, and institutional ( CII ) parcels ( 1, 768 in Hillsborough County Water Resources Servi ces ( H CW RS ) and 1,437 in Gainesville Regional Utilities ( GRU ) ) to develop water use coefficients normalized by heated building area. HCWRS provided four complete years of monthly water billing from January 2003 through December 2006, while GRU supplied two complete years of monthly water billing from January 2008 to December 2009. Hillsborough County Water Resources Services HCW RS data provided utility billing data for 1,768 CII accounts (67% commercial, 9% industrial, and 24% institutional) for 48 months b eginning in January 2003 The FDOR and Hillsborough County Property Appraiser (HCPA) property attribute data are updated annually. These databases include the year built so it is possible to show how the size of any sector has changed over time. Thus, it i s possible to adjust the monthly water use statistics to the number of accounts that existed in each of the years from 2003 to 2006. By using FDOR land use codes, CII water customers were grouped into the commercial industrial and institutional sector s HCW RS provided the crucial link to FDOR via a parcel ID. Parcel ID is th e common identifier which allows parcel attributes from FDOR to be related to water use. HCW RS data also provide d valuable time series information about the nature of water use in its CII sectors. By

PAGE 67

67 using FDOR land use codes, CII water customers can be aggregated into either the commercial, industrial, or institutional sector ( Figure 3 1 ) The time series plots of these sectors, presented in Figure 4 1 show s minimal trend ing and mont hly variability among the sectors. By obtaining the average water use for each month within the four year time series, the minimum and maximum average monthly water use is taken to be the base and peak usage, respectively. The seasonal component of the tim e series is extracted by subtracting the base usage from the overall monthly average water use as shown in Equation 4 1 ( 4 1 ) Where: Q Seasonal = a verage monthly seasonal water use Q Avg = a verage monthly water use Q Base = m inimum month water use The seasonal component divided by the average usage ( Equation 4 2 ) results in an estimate of percent seasonality, a measure of the significance of seasonal water use for the given system or sector ( 4 2 ) W here: % Seasonality = e stimate of percent of average monthly water use that is seasonal Q Seasonal = a verage monthly seasonal water use Q Avg = a verage monthly water use These water use statistics per sector are presented in Table 4 1 for HCWRS. The percent seasonality estimates are 3.2%, 7.8%, and 11.0% for the commercial, industrial and institutional sectors, respectively. Such seasonal estimates indicate that point estimate s of water use coefficients are reasonable for the CII sectors. The lack of seasonality within the CII sectors of

PAGE 68

68 HCWRS is also apparent in the time series information presented in Figure 4 1 This figure also indicates minimal trending within the sectors following a least square regression trend line. The trend lines in Figure 4 1 indicate a 0.004%, 0.007%, and 0.005% monthly change in total water use amongst the commercial, industrial, and institutional sectors, respectively. Such minimal trending, given that fact that only customers with water billing throughout the entirety of the billing period were included in the time series, once again indicated that point estimates of water use are appropriate for the CII sectors. To develop the water use coefficients, the HCW RS water billing records were adjusted so that all customer reco rds on a given parcel in a given month were summed. This procedure aggregated the water billing to the parcel level, and accounted for customers with multiple water use entries for a given month. Multiple entries are common in billing records, where a util ity seeks to correct for a billing over/under charge with a separate billing entry. This method of aggregating the billing records maintains the water use in a time series for each parcel, and allows for parcel land use classification via FDOR. Only parcel s reporting monthly water use through the entire study period were included. The adjusted billing records for 1,177 commercial parcels, 163 industrial parcels, and 428 institutional parcels were then linked to FDOR and HCPA via the unique parcel ID. Gaines ville Regional Utilities GRU supplied two complete years of monthly water billing from January 2008 to December 2009. Unlike HCWRS, the GRU billing records did not include a Parcel ID field. To arrive at this field, a geocoded GIS point shapefile which inc luded the utility specific, [Premise_Key], was provided by GRU. This file was spatially joined to the FDOR parcel polygon shapefile, so that every [Premise_ Key] was tagged with their corresponding Parcel ID. This spatial join resulted in a look up table l inking [Premise_Key] to [Parcel_ID] which allowed

PAGE 69

69 billing records provided by GRU to be joined with the parcel attributes from FDOR and Alachua County Property Appraiser. The GRU water billing records were also not rectified, meaning that each customer ha d billing periods which differed depending on the date in which each meter was read. In order to compare and aggregate the water billing within sectors, the water billing records required rectification. The water billing records included several key fields that made rectification possible, these included: Usage quantity Usage units Bill period length Bill period end date [USAGE_QUANTITY] is the total water billed to a customer for a given billing period. [ USAGE_ U NITS] is the units in which this water bill ing is reported. [BILL_PERIOD_LENGTH] is the number of days which encompass the billing period. This field allows water use per day to be calculated for each billing period, which is key to rectifying billing records. [BILL_ PERIOD END DATE] is the date in which a given billing period ends. all the required information to rectify billing records following Equation 4 3

PAGE 70

70 ( 4 3 ) Where: Q rectified = rectified month water use M = number of days in rectification month x = water y = water use per day for following billing period (M B) is less than the following billing period With the billing rectified, the water billing records were then grouped so that all customer records on a given parc el in a given month were summed. The time series plots for the aggregated 1,037 commercial parcels, 144 industrial parcels and 256 institutional parcels in GRU are presented in Figure 4 2 This figure show s the lack of monthly variability amon g the CII sectors as well as minimal trending The trend lines in Figure 4 2 indicate a 0.02%, 0.04%, and 0.01% monthly change in total water use amongst the commercial, industrial, and institutional sectors, respectively. O nly parcels reporting monthl y water use through the entirety of the study period were included in the analysis. These water use statistics per sector for GRU are shown in Table 4 2 The percent seasonality estimates are 7.2%, 17.8%, and 10.1% for the commercial, industrial and instit utional sectors, respectively. Such small percent seasonal estimates once again demonstrate that point estimates of water use are reasonable for the CII sectors. Combined Utilities In order to increase the overall sample size and include as many different customers with dissimilar building and water use characteristics, the billing records from HCWRS and GRU were combined. The average monthly water use per account time series for the commercial sector in HCWRS, GRU and the two utilities combined is present ed in Figure 4 3 Similar plots are available for the industrial, and institutional sectors ( Figure 4 4 Figure 4 5 ). The presentation

PAGE 71

71 of this time series information once again shows the lack of seasonality within the CII sectors of water use. These figur es also demonstrate that the average water use per account across the utilities is relatively similar, despite such sectors being composed of varied subsectors. In order to ease the bias of different land use mixes within the two utilities Table 4 3 Tabl e 4 4 and Table 4 5 present the average gallons per account per day (gpad) for HCWRS, GRU, and the two utilities combined, for the commercial, industrial, and institutional subsectors, respectively. The percent difference in gpad between HCWRS and GRU is calculated to provide a measure of the water use similarity of water users across utilities. Subsector percent differences in gpad across utilities ranged from 92% to 274%. Differences in the type of water users and measures of size, as shown in the previ ous chapter, are evident across utilities. Normalization through a measure of size is r equired to compensate for the heterogeneous nature of CII water users. The measure of size proposed in this methodology is heated building area as was discussed in the p revious chapter The relationship between water use and heated building area is evaluated in the following section. Relationship of Heated Area to Water Use The ability for heated area to be a good estimator of water use is cr itical for it s utiliz ation a s a measure of size to normalize water use. By linking the water billing databases of HCWRS and GRU with the land use databases of FDOR and FCPA, the relationship between property attributes and CII water use can be evaluated. T he strong correlation between heated area and water use for all 3,205 CII parcels in HCWRS and GRU shown in Table 4 6 indicates that heated area with a correlation coefficient of 0.63 is good predictor of water use within the CII sector Other property attributes such as parcel are a and effective year built, can be evaluated alongside heated area through stepwise multivariate regression (Neter et al. 1996) The stepwise

PAGE 72

72 regression was carried out using StatTools (Palisade Tools 2010), which employs a method that model s the choice of entering predictive variables based on their p value, whereby if the p value is less than 0.05, the variable is entered in the regression, and if the p value is greater than 0.1, then the variable leaves the regression. The resul t for this stepwise regres sion is presented in Table 4 7 T he table shows that all three predictive variables are entered in the regression and provides their coefficient s in the regression equation, which passed through the origin The adjusted R 2 value of the stepwise regression equation ( Equation 4 4 ) is shown in Table 4 8 to equal 0.41. The ANOVA ( Table 4 9 ) shows that the p value for the overall regression equation is less than 0.0001, which indicates that the relationship between these variables in predicting water use is hig hly statistically significant. ( 4 4 ) Where: Q i = a verage gallons per day water use for parcel i HA i = h eated square footage of all buildings on parcel i EYBi = e ffective year built of major improvements on parcel i (e.g., 1 984) T A i = p arcel area of parcel i in acre s The order that the predictive variables enter the regression model is dependent on both their correlation to water use and to each other, and is critical in determining the best fit regression model. From Tab le 4 10 it is shown that heated square footage is the first predictive variable entered into the regression, given that it is the most highly correlated variable to water use. The Adjusted R 2 value for the regression model of water use solely using heated square footage as the predictive variable is 0.39. Hence, by adding effective year built, and parcel acreage, little predictive power is gained, since the overall regression equation produces an adjusted R 2 of 0.41.

PAGE 73

73 Since heated area is the best predictor of water use available from the property attributes evaluated, and little is gained from the other variables, this methodology proposes water use relationships be based solely on heated square footage of buildings on a parcel. The regression statistics th rough the origin between heated square footage and average daily water use for the individual FDOR CII subsectors in the combined utilities is presented in Table 4 11 Though p values across subsectors vary from <0.0001 to 0.6352, for the most part the rel ationship between these two variables is highly significant. Heated square footage is also a good measure of size for CII parcels since it is available free for every parcel in the State through the FDOR database, and relationships between effective and he ated areas presented in this thesis. This database is of high quality since it is used for setting property taxes and is updated annually. Heated area is also a standardized area across most fields and outside the state of Florida, whereby heated area is a ll building areas under climate control. Such a standardized metric as a measure of size allows water use coefficients normalized by heated area to be readily applied to other property databases outside of the State. Property databases such as FDOR and FCP A, also provide an added benefit in that they provide heated area at the parcel level, which is a finer spatial resolution than Traffic Analysis Zones (TAZ). TAZ is the finest geographical area by which the U.S. Census aggregates employment figures. In the state of Florida, there are 12,747 TAZs and nearly 9 million parcels. Parcel level data allows for greater precision in estimating water use, and identifying sectors and drivers of demand. The following chapter will address the development of CII water u se coefficients, carried out by using heated area to normalize water use.

PAGE 74

74 Table 4 1 Summary statistics for CII sectors in Hillsborough C ounty Water Resources Services Commercial Industrial Institutional Sample size 1,177 163 428 Total (MGD) 2.64 0 .29 0.86 Base (MGD) 2.56 0.27 0.77 Base month September January January Peak (MGD) 2.70 0.30 0.97 Peak month May August August Seasonal (MGD) 0.09 0.02 0.10 % Seasonal 3.2% 7.8% 11.0% Table 4 2 Summary statistics for CII sectors in Gainesville Reg ional Utilities Commercial Industrial Institutional Sample size 1,037 144 256 Total (MGD) 1.50 0.18 0.57 Base (MGD) 1.39 0.15 0.51 Base m onth December December December Peak (MGD) 1.62 0.21 0.63 Peak m onth May July September Seasonal (MGD) 0.11 0. 03 0.06 % Seasonal 7.2% 17.8% 10.1%

PAGE 75

75 Table 4 3 Comparison of water use per commercial account for Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined FDOR c ode Description HCWRS GRU Combined u tility n Total water use (gal/d) Water use per account (gpa d) n Total water use (gal/d) Water use per account (gp a d) Utility % diff in gp a d n Total water use (gal/d) Water use per account (gp a d) 11 Stores, one story 113 78,275 693 176 137,392 781 13 % 289 215,668 746 12 Mixed use 101 131,528 1,302 42 19,957 475 64% 143 151,485 1,059 13 Department stores 18 148,133 8,230 1 2,344 2,344 72% 19 150,477 7,920 14 Supermarkets / conv. s tores 117 184,474 1,577 6 7,833 1,305 17% 123 192,306 1,563 15 Re gional malls 2 143,501 71,750 1 43,354 43,354 40% 3 186,855 62,285 16 Community shopping centers 161 733,342 4,555 78 193,460 2,480 46% 239 926,802 3,878 17 Office, one story 152 148,692 978 232 147,634 636 35% 384 296,327 772 18 Office, multi story 41 124,393 3,034 32 30,168 943 69% 73 154,561 2,117 19 Medical office 107 94,648 885 157 219,951 1,401 58% 264 314,599 1,192 20 Transit terminals 5 18,622 3,724 1 217 217 94% 6 18,840 3,140 21 Restaurants 68 285,242 4,195 52 150,690 2,898 31% 120 435 ,932 3,633 22 Fast food restaurant s 55 116,724 2,122 50 84,187 1,684 21% 105 200,911 1,913 23 Financial institution s 63 68,994 1,095 35 117,807 3,366 207% 98 186,801 1,906 24 Insurance company offices 11 8,604 782 11 8,604 782 25 Service shops 18 21,753 1,208 31 24,744 798 34% 49 46,496 949 26 Service stations 5 1,555 311 5 1,555 311 27 Auto sales / repair 116 138,436 1,193 58 34,057 587 51% 174 172,493 991

PAGE 76

76 Table 4 3. Continued FDOR c ode Description HCWRS GRU Combined u tility n Total water use (gal/d) Water use per account (gpad) n Total water use (gal/d) Water use per account (gpad) Utility % diff in gpad n Total water use (gal/d) Water use per account (gpad) 29 Wholesale outlets 5 2,991 598 5 2,991 598 30 Florist / greenhouses 2 1,459 730 2 1,459 730 32 Enclosed theaters / auditoriums 1 11,802 11,802 2 6,653 3,326 72% 3 18,455 6,152 33 Nightclubs / bars 8 7,848 981 12 10,668 889 9% 20 18,516 926 34 Bowling alleys / skating rinks 1 1,228 1,228 2 2,532 1,266 3% 3 3,759 1,253 38 Golf courses / driving ranges 20 54,602 2,730 6 23,430 3,905 43% 26 78,032 3,001 39 Hotels / motels 10 99,301 9,930 40 278,572 6,964 30% 50 377,873 7,557 Total commercial 1,177 2,611,538 2,219 1,037 1,550,259 1,495 33% 2,214 4,161,797 1,880

PAGE 77

77 Table 4 4 Comparison of water use per industrial account for Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined FDOR c ode Description HCWRS GRU Combined u tility n Total wat er use (gal/d) Water use per account (gpad) n Total water use (gal/d) Water use per account (gpad) Utility % diff in gpad n Total water use (gal/d) Water use per account (gpad) 41 Light manufacturing 10 14,866 1,487 23 55,852 2,428 63% 33 70,718 2,143 42 Heavy industrial 2 3,515 1,758 1 1,397 1,397 21% 3 4,912 1,637 43 Lumber yards 1 10,765 10,765 1 10,765 10,765 45 Bottler / canneries 1 339 339 1 339 339 47 Mineral processing 8 140,064 17,508 2 280 140 99% 10 140,344 14,034 48 War ehousing / distribution 118 120,308 1,020 110 113,492 1,032 1% 228 233,800 1,025 49 Open storage 17 6,410 377 2 703 352 7% 19 7,113 374 91 Utility, gas & elec. 8 2,836 354 4 5,303 1,326 274% 12 8,139 678 Total industrial 163 287,998 1,767 144 188,13 2 1,306 26% 307 405,412 1,321

PAGE 78

78 Table 4 5 Comparison of water use per institutional account for Hillsborough County Water Resources Services, Gainesville Regional Utilities, a nd the two utilities combined FDOR c ode Description HCWRS GRU Combined u tilit y n Total water use (gal/d) Water use per account (gpad) n Total water use (gal/d) Water use per account (gpad) Utility % diff in gpad n Total water use (gal/d) Water use per account (gpad) 71 Churches 221 118,775 537 116 98,282 847 58% 337 217,057 6 44 72 Private schools & colleges 65 70,238 1,081 46 41,872 910 16% 111 112,110 1,010 73 Private hospitals 2 73,145 36,572 4 77,174 19,294 47% 6 150,319 25,053 74 Homes for the a ged 12 140,993 11,749 12 140,993 11,749 75 Orphanages / n on profi ts 67 122,526 1,829 15 20,027 1,335 27% 82 142,552 1,738 76 Mortuaries / cemeteries 7 3,344 478 6 9,387 1,565 228% 13 12,731 979 77 Clubs / union halls 15 25,422 1,695 40 105,102 2,628 55% 55 130,524 2,373 78 Sanitariums / convalescents 1 16,464 16,464 1 16,464 16,464 79 Cultural organizations 1 375 375 1 375 375 82 Parks and r ecreation 1 1,209 1,209 1 1,209 1,209 83 Public county schools 51 448,492 8,794 1 1,474 1,474 83% 52 449,966 8,653 84 Colleges 9 54,621 6,069 9 54,621 6,069 85 Hospitals 3 2,208 736 3 2,208 736 97 Outdoor recreational 1 239 239 1 239 239 Total institutional 428 861,941 2,014 256 569,429 2,224 10% 684 1,431,370 2,093

PAGE 79

79 Table 4 6 Correlation matrix of Florida Department of Re venue (FDOR) and Florida County Property Appraiser (FCPA) property attributes and water use for CII parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities Heated a rea (sf) Effective a rea (sf) Parcel a rea (acres) Effect ive year b uilt Average monthly w ater u se Heated a rea (sf) 1.000 Effective a rea (sf) 0.996 1.000 Parcel a rea (acres) 0.347 0.356 1.000 Effective year built 0.028 0.030 0.003 1.000 Average monthly water use 0.631 0.639 0.096 0.021 1. 000 Table 4 7 Stepwise regression results for CII parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities Coefficient Standard e rror t v alue p v alue Confidence i nterval 95% Lower Upper Constant 0 N/A N/A N/A N /A N/A Heated a rea (sf) 0.0639 0.0014 46.9384 < 0.0001 0.0612 0.0665 Effective year built 0.3727 0.0355 10.4866 < 0.0001 0.3030 0.4424 Parcel a rea (acres) 0.6461 0.0652 9.9052 < 0.0001 0.7740 0.5182 Table 4 8 of measures for CII parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities Multiple R R 2 Adjusted R 2 StErr of estimate 0.6439 0.4146 0.4142 3738.1719 Table 4 9 ANOVA table for stepwise regression of CII parcels in Hill sborough County Water Resources Services and Gainesville Regional Utilities Degrees of f reedom Sum of s quares Mean of s quares F r atio p v alue Explained 3 31519189783 10506396594 751.8570 < 0.0001 Unexplained 3185 44506963976 13973929

PAGE 80

80 Table 4 10 S tep information for stepwise regression of CII parcels in Hillsborough County Water Resources Services and Gai nesville Regional Utilities Multiple R R 2 Adjusted R 2 StErr of e stimate Enter or e xit Heated a rea (sf) 0.6275 0.3938 0.3938 3802.7815 Enter Eff ective year b uilt 0.6297 0.3966 0.3964 3794.7158 Enter Parcel a rea (acres) 0.6439 0.4146 0.4142 3738.1719 Enter

PAGE 81

81 Table 4 11 Regression statistics between heated square footage and average daily water use for the FDOR CII subsectors in Hillsborough Count y Water Resources Services and Gainesville Regional Utilit ies FDOR c ode Description p value FDOR c ode Description p value 11 Stores, one story 0.0126 41 Light manufacturing 0.6252 12 Mixed use 0.0103 42 Heavy industrial 0.0302 13 Department store s 0.3288 43 Lumber yards 14 Supermarkets / conv. s tores <0.0001 44 Packing plants 15 Regional malls <0.0001 45 Bottler / canneries 16 Community shopping centers <0.0001 46 Food processing 17 Office, one story 0.0042 47 Mineral processing 0.0006 18 Office, multi story <0.0001 48 Warehousing / distribution 0.3862 19 Medical office <0.0001 49 Open storage <0.0001 20 Transit terminals <0.0001 91 Utility, gas & elec. <0.0001 21 Restaurants <0.0001 92 Mining / petroleum 22 Fast food restaurants <0.0 001 Total industrial <0.0001 23 Financial institutions <0.0001 71 Churches <0.0001 24 Insurance company offices 0.063 72 Private schools & colleges <0.0001 25 Service shops <0.0001 73 Private hospitals 0.0004 26 Service stations 0.6087 74 Homes for th e a ged 0.424 27 Auto sales / repair <0.0001 75 Orphanages / n on profits 0.0005 29 Wholesale outlets <0.0001 76 Mortuaries / cemeteries <0.0001 30 Florist / greenhouses 77 Clubs / union halls <0.0001 31 Drive in t heaters / o pen s tadiums 78 Sanitariums / convalescents 32 Enclosed theaters / auditoriums 0.0086 79 Cultural organizations 33 Nightclubs / bars 0.0007 81 Military 34 Bowling alleys / skating rinks 0.1708 82 Parks and r ecreation 35 Tourist attractions 83 Public county schools <0.0001 36 Camps 84 Colleges 0.0009 37 Race tracks 85 Hospitals <0.0001 38 Golf courses / driving ranges 0.0223 90 Gov. l eased interests 39 Hotels / motels <0.0001 97 Outdoor recreational Total commercial <0.0001 Total institutional <0.0001

PAGE 82

82 Figure 4 1 Time series plots of monthly water use for 1,177 commercial parcels, 163 industrial parcels, and 428 institutional parcels in Hillsborough County Water Resources Services y = 0.0001x 1.9135 R = 0.2506 y = 2E 05x + 0.9238 R = 0.0562 y = 4E 05x 0.7445 R = 0.0422 0.00 0.60 1.20 1.80 2.40 3.00 Jan 03 Mar 03 May 03 Jul 03 Sep 03 Nov 03 Jan 04 Mar 04 May 04 Jul 04 Sep 04 Nov 04 Jan 05 Mar 05 May 05 Jul 05 Sep 05 Nov 05 Jan 06 Mar 06 May 06 Jul 06 Sep 06 Nov 06 Water Use (MGD) Commercial Industrial Institutional

PAGE 83

83 Figure 4 2 Time series plots of monthly water use for 1,037 commercial parcel s, 144 industrial parcels, and 256 institutional parcels in Gainesville Regional Utilities y = 0.0003x + 14.594 R = 0.5446 y = 8E 05x + 3.425 R = 0.5297 y = 4E 05x + 1.9954 R = 0.0236 0.00 0.40 0.80 1.20 1.60 2.00 Water Use (MGD) Commercial Industrial Institutional

PAGE 84

84 Figure 4 3 Average monthly water use per commercial account in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utiliti es combined 0 600 1,200 1,800 2,400 3,000 Gallons per Month per Account GRU HCWRS Combined

PAGE 85

85 Figure 4 4 Average monthly water use per industrial account in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined 0 500 1,000 1,500 2,000 2,500 Gallons per Account per Month GRU HCWRS Combined

PAGE 86

86 Figure 4 5 Average monthly water use per institutional accoun t in Hillsborough County Water Resources Services, Gainesville Regional Utilities, and the two utilities combined 0 600 1,200 1,800 2,400 3,000 Gallons per Account per Month GRU HCWRS Combined

PAGE 87

87 CHAPTER 5 COMMERCIAL, INDUSTRI AL, AND INSTITUTIONA L WATER USE COEFFICI ENTS Introduction Commercial, industrial and institutional ( CII ) public water use activity coefficients were developed by linking the parcel level property attribute data with the water billing data from Hillsborough County Water Resources Services ( HCWRS ) and Gainesville Regional Utilities ( GRU ) Because the Florida Department of R evenue ( FDOR ) database identifies the classification of land use for each parcel, this database provides the means to aggregate or disaggregate the water use activity coefficients for sectors by different levels of specification. The most aggregated coeffi cients are for the general CII sectors. These coefficients can be disaggregated into 5 5 sub sectors based on the FDOR two digit land use categories. By this method, every utility in the s tate of Florida can determine the relative water use by different sub s ectors of customers in their service area. The relative use of each sector can be calibrated with known total water use in order to identify how the water is used at present and which sub sectors are the more significant water users. Four water use coeffi cients are presented in this chapter: average, base, seasonal, and May peak. The average water use coefficients were developed by summing the average monthly water use of all parcels within a given sub sector and dividing by their total heated area and the average number of days in the months billed ( Equation 5 1 ) This method of calculating the coefficients provides a weighted average which compensates for the skewness often found in the distribution of CII water users.

PAGE 88

88 ( 5 1 ) Where: AWU = a verage weighted water use coefficient (gallons/heated sq. ft./day) Q i = a verage monthly water use of parcel i (gallons/month) HA i = h eated square footage of all buildings on parcel i (square feet) AD = a verage number of days in months b illed (days) Peak ( Equation 5 2 ) and base ( Equation 5 3 ) water use coefficients were also developed by correspondingly summing the average May and average minimum monthly water use of all parcels in a sub sector, and dividing by the total heated area of th e sub sector. The average May usage is the peak month use for most water utilities in Florida. Thus, it is appropriate to use May as the peak water use of interest. Unlike the peak coefficient, where the overall system peak is of concern, the base coefficie nt provides a measure of the seasonality of a given sub sector, and is dependent on that given sub Only parcels reporting monthly water use through the entirety of the study period were included in the analysis. ( 5 2 ) Where: P WU = May peak weighted water use coefficient (gallons/heated sq. ft./day) Q May,i = a verage May monthly water use of parcel i (gallons/month) HA i = h eated square footage of all buildings on parcel i (square feet) AD May = a verage number of days in May months billed (days) ( 5 3 ) Where: BWU = b ase weighted water use coefficient (gallons/heated sq. ft./day) Q Min,i = a verage minimum monthly water use of parcel i (gallons/month) HA i = h eated square foota ge of all buildings on parcel i (square feet) AD M in = a verage number of days in minimum month months billed (days)

PAGE 89

89 T he measure of size used to normalize the water use data and develop the activity coeff icients is heated area that is available from the FCP A. As documented in Chapter 3, i n order to use these water use coefficients directly with the FDOR statewide database, effective building area must be converted to heated building area. The coefficients to convert from effective area (EA) to heated area (H A) are presented under the subheading of HA/EA. Commercial Sector The developed water use coefficients for the available commercial FDOR sub sector categories in HCWRS and GRU are shown in Table 5 1 and Table 5 2 respectively Th ese table s include the samp le sizes from which the coefficients were derived, the average effective year built and heated building areas, and percent seasonal, a measure of the significance of seasonal water use. This measure is obtained by subtracting the base water use coefficient from the average water use coefficient to arrive at the seasonal water use coefficient ( Equation 5 4 ) ( 5 4 ) Where: SWU = s easonal weighted water use coefficient (gallons/heated sq. ft./day) AWU = a verage weighted water use coefficient (gallons/heated sq. ft./day) BWU = b ase weighted water use coefficient (gallons/heated sq. ft./day) The seasonal weighted water use coefficient is then divided by the average water use coefficient to arrive at an estimate of the percentage of total water use that is seasonal ( Equation 5 5 ) ( 5 5 ) Where: % Seasonal = p ercentage of total water use that is seasonal SWU = s easonal weighted water use coefficient (gallons/heated sq. ft./day) AWU = a verage weighted wa ter use coefficient (gallons/heated sq. ft./day)

PAGE 90

90 Also included in th ese table s is the percent of total heated ar ea and water use in each sub sector. T hese columns provide a measure of the relative importance of each subsector when it come s to CII land use a nd water use for the given utility. Not shown in Table 5 1 are insurance company offices (FDOR 24), service stations (FDOR 26), wholesale outlets (FDOR 29), florists/greenhouses (FDOR 30), open stadiums (FDOR 31), enclosed theaters (FDOR 32), tourist attra ctions (FDOR 35), camps (FDOR 36), and race tracks (FDOR 37), since data on these commercial sub sectors was unavailable from HCWRS. The absent sub sectors for GRU are open stadiums (FDOR 31), tourist attractions (FDOR 35), camps (FDOR 36), and race tracks ( FDOR 37). Total Commercial Coefficient Calculation The total average commercial water use coefficient, located at the bottom of Table 5 1 and Table 5 2 is area weighted, and is thus heavily dependent on the commercial land use mix for a given utility. As shown in E quation 5 6, for the total average water use coefficient, a weighted average based on the total heated area of the two digit FDOR sub sectors is used For the area conversion coefficients, a similar operation follows, though the total effective area is used in the weighting ( Equation 5 7 ) ( 5 6 ) Where: A WU Sector = t otal sector w eighted average water use coefficient ( gallons/heated sq. ft./day) A WU j = w eighted average water use coefficient of subsector j (gall ons/heated sq. ft./day) HA Total,j = t otal heated square footage of all parcels in subsector j (sq. ft.)

PAGE 91

91 ( 5 7 ) Where: HA/EA Sector = t otal sector effect i ve area to heated area conversion coefficient HA/EA j = e ffect ive area to heated area conversion coefficient of subsector j EA Total,j = t otal effective square footage of all parcels in subsector j (sq. ft.) I f the same subsector water use coefficients a re applied to another utility where only the heated area is known, then the area weighted average for the commercial sector would reflect the relative importance of the commercial subsectors. Indeed, the area weighting provides an important improvement in the accuracy of CII estimates since the sizes of the various activities are included directly in the calculations. In the case of HCW RS the point estimate for total commercial average water use is the sum of the water use, divided by the summed heated buildin g va lues over the 1,17 7 commercial parcels in H CWRS. Following this calculation, the HCWRS commercial sector uses an area weighted average of 0.135 gallons of water per square foot of heated building area per day (gal/hsf/d), which is close to the GRU total co mmercial average water use coefficient of 0.130 gal/ h sf/d Total base and seasonal water use coefficients are dependent on the overall sectoral time series. Commercial Subsector Analysis at the Utility Level The two digit breakdown in Table 5 1 and Table 5 2 allows for the evaluation of which subsectors are the most important as judged by the combination of their water use rate and their size as measured by heated area. If HCWRS was to pursue water conservation in the commercial sector, for example, the level of disaggregation in Table 5 1 can justify target ing a specific class of customers. Restaurants (FDOR 21) have the highest rate of water use per square foot of heated area Though their h eated area only accounts for 1.8 % of the heated area for the co mmercial sector, the ir overall water use totals 10.9 % of the commercial water use. Supermarkets/

PAGE 92

92 convenience stores (FDOR 14) have many customers in that subsector and a relatively high water use rate. For GRU ( Table 5 2 ), restaurants (FDOR 21) and fast fo od restaurants (FDOR 22) once again have a high water use rate, and make up a significant fraction (15.1%) of commercial water use. The largest commercial water use subsector for GRU is hotels/motels (FDOR 39), which utilizes close to 18% of total commerci al water use for the utility, at an average rate of 0.220 gal/hsf/d. Such s ectors c ould be analyzed through further disaggregation for water conservation potential. The percent seasonal column in Table 5 1 and Table 5 2 allows for the evaluation of seasona lity at the subsector level. In Chapter 3 it was determined that the aggregate CII sectors of water use lack significant variability, and thus the argument was made that point estimates of water use were reasonable. At the commercial, industrial, and insti tutional level of aggregation however, it is impossible to determine whether the individual subsectors within the aggregated sectors actually lack seasonal variability. Each subsector might be highly variable, but at different points in the time series. Th us, the aggregation of the individual subsector time series might mute the overall sector time series. The percent seasonal estimates for the commercial subsectors in Table 5 1 and Table 5 2 indicate small seasonal components of water use throughout the ma jority of subsectors, further justifying the use of point estimates of water use. The May peak provides a good indication of the extent to which a sector impacts the utility wide peak. The May peak is caused primarily by irrigation needs during the spring dry season. CII use may be lower in May since a significant number of winte r residents have left Florida. In this case, the CII users may not be significant contributors to the May peak. Commercial Subsector Analysis across Utilities Through the availabi lity of water use and property attribute data from multiple utilities, evaluation of the inter utility water use coefficient differences between subsectors is possible.

PAGE 93

93 Heated area and water use comparisons across utilities were made in Chapters 3 and 4, r espectively. Inter utility differences between commercial water use coefficients at the subsector level show that for subsectors where both utilities had a significant sample size, water use co efficients were largely within a n order of magnitude ( Table 5 3 ) A significant sample size is important in analyzing CII water use given its heterogeneous nature of customers, and skewed measures of size. These water use coefficients and heated area statistics would greatly benefit from increased sample sizes provid ed by other utilities, particularly for the insignificantly sampled subsectors. Heated area s, as shown in Chapter 3, may vary widely within an FDOR subsector The limited number of 2 digit CII FDOR subsectors ensures grouping of multiple facility types wit h differing drivers of water use. Disaggregated groupings within subsectors are possible by developing size categories based on heated building area. Year built of a facility might also affect water use, given the requirement or availability of certain en d use devices at the time of construction. For example, the residential sector is broken up into three age groups (pre 1983, 1983 1994, 1995 present) corresponding with State regulations requiring minimum plumbing fixture water efficiencies. Even facilitie s within the same sub sector might offer new services requiring different end use water devices as they respond to changing conditions. Predicting what fixture types are prevalent in certain customer groups greatly improves estimates of water use, as well a s facilitates the weighing of water conservation options. Two sub sectors of note in Table 5 1 Table 5 2 and Table 5 3 are financial institutions (FDOR 23) and service shops (FDOR 25). The average water use coefficient for financial institutions in GRU i s over two times larger than that of HCWRS. Though their average year built is nearly identical, GRU financial institutions are close to 2,000 square feet larger than

PAGE 94

94 those in HCWRS, and thus likely represent a different customer subset of financial instit utions. Service shops in HCWRS use nearly four times as much water per square foot on the average as those in GRU. Service shops in HCWRS however, are slightly older than and nearly two and a half times as large as those in GRU. Thus, once again, service s tations across these two utilities are dissimilar in both their property attributes and in the way they use water. As described in Chapter 3, the a ppendix presents the subsector distributions of heated areas fo r the State and combined utilities which can be used to gauge the representativeness of the subsector samples total of CII parcels Aggregation of Commercial Utility Data Given that the water use and property attributes of the subsectors are similar across ut ilities, in order to increase our overall sample size and include as many different customer types as possible, the combined commercial coefficients including all sampled parcels in HCWRS and GRU are presented in Table 5 4 The top commercial water users f or the combined utilities are community shopping centers (FDOR 16), restaurants (FDOR 21), and hotels/motels. Data for the following commercial categories was unavailable from HCWRS or GRU: open stadiums (FDOR 31), tourist attractions (FDOR 35), camps (FDO R 36), and race tracks (FDOR 37). The combined parcel level water use and property attribute statistics presented in Table 5 4 provide a more comprehensive view of the commercial water use sector in Florida. Industrial Sector The available industrial sub s ector s along with their water use and area conversion coefficients are presented in Table 5 5 and Table 5 6 for HCWRS and GRU, respectively Like the commercial sector, the total industrial water use coefficient s are a calculated heated area weighted aver age of the eleven FDOR codes that make up that sector. The industrial sector accounts for the least amount of CII water users in both HCWRS and GRU With a total

PAGE 95

95 weighted average water use coefficient of 0.042 gal/hsf/d (HCWRS) and 0.070 gal/hsf/d (GRU) t his sector also presents the smallest rate of water use amongst the CII sectors. Not shown in Table 5 5 are lumber yards (FDOR 43), packing plants (FDOR 44), bottlers/canneries (FDOR 45), food processing (FDOR 46), and mining/petroleum (FDOR 92). Packing p lants (FDOR 44), food processing (FDOR 46), and mining/petroleum (FDOR 92) were the unavailable industrial subsectors from GRU. Industrial water use statistics are only for those industries that are served by the public water supply system. Many larger ind ustries are self supplied and need to be evaluated separately. Industrial Subsector Analysis at the Utility Level The largest industrial subsector in terms of parcel count and heated building area for both HCWRS and GRU is warehousing/distribution (FDOR 48 ), despite its relatively small average water use coefficient. Given that the water use calculation is a product of a sectors size (total heated building area) and water use coefficient, it is not surprising that warehousing/distribution is by far the larg est and second largest industrial water user in HCWRS and GRU at 60% and 42% of total industrial water use, respectively. Throughout the industrial subsectors, large average heated areas and small water use coefficients are prevalent Given this fact, it s eems fair to infer that these customers likely do not utilize their potable water connections for industrial processes. Industrial Subsector Analysis across Utilities The industrial subsectors lacked significant sample sizes for inter utility comparisons of water use and property attributes. For the two subsectors with significant sample sizes across utilities, light manufacturing (FDOR 23) and warehousing/distribution (FDOR 48), water use and property attributes vary considerably, evidence of the even gre ater heterogeneity of industrial water users ( Table 5 7 ) The industrial sector has the fewest number of FDOR

PAGE 96

96 categories across CII with only 11 subsectors. Despite having such few subsectors, the industrial sector comprises the most diverse number of cust omer types and drivers of demand in CII. Light manufacturing (FDOR 23) facilities in GRU use on the average 6.3 times more water per square foot than those in HCWRS. In addition, GRU light manufacturing customers are nearly 60,000 square feet smaller than those in HCWRS, a clear indication that these facilities across the two utilities vary significantly. Warehousing/distribution facilities in HCWRS are over twice the size as those in GRU, while utilizing nearly half as much water use per square foot on the average. The lack of disaggregation inherent in the FDOR classification of CII ensure that multiple customer types with drastically different property attributes and drivers of demand are grouped within single 2 digit FDOR subsectors Aggregation of Indus trial Utility Data The combined industrial coefficients including all sampled parcels in HCWRS and GRU are shown in Table 5 8 The top industrial water users for the combined utilities are warehousing/distribution (FDOR 48), mineral processing (FDOR 46), a nd light manufacturing (FDOR 41). Data for the following industrial categories was unavailable from HCWRS or GRU: packing plants (FDOR 44), food processing (FDOR 46), and mining/petroleum (FDOR 92). In this more comprehensive view of industrial public wat er use in the state of Florida, it is once again apparen t that small water use coefficients and large heated areas characterize industrial customers Institutional Sector The institutional sector is disaggregated into 18 FDOR subsectors The water use and area conversion coefficients for these institutional sub sectors are presented in Table 5 9 for HCWRS, and Table 5 10 for GRU Not shown in Table 5 9 are homes for the aged (FDOR 74), sanitariums/convalescents (FDOR 78), cultural organizations (FDOR 79), m ilitary (FDOR 81),

PAGE 97

97 parks and recreation (FDOR 82), colleges (FDOR 84), hospitals (FDOR 85), government leased interested (FDOR 90), and outdoor recreational (FDOR 97). Military (FDOR 81), and government leased interested (FDOR 90) were the unavailable inst itutional subsectors from GRU. Institutional Subsector Analysis at the Utility Level The largest institutional water user for HCWRS is public county schools (FDOR 83). Though this subsector has a relatively small average water use coefficient of 0.069 gal /hsf/d, public county schools in the utility have a large average heated building area of 127,905 square feet and there are many of them. Another subsector of note is churches (FDOR 71), which make up the largest institutional subsector for both HCWRS and GRU in terms of parcel count, while also being a significant institutional water user for both utilities. Institutional Subsector Analysis across Utilities The institutional subsectors largely share similar water use coefficients, largely within a measure of magnitude ( Table 5 11 ) Throughout the subsectors, average water use coefficients are stable towards the small side, and seasonality appears to be minimal. The relatively small water use coefficients indicate several possibilities in terms of end uses, such as utilization of private wells for irrigation. The small seasonal measures also justify the use of point estimates of water use Aggregation of Institutional Utility Data Given that the water use and property attributes of the subsectors are simil ar across utilities, in order to increase our overall sample size and include as many different customer types as possible, the combined institutional coefficients including all sampled parcels in HCWRS and GRU are presented in Table 5 12 The top institut ional water users for the combined utilities are public county schools (FDOR 83), churches (FDOR 71), and private

PAGE 98

98 hospitals (FDOR 73). Data for the following institutional categories was unavailable from HCWRS or GRU: military (FDOR 81), and government lea sed interested (FDOR 90). Coefficient Comparison with other Studies O ther studies have also developed CII water use coefficients. The American Water Works CI End Uses of Water (2000) analyzed the water use patterns of fiv e major CII subsectors, including supermarkets, office buildings, restaurants, hotels, and s Industrial, Commercial & Institutional Water Conservation (2007) report, presented water use coefficients or benchmarks for restaurants, hotels, schools and homes for the aged. Comparison across the coefficients presented in this thesis and others developed in previous studies can be carried out by mapping the FDOR subsectors used in this thesis accordingly: supermarkets (FDOR 14), office buildings (FDOR 17 & 18), restaurants (FDOR 21), hotels (FDOR 39), schools (FDOR 83), and homes for the aged (FDOR 74). The comparison of CII coefficients from this and previous studies is shown in Table 5 13 For the most part, the coefficien ts are comparable. The percent difference between coefficients presented in this thesis and other studies ranged from 38 % to 348% across the available subsectors. Large discrepancies in coefficients can be attributed to the other studies being specific t o the south western region of the United States where other factors such as varied climatic conditions can affect water use. Incorporation of Results into EZ Guide 2 By employing a measure of size that is standard and reliable across the CII sub sectors, al ong with default water use coefficients, any utility within the State can estimate the subsectoral breakdown of CII water use within their service boundary. The FDOR database is accompanied by polygon shapefiles that delineate every parcel in the State. Th is database can be queried to determine which parcels are within the service boundaries of a given utility South Florida Water

PAGE 99

99 Management District ( WMD ), St. Johns River WMD, and Southwest Florida WMD provide the water service area boundaries of utilities in their d istricts as polygon shapefiles available in their respective websites to be viewed in GIS The parc els are identified by a unique p arcel identification number which can be related to the FDOR database to find the attributes for the parcels in th e utility being analyzed. These shapefiles need to be checked to verify their accuracy. EZ Guide 2 utilizes the coefficients shown in Table 5 4 Table 5 8 and Table 5 12 to estimate CII water use for any utility in the State. These coefficients are applie d within the wat er budget section of EZ Guide 2 ( Figure 5 1 ). By estimating the individual water use for each CII subsector a utility or planning agency can develop a conservation strategy according to the relative importance and water use intensity of it s sub sectors. To estimate the amount of water use for the single and multi family residential sector s a similar data driven measure of size approach is also taken EZ Guide 2 is available free online, and the Conserve Florida Water Clearinghouse can assis t water utilities and water management districts in generating the necessary information ( http://www.conservefloridawater.org/ ).

PAGE 100

100 Table 5 1 Water use coefficients and sector statistics based on sample of 1,177 commercial parcels and four years of billing records from Hillsborough County Water Resources Services FDOR c ode Description Sample s ize Average effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square fo ot/day) Percent seasonal % H eated area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 11 Stores, one story 113 1989 9,375 0.96 0.074 0.067 0.007 0.078 9.0% 5.47% 3.00% 12 Mixed use 101 1975 13,601 0.96 0.096 0.092 0 .003 0.099 3.5% 7.10% 5.04% 13 Department stores 18 1995 130,076 0.96 0.063 0.056 0.008 0.065 11.9% 12.10% 5.67% 14 Supermarkets / conv. s tores 117 1991 5,576 0.97 0.283 0.248 0.035 0.306 12.2% 3.37% 7.06% 15 Regional malls 2 1997 820,551 0.87 0.087 0.0 80 0.007 0.081 8.4% 8.48% 5.49% 16 Community shopping centers 161 1989 39,269 0.95 0.116 0.114 0.002 0.116 1.5% 32.67% 28.08% 17 Office, one story 152 1985 5,446 0.96 0.180 0.170 0.010 0.184 5.4% 4.28% 5.69% 18 Office, multi story 41 1988 38,283 0.98 0. 079 0.075 0.004 0.084 5.1% 8.11% 4.76% 19 Medical office 107 1989 6,770 0.96 0.131 0.120 0.011 0.143 8.5% 3.74% 3.62% 20 Transit terminals 5 1979 10,670 0.96 0.349 0.262 0.087 0.358 25.1% 0.28% 0.71% 21 Restaurants 68 1993 5,084 0.96 0.825 0.778 0.047 0 .846 5.7% 1.79% 10.92% 22 Fast food restaurants 55 1998 3,105 0.97 0.684 0.647 0.037 0.719 5.4% 0.88% 4.47% 23 Financial institutions 63 1991 4,424 0.87 0.248 0.221 0.027 0.259 10.9% 1.44% 2.64% 25 Service shops 18 1984 2,787 0.78 0.434 0.393 0.041 0.43 8 9.4% 0.26% 0.83% 27 Auto sales / repair 116 1985 9,095 0.88 0.131 0.116 0.015 0.133 11.5% 5.45% 5.30% 32 Enclosed theaters / auditorium 1 1999 97,632 0.92 0.121 0.080 0.041 0.131 33.6% 0.50% 0.45%

PAGE 101

101 Table 5 1. Continued FDOR code Description Sample s iz e Average effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent seasonal % H eated area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 33 Nightclubs / ba rs 8 1981 3,686 1.00 0.266 0.226 0.040 0.328 15.0% 0.15% 0.30% 34 Bowling alleys / skating rinks 1 1983 30,784 0.98 0.040 0.036 0.004 0.036 8.9% 0.16% 0.05% 38 Golf courses / driving ranges 20 1991 18,091 0.88 0.151 0.142 0.009 0.150 6.2% 1.87% 2.09% 39 Hotels / motels 10 1986 36,875 0.95 0.269 0.234 0.035 0.276 12.9% 1.91% 3.80% Total commercial 1,177 1988 16,444 0.90 0.135 0.131 0.004 0.138 3.0% 100.00% 100.00%

PAGE 102

102 Table 5 2 Water use coefficients and sector statistics based on sample of 1,037 commer cial parcels and two years of billing records from Gainesville Regional Utilities FDOR code Description Sample s ize Average effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent seasonal % H ea ted area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 11 Stores, one story 176 1982 6,532 0.90 0.120 0.109 0.010 0.127 8.6% 9.68% 8.86% 12 Mixed use 42 1978 6,390 0.84 0.074 0.069 0.005 0.072 7.0% 2.26% 1.29% 13 Department stores 1 1982 94,109 0.80 0.025 0.014 0.011 0.027 43.8% 0.79% 0.15% 14 Supermarkets / conv. Stores 6 1984 10,065 0.90 0.130 0.106 0.024 0.132 18.2% 0.51% 0.51% 15 Regional malls 1 1995 928,070 0.99 0.047 0.033 0.014 0.058 29.9% 7.81% 2.80% 16 Community shopping centers 78 1985 39,269 0.95 0.063 0.058 0.005 0.069 8.0% 25.78% 12.48% 17 Office, one story 232 1984 6,334 0.96 0.100 0.085 0.016 0.112 15.6% 12.37% 9.52% 18 Office, multi story 32 1986 20,700 0.95 0.046 0.040 0.006 0.059 13.1% 5.58% 1.95% 19 Medical office 157 1991 8,070 0.98 0.174 0.158 0.016 0.183 9.2% 10.66% 14.19% 20 Transit terminals 1 2000 2,193 0.98 0.099 0.077 0.023 0.121 22.8% 0.02% 0.01% 21 Restaurants 52 1981 4,664 0.97 0.621 0.582 0.039 0.631 6.3% 2.04% 9.72% 22 Fast f ood restaurants 50 1989 2,697 0.96 0.624 0.608 0.016 0.631 2.6% 1.13% 5.43% 23 Financial institutions 35 1992 6,338 0.93 0.531 0.464 0.067 0.570 12.7% 1.87% 7.60% 24 Insurance company offices 11 1988 10,736 0.94 0.073 0.060 0.012 0.086 17.0% 0.99% 0.55% 25 Service shops 31 1979 6,906 0.81 0.116 0.100 0.016 0.129 13.8% 1.80% 1.60%

PAGE 103

103 Table 5 2. Continued FDOR code Description Sample s ize Average effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent seasonal % H eated area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 26 Service stations 5 1986 1,829 0.71 0.170 0.145 0.025 0.213 14.7% 0.08% 0.10% 27 Auto sales / repair 58 1983 5,836 0.83 0.101 0.085 0.0 15 0.106 15.1% 2.85% 2.20% 29 Wholesale out lets 5 1972 23,700 0.76 0.025 0.021 0.004 0.030 17.7% 1.00% 0.19% 30 Florist / greenhouses 2 1966 3,376 0.92 0.216 0.144 0.072 0.250 33.5% 0.06% 0.09% 32 Enclosed theaters / auditoriums 2 2001 27,989 1.00 0.119 0.097 0.022 0.115 18.7% 0.47% 0.43% 33 Nig htclubs / bars 12 1966 5,336 0.92 0.167 0.134 0.033 0.209 19.6% 0.54% 0.69% 34 Bowling alleys / skating rinks 2 1988 34,410 0.93 0.037 0.032 0.005 0.033 14.2% 0.58% 0.16% 38 Golf courses / driving ranges 6 1986 9,633 0.90 0.405 0.219 0.186 0.452 46.0% 0.49% 1.51% 39 Hotels / motels 40 1981 31,626 0.94 0.220 0.186 0.034 0.236 15.7% 10.65% 17.97% Total commercial 1,037 1984 11,457 0.96 0.130 0.121 0.010 0.141 7.6% 100.00% 100.00%

PAGE 104

104 Table 5 3 Percent differenc e between commercial water use coefficients of Hillsborough County Water Resources Services and Gainesville Regional Utilities FDOR c ode Description Percent d ifference Avg. Base Seasonal May p eak 11 Stores, one story 62% 63% 43% 63% 12 Mixed use 23% 25% 67% 27% 13 Department stores 60% 75% 38% 58% 14 Supermarkets / conv. s tores 54% 57% 31% 57% 15 Regional malls 46% 59% 100% 28% 16 Community shopping centers 46% 49% 150% 41% 17 Office, one story 44% 50% 60% 39% 18 Office, multi story 42% 47% 50% 30% 19 Medical office 33% 32% 45% 28% 20 Transit terminals 72% 71% 74% 66% 21 Restaurants 25% 25% 17% 25% 22 Fast food restaurants 9% 6% 57% 12% 23 Financial institutions 114% 110% 148% 120% 24 Insurance company off ic es 25 Service shops 73% 75% 61% 71% 26 Service stations 27 Auto sales / repair 23% 27% 0% 20% 29 Wholesale outlets 30 Florist / greenhouses 31 Drive in theaters / open s tadiums 32 Enclosed theaters / auditoriums 2% 21% 4 6% 12% 33 Nightclubs / bars 37% 41% 18% 36% 34 Bowling alleys / skating rinks 8% 11% 25% 8% 35 Tourist attractions 36 Camps 37 Race tracks 38 Golf courses / driving ranges 168% 54% 1967% 201% 39 Hotels / motels 18% 21% 3% 14% Total commercial 4% 8% 150% 2%

PAGE 105

105 Table 5 4 Water use coefficients and sector statistics based on sample of 2,214 commercial parcels in Hillsborough County Water Resources Services and Gainesville Regio nal Utilities FDOR code Description Sample s ize A verage effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent seasonal % H eated area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 11 Stores, One Story 289 1985 7,644 0.93 0.098 0.093 0.004 0.104 4.2% 7.07% 5.18% 12 Mixed Use 143 1976 11,483 0.92 0.092 0.089 0.003 0.095 3.4% 5.26% 3.64% 13 Department Stores 19 1994 128,183 0.89 0.062 0.054 0.008 0.063 12.2% 7.80% 3.62% 14 Supermarkets / Conv. Stores 12 3 1991 5,795 0.93 0.270 0.238 0.032 0.291 11.8% 2.28% 4.62% 15 Regional Malls 3 1996 856,391 0.93 0.073 0.065 0.008 0.073 10.6% 8.23% 4.49% 16 Community Shopping Centers 239 1988 39,269 0.95 0.099 0.098 0.001 0.101 0.9% 30.05% 22.27% 17 Office, One Stor y 384 1984 5,983 0.96 0.129 0.117 0.012 0.138 9.0% 7.36% 7.12% 18 Office, Multi Story 73 1987 30,576 0.97 0.069 0.065 0.005 0.077 6.7% 7.15% 3.71% 19 Medical Office 264 1990 7,543 0.97 0.158 0.144 0.014 0.168 8.7% 6.38% 7.56% 20 Transit Terminals 6 1982 9,257 0.97 0.339 0.254 0.085 0.349 25.0% 0.18% 0.45% 21 Restaurants 120 1988 4,902 0.96 0.741 0.711 0.030 0.757 4.0% 1.88% 10.47% 22 Fast Food Restaurants 105 1994 2,910 0.96 0.657 0.636 0.021 0.680 3.3% 0.98% 4.83% 23 Financial Institutions 98 1992 5, 108 0.91 0.373 0.349 0.024 0.397 6.6% 1.60% 4.49% 24 Insurance Company Offices 11 1988 10,736 0.94 0.073 0.060 0.012 0.086 17.0% 0.38% 0.21% 25 Service Shops 49 1981 5,393 0.80 0.176 0.159 0.017 0.187 9.9% 0.85% 1.12%

PAGE 106

106 Table 5 4. Continued FDOR code D escription Sample s ize Average effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent seasonal % H eated area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 26 Service stations 5 1986 1,829 0.71 0.170 0.145 0.025 0.213 14.7% 0.03% 0.04% 27 Auto sales / repair 174 1984 8,009 0.86 0.124 0.110 0.014 0.126 11.3% 4.46% 4.14% 29 Wholesale outlets 5 1972 23,700 0.76 0.025 0.021 0.004 0.030 17.7% 0.38% 0.07% 30 Florist / greenhouses 2 1966 3,376 0.92 0.216 0.144 0.072 0.250 33.5% 0.02% 0.04% 32 Enclosed theaters / auditoriums 3 2000 51,203 0.94 0.120 0.095 0.025 0.125 20.9% 0.49% 0.44% 33 Nightclubs / bars 20 1972 4,676 0.95 0.198 0.164 0.034 0.247 17.3% 0.30% 0.44% 34 Bowling alleys / skating rinks 3 1986 33,201 0.96 0.038 0.033 0.004 0.034 11.3% 0.32% 0.09% 38 Golf courses / driving ranges 26 1990 16,139 0.89 0.186 0.161 0.025 0.191 13.4% 1.34% 1.87% 39 Hotels / motels 50 1982 32,676 0.95 0.231 0.206 0.025 0.245 10.9% 5.23% 9.08% Total commercial 2,214 1986 14,108 0.93 0.133 0.129 0.004 0.139 2.8% 100.00% 100.00%

PAGE 107

107 Table 5 5 Water use coefficients and sector statistics based on sample of 163 industrial parcels and four years of billing records from Hills borough County Water Resources Services FDOR code Description Sample s ize Average effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent seasonal % H eated area in sector % A vg. Water use in sec tor Avg. Base Seasonal May p eak 41 Light manufacturing 10 1982 80,659 0.90 0.018 0.014 0.004 0.024 23.7% 11.90% 5.16% 42 Heavy industrial 2 1995 44,432 0.98 0.040 0.028 0.012 0.050 30.2% 1.31% 1.22% 47 Mineral processing 8 1972 102, 201 0.98 0.171 0.162 0.009 0.162 5.5% 12.06% 48.63% 48 Warehousing / distribution 118 1990 41,608 0.97 0.025 0.022 0.003 0.025 10.9% 72.41% 41.77% 49 Open storage 17 1979 1,860 0.97 0.203 0.175 0.027 0.213 13.5% 0.47% 2.23% 91 Utility, gas & elec. 8 197 4 15,714 0.96 0.023 0.021 0.002 0.025 7.0% 1.85% 0.98% Total industrial 163 1987 41,596 0.96 0.042 0.039 0.003 0.043 7.8% 100.00% 100.00%

PAGE 108

108 Table 5 6 Water use coefficients and sector statistics based on sample of 144 industrial parcels and two years of billing records from Gainesville Regional Utilities FDOR code Description Sample s ize Average effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent seasonal % H eated area in sector % A vg. W ater use in sector Avg. Base Seasonal May p eak 41 Light manufacturing 23 1979 20,851 0.90 0.116 0.095 0.021 0.112 18.3% 17.74% 29.69% 42 Heavy indus trial 1 1998 41,893 0.72 0.033 0.024 0.009 0.031 26.7% 1.55% 0.74% 43 Lumber yards 1 1990 18,488 0.96 0.582 0.361 0.222 0.639 38.1% 0.68% 5.72% 45 Bottler / canneries 1 1969 34,541 0.79 0.010 0.007 0.003 0.010 31.9% 1.28% 0.18% 47 Mineral processing 2 1984 23,001 0.95 0.006 0.003 0.003 0.007 42.9% 1.70% 0.15% 48 Warehousing / distribut ion 110 1985 16,966 0.90 0.061 0.050 0.011 0.069 17.6% 69.02% 60.33% 49 Open storage 2 1999 7,589 0.97 0.046 0.022 0.025 0.078 53.0% 0.56% 0.37% 91 Utility, gas & elec. 4 1983 50,505 0.89 0.026 0.015 0.011 0.020 42.1% 7.47% 2.82% Total industrial 144 1 984 18,777 0.85 0.070 0.058 0.012 0.075 17.2% 100.00% 100.00%

PAGE 109

109 Table 5 7 Percent difference between industrial water use coefficients of Hillsb o rough County Water Resources Services and Gainesville Regional Utilities FDOR c ode Description Percent d iffe rence Avg. Base Seasonal May p eak 41 Light manufacturing 544% 579% 425% 367% 42 Heavy industrial 18% 14% 25% 38% 43 Lumber yards 44 Packing plants 45 Bottler / canneries 46 Food processing 47 Mineral processing 96% 98% 67% 96% 48 Warehousing / distribution 144% 127% 267% 176% 49 Open storage 77% 87% 7% 63% 91 Utility, gas & elec. 13% 29% 450% 20% 92 Mining / petroleum Total industrial 67% 49% 300% 74%

PAGE 110

110 Table 5 8 Water use coefficient s and sector statistics based on sample of 307 industrial parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities FDOR code Description Sample s ize Average effective year built Average heated area (sf) H A E A Weighted wate r use coef. (gallons/heated square foot/day) Percent seasonal % H eated area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 41 Light manufacturing 33 1979 38,975 0.90 0.055 0.044 0.011 0.057 19.4% 13.56% 14.85% 42 He avy industrial 3 1996 43,586 0.83 0.038 0.031 0.006 0.044 16.6% 1.38% 1.03% 43 Lumber yards 1 1990 18,488 0.96 0.582 0.361 0.222 0.639 38.1% 0.19% 2.26% 45 Bottler / canneries 1 1969 34,541 0.79 0.010 0.007 0.003 0.010 31.9% 0.36% 0.07% 47 Mineral proce ssing 10 1974 86,361 0.98 0.163 0.154 0.009 0.154 5.5% 9.11% 29.48% 48 Warehousing / distribution 228 1988 29,719 0.95 0.035 0.031 0.003 0.037 9.1% 71.45% 49.10% 49 Open storage 19 1981 2,463 0.97 0.152 0.132 0.020 0.169 13.2% 0.49% 1.49% 91 Utility, ga s & elec. 12 1977 27,311 0.90 0.025 0.018 0.007 0.022 26.3% 3.46% 1.71% Total industrial 307 1986 30,893 0.91 0.050 0.045 0.005 0.052 10.6% 100.00% 100.00%

PAGE 111

111 Table 5 9 Water use coefficients and sector statistics based on sample of 428 institutional pa rcels and four years of billing records from Hillsborough County Water Resources Services FDOR code Description Sample s ize Average effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent season al % H eated area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 71 Churches 221 1957 12,838 0.94 0.042 0.038 0.004 0.047 9.2% 24.56% 13.78% 72 Private schools & colleges 65 1986 7,309 0.95 0.148 0.131 0.016 0.166 11 .1% 4.11% 8.15% 73 Private hospitals 2 1997 424,916 0.98 0.086 0.078 0.008 0.083 9.2% 7.36% 8.49% 75 Orphanages / n on profits 67 1987 9,741 0.95 0.188 0.163 0.025 0.204 13.1% 5.65% 14.22% 76 Mortuaries / cemeteries 7 1989 7,560 0.88 0.063 0.048 0.015 0. 073 23.3% 0.46% 0.39% 77 Clubs / union halls 15 1981 10,654 0.83 0.159 0.141 0.018 0.166 11.5% 1.38% 2.95% 83 Public county schools 51 1987 127,905 0.98 0.069 0.059 0.010 0.075 14.8% 56.47% 52.03% Total institutional 428 1971 26,988 0.94 0.075 0.066 0. 008 0.081 11.1% 100.00% 100.00%

PAGE 112

112 Table 5 10 Water use coefficients and sector statistics based on sample of 256 institutional parcels and two years of billing records from Gainesville Regional Utilities FDOR code Description Sample s ize Average effecti ve year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent seasonal % H eated area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 71 Churches 116 1979 13,555 0.96 0. 063 0.054 0.009 0.070 14.0% 23.27% 17.26% 72 Private schools & colleges 46 1981 9,149 0.95 0.099 0.085 0.014 0.118 14.1% 6.23% 7.35% 73 Private hospitals 4 1988 141,224 0.96 0.137 0.112 0.025 0.141 18.0% 8.36% 13.55% 74 Homes for the a ged 12 1993 116,67 5 0.92 0.101 0.087 0.013 0.108 13.4% 20.72% 24.76% 75 Orphanages / n on profits 15 1985 10,894 0.96 0.123 0.104 0.019 0.137 15.5% 2.42% 3.52% 76 Mortuaries / cemeteries 6 1974 5,286 0.89 0.296 0.265 0.031 0.294 10.5% 0.47% 1.65% 77 Clubs / union halls 40 1935 13,308 0.99 0.197 0.154 0.044 0.178 22.1% 7.88% 18.46% 78 Sanitariums / convalescents 1 1975 43,505 1.00 0.378 0.328 0.050 0.360 13.3% 0.64% 2.89% 79 Cultural organizations 1 1989 2,302 0.74 0.163 0.154 0.009 0.159 5.7% 0.03% 0.07% 82 Parks and r e creation 1 1998 5,288 0.92 0.229 0.174 0.055 0.258 24.0% 0.08% 0.21% 83 Public county schools 1 1980 59,448 0.98 0.025 0.015 0.010 0.026 38.9% 0.88% 0.26% 84 Colleges 9 1970 175,786 1.00 0.035 0.026 0.008 0.027 23.5% 23.41% 9.59% 85 Hospitals 3 1988 123 ,175 0.96 0.006 0.002 0.004 0.009 66.9% 5.47% 0.39% 97 Outdoor recreational 1 1950 9,233 0.92 0.026 0.015 0.011 0.037 42.6% 0.14% 0.04% Total institutional 256 1973 26,395 0.96 0.084 0.076 0.009 0.086 10.1% 100.00% 100.00%

PAGE 113

113 Table 5 11 Percent differe nce between in stitutional water use coefficients of Hillsborough County Water Resources Services and Gainesville Regional Utilities FDOR c ode Description Percent d ifference Avg. Base Seasonal May p eak 71 Churches 50% 42% 125% 49% 72 Private schools & colleges 33% 35% 13% 29% 73 Private hospitals 59% 44% 213% 70% 74 Homes for the aged 75 Orphanages / non profits 35% 36% 24% 33% 76 Mortuaries / cemeteries 370% 452% 107% 303% 77 Clubs / union halls 24% 9% 144% 7% 78 Sanitariums / con valescents 79 Cultural organizations 81 Military 82 Parks and recreation 83 Public county schools 64% 75% 0% 65% 84 Colleges 85 Hospitals 90 Gov. leased interests 97 Outdoor recreational Total institutional 12% 15% 13% 6%

PAGE 114

114 Table 5 12 Water use coefficients and sector statistics based on sample of 684 institutional parcels in Hillsborough County Water Resources Services and Gainesville Regional Utilities FDOR code Description Sa mple s ize Average effective year built Average heated area (sf) H A E A Weighted water use coef. (gallons/heated square foot/day) Percent seasonal % H eated area in sector % A vg. Water use in sector Avg. Base Seasonal May p eak 71 Churche s 337 1965 13,085 0.95 0.049 0.044 0.005 0.055 10.0% 24.09% 15.16% 72 Private schools & colleges 111 1984 8,071 0.95 0.125 0.110 0.015 0.143 12.2% 4.89% 7.83% 73 Private hospitals 6 1991 235,788 0.97 0.106 0.096 0.010 0.106 9.6% 7.73% 10.50% 74 Homes fo r the a ged 12 1993 116,675 0.92 0.101 0.087 0.013 0.108 13.4% 7.65% 9.85% 75 Orphanages / n on profits 82 1986 9,952 0.91 0.175 0.151 0.023 0.191 13.4% 4.46% 9.96% 76 Mortuaries / cemeteries 13 1982 6,511 0.88 0.150 0.130 0.021 0.156 13.9% 0.46% 0.89% 77 Clubs / union halls 55 1947 12,584 0.91 0.189 0.152 0.037 0.175 19.5% 3.78% 9.12% 78 Sanitariums / convalescents 1 1975 43,505 1.00 0.378 0.328 0.050 0.360 13.3% 0.24% 1.15% 79 Cultural organizations 1 1989 2,302 0.74 0.163 0.154 0.009 0.159 5.7% 0.01% 0.03% 82 Parks and r ecreation 1 1998 5,288 0.92 0.229 0.174 0.055 0.258 24.0% 0.03% 0.08% 83 Public county schools 52 1987 126,588 0.98 0.068 0.058 0.010 0.074 14.7% 35.96% 31.44% 84 Colleges 9 1970 175,786 1.00 0.035 0.026 0.008 0.027 23.5% 8.64% 3.82% 85 Hospitals 3 1988 123,175 0.96 0.006 0.002 0.004 0.009 66.9% 2.02% 0.15% 97 Outdoor recreational 1 1950 9,233 0.92 0.026 0.015 0.011 0.037 42.6% 0.05% 0.02% Total institutional 684 1972 26,766 0.95 0.078 0.070 0.008 0.083 9.8% 100.00% 100.00%

PAGE 115

115 Ta ble 5 13 Water use coefficient comparison to other studies on CII water use Florida study coefficients (gal/hsf/d) AWWARF CI end uses study (gal/hsf/d) % Difference 2007 Colorado Waterwise ** (gal/hsf/d) % Difference Supermarkets (FDOR 14) 0.270 0.223 17% Office buildings (FDOR 17& 18) 0.100 0.103 3% Restaurants (FDOR 21) 0.741 0.845 14% 0.526 29% Hotels (FDOR 39) 0.231 0.248 7% 0.329 42% Schools (FDOR 83) 0.068 0.306 348% 0.042 38% Homes for the aged (FDOR 74) 0.101 0.219 118% Dziegiele wski et al. 2000 ** Colorado Waterwise 2007

PAGE 116

116 Figure 5 1 EZ Guide 2 water budget summary for a utility in South Florida

PAGE 117

117 CHAPTER 6 SUMMARY, CONCLUSIONS AND NEED FOR ADDITIO NAL RESEARCH This commercial, in dustrial, and institutional ( CII ) water use estimatin g method should offer a significant improvement o ver traditional methods of estimating CII water use by combining water billing records with parcel level land use databases, principally Florida Department of Revenue ( FDOR ) These databases allow for the si ze of sub sectors and their activity coefficients to be developed by parcel level data, which is a finer resolution than Traffic Analysis Zones ( TAZ ) or Census block data. They also provide a standardized classification system to categorize land uses across the State. The 5 5 CII FDOR land use subsectors allow water users to be classified within various degrees of disaggregation based on the level of homogeneity desired in a sector. The FDOR database is public information and is capable of being linked to any utility billing records through the p arcel identification number. Existing utility linkages between the FDOR and water billing records are available for only a few utilities in the State. This thesis develops activity water use coefficients and presents a methodology to carry out water budgets regardless of this link. By employing a measure of size that is standard and reliable across the CII sectors, along with default water use coefficients, any utility within the State is able to develop a water budget. The water budget is an essential tool to investigate the gross gpcd of a utility by sectors. It also allows a utility to estimate how the water in their service boundary is being used. This information about water users is crucial to develop an accurate f orecast for water use or in planning and evaluating conservation efforts. Water use coefficients presented in this thesis were calculated from historical billing records from Hillsborough County Water Resources Services (HCWRS) and Gainesville Regional Ut ilities (GRU), and heated areas from Florida County Property Appraisers (FCPA) H CWRS and GRU provided for a relatively large cross section of customers and long time

PAGE 118

118 series. More utilities from across the State should be incorporated into the analysis to account for regional differences in water use, as well as increase the sample size for the various water use sub sectors. The extensive amount of data available from FDOR/ F CPA allow for substantial amounts of future refinements The available data can be u sed to improve the accuracy of water estimates, as well as further disaggregate water use to the end use or process level. For example, FDOR incorporates year built in its database T his information can be used to carry out time series analysis and find tr ends for both heated areas and activity water use coefficients over time. This analysis improves the accuracy of water use estimates and forecasts, and could provide insight into end uses. Also, by estimating irrigated area and analyzing the billing record s for seasonal components, it should be possible to break down water use further into the seasonally dependent irrigation and cooling end uses For this reason, seasonal water use might best be normalized by a measure of size such as irrigable area, which is available through the FCPA as presented in Chapter 3. Previous studies (Dziegielewski et al. 2000, U.S. EPA 2009, 1997) present valuable information on the breakdown of end uses within CII water use. Future work should include estimates on the number, e fficiency, and frequency of use of water using devices in the CII subsectors. Such estimates should include both indoor domestic uses such as toilets, urinals, and faucets, as well as outdoor uses such as irrigation application rates based on estimates of irrigable area and cooling water use for CII subsectors where applicable. End use estimates should be linked with available best management practices, and incorporated with cost/benefit data to optimize for the best blend of water conservation controls. Fu ture work should also include a study to analyze the accuracy of the water use estimates described in this thesis and the reliability

PAGE 119

119 of savings with conservation efforts. Such a study would allow for a measure of un certainty to be associated with these es timates. The availability of the FDOR database provides a major improvement in our ability to estimate CII water use. The Conserve Florida Water Clearinghouse ( http://www.conservefloridawater.org/ ) is developing these water use coefficients and heated area statistics and will make them available to interested utilities. Though seasonal components of water use across CII subsectors were found to be minimal, base water use coefficients presented in this t hesis can be used to arrive at an estimate of non seasonal water use for CII customers. These non seasonal estimates of water use should be relatively stable outside of Florida, and thus t hese coefficients should provide good estimates for CII users elsewh ere. Average water use coefficients should also be applicable outside of Florida except where landscape irrigation is a significant component of water use. Utilities should link their billing data with the FDOR/CPA databases and share this information so t hat these estimates can be improved over time.

PAGE 120

120 Appendix HEATED AREA SUBSECTO R DISTRIBUTIONS This appendix presents the histograms for each of the 55 commercial, industrial, and institutional (CII) subsectors in the Florida Department of Revenue (FDOR) da tabase. Within each figure, histograms of both the state of Florida total of heated areas per parcel and the combined sample parcel heated areas from Hillsborough County Water Resources Services and Gainesville Regional Utilities are presented. These histo grams along with the cumulative frequency curve, and fitted lognormal distribution equations, can be used to gauge to what extent the sample data presented in this thesis is applicable to the rest of the State. For example, Figure A 1 presents the heated a rea histograms and cumulative frequency curves for the State and combined utility FDOR 11 (store, one story) parcels. Figure A 1, with its matching histograms, cumulative frequency curve, and similar fitted lognormal equations, presents a CII subsector wer e the combined utility sample is highly representative of the State total of FDOR 11 parcels. A differing circumstance is presented in Figure A 3 where the sample data only appears to be representative of a segment of the total distribution of FDOR 13 par cels in the State (i.e., department stores larger than 60,000 square feet).

PAGE 121

121 Figure A 1. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FD OR 11 (s tores, o ne story) parcels Figure A 2. H eated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 12 ( m ixed use) parcels 0% 20% 40% 60% 80% 100% 0% 5% 10% 15% 20% 25% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 39,002) Lognorm(6205,9267.4) Lognorm(6191.4,9148.7,Shift( 9.8419)) Combined Utilities (n = 289) Lognorm(6915.4,9009.3) Lognorm(6879.1,10162.8,Shift(219.34)) 0% 20% 40% 60% 80% 100% 0% 7% 14% 21% 28% 35% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 18,931) Lognorm(3979.6,4892.6) Lognorm(3971.2,4809.7,Shift( 10.617)) Combined Utilities (n = 143) Lognorm(7664.1,8870.4) Lognorm Shift N/A

PAGE 122

122 Figure A 3. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 13 (department stores) parcels Figure A 4. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 14 (supermarkets / conv. stores) parcels 0% 20% 40% 60% 80% 100% 0% 4% 8% 12% 16% 20% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 829) Lognorm(130417.5,226144.1) Lognorm Shift N/A Combined Utilities (n = 19) Lognorm(128058.9,44261.3) Lognorm(105097.2,46236,Shift(23164)) 0% 20% 40% 60% 80% 100% 0% 13% 26% 39% 52% 65% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 2,571) Lognorm(10466.9,18876) Lognorm(10491,19976.4,Shift(112.83)) Combined Utilities (n = 123) Lognorm(4533.9,4488) Lognorm(3851.9,6063.4,Shift(835.36))

PAGE 123

123 Figure A 5. Heated area histograms and fitted logno rmal equations for State total and combi ned utility sample of FDOR 15 (regional malls) parcels Figure A 6. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 16 (community s hopping centers) pa rcels 0% 20% 40% 60% 80% 100% 0% 9% 18% 27% 36% 45% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 449) Lognorm(504479.3,5563273.4) Lognorm(677996.5,11046462.1,Shift(320.63)) Combined Utilities (n = 3) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 5% 10% 15% 20% 25% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 7,726) Lognorm(42572.1,101645) Lognorm(43500.7,109776.5,Shift(162.45)) Combined Utilities (n = 239) Lognorm(41189.7,72650) Lognorm(42574.9,88032.7,Shift(871.69))

PAGE 124

124 Figure A 7. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 17 (office, one story) parcels Figure A 8. Heated area histograms and fitted lognormal equations for State total an d combi ned utility sample of FDOR 18 (office, multi story) parcels 0% 20% 40% 60% 80% 100% 0% 7% 14% 21% 28% 35% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 37,629) Lognorm(4189.8,6972.1) Lognorm Shift N/A Combined Utilities (n = 384) Lognorm(5592.5,6137.9) Lognorm(5363.2,9389.2,Shift(702.9)) 0% 20% 40% 60% 80% 100% 0% 13% 26% 39% 52% 65% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 16,033) Lognorm(13586.8,56193.4) Lognorm(13597.5,56316,Shift(0.55288)) Combined Utilities (n = 73) Lognorm(31701.2,48051.2) Lognorm(33856.1,73791.5,Shift(1683.9))

PAGE 125

125 Figure A 9. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 19 (medical office) parcels Figure A 10. Heated area his tograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 20 (transit terminals) parcels 0% 20% 40% 60% 80% 100% 0% 6% 12% 18% 24% 30% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 20,953) Lognorm(5223.1,9767.9) Lognorm(4878.2,7364.5,Shift( 108.5)) Combined Utilities (n = 264) Lognorm(6619.7,6010.2) Lognorm(5952.2,7472.9,Shift(868.41)) 0% 20% 40% 60% 80% 100% 0% 16% 32% 48% 64% 80% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 2,825) Lognorm(5701,102409.2) Lognorm(6270.7,127055.2,Shift(0.71497)) Combined Utilities (n = 6) Lognorm(8147.7,16823.5) Lognorm Shift N/A

PAGE 126

126 Figure A 11. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 21 (resta urants) parcels Figure A 12. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 22 (fast food restaurants) parcels 0% 20% 40% 60% 80% 100% 0% 3% 6% 9% 12% 15% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 7,629) Lognorm(4487.7,3618.4) Lognorm Shift N/A Combined Utilities (n = 120) Lognorm(5078.8,3785.5) Lognorm(8645,2775.3,Shift( 3737.2)) 0% 20% 40% 60% 80% 100% 0% 3% 6% 9% 12% 15% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 4,377) Lognorm(3224.7,2156.5) Lognorm Shift N/A Combined Utilities (n = 105) Lognorm(2992.3,1775.5) Lognorm(22534,1247.4,Shift( 19623))

PAGE 127

127 Figure A 13. Heated area histograms and fitted lognormal equations for St ate total and combi ned utility sample of FDOR 23 (financial institutions) parcels Figure A 14. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 24 (insurance company offices) parcels 0% 20% 40% 60% 80% 100% 0% 7% 14% 21% 28% 35% Cumulative Fequency Relative Frequency Heated Area (sf) State (n = 4,732) Lognorm(6571.5,6560.8) Lognorm Shift N/A Combined Utilities (n = 98) Lognorm(5030.5,2929.5) Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 9% 18% 27% 36% 45% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 283) Lognorm(4570.4,6424.4) Lognorm(4425.7,7098.1,Shift(208.47)) Combined Utilities (n = 11) Lognorm(10121.5,17828.6) Lognorm(15285.7,77251.8,Shift(831.87))

PAGE 128

128 Figure A 15. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 25 (service shops) parcels Figure A 16. Heated area histograms and fitted lognormal equations for State total and combi ned utility sa mple of FDOR 26 (service stations) parcels 0% 20% 40% 60% 80% 100% 0% 6% 12% 18% 24% 30% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 4,309) Lognorm(4647.3,5794.8) Lognorm(4629,6006.4,Shift(53.395)) Combined Utilities (n = 49) Lognorm(5224.4,5678.8) Lognorm(5224.9,5675.9,Shift( 1.0112)) 0% 20% 40% 60% 80% 100% 0% 13% 26% 39% 52% 65% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 3,434) Lognorm(2487.9,1936.4) Lognorm Shift N/A Combined Utilities (n = 5) Lognorm(1829.1,256.26) Lognorm(2263.2,254.85,Shift( 434.2))

PAGE 129

129 Figure A 17. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 27 (auto sales / repair) parcels Figure A 18. Heated area histograms and fitted l ognormal equations for State total and combi ned utility sample of FDOR 29 (wholesale o utlets) parc els 0% 20% 40% 60% 80% 100% 0% 5% 10% 15% 20% 25% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 15,129) Lognorm(6238.3,9484.1) Lognorm(6230.3,9407.6,Shift( 6.5661)) Combined Utilities (n = 174) Lognorm(6043.8,7416.4) Lognorm(6048.8,7296.1,Shift( 27.336)) 0% 20% 40% 60% 80% 100% 0% 5% 10% 15% 20% 25% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 729) Lognorm(14316,23222.4) Lognorm(14417.7,24479.5, Shift(103.52)) Combined Utilities (n = 5) Lognorm(24295.6,17153.8) Lognorm(41463.5,13411.9, Shift( 17713))

PAGE 130

130 Figure A 19. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 30 (florist / greenhouses) pa rcels Figure A 20. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 31 (drive in theaters / open stadiums) parcels 0% 20% 40% 60% 80% 100% 0% 11% 22% 33% 44% 55% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 352) Lognorm(3075.9,3956.8) Lognorm(3071.4,4261.6,Shift(56.464)) Combined Utilities (n = 2) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 12% 24% 36% 48% 60% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 28) Lognorm(39865,187722.3) Lognorm(66864.5,776190.2,Shift(702.06)) Combined Utilities (n = 0) Lognorm N/A Lognorm Shift N/A

PAGE 131

131 Figure A 21. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 32 (enclosed theaters / a uditori ums) parcels Figure A 22. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 33 (nightclubs / bars) parcels 0% 20% 40% 60% 80% 100% 0% 7% 14% 21% 28% 35% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 242) Lognorm(28010.4,54268.4) Lognorm(29357.2,68186.4,Shift(573.64)) Combined Utilities (n = 3) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 6% 12% 18% 24% 30% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 1,914) Lognorm(4346.6,3742.9) Lognorm Shift N/A Combined Utilities (n = 20) Lognorm(4556,3802.7) Lognorm(3892.9,6069.7,Shift(985.81))

PAGE 132

132 Figu re A 23. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 34 (bowling alleys / s kating rinks) parcels Figure A 24. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 35 (tourist a ttractions) parc els 0% 20% 40% 60% 80% 100% 0% 14% 28% 42% 56% 70% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 492) Lognorm(30297.7,81922.8) Lognorm(24285.3,38140.6,Shift( 1017.9)) Combined Utilities (n = 3) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 8% 16% 24% 32% 40% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 742) Lognorm(18060.9,73944.1) Lognorm(19303.9,88847.3,Shift(49.97)) Combined Utilities (n = 0) Lognorm N/A Lognorm Shift N/A

PAGE 133

133 Figure A 25. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 36 (camps) parcels Figure A 26. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 37 ( race tracks) parcels 0% 20% 40% 60% 80% 100% 0% 3% 6% 9% 12% 15% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 330) Lognorm(8234,14388.4) Lognorm(8076.2,12907.7,Shift( 96.975)) Combined Utilities (n = 0) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 15% 30% 45% 60% 75% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 129) Lognorm(28736.7,139346.2) Lognorm(33492.9,215636.3,Shift(195.85)) Combined Utilities (n = 0) Lognorm N/A Lognorm Shift N/A

PAGE 134

134 Figure A 27. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 38 (golf c ourses / driving ranges) parcels Figure A 28. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 39 (hotels / motels) parcels 0% 20% 40% 60% 80% 100% 0% 5% 10% 15% 20% 25% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 1,392) Lognorm(25060.1,84420.3) Lognorm(20406.4,45509.6,Shift( 373.38)) Combined Utilities (n = 26) Lognorm(20768.6,37549) Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 17% 34% 51% 68% 85% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 21,702) Lognorm(4718,18708.6) Lognorm(4693.3,18461.7,Shift( 0.81817)) Combined Utilities (n = 50) Lognorm(37478.7,54315.1) Lognorm(38324.6,33788.4,Shift( 4828.5))

PAGE 135

135 Figure A 29. Heated area histograms and fitted lognormal equations for State t otal and combined utility sample of FDOR 41 (light manufacturing) parcels Figure A 30. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 42 (heavy industrial) parcels 0% 20% 40% 60% 80% 100% 0% 11% 22% 33% 44% 55% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 18,398) Lognorm(13375.8,32016.2) Lognorm(13469,33184,Shift(41.897)) Combined Utilities (n = 33) Lognorm(35589.2,82016.2) Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 8% 16% 24% 32% 40% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 677) Lognorm(70712.4,196744.3) Lognorm(69618.8,187617.9,Shift( 100.43)) Combined Utilities (n = 3) Lognorm N/A Lognorm Shift N/A

PAGE 136

136 Figure A 31. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 43 (lumber yards) parcels Figure A 32. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 44 (p acking p lants) parcels 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 471) Lognorm(33928.7,86991.8) Lognorm(33597.4,84239.6,Shift( 41.52)) Combined Utilities (n = 1) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 5% 10% 15% 20% 25% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 526) Lognorm(39030,99836.6) Lognorm(37471.6,87439.9,Shift( 195.3)) Combined Utilities (n = 0) Lognorm N/A Lognorm Shift N/A

PAGE 137

137 Figure A 33. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 45 (bottler / canneries) parcels Figure A 34. Heated area histograms and fitted lognormal equations f or State total and combi ned utility sample of FDOR 46 (food processing) parcels 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 90) Lognorm(121878.7,494517.2) Lognorm(115994.8,434806.5,Shift( 164.33)) Combined Utilities (n = 1) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 11% 22% 33% 44% 55% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 295) Lognorm(37283.9,113069.3) Lognorm(39449,135557,Shift(229)) Combined Utilities (n = 0) Lognorm N/A Lognorm Shift N/A

PAGE 138

138 Figure A 35. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 47 (mineral processing) parcels Figure A 36. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 48 (warehousing / distribution ) parcels 0% 20% 40% 60% 80% 100% 0% 17% 34% 51% 68% 85% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 598) Lognorm(16233,50857.9) Lognorm(16955,58041.8,Shift(58.741)) Combined Utilities (n = 10) Lognorm(128958.6,1274954) Lognorm(15291500000,181412.3,Shift( 15291500000)) 0% 20% 40% 60% 80% 100% 0% 8% 16% 24% 32% 40% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 42,406) Lognorm(17258.9,39654.1) Lognorm(17209.7,39262.3,Shift( 8.7696)) Combined Utilities (n = 228) Lognorm(31655.2,54270.5) Lognorm(31697.6,54708.6,Shift(29.426))

PAGE 139

139 Figure A 37. Heated area histograms and fitted lognormal equations for State total and combi ned utilit y sample of FDOR 49 (open storage) parcels Figure A 38. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 91 (utility, gas & elec.) parcels 0% 20% 40% 60% 80% 100% 0% 12% 24% 36% 48% 60% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 2,190) Lognorm(12588.5,66705) Lognorm(7474.9,18983.8,Shift( 72.531)) Combined Utilities (n = 19) Lognorm(2249.2,2004.3) Lognorm(66065071.7,3037.2,Shift( 66062608)) 0% 20% 40% 60% 80% 100% 0% 17% 34% 51% 68% 85% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 3,185) Lognorm(6195.7,24310.6) Lognorm(6754.5,31730.1,Shift(40.436)) Combined Utilities (n = 12) Lognorm(23727.1,22734.2) Lognorm(30900.2,243072.7,Shift(8314.8))

PAGE 140

140 Figure A 39. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 92 (mining / petroleum) parcels Figure A 40. Heated area histograms and fitted lognormal equations for State total and combined utility sample of FDOR 71 (churches) parcels 0% 20% 40% 60% 80% 100% 0% 12% 24% 36% 48% 60% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 112) Lognorm(14980.6,58563.9) Lognorm(1843610000,49323.8,Shift( 1843590000)) Combined Utilities (n = 0) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 6% 12% 18% 24% 30% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 18,886) Lognorm(9755.6,15478.2) Lognorm(9492.5,14051.4,Shift( 52.418)) Combined Utilities (n = 337) Lognorm(12941.2,18379.4) Lognorm Shift N/A

PAGE 141

141 Fi gure A 41. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 72 (p rivate s chools & colleges) parcels Figure A 42. Heated area histograms and fitted lognormal equations for State total and c ombi ned utility sample of FDOR 73 (private hospitals) parcels 0% 20% 40% 60% 80% 100% 0% 8% 16% 24% 32% 40% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 3,243) Lognorm(12686.5,25968.3) Lognorm(12106.6,22765,Shift( 41.379)) Combined Utilities (n = 111) Lognorm(7458.2,7491) Lognorm(7117.8,8751.9,Shift(537.89)) 0% 20% 40% 60% 80% 100% 0% 9% 18% 27% 36% 45% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 570) Lognorm(149847.5,612156.5) Lognorm Shift N/A Combined Utilities (n = 6) Lognorm(301959.4,981345.9) Lognorm Shift N/A

PAGE 142

142 Figure A 43. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 74 (h omes for the aged) parcels Figure A 44. Heated area histo grams and fitted lognormal equations for State total and combi ned utility sample of FDOR 75 (orphanages / non profits) parcels 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 4,447) Lognorm(6225.8,24872.7) Lognorm(8023.4,54233.7,Shift(117.32)) Combined Utilities (n = 12) Lognorm(127980.6,448598.9) Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 12% 24% 36% 48% 60% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 2,214) Lognorm(9998.3,20236.8) Lognorm(10167.6,21827.5,Shift(62.384)) Combined Utilities (n = 82) Lognorm(9547.1,28900.4) Lognorm(9852.3,31747.8,Shift(23.7))

PAGE 143

143 Figure A 45. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 76 ( mortuaries / cemeteries) parcels Figure A 46. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 77 (clubs / union h alls) pa rcels 0% 20% 40% 60% 80% 100% 0% 6% 12% 18% 24% 30% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 887) Lognorm(6442.2,7980.9) Lognorm Shift N/A Combined Utilities (n = 13) Lognorm(6581.3,3300.4) Lognorm(18004.1,2742.8,Shift( 11492)) 0% 20% 40% 60% 80% 100% 0% 5% 10% 15% 20% 25% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 3,244) Lognorm(6931.3,8460.6) Lognorm Shift N/A Combined Utilities (n = 55) Lognorm(13045.8,14582.2) Lognorm(13149.6,13893.8,Shift( 228.41))

PAGE 144

144 Figure A 47. Heated area histograms and fitted lognormal e quations for State total and combi ned utility sample of FDOR 78 (sanitariums / convalescents) parcels Figure A 48. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 79 (cultural organizations) parcels 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 457) Lognorm(48378,98509.5) Lognorm(48081.6,95982.2,Shift( 88.963)) Combined Utilities (n = 1) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 234) Lognorm(9926.2,20234.2) Lognorm(10208.5,23085.5,Shift(113.43)) Combined Utilities (n = 1) Lognorm N/A Lognorm Shift N/A

PAGE 145

145 Figure A 49. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 81 (military) parcels Figure A 50. Heated area histograms and fitted lognormal equations for State total and combi ne d utility sample of FDOR 82 (parks and recreation) parcels 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 43) Lognorm(68159.2,204065.6) Lognorm(85496900000,180027.1,Shift( 85496900000)) Combined Utilities (n = 0) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 850) Lognorm(6238.2,15500.5) Lognorm(6588.5,18670.1,Shift(50.818)) Combined Utilities (n = 1) Lognorm N/A Lognorm Shift N/A

PAGE 146

146 Figure A 51. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 83 (public c ounty schools) parcels Figure A 52. Heated area histo grams and fitted lognormal equations for State total and combi ned utility sample of FDOR 84 (colleges) parcels 0% 20% 40% 60% 80% 100% 0% 6% 12% 18% 24% 30% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 2,953) Lognorm(465926.8,5919503.5) Lognorm(110758.3,142790.4,Shift( 10436)) Combined Utilities (n = 52) Lognorm(130157.5,94366.6) Lognorm(150120.7,73721.4,Shift( 24686)) 0% 20% 40% 60% 80% 100% 0% 14% 28% 42% 56% 70% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 288) Lognorm(423964.5,5902238.9) Lognorm(271500.7,2238345.1,Shift( 96.53)) Combined Utilities (n = 9) Lognorm(199478.1,1154125.5) Lognorm Shift N/A

PAGE 147

147 Figure A 53. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 85 (hospitals) parce ls Figure A 54. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 90 (g ov leased interests) parcels 0% 20% 40% 60% 80% 100% 0% 14% 28% 42% 56% 70% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 173) Lognorm(231446.7,1827373.5) Lognorm(293003.6,3293062.8,Shift(523.62)) Combined Utilities (n = 3) Lognorm N/A Lognorm Shift N/A 0% 20% 40% 60% 80% 100% 0% 11% 22% 33% 44% 55% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 1,276) Lognorm(27445.3,149544.1) Lognorm(27953.8,156847,Shift(11.64)) Combined Utilities (n = 0) Lognorm N/A Lognorm Shift N/A

PAGE 148

148 Figure A 55. Heated area histograms and fitted lognormal equations for State total and combi ned utility sample of FDOR 97 (outdoor recreational) parcels 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Cumulative Frequency Relative Frequency Heated Area (sf) State (n = 2,090) Lognorm(2856.7,6198.6) Lognorm(2841,6084.1,Shift( 2.6213)) Combined Utilities (n = 1) Lognorm N/A Lognorm Shift N/A

PAGE 149

149 L IST OF REFERENCES Baumann, D., Boland, J., and Hanemann, D. (1998). Urban Water Demand Management and Planning McGraw Hill, New York. Boland, J. (1997). "Assessing Urban Water Use and t he Role of Water Conservation Measures Under Climate Uncertainty." Climatic Change 37(1), 157 176. CDM. (2008). "IWR MAIN." < http://www.iwrmain.com > (June 9, 2009). Colorado WaterWise Council Benchmarking Task Force (2007). "Industrial, Commercial & Institutional Water Conservation." Dziegielewski, B., and Boland, J. (1989). "Forecasting Urban Water Use The IWR MAIN Model." Water Resources Bulletin 25(1), 101 109. Dziegielewski, B., Kiefer, J., Opitz, E., Porter, G., Lantz, G., DeOreo, W., Mayer, P., and Nelson, J. (2000). Commercial and Institutional End Uses of Water AWWARF, Denver, CO. East Bay Municipal Utility District. (2008). Watersmart Guidebook: A Water Use Efficiency Plan Review Guide EBMUD, Oakland. F lorida Department of Renevue. (2009). "Property Tax Oversight." < http://dor.myflorida.com/dor/property/ > (June 10, 2009). Friedman, K. (2009). "Evaluation of Indoor Urban Water Use and Water Loss Mana gement as Conservation Options in Florida." Master of Engineering Thesis, Dept. of Environmental Engineering Sciences, University of Florida, Gainesville. Hazen & Sawyer and PMCL. (2004). "The Tampa Bay Water Long Term Demand Forecasting Model." < http://www.tampabaywater.org/documents/conservation/TBWLTDFS10 11 2004.pdf > Clearwater, FL (June 9, 2009). Kim, J., and McCuen, R. (1979). "Factors for Predicting Commercial Water Use." Water Resources Bulletin 15(4), 1073 1080. Maddaus, W., and Maddaus, M. (2004). "Evaluating Water Conservation Cost Effectiveness with an End Use Model." Water Sources Conference Austin, TX, 13. Marella, R. (2009). "Water Withdrawals, Use, a nd Trends in Florida, 2005." Dept. of the Interior, ed., U.S. Geological Survey, Reston, 49. Mayer, P. at al. (2005). "Water and Energy Savings from High Efficiency Fixtures and Appliances in Single Family Homes." US EPA. Mayer, P. W., and DeOreo, W. B. (1 999). Residential End Uses of Water, American Water Works Association Research Foundation, Denver, Colorado.

PAGE 150

150 Mercer, L., and Morgan, D. (1974). "Estimation of Commercial, Institutional and Governmental Water Use for Local Areas." Water Resources Bulletin 10(4), 794 801. Neter, J., Kutner, M., Nachtsheim, C., and Wasserman, W. (1996). Applied Linear Statistical Models McGraw Hill. Opitz, E., Langowski, J., Dziegielewski, B., Hanna Somers, N., Willet, J., and Hauer, R. (1998). "Forecasting Urban Water Use: Models and Applications." Urban Water Demand Management and Planning, D. Baumann, J. Boland, and W. M. Haneman, eds., McGraw Hill, Inc., New York, 95 135. Palisade Corporation (2010). "Palisade DecisionTools." Southwest Florida Water Management District ( 2006). "Regional Water Supply Plan." < http://www.swfwmd.state.fl.us/documents/plans/RWSP/rwsp.pdf > (June 10, 2009). Southwest Florida Water Management District (1997). "ICI Conser vation in the Tri County Area of the SWFWMD." SWFWMD, Brooksville. < http://www.weap21.org/ > (September 17, 2009). Swank, W. T., and Schreude.H.T. (1974). Comparison o f Three Methods o f Estimating Surface Area a nd Biomass for a Forest o f Young Eastern White Pine." Forest Science, 20(1), 91 100. U.S. Census (2010). "2007 Economic Census." < http://www.census.gov/econ/census07/ >. (June 21, 2010). U.S. Census. (2009 a ). "County Business Patterns." < http://www.census.gov/econ/cbp/index.html >. (June 21, 2010). U.S. Census. (200 9b). "Longitudinal Employer Household Dynamics." < http://lehd.did.census.gov/led/index.php >. (June 21, 2010). U.S. Environmental Protection Agency (1997) Study of Potential Water Efficiency Improvem ents in Commercial Businesses. Final Report to State of California Department of Water Resources, Sacramento. U.S. Environmental Protection Agency (2009) Water Efficiency in the Commercial and Institutional Sector: Considerations for a WaterSense Program. US EPA WaterSense Program, Washington, D.C. Wurbs, R. (1995). Water Management Models: A Guide to Software Prentice Hall, Englewood Cliffs.

PAGE 151

151 BIOGRAPHICAL SKETCH Miguel Morales was born in Caracas, Venezuela in 1986. At the age of seven his family relocated to the United States seeking greater educational and professional opportunities. When engineering. His environmental studies began in the Enviro nmental Engineering Sciences (EES) department at the University of Florida. EES provided him with a grueling and comprehensive curriculum that quickly sparked his interest in water engineering. Miguel fueled this interest by conducting undergraduate resear ch on novel water treatment technologies under the advisement of Dr. David W. Mazyck. His undergraduate research included involvement in the University Scholars Program, and culminated with his senior thesis: Development of Silica Titania Coated Packing Ma terial for use in Photocatalytic Reactors. Miguel graduated summa cum laude with a Bachelor of Science in environmental engineering in the fall of 2008. Having developed a passion for research, the semester following graduation Miguel joined the Conserve F lorida Water Clearinghouse under the advisement of Dr. James P. Heaney. His studies towards a Master of Engineering degree began in evaluating the commercial, industrial and institutional sectors of water use for water conservation potential. Miguel recei ved his Master of Engineering from the University of Florida in the summer of 2010 and will continue his doctoral work at the University of Florida.