University of Southern California Dissertations and Theses

A fire insurance map geocoder for pre-earthquake San Francisco

(USC Thesis Other)

A fire insurance map geocoder for pre-earthquake San Francisco

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
A FIRE INSURANCE MAP GEOCODER FOR PRE-EARTHQUAKE SAN FRANCISCO

by

Yonatan Rosen

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(GEOGRAPHIC INFORMATION SCIENCE AND TECHNOLOGY)

May 2015

Copyright 2015 Yonatan Rosen
ii
ACKNOWLEDGEMENTS
I am grateful for all the guidance and advice given to me by my professors at the USC
Spatial Sciences Institute, including Dr. Darren Ruddell, Dr. Jennifer Swift, Dr. Yao-Yi Chiang,
Dr. Robert Vos, and my advisors, Dr. Katsuhiko Oda and Dr. Karen Kemp.
GIS Librarians Susan Powell of the Earth Sciences and Map Library at U.C. Berkeley and
Andrzej Rutkowski of USC provided me with valuable feedback, and directed me towards many
Sanborn-related resources.
Images of Sanborn maps were downloaded from the David Rumsey Historical Map
Collection.
iii
1 TABLE OF CONTENTS
Acknowledgements .................................................................................................................. ii
Table of Contents ................................................................................................................... iii
List of Tables .......................................................................................................................... vii
List of Figures ....................................................................................................................... viii
Abstract .................................................................................................................................... x
Chapter One: Introduction ........................................................................................................ 1
1.1 Motivation: Finding Context..................................................................................... 1
1.2 Problem Statement .................................................................................................... 2
1.3 Research Questions ................................................................................................... 2
1.4 Objective ................................................................................................................... 3
1.5 Methodology ............................................................................................................. 3
1.6 Outline of this Document .......................................................................................... 3
Chapter Two: Background ....................................................................................................... 5
2.1 History and GIScience .............................................................................................. 5
2.1.1 Mapping Texts ................................................................................................... 6
2.2 Fire Insurance Maps .................................................................................................. 7
2.2.1 Sanborn Maps of San Francisco ........................................................................ 7
2.2.2 Historical Development of Insurance Maps ...................................................... 8
2.2.3 Scholarly Access to Sanborn Fire Insurance Maps ......................................... 10
iv
2.3 Exploring Fire Insurance Maps with GISystems .................................................... 11
2.3.1 Web Access to Insurance Maps of San Francisco ........................................... 13
2.4 Geocoding ............................................................................................................... 14
2.4.1 Geocoder Data Models for Street Addresses ................................................... 14
2.5 ArcGIS Address Locator Styles .............................................................................. 15
2.6 Assessing Geocoding Error ..................................................................................... 16
2.7 Developing a Historic Geocoder ............................................................................. 17
2.8 Geocoding with Insurance Maps ............................................................................ 17
Chapter Three: Design and Implementation of Insurance Map Geocoder ............................. 18
3.1 Conceptual Model of the Sanborn Insurance Map Indexes .................................... 19
3.1.1 Transforming the Index Structure into an Address Locator ............................ 20
3.2 Capturing Street Segment Address Ranges ............................................................ 23
3.2.1 Assigning Street Type ..................................................................................... 25
3.2.2 Assigning Sides to Address Ranges ................................................................ 26
3.2.3 Text Descriptions of Street Ranges ................................................................. 26
3.3 Index Map Georectification and Creation of Map Sheet Footprints ....................... 27
3.3.1 Creating Map Sheet Footprints ........................................................................ 29
3.3.2 Joining Sheets to Links of Map Images........................................................... 31
3.4 Creating a Dummy Street Grid ............................................................................... 31
3.5 Building the Address Locator in ArcMap ............................................................... 34
v
3.5.1 Alias Table ....................................................................................................... 34
3.6 Employing the Address Locator ............................................................................. 34
3.7 Conclusion .............................................................................................................. 43
Chapter Four: Application and Evaluation of Geocoder ........................................................ 44
4.1 Bakeries of San Francisco ....................................................................................... 44
4.1.1 Mapping Bakeries with the Sanborn Geocoder ............................................... 45
4.2 Assessing Geocode Errors ...................................................................................... 46
4.2.1 Range Overlaps................................................................................................ 47
4.2.2 Renumbering of Houses .................................................................................. 48
4.2.3 Classifying Bakeries ........................................................................................ 50
4.3 Female Proprietorship ............................................................................................. 52
4.4 Temporal Reliability of Sanborn Maps and Business Directories .......................... 52
4.5 Comparison of Sanborn-based Geocoder and Contemporary Address Locator ..... 53
4.5.1 Renaming Errors .............................................................................................. 54
4.5.2 Renumbering Errors ........................................................................................ 54
4.5.3 Changes to Streets ........................................................................................... 54
4.5.4 Intersections ..................................................................................................... 55
4.6 Assessment of Geocoders ....................................................................................... 55
4.7 Summary of Geocoding Test .................................................................................. 56
vi
Chapter Five: Discussion ........................................................................................................ 57
5.1 Reading Insurance Maps through a GISystem ........................................................ 57
5.1.1 Overlaying Insurance Maps with Contemporary Data .................................... 58
5.1.2 Overlaying Historical Data on Insurance Maps .............................................. 59
5.2 Extracting Data from Sanborn Maps ...................................................................... 60
5.2.1 Infrastructure ................................................................................................... 60
5.2.2 Environment .................................................................................................... 60
5.2.3 Buildings .......................................................................................................... 61
5.3 Assessment of Geocoding Approach ...................................................................... 62
5.4 Further Work ........................................................................................................... 62
5.4.1 Improving Geocoder Precision ........................................................................ 63
5.5 Conclusions ............................................................................................................. 64
5.5.1 Exploiting the Structure of Insurance Map Indexes ........................................ 64
5.5.2 Comparison of Approaches to Historic Geocoding ......................................... 64
5.5.3 Costs and Benefits of an Insurance Map Geocoder ......................................... 65
5.6 Final Remarks ......................................................................................................... 65
References .............................................................................................................................. 67
Attributions ............................................................................................................................. 71
Appendix A: R Code .............................................................................................................. 74
vii

LIST OF TABLES
Table 3.1: Street segments found on Sheet 174 33
Table 3.2: List of test addresses 37
Table 4.1: Comparison of Insurance Map and Centerline Geocoders 55

viii

LIST OF FIGURES
Figure 3.1 Sanborn Index Map for Volume 1 19
Figure 3.2 Conceptual model of the insurance maps 20
Figure 3.3 Process of development of address locator reference data 22
Figure 3.4 The table tool in ABBYY FineReader 23
Figure 3.5 Sample scale and resolution of recognized text in FineReader 23
Figure 3.6 The Table toolbar in FineReader 24
Figure 3.7 Screenshot of ABBYY FineReader Correction process 24
Figure 3.8 Text descriptions of Coso Avenue 26
Figure 3.9 Detail of Volume II Index Map in ArcMap 28
Figure 3.10 Composite image of map index pages. 28
Figure 3.11 Selecting parcel data and recoding 30
Figure 3.12 Overview of map footprints 30
Figure 3.13 Intersected lines for sheet 174 32
Figure 3.14 Editing the Address Locator style in XML 34
Figure 3.15 Study area 35
Figure 3.16 Composite of rectified map sheets 36
Figure 3.17 Study area in context 36
Figure 3.18 Interactive rematch for 1499 Dupont Street 37
Figure 3.19 Dupont Street, shown in pink, and corresponding dummy lines, in blue. 38
Figure 3.20 Detail of sheets 42 and 43, 600 Block of Green Street 39
Figure 3.21 The positions of 635 and 622 Green, sheet as identified by geocoder 40
Figure 3.22 Tied candidate matches 40

ix

Figure 3.23 Using the Identify Tool to access map sheet image 41
Figure 3.24 Matching Union Place 41
Figure 3.25 Distinguishing Montgomery Avenue from Montgomery Street 42
Figure 3.26 Detail of Sheet 43, Sanborn Fire insurance map 42
Figure 4.1 Detail of Sheet 103, Sanborn fire insurance map 45
Figure 4.2 A sample of listings from the 1904 directory shows their tabular structure. 45
Figure 4.3: Detail of Map sheet 144 shows inconsistent numbering on Market Street. 47
Figure 4.4: Detail of map sheet 146 shows 1587 Market Street. 47
Figure 4.5 Detail of sheet 639 48
Figures 4.6a and 4.6b Details of sheet 145 49
Figure 4.7 1794 Haight Street, Detail sheet 425 50
Figure 4.8 Distribution of bakeries within San Francisco 51
Figure 4.9 New bakeries found in 1905 directory 53
Figure 5.1 Schematic depiction of hill contours, Sheet 175, Sanborn Insurance Map 61

x
ABSTRACT
In the years following the 1906 earthquake and fires, the streets of San Francisco were
renamed, renumbered, and reshaped. These changes make it challenging to locate addresses
found in historical directories, newspapers, and archives. Fire insurance maps produced by the
Sanborn Map Company represent some of the most detailed sources of spatial information about
early twentieth century San Francisco, but they are cumbersome to navigate.
Insurance maps contain detailed street indexes that mirror address geocoders in content and
function—listing street names and address ranges. Exploiting their structure, the text of these
street indexes was transcribed in order to create a geocoder that identifies map sheets. The
Sanborn indexes served as reference data for an ArcMap address locator. The geocoder makes
the insurance maps more navigable and provides historical context for addresses .
1

CHAPTER ONE: INTRODUCTION
At the turn of the twentieth century, San Francisco was a rapidly growing metropolis. The 1906
earthquake and fires destroyed the city’s urban core, erasing the city’s fabric and much of its
architectural character. It is challenging to find context for historical address records like
directories and business ephemera. Fire insurance maps, like those produced by the Sanborn Map
Company, depict the built environment with considerable detail, helping to provide this context.
Insurance maps are a means to understand the spatial characteristics of the pre-earthquake city,
but they are cumbersome to navigate. GISystems serve as a spatial framework to organize the
maps, making it possible to overlay the maps with other data sources, and facilitating the process
of navigation.
1.1 Motivation: Finding Context
Late nineteenth and early twentieth century texts like newspapers, diaries, and directories are
awash in spatial information. Before the widespread adoption of the telephone, the postal address
served both as a means of contact and as locational information. The presence of addresses in
these disparate sources makes it possible to conduct research on individual buildings, blocks and
neighborhoods, by employing the address as a spatial attribute. An address can be used to link an
anonymous classified advertisement to other records, such as directory listings, in order to create
a richer story—was the address a boarding house or a the home of a prominent family? In short,
postal address can serve to contextualize historical sources. Fire insurance maps produced by the
Sanborn Map Company can expand the context beyond the individual address, helping users
explore the broader urban environment, including the surrounding blocks and neighborhoods. In
a city like San Francisco, where much of the historical context is missing, insurance maps can
help to bring that environment into focus.

2

1.2 Problem Statement
Street addresses provide a tangible and discrete geographical reference point. However, in
San Francisco, renaming, renumbering and physical changes to streets made historical addresses
ambiguous. In the first decade of the twentieth century, whole blocks were renumbered, streets
renamed, and roads closed or rebuilt. Identifying the precise location of a historical address can
be a laborious process, requiring research and careful reading of street descriptions in street
directories. Reference data from the appropriate time period can help to resolve ambiguities.
Street maps, directories and indexes vary in their reliability. Insurance maps contain the greatest
level of detail. They show individual structures, documenting building types, materials and uses
on the basis of careful ground surveys. However, working with the multitude of individual large
scale Sanborn maps is awkward and time consuming. A more efficient way to navigate these
maps is needed.
1.3 Research Questions
Street addresses provide fine-grain geographical information, but, as stated above, historical
street addresses in San Francisco are sometimes ambiguous. By developing an address locator
based on fire insurance map indexes it is possible to verify that a geocode is correct and also
provide the geocode with architectural context. This research endeavored to exploit the structure
of the insurance map indexes to develop a geocoder that relies on this historical reference data.
This task prompted three fundamental questions. First, can insurance map directory data meet the
technical requirements of an address geocoder? Second, would such a geocoder represent an
improvement over other methods of developing a historical address geocoder? Finally, what are
the costs and benefits of using this approach?

3

1.4 Objective
Sanborn insurance maps provide a means of visually substantiating the presence of
individual buildings and their listed addresses. Taking advantage of the structure of the insurance
map indexes, an address locator can be developed to reference individual map sheets. The
objective of this thesis was to develop a historical address geocoder that relies on this historical
reference data as a means of contextualizing historical addresses. Thus, the intent was to create a
geocoder that takes a street address as input and produces as output a specific Sanborn map sheet
number, represented in ArcMap as a map sheet footprint, so that the location of a historic address
can be examined in its contemporary context.
1.5 Methodology
Developing an insurance map-based geocoder requires four principal steps. First, the text
from street indexes found in insurance maps must be recognized, corrected, and brought into a
tabular form suitable for use within a database. Second, data from index maps must be
vectorized. Third, the vector representation of map sheets must be linked to the text descriptions
found in the street indexes. Finally, this data must be developed into a geocoder using a
GISystem.
1.6 Outline of this Document
The following four chapters explain the design and implementation of the insurance map
geocoder, and explore the implications of the tool. Chapter Two provides historical background
on fire insurance maps and how historians have made use of them as a resource, as well as
theoretical background on postal address-based geocoding. Chapter Three explains the technical
process of converting the text and image based insurance maps for use as reference information
in a historic geocoder, and illustrates the way that the tool functions. In Chapter Four, directory

4

listings of bakeries are mapped using the insurance map-based geocoder, demonstrating how the
tool reveals patterns in textual sources. The concluding chapter reflects on the success of the
project and explores the implications of using historic resources in this manner.

5

CHAPTER TWO: BACKGROUND
Historical geographical information exist on a spectrum between structured, explicitly
geographical documents and unstructured information, such as text references to place names,
trade routes and visual depictions of place and landscape. Historical Geographical Information
Science (HGIS) can bring together such structured and unstructured sources to make sense of
patterns and make historical information easier to access and analyze. Structured historical
resources include maps, gazetteers and directories, which organize and store geographical
knowledge. Among these, fire insurance maps constitute some of the most detailed sources of
geographical information about nineteenth century American cities. For urban historians,
preservationists and genealogists, they provide a snapshot of cities at various stages in their
development and contain valuable information to bolster research on historic buildings and land
uses. The structure of the insurance map indexes parallels the data requirements of an address
geocoder.
This chapter discusses how historians and other humanities researchers have employed
GIScience. Then, it outlines the history of insurance maps and their use by historians and other
researchers as public and scholarly access to Sanborn maps has grown. Finally, some of the
technical underpinnings of geocoding, particularly as it is implemented in ArcGIS, are explained
to establish the requirements for the development of an address locator using data from Sanborn
indexes.
2.1 History and GIScience
GIScience grew out of the natural sciences, which view the world through an empirical lens
(Gregory and Ell 2007). Historians and other researchers in the humanities rarely have the luxury
of large datasets. Instead, they make sense a fragmentary evidence—material artifacts, textual

6

sources—to construct history. GISystems allow researchers to employ location as a framework
to organize disparate sources. GISystems have been deployed to organize urban archives in
Atlanta (Page et al. 2013), and create a national resource for genealogists in Scotland
(MacDonald and Osborne 2013). These projects underscore the utility of GISystems in helping
researchers to identify locally-relevant information.
Somewhat paradoxically, GISystems are commonly employed to identify sweeping spatial
patterns, but they are equally adept at finding location-specific information. Geocoding historical
addresses is only useful insofar as there are other pieces of information to be gleaned from
plotting the location on a map. Overlay is the process by which two or more data sources can be
integrated within a GISystem. The task of comparing two maps is extremely challenging to
conduct manually. It requires an understanding of each source’s projection and scale. However,
overlay is one of the most basic functions in a GISystem, makes a consuming manual task trivial
(Gregory and Ell 2007).
2.1.1 Mapping Texts
Gregory and Hardie (2011) used corpus linguistics techniques to illustrate attitudes in the
sixteenth and seventeenth century British press towards cities in Britain and continental Europe
by mapping place names mentioned in texts. Similarly, the Mapping Texts project developed a
GIS interface to understand regional and temporal variation in Texas newspapers, employing
topic modeling techniques to determine the subjects of articles (Torget et al., 2011). Use of place
names a geographical information can illustrate regional patterns. Within a city, however, street
addresses can be employed to find fine-grain patterns in the distribution of heterogeneous
phenomena described in newspaper articles, like crime, employment and industry.

7

2.2 Fire Insurance Maps
Insurance maps represent nineteenth century cities vividly, with an engaging level detail that
appears almost comprehensive. However, insurance maps were created to suit insurance
underwriters, not modern historians or the broader public. In fact, prior to the early 1960s, the
maps were rarely accessible outside of the insurance industry (Lamb 1961; Keller 1993). Despite
their prosaic purposes, map surveyors recorded details that went beyond infrastructure and risk
assessment. They depict saloons, stores and hotels—recording buildings as they were used,
whether this use was “genteel or disreputable,” as an early Sanborn surveyor wrote (Ristow
1968, 202). Few cartographic resources approach fire insurance maps in their ability to help
modern users visualize the architectural and commercial character of nineteenth century streets.
2.2.1 Sanborn Maps of San Francisco
The Sanborn maps are not the only available large scale maps available of San Francisco
before the 1906 earthquake. Plat maps are a legal record of property boundaries (Patton et al.
2005). Historical plat maps are available through the website of the San Francisco Department of
Public Works (2014). However, Plat maps provide little additional information beyond block and
lot numbers, dates of registrations and dimensions of lots. Real estate atlases, such as the block
books published by Hicks-Judd in San Francisco, annotate lots with names of owners (1901).
Plat maps and block books provide no indication of the existence of buildings on the site, or even
if the street was had been constructed. In fact, many recorded lots were sited on sand dunes,
steep hillsides and or under shallow bay water, land that was only developed decades later. In
theory, plat maps are temporally and spatially is continuous, because all changes to properties
were required to be registered to be made official. By contrast, Sanborn maps are discontinuous;
they represent relevant sections of cities at discrete periods (Patton et al. 2005).

8

If plat maps represent property in the abstract, fire insurance maps represent the material
environment, as observed by surveyors on the ground. Sanborn maps team with information
about building use and construction, including building footprints, heights, materials and names
of businesses. Street addresses of each building were carefully recorded—a detail lacking in both
plat maps and block books.
Sanborn Fire Insurance maps of San Francisco dated between 1875 and 1991 are available at
various libraries throughout the United States. The Sanborn Company produced lithographed
map volumes in four major editions, 1875, 1893, 1899 and 1915. These editions were
periodically updated with pasted inserts, making each surviving copy a unique source. Scans of
the 1899 microfilms are available online through a ProQuest database and the San Francisco
Genealogy Website (2014). The black and white microfilm scans deprive the maps of the color-
coded symbology. Color photographs of a 1905 update of the 1899 maps are available through
the David Rumsey collection. The color images create a clearer sense of documents’ materiality.
Structures are easier to distinguish and distortions are minimized. The 1899 Sanborn insurance
maps of San Francisco consist of six volumes of lithographed map sheets (Rumsey 2011; Hoehn
2014). Each sheet measures twenty-five by twenty-one inches, recorded at 1”:50’ scale, the
typical scale employed for dense urban centers (Keller 1993).
2.2.2 Historical Development of Insurance Maps
The development of fire insurance maps reflected needs and growth of the fire insurance
industry. Fire insurance maps originated in seventeenth century London as a means for insurers
to assess the vulnerability of their customers’ properties to hazardous conditions (Keller 1993).
In the early nineteenth century, fire insurance was largely conducted by local companies. Local
outfits relied less on maps, because insurers were able to inspect sites themselves. By the mid-

9

nineteenth century, as urban centers grew and new cities developed, insurance maps helped to
provide insurers with information about properties in places they may have never been (Ristow
1968). The 1850 Perris-Hope map of New York City, a multi-volume large scale atlas of New
York City, marks the beginning of the industry, and established many of the conventions of the
genre. The industry grew steadily, although map production was slowed by the Civil War. In
1867, Aetna insurance company hired William Sanborn to survey cities in Tennessee. The next
year, Sanborn founded his own operation, which quickly dominated the insurance mapping
industry (Ibid).
Between 1868 and the early 1950s, the Sanborn Company produced maps for over 12,000
cities and towns in the United States (Ristow 1968). In small towns, the maps concentrated on
industrial and commercial areas where there was greater risk of fire (and potential customers for
insurance companies). In large cities like San Francisco, residential districts were mapped
extensively, although neighborhoods in the periphery of the city may have been omitted.
In part because of the dominance of the Sanborn Company, the maps exhibit a remarkable
uniformity in scale and representation, making it fairly straightforward to read and interpret
them. Most maps were drawn at 1 inch to 50 feet. Building materials are indicated with color—
yellow for wood, red for stone or brick, grey for metal, and green for special hazards (Keister
1993). Building uses were either written out or somewhat inconsistently abbreviated. Single-
family residences were indicated with a “D” for “Dwelling”. Multiple families living on separate
floors were indicated with an “F” for “Flats”. Buildings with multiple families living on the same
floor were classified as “A” for “Apartments” (Lamb 1961; Grim and Narrow 1990). Hotels,
Lodging houses, boarding houses and other housing functions like dormitories were typically
written out.

10

Attention to commercial functions depended largely on their relevance to fire insurance
underwriting. Uses that posed a hazard were given more attention, while other uses are ignored.
Stores are generally indicated with an “S”. Insurance risks informed the level of detail provided
about industrial uses. Uses that had implications for fire such as paint shops, blacksmiths,
bakeries and received more attention than less risky functions like book binding. This focus on
hazards makes the maps less reliable or comprehensive for functions outside of this purview.
Insurers subscribed to map correction services through the Sanborn Company. Maps were
updated with correction stickers pasted over the original to reflect updates and new construction.
Many surviving maps contain several layers of correction stickers. Researchers seeking to
uncover previous layers sometimes removed the correction stickers (Lamb 1961).
2.2.3 Scholarly Access to Sanborn Fire Insurance Maps
Sanborns were a highly specialized proprietary form of geographical information, akin to
high resolution satellite imagery today. Insurers paid dearly for the maps, and the heavy tomes
were costly to store and maintain (Ristow 1968). A single volume of the San Francisco map cost
seventy five dollars in 1893, meaning that the six volumes set was worth nearly $12,000 in 2013
dollars (Sanborn Map Company 1893). In 1925, maps of Chicago cost $1,500 and $500 a year to
maintain updates (Keller 1993). Insurers also paid the Sanborn Company a subscription fee to
keep the maps updated with pasted in updates as new buildings were built. Their cost and rarity
limited scholarly access to the Sanborn Maps (Lamb 1961). Ninety five percent of Sanborn
clients were in the insurance industry (Wrigley 1949).
By the 1930s, changes to insurance underwriting practices made maps less essential to
insurers. To lower costs for their customer base, the Sanborn Company released new maps at
reduced scales (Keister 1993). As Sanborn maps became outdated, they were sometimes donated

11

to libraries and universities (Lamb 1961). However, these piecemeal collections of fire insurance
maps remained fairly cumbersome to access and did not lend themselves to large-scale analysis.
Beginning in the 1860s, the Sanborn Company deposited new maps at the Library of
Congress for copyright purposes, making their collection the most comprehensive. The Census
Bureau accumulated a collection of 1,804 map volumes in the 1940s. In 1967, the Census Bureau
transferred its collection to the Library of Congress. Following this acquisition, the Library of
Congress began to distribute duplicate maps to research libraries (Ristow 1968). In 1983,
Chadwyck-Healey, Inc. released microfilms copies of many of the Sanborn maps held in the
Library of Congress collection. While microfilming led to wider access to the maps, limitations
of the microfilming medium made them difficult to interpret. Maps were filmed at a reduced
scale, and sheets were only partially visible on the screen. Microfilming also introduced
distortions at edges, and the lacked the colors indicating building materials (Keller 1993). Users
had to rely on index maps to assess the orientation of fragmentary map sheets. In 2001, ProQuest
digitized the Chadwyck-Healey microfilms, providing online access to maps by subscription or
for purchase by libraries (Lutkenhaus 2002). The digital database provided modest improvement
over microfilm machine as thumbnails could be viewed simultaneously. However, the distortions
and other flaws of the original microfilms remained.
2.3 Exploring Fire Insurance Maps with GISystems
GISystems have been used widely to integrate information from fire insurance maps with
contemporary data for research and public use. GIScience has been applied both as means to
facilitate access to maps and as a source of data. As early as 1992, the Illinois State Museum
developed the Historical Hazards Geographic Information System, which took advantage of the
Sanborn maps to help identify industrial hazards throughout Illinois. (Keller 1993; Colten 1992).

12

A number of research libraries have employed GISystems to make it easier to navigate their
historic Sanborn collections. In 2010, Yale University Library digitized its large collection of
insurance maps of cities throughout Connecticut. To make the maps easier to access, Yale
digitized map footprints on the basis of index maps. The indexes are available as KMZ files on
the library website. Additionally, they digitized building footprints from the maps of Yale’s
campus, using building heights indicated on the map to create a three dimensional visualization
of the campus (Yale University Library 2013).
The map division of the New York Public Library has developed two excellent public
participation GISystems that made insurance maps of New York City easier for the public to
access. In 2010, they released the Map Warper, a map image georectification tool for the
thousands of maps in their collection, including fire insurance maps of New York City dating to
the 1850s.The Map Warper allows users to identify control points on map images based on
landmarks. Users can also search for maps by location (Vershbow 2013; New York Public
Library 2014). Georectified raster images of historical maps can be used for visual overlay
within a contemporary GISystem, but the images must be vectorized for the data to be extracted.
To facilitate this process, NYPL Labs, a group at the New York Public Library that creates web
apps to help the public interact with its collections, developed Building Inspector. Building
Inspector applies image recognition software to vectorize building footprints. Users then help to
correct vectorization errors, and assign attribute data that cannot be automatically extracted,
including color and building address. The data produced by Building Inspector can then be
exported through a public API.
Insurance maps are commonly employed a source data for studies related to the built
environment, particularly when Raymond (2011) employed Sanborn maps to reconstruct a

13

neighborhood in Seattle that was destroyed by urban redevelopment. Leonard and Spellane
(2013) made use of fire insurance maps dating to 1850 to identify potential historical sources of
contamination to Newtown Creek, which straddles Queens and Brooklyn.
2.3.1 Web Access to Insurance Maps of San Francisco
In 2011, the map collector David Rumsey worked with the San Francisco Public Library to
digitize its 1905 revision of Sanborn maps for San Francisco. The maps were partially damaged
in the fires of 1906. The maps represent a vast improvement on previous microfilm scans. The
high-resolution images are clear, and contain marginalia that dramatize the use of the documents,
including pencil markings and notes. In response to the release of the color insurance maps,
Michal Migurski developed a web interface to allow members of the public to georectify map
scans (Sommer 2011). The resulting website, maptcha.org, has a clickable webmap, allowing
access to large scale map sheets. Map sheets are represented by a thumbnail graphic on the map
and georectified map sheets are not visible on the site (Migurski 2011). Separate from the
maptcha.org site, the David Rumsey site also provides an interface for georectification of map
images through georeferencer.org.
Rectification of the insurance map images has great potential, as demonstrated by the New
York Public Library project. However, the maptcha.org project only modestly improves the
navigability of the insurance maps. In part, the problem stems from a quirk in the San Francisco
map sheets: street widths are not drawn to scale. This means that each block would need to be
independently georectified to minimize distortions. With at least four blocks depicted on each
map, this would require a more intensive process to create accurately rectified maps.

14

2.4 Geocoding
Goldberg, Wilson and Knoblock (2007) offer the most comprehensive review article
covering the state of the art in geocoding research. They define geocoding, in its most basic
form, as the process of transforming textual geographical references into a spatial representation
within a GISystem. These textual references can include place names, relative location
description, or areal units like postal codes, but the most common form of input data is the postal
address (Goldberg, Wilson and Knoblock 2007). Most address systems in the United States
employ house number, street name, as well as city and state to identify location.
Geocoders must accomplish three distinct tasks to correctly identify the location of an
address. First, they must parse the address, distinguishing, for example, the house number from
the street name, street type and directional suffixes. Second, they must use these elements to
identify a geographic feature with matching attributes. Third, they must create a geocode—the
spatial representation of the address (Goldberg, Wilson and Knoblock 2007). The latter two steps
introduce much of the potential for error.
2.4.1 Geocoder Data Models for Street Addresses
Geocoders can be used to identify states, geographical zones like zip codes, or even
geographical features. Each type of geocoder has distinct data models and outputs. There are
three data models that are used for street address geocoders. The first, most commonly employed
model uses street centerlines—line data that represents the street system of a city. An address
range is associated with each line segment—corresponding to a block range, with an odd or even
range assigned to either side of each segment. Once the correct segment is identified, linear
interpolation is employed to identify the location of the address along the segment. By this
method, on a street segment with the range 800 to 899 (i.e. the 800 block), the address 875 would

15

appear at a position three quarters of the way along the segment. In this method, all of the
addresses within a range could potentially geocode correctly—all addresses from 800 to 899 are
assumed to have the same dimensions. In an urban environment, the potential for error is limited
by the relatively short street segments. The magnitude of error is equal to half of the length of a
segment (Goldberg et al, 2007)
Two alternatives to the centerline-based data model have been developed to introduce greater
precision into the geocoding process. Parcel geocoders employ the polygon geometry of parcel
data, with a discrete address assigned to each lot. By this method, an address must match
correctly with both the lot number and street name to be located. Similarly, address point
geocoders employ point geometry to offer more precision on large lots, or when multiple
addresses are associated with a single lot (Goldberg et al. 2007; Zandbergen 2008). While a
centerline geocoder is more flexible and forgiving; address point and parcel geocoders are more
precise. A centerline geocoder will return more false positives—geocoding for addresses that
may not exist in reality, while address point or parcel geocoders are more likely to return false
negatives (Zandbergen 2008).
2.5 ArcGIS Address Locator Styles
Customized geocoders in ArcGIS are termed “Address Locators”. ArcMap version 10.2
offers users twelve basic address locator styles, reflecting distinct reference data requirements
and specific outputs. In addition to street address locators, these styles include city/state, zip code
and place name locators. An address locator style designed to identify the location of a city and
state could not be used to locate house numbers. Among these twelve choices, only three styles
can be used to geocode street addresses: “US Address—Dual Ranges”, “US Address—One
Range”, and “US Address—Single House” (Esri 2015). The Single House locator style is used

16

with reference data that links addresses to discrete objects, represented as points or polygons.
This style is suitable for address point or parcel geocoders described by Goldberg et al. (2007).
The Dual Ranges and One Range locator styles are distinguished by how they account for
polarity of addresses. The Dual Ranges style requires separate ranges for each side of a street
segment, meaning that one line segment has two pairs of range attributes for the left and right
sides of the streets. By contrast, the One Range locator style requires that the each street segment
be designated either Left or Right. In this way, the address locator is able to distinguish odd-
numbered ranges from even-numbered ranges (Esri 2015).
2.6 Assessing Geocoding Error
As shown in Section 2.4, mitigating and quantifying error is a central concern in the
geocoding literature. Incorrect reference data contribute to ambiguous results. In this vein, an
important distinction exists between precision and accuracy. Precision is a technical measure of
an instrument. Just as a precision watch can be trusted to tell time to the closest millisecond,
precision of a geocoder relates to specificity of the measurement. Can a geocoder be trusted to
identify a feature to the nearest inch or the nearest yard? A linear interpolation geocoder may
return the correct street segment for an address, but limitations in the data can hobble its
reliability for finding the correct position on that segment. By contrast, accuracy is the degree to
which points correspond to the real world position being represented (Bolstad et al. 1990).
Precision is an important goal but the emphasis on precision is misplaced for historical
addresses, because reliable reference data are difficult to acquire. It is more important to have a
method of verifying geocode than having fine-level geocode.

17

2.7 Developing a Historic Geocoder
The development of historical address locators has received limited scholarly attention, even
among researchers in HGIS who have employed addresses in their geocodes. The requirements
of address locators are context specific. Debats and Lethbridge (2007) created a geocoder to
accommodate sequentially recoded tax records from Alexandria, Virginia, from a time when
there were no house numbers. In some instances, historic address ranges and street names have
not changed enough to warrant the development of new address locators. Contemporary street
centerline data can be used, employing alias tables to reference modifications to street names.
However, physical changes to the streets stemming from redevelopment, landfill and other
infrastructure modification make editing of the centerline data necessary. Editing street
centerline data is a laborious process, and potentially introduces error.
2.8 Geocoding with Insurance Maps
The significance of Sanborn insurance maps as a historical geographic resource is evidenced
by the extensive scholarly engagement detailed in this chapter. Insurance maps began as a
rarefied resource, but as they became outdated they became more accessible to scholars. Efforts
to reproduce and digitize them have increased availability and public and scholarly interest.
GIScience has been deployed to extract information from insurance maps, and to help make map
collections more navigable. Address geocoding allows users to transform text into spatial
representation in a GISystem. The following chapter demonstrates how a geocoder can be
developed in order to improve the usability of insurance maps within a GISystem to provide
historical context.

18

CHAPTER THREE: DESIGN AND IMPLEMENTATION OF INSURANCE MAP GEOCODER
Fire insurance maps present spatial information to users with varying degrees of on-the-ground
knowledge. Indexes of streets and key maps help users navigate the volumes. These indexes
create a structure that allow users to find spatially relevant information. Making use of this
structure facilitates the process of data development and capture for use within a GISystem. This
project required identifying the elements of the indexes that could be harvested for use in a
GISystem.
As stated above, the objective of this process is to create a geocoder that takes as input a
street address and produces as output a specific Sanborn map sheet number, represented in
ArcMap as a map sheet footprint so that a historic address can be examined in its contemporary
context. As such, the address ranges found in the Sanborn indexes had to be adapted to meet the
requirements for reference data of US Address – One Range style locator, discussed in Section
2.5. Section 3.1 describes the navigational elements of the Sanborn map sheet and street indexes,
and outlines how these elements have been adapted to suit the requirements of an address
locator. The section ends with a flow chart that summarizes the process of development of the
address locator reference data that is described in greater detail in the following three sections.
Section 3.2 explains how the Sanborn street indexes were digitized and restructured to serve as
reference data for the locator. Section 3.3 outlines the process used to georeference the index
maps and create map sheet footprints. Section 3.4 explains how a dummy grid was created to
link the address ranges to the geometry of the map sheet footprints. In Section 3.5, the creation of
the address locator in ArcMap is described. Finally, Section 3.6 demonstrates the functionality of
the address locator.

19

3.1 Conceptual Model of the Sanborn Insurance Map Indexes
In order to adapt the index structure of insurance maps to GISystems tools, it is worthwhile
to delineate their basic elements. Each of the six volumes of the San Francisco Sanborn maps
contains map sheets for a roughly contiguous region of the city. These regions have no explicit
social or political significance, although their boundaries tend to be defined by major streets.
Each volume contained an index or key map, an example is shown in Figure 3.1, which provides
a visual means of identifying the location of a map sheet in relation to other sheets. Each volume
contains over one hundred sheets.

Figure 3.1 Sanborn Index Map for Volume 1
1

Two index pages consisting of a street index, specials index, block index, and
miscellaneous report is found at the front of each volume. The street index lists the streets found
in the volume, their address ranges and the corresponding map sheet. A table titled, “List of
Streets on Old Maps Appearing under New Official Names” details the name changes for streets
between the 1893 edition and the 1899 maps. The specials index identifies sheets for major
landmarks, buildings and significant sites. Figure 3.2 illustrates the various navigational tools
1
This figure from the David Rumsey Map Collection is reused under the Creative Commons Attribution-
NonCommercial-ShareAlike 3.0 License. Attributions for subsequent figures can be found on page 71.

20
included in each map volume used to identify each map sheet. The structure present in the
insurance maps requires a user to take a linear set of steps to find a map of a specific address.
First, she must identify the volume where the address was found. She can then look to the index
map to find a map sheet visually, or consult the index of streets. The index of street simply lists
the street names, the address ranges and their corresponding map sheet. If a street name is
missing from the index of streets, she can look to the list of old street names to determine if the
street name had changed. However, a missing street may simply be found in a different volume.
The separation among the volumes complicates the process of identifying locations, because
each volume has its own index page and index map.
Figure 3.2 Conceptual model of the Insurance maps
3.1.1 Transforming the index structure into an address locator
The San Francisco fire insurance maps comprise 688 sheets. Each sheet depicts an area of
four city blocks, or roughly twelve acres. The regularity in size of maps sheets makes it possible
to think of them as a small areal unit. The footprints can be represented in a GISystem as
polygons which correspond to the location of the map sheet numbers referenced in the street
index.
Volume
Index Page
Index of Streets
Street Name
Address Range
List of Old Street
Names
Block Index Block Number
Index of Specials Names of locations
Index Map Sheet Footprints
21

The navigational elements found in each volume of the insurance map create a robust
means of identifying locations depicted in the insurance maps by hand, but they lack the
consistency and data integrity for computational interpretation. The elements of the index of
streets have clear analogues in the data model of a street address geocoder. By digitizing and
manipulating these elements, they can be transformed into an address locator.
The street indexes share three of the required elements of an ArcGIS street Address
Locator: A street name, followed by an address range, followed by a sheet number. Several steps
are required to fulfil data model requirements of the “U.S. Address—One Range” Address
Locator style. First, the street name field must be divided into three attributes StreetName,
StreetType, and Directional Suffix. Second, the numerical address ranges must be separated into
‘From’ and ‘To’ values. Third, the polarity—the side of the street—must be assigned.
Additionally, the list of old street names can be used to create an alternate name table, providing
aliases for names that have changed. Finally, the sheet number can be used to link the nominal
attributes of the locator to their spatial representations in the GISystem.
The “U.S. Address—One Range” style requires lines as reference data, because it uses
linear interpolation to estimate the position of an address along a line segment. The data derived
from the street indexes refer to sheets, which would be better represented by polygons. However,
in order to develop reference data that would function within the requirements the street address
locator style, a pseudo-grid consisting of multiple line segments falling within each map sheet
footprint was created (the process for developing this grid is described in Section 3.4). The One
Range locator style is preferable here to Dual Range style because it makes it possible to
distinguish odd-numbered segments from even numbered segments, which often occur on
separate map sheets. This workaround made it possible to employ ArcMap’s geocoding

22

interface. The address range transcription and index map digitization processes are both labor-
intensive. While they exploit a robust structure, the incongruities of the source documents and
imprecision of digitization tools mean that neither process can be entirely automated.
Figure 3.3, below, depicts the how reference data for the address locator was developed by
combining the text address ranges from the street indexes (highlighted in blue) with the map
sheet footprints from the index maps (highlighted in green). The street indexes were transcribed
automatically to create a table of street names and address ranges associated with each map
sheet. The index maps were brought into ArcMap and used to assign sheet numbers to modern
parcel data. Map sheet footprints were created, and a false grid of lines was created to represent
street segments. Finally, the address ranges and dummy line segments were each assigned a
unique identification number so that they could be joined in ArcMap, creating reference data for
the address locator. The steps of this process are detailed in Sections 3.2 through 3.4. Sections
3.5 and 3.6 detail the development and implementation of the address locator in ArcMap.

Figure 3.3 Process of development of Address Locator reference data

23

3.2 Capturing Street Segment Address Ranges
At the front of each volume, an index page lists street names and address ranges and the
sheet where that address can be found. The index pages have a roughly tabular format that can be
exploited for use within a geocoder. The tables are not entirely dissimilar to the attributes found
in a modern geocoder. They contain a street name, street type, address range and a spatial
attribute in the form of the sheet number, shown in Figure 3.4.

Figure 3.4 The table tool in ABBYY FineReader
ABBYY FineReader is a document management and Optical Character Recognition (OCR)
software tool. Like Adobe Acrobat Pro, it can be used to convert images of text into machine
readable text, but it is also able to identify page structure, and distinguish page element including
text, tables, and images. Scans of the index pages were downloaded from the David Rumsey
website at 490 dots per inch (dpi). The images were loaded into FineReader, and automatically
preprocessed according to the default settings, which reduced the resolution to 350 dpi. Figure
3.5 shows a sample of the scale and resolution of the recognized text.

Figure 3.5 Sample scale and resolution of recognized text in FineReader
The Analyze Page tool can identify page elements automatically, but it was more reliable to
draw a table over each column on the index page using the Draw Table Area tool. Once each
column was defined, the Analyze Table Structure tool was used to identify the elements of the
table, as shown in Figure 3.6, on the following page. The table structure had to be touched up
using the Delete separator tools.

24

Figure 3.6 The table toolbar in FineReader
Once the table structure was properly drawn, the text could be identified using FineReader’s
OCR capabilities. The results of the OCR must be reviewed for errors. FineReader provides a
means to compare the text image to the recognized text in two windows, as shown in Figure 3.7.
The recognized text are displayed next to the text image. FineReader identifies text characters
based on their similarity to known fonts. Characters are assigned a confidence score. “Low
confidence” characters are highlighted in blue, which facilitates the manual correction process.
Extra attention was paid to numerals, because errors in recognition of numeric characters are
more difficult to identify than errors in alphabetical characters that affect the spelling of words.

Figure 3.7 Screenshot of ABBYY FineReader correction process
Images in the David Rumsey Collection of index pages for volumes one through four have
missing sections due to damage at the edges. These lacunae were supplemented with text from
microfilmed copies available through the ProQuest database. However, due to poor image
quality, the microfilmed portions were transcribed by hand.

25

After completing optical character recognition for the index pages of each volume, the files
were exported in Rich Text Format (RTF). RTF preserves the tables identified by ABBYY
FineReader. In Microsoft Word, the six separate files were combined manually into a single
document. The resulting tables contained four columns: the street name, text descriptions of
ranges, numerical ranges, and sheet number. In order to develop an address locator in ArcMap,
the four columns needed to be further subdivided into Street Name, Street Type, Directional
Suffix, From, To, and Sheet number.
Extraneous formatting was removed using the find and replace function. The hyphen
character between the numerical ranges was replaced with a tab character. The tables were then
converted to tab delimited values (TDV) and the file was saved as plain text. The TDV file was
then imported into Excel. In Excel, the resulting file contained four columns: Street, From, To,
and Sheet number. A combined total of 5,153 street segments are listed in the street indexes.
The structure of the street indexes lends itself to digitization, but the format was designed for
human interpretation. Inconsistencies in formatting had to be manually corrected to meet the data
requirements of an address locator. These inconsistencies varied enough to make automated
correction impractical. However, simple functions in Excel make it possible to correct values
that fail to meet data requirements. In the street index, repeated street names were represented
with ditto marks. In Excel, the names of repeated streets were filled by dragging the fill handle.
Using column filters, it is straightforward to identify incorrect values found in a column.
3.2.1 Assigning Street Type
The street type value is easy to assign using Column filters. The term “Street” was omitted
from most streets in the directory. By filtering the Address column by other street types (i.e.
Avenue, Way, Alley, Lane, Place), it was possible to assign to correct street types to large groups

26

of streets at once. The remainder (3,426 of 5,151) were filled with the term “Street”. A small
number of streets (151) also included a directional suffix, which were identified by filtering.
3.2.2 Assigning Sides to Address Ranges
Single-range Address Locators are also required to have an attribute that denotes the side of
the street where the address is found. Most ranges were arbitrarily assigned the value of L,
corresponding to the left side of the street. In the street indexes, streets that were split across
multiple sheets were demarcated with an asterisk next to the sheet number. Using column filters
in Excel, one-sided street numbers were isolated. Columns were sorted by address name and
numerical range. Odd numbered ranges were assigned the letter R for the right side.
3.2.3 Text Descriptions of Street Ranges
Not all of the street segments included in the street indexes include a numerical address
range. Address ranges for smaller streets and alleys were often omitted. Some 763 street
segments contain no address range or text description. Most of the names of the segments do not
occur on multiple pages. These segments correspond to smaller streets. Other street segments
lack a numerical address range, but contain text descriptions of streets bounding the segment. For
example, Coso Avenue appears on three sheets, shown in Figure 3.8.

Figure 3.8 Text descriptions of Coso Avenue
The first segment depicts the north side of Coso Avenue between California Avenue and Buena
Vista. Just 341 of the street segments contain such text descriptions. Examination of these
occurrences reveals that these street segments appeared either undeveloped or lacked numbered
buildings represented on the map sheets. To deal with these missing numerical ranges, it is

27

possible to find the corresponding address values using other data sources like street directories.
However, it is not possible to automate this process, because available street directories from San
Francisco of this period lack necessary structure. The fact that these address ranges tend to
correspond to unbuilt (or unrepresented) streets on the insurance maps suggests that addresses in
these ranges would occur infrequently. As an alternative measure, the Address Locator can be set
to match addresses without a house number in the Locator Preferences. Selecting this option
slows the performance of the Locator slightly, because it creates a greater number of possible
candidate matches. However, the tool makes it easy to check candidate map sheets to identify the
correct sheet.
3.3 Index Map Georectification and Creation of Map Sheet Footprints
At the front of each volume, an index or key map appears in order to help users navigate the
map sheets. The six images of the index maps were downloaded from the David Rumsey
website. Using the slice tool in Adobe Photoshop, images of the index maps were divided into
smaller tiles, in order to make them easier to manipulate within ArcGIS.
The tiled images were loaded into ArcMap. Employing the Georeferencing Tool, control
points were assigned to each map image tile. Georeferencing is an imprecise process, requiring
trial and error to adjust map images to fit the coordinate system. By georeferencing tiled areas of
the images, distortions caused by discrepancies in projections could be minimized.
Next, parcel data from the City of San Francisco data portal were loaded into ArcMap. While
some individual lots may have changed, the shape and dimensions of the blocks have remained
consistent for the most part. This allowed the corners of blocks to be used as control points.
Street centerline data were loaded to provide labels to streets. Figure 3.9, on the next page, shows
the alignment of one index map tile with modern parcel data.

28

Figure 3.9 Detail of Volume II Index Map in ArcMap
The process of georectification was repeated for each of the six volumes of the map sheets.
Figure 3.10 shows a composite of the index pages. There is no map coverage for large portions
of the city, including the Presidio, Golden Gate Park, and large areas of the Sunset and
Richmond Districts on the city’s west. These absences tend to correspond to areas that were not
relevant to insurance mappers—parks, cemeteries, military bases and undeveloped land.

Figure 3.10 Composite image of map index pages.

29

On their own, the georectified index map images cannot be used for analysis. In order to be
used as part of the address locator data model, the illustrated map sheet footprints must be
represented with vector data to create discrete objects that can be manipulated within the
database.
3.3.1 Creating Map Sheet Footprints
Once the task of georeferencing of index maps was complete, a vector representation of the
map sheets needed to be created. Chiang et al. (2009) employed raster analysis techniques with
historical maps to automate the process of digitization. While the task of digitization could have
been accomplished partially by identifying regions through raster analysis, the scale and
generalization of the index maps make them unreliable for overlay with modern data. Instead,
map sheet numbers were manually assigned to the modern parcel data geometry, insuring that
the referenced map footprints reflect the geography of the city.
In ArcMap, parcel data were overlaid on top of the tiled images of the index maps, as shown
in Figure 3.11. All parcels falling within a single color-coded map sheet were selected. The
selected parcels were then assigned the corresponding sheet number as an additional attribute.
Some blocks of the city aligned cleanly with parcel data, but in other cases, careful examination
of the map sheets themselves was required to determine which parcels to code to which map
sheet. This was particularly true in the peripheral tracts of the city where the orthogonal structure
of the city’s street grids were not maintained, such as in Bernal Heights or the Fairmount Tract.

30

Figure 3.11 Selecting parcel data and recoding
In limited regions of the city, the geometry of the parcels did not correspond properly to the
maps sheets. For example, the Marina District, which was not developed until after 1915, did not
continue the street grid of the surrounding streets. In these instances, the parcels were edited to
fit the historical street grid structure. These areas were undeveloped in the 1899 and 1905
Sanborn Editions. In fact, much of the property in these parcels was still unfilled bay and
marshland. Figure 3.12 depicts the map sheet footprints generated by the coded parcel data. The
spatial regularity and relative continuity of the sheets is evident.

Figure 3.12 Overview of map footprints

31

3.3.2 Joining Sheets to Links of Map Images
While not a part of the process to develop an address geocoder, eventual use of the geocoder
required the map footprint data to be related to images of the original Sanborn maps. Images of
the map sheets are available on three different databases. ProQuest’s Digital Sanborn Maps,
1867-1970 provides access to scans of the Chadwick-Healy microfilm of the 1899 edition of the
maps in PDF format. The ProQuest database requires a subscription for access, and the database
interface makes linking impractical. The website SFGenealogy.com provides public access to the
same 1899 edition with scans of a distinct microfilm. The SFGenealogy scans are superior in
clarity and legibility, but evidence substantial distortion at page edges. The full-color images
available from on the David Rumsey website supersede both microfilm sources in quality,
notwithstanding the fire-damaged sections of the pages. Contrasts between the 1899 edition and
the 1905 update also provide further insights into the development of the city during that period.
Links to the images from SFGenealogy.com and the David Rumsey website were extracted
by editing the relevant index HTML pages in a text editor. In both cases, the links to the map
sheets contained the sheet number in the URL. Using the sheet number attribute, a table
containing links was joined to the map sheet footprints. This allows the relevant map image to be
opened within ArcMap by using the Identify tool on a particular map sheet footprint.
3.4 Creating a Dummy Street Grid
Theoretically, a geocoder can use an address to identify any type of object. A geocoder can
return a polygon, corresponding to a zip code region, for example. However, ArcGIS requires
line data for Address Locator styles that employ address ranges, like the “U.S. Address- One
Range” style. In order to meet the data requirements, arbitrary line street segments were created

32

to correspond to map sheets. The objective in creating this locator was to make geocodes that
identify the correct map sheet.
After examining options for creating lines within a polygon, a more straightforward method
was selected. A fishnet consisting of 1000 columns and no rows was generated, using the extent
of the map sheet footprints layer. The fishnet was intersected with map sheet footprints creating
14,777 segments (many more than the 5,151 segments in the directory), each coded with the
number of a corresponding map sheet. Figure 3.13 shows the grid intersected with the fishnet.

Figure 3.13 Intersected lines for sheet 174
Line segments in each map sheet footprint needed to be numbered sequentially to serve as a
unique identifier, in order to be able to join the address range data. Using Feature to ASCII tool
in ArcMap, the attributes of the intersected line features were exported into a text file. The
Feature to ASCII tool automatically includes a coordinate pair and length attribute for each
feature, but only the ObjectID and Sheet number attribute are necessary. The text file was loaded
into RStudio. A short code, shown in Appendix A, was run. The code sorts the lines by their
sheet number, and creates a sequence number for each feature. The sequence number is then

33

concatenated with the sheet number to create a unique id number, separated by an underscore
character. The resulting table is saved as a CSV for import into ArcMap. Using the ObjectID
attribute, the resulting table is joined to the intersected fishnet lines in ArcMap. A new feature
class is created.
A similar code was used for the table containing address ranges, also shown in Appendix A.
The lines are sorted based on sheet number, then a sequence number is generated, and a unique
identifier is created for each feature. The resulting table is imported into ArcMap and joined to
the numbered line segments, based on the unique identifier previously created. While the line
number tables and address description tables could be merged in R, joining in ArcMap allows for
more flexibility as street description files must be edited periodically.
The footprint for sheet 174, depicted in Figure 3.13, above, contains twenty nine line
segments. Each line segment is numbered sequentially: 174_1, 174_2, up to 174_29. In the street
directories, nine address ranges are associated with the Sheet 174, shown in Table 3.1 on the next
page. Employing the sequentially numbered value generated using the R code, the attributes
developed from the index maps can finally be associated with a spatial feature in ArcMap.
Table 3.1 Street segments found on Sheet 174
UniqueID ST_NAME ST_TYPE FROM TO JoinID
174_1 BOND STREET 1390
174_2 FRANKFORT AVENUE 1392
174_3 GLEN PARK STREET 1393
174_4 TONNINGSEN STREET 1397
174_5 HOWARD STREET 1600 1699 1394
174_6 MISSION STREET 1601 1699 1395
174_7 FOLSOM STREET 1699 1640 1391
174_8 THIRTEENTH STREET 100 290 1396
174_9 TWELFTH STREET 100 256 1398

34

3.5 Building the Address Locator in ArcMap
In ArcMap, a US Address locator requires linear features with these attributes: name, type,
direction, joinID. The fishnet line features with the address ranges assigned to them meet these
requirements. Default setting for the one-range U.S. Address locator functioned sufficiently. A
slight modification was required to allow for matching addresses without an address.
3.5.1 Alias Table
Changes to street names are listed in on index pages and within street directories. An alias
table was developed by digitizing the street list provided in the fire insurance map indexes and
finding the corresponding street segments. Just eighty two street name changes were identified in
this manner. Another group of changed names were identified by consulting street directories.
The alias table simply requires a join ID for each segment and the modified name. Some aliases
were created to correspond to frequently seen abbreviations. These included the ordinals
“second” and “third”, which are abbreviated “2d” and “3d”, and other abbreviations unique to
this period. Figure 3.14 shows how to edit the alias of the Address Locator style in the XML file
“USAddress.lot”, which is found in the Geocode folder of ArcMap system folders.

Figure 3.14 Editing the Address Locator style in XML
3.6 Employing the Address Locator
A group of addresses reflecting a discrete area of the city can illustrate the utility of the
Address Locator in identifying correct map sheets. Sheet 43, which includes Washington Square
in the district now known as North Beach was selected as the focus of the study area. The region

35

contains many of the prominent streets of San Francisco of the period, including Montgomery
Avenue, now known as Columbus Avenue and Dupont, now known as Grant Street. Seven
adjacent map sheets, Sheets 31, 32, 42, 44, 55, 56, and 57, were also included in the study area,
shown in Figure 3.15. Figure 3.16, shown on the next page, is a composite of georectified map
sheets for the study area. Figure 3.17, also on the next page, shows the study area in relation to
the city as a whole.

Figure 3.15 Study area

36

Figure 3.16 Composite of rectified map sheets

Figure 3.17 Study area in context

37

A list of addresses, shown in Table 3.2, were created as a table and added to ArcMap. The
addresses are designed to illustrate the way that the locator responds to various conditions,
including, addresses falling within a known address range, addresses outside of known ranges,
and addresses that match multiple ranges.
Table 3.2 List of Test Addresses
No. Address Geocoding Result
1 1499 Dupont Street Matched to Sheet 43
2 622 Green Street Matched to Sheet 43
3 650 Green Street Not Matched
4 635 Green Street Matched to Sheet 42
5 501 Union Tied Candidates
6 502 Union Tied Candidates
7 8 Union Place Matched to Sheet 43
8 541 Montgomery Ave Matched to Sheet 43
9 543 Montgomery Ave Tied Candidates
10 1001 Jasper Place Matched to Sheet 43
The table was geocoded in ArcMap. Five of the addresses were matched correctly, and the
other five matched but had other candidate matches. The address 1499 Dupont Street was
matched readily, and the correct map sheet identified, shown in Figure 3.18.

Figure 3.18 Interactive rematch for 1499 Dupont Street

38

Note that the candidate matches, shown as blue points, are scattered across the study area.
This is an artifact of the dummy lines that fall arbitrarily on each map sheet footprint. The
matched point, shown in yellow, is more than two blocks away from Dupont Street itself. Figure
3.19 shows Dupont Street in pink, and the corresponding dummy lines found on each of the
adjacent map sheets, in blue.

Figure 3.19 Dupont Street, shown in pink, and corresponding dummy lines, in blue.

39

The three addresses on Green Street are illustrative of how the locator deals with addresses
falling within and outside of the address range associated with a street segment. The 500 and 600
Blocks of Green Street are divided between sheets 42 and 43. The upper limit of the 600 Block is
640, reflecting the numbering of buildings shown in the Sanborn map, shown in figure 3.20.
Note that this figure is oriented with north at the bottom.

Figure 3.20 Detail of sheets 42 and 43, 600 Block of Green Street
The address 622 Green Street falls within the correct range. However, 650 Green Street is
not found by the locator. Instead, the locator offered the twenty line segments associated with
Green Street, as it did with Dupont Street. The task of identifying the correct range falls on the
user. The odd range of the line segment, 501-635 Green Street, falls on sheet 42. Figure 3.21
illustrates the geocoded coordinates of 622 and 635 Green Streets. The correct locations on each
map sheet is outlined in black. The locator is able to correctly identify the correct map sheet for
odd and even ranges appearing on separate sheets.

40

Figure 3.21 The positions of 635 and 622 Green, sheet as identified by geocoder
Addresses on Union Street illustrate a different problem. An entirely different street called
Union also existed in Bernal Heights. For this reason, there are two equal candidate matches in
the locator for “501 Union” without a street type specified: the Union Street falling on sheet 42,
and the one falling on sheet 588, as seen in Figure 3.22.

Figure 3.22 Tied candidate matches

41

To decide between these matches, the user can examine the insurance maps themselves,
which provide more evidence to corroborate the presence of an address at the given location.
Using the identify tool, candidate links to images of candidate map sheets can be accessed,
shown in Figure 3.23.

Figure 3.23 Using the Identify Tool to access map sheet image
Examination of the map sheet revealed that no location existed at 501 Union in Bernal
Heights. This context helps to resolve some of the ambiguities that the locator itself is unable to.
However, addresses with similar names but different street types (e.g. Street, Avenue, Alley)
match correctly. Despite the ambiguity between the different Union addresses, the address 8
Union Place matched to the correct segment, as shown in Figure 3.24. The match takes place
without regard for the address number.

Figure 3.24 Matching Union Place

42

Montgomery Avenue (now called Columbus Street) is a diagonal street that shares many
address ranges with Montgomery Street. The locator correctly identifies the sheet for the address
that falls within the correct address range (541 Montgomery Avenue), as shown in Figure 3.25,
but it cannot distinguish between the ranges for 543 Montgomery Avenue, which falls outside of
the correct range.

Figure 3.25 Distinguishing Montgomery Avenue from Montgomery Street
Jasper Place, a short street between Union and Filbert Streets, shown in Figure 3.26, does not
have an address range associated with it in the street index, although the map shows addresses
are assigned in this block. An address far outside of the appropriate range (1001) still matches to
the correct map sheet, because the locator can match addresses without house numbers.

Figure 3.26 Detail of sheet 43, Sanborn Fire insurance map

43

3.7 Conclusion
Developing a geocoder based on insurance maps required the use of sophisticated Optical
Character Recognition software to transcribe and correct the text of the address ranges. While the
street indexes had a roughly tabular form, considerable effort was required to manipulate the text
into a form that would match stringent data requirements of a contemporary address locator.
Additionally, the map footprints were created by georeferencing index maps and recoding parcel
data. The resulting geocoder allows users to visualize the rough position of geocodes within
ArcMap, as well as a means to quickly find and inspect an image of the original insurance map
sheets.

44

CHAPTER FOUR: APPLICATION AND EVALUATION OF GEOCODER
The objective in creating a historic address geocoder is to correctly identify the location of
historic addresses. Mapping a large group of historic addresses can demonstrate the strengths of
a geocoder, and the types of addresses that it fails to recognize. Errors can result from problems
in the geocoding method, errors in transcription or problems in the source materials. Mapping a
large set of addresses can shed light on the nature and characteristics of source materials that
would otherwise be obscure. Using directory listings for bakeries reveals insights into the utility
of business directories as well as fire insurance maps in research.
4.1 Bakeries of San Francisco
Listings in business directories are a convenient source of historical addresses, because they
are structured in a way that is machine readable. The Crocker-Langley Directory was published
annually, with alphabetical and classified listings within San Francisco. Researchers have relied
on the Crocker-Langley business directories to identify locations of businesses or residences.
Paul Groth (1994) employed listings for residential hotels and boarding houses to illustrate their
distribution throughout San Francisco. Edith Sparks (2006) and Jessica Sewell (2011) mapped
listings of groceries and other female-headed businesses to explain their role in commerce during
the turn of the century.
Bakeries demonstrate both the limitations and the merits of the Sanborn maps as a data
source. Bakeries were fire hazards, but they did not always bear the same attention paid to larger
industrial hazards. They were also more dispersed throughout the city than industrial functions
like paint production. Figure 4.1 shows a typical bakery found on a Sanborn map. Brick bakery
ovens shown in red look distinct against the mostly timber-framed construction in San Francisco,
coded yellow, making bakeries easy to identify visually.

45

Figure 4.1 Detail of Sheet 103, Sanborn fire insurance map
4.1.1 Mapping Bakeries with the Sanborn Geocoder
To identify bakeries, listings under the heading “Bakeries” were copied from the digitized
1904 and 1905 editions of the Crocker-Langley San Francisco City Directory. The digitized text
is available for download through the Internet Archive. Text was recognized using ABBYY
FineReader, and reviewed for character recognition errors. Figure 4.2 shows that the listings
follow a consistent structure: last name and first name, followed by the addresses separated from
the name with a comma. An additional comma separated address numbers from numbered
streets. The recognized document was saved as plain text, and opened within Microsoft Excel.
The commas and line breaks of the listings parallel the structure of a comma-separated values
(CSV) file.

Figure 4.2 A sample of listings from the 1904 directory shows their tabular structure.
Excel interprets the commas as column breaks when a text file is loaded. To remove the
comma used as separation between numbered streets, the columns containing addresses and
numbers were concatenated, forming a new, corrected address column. New lines were manually
inserted for listings with multiple addresses. The names contained some clue as to the ownership

46

of the bakeries, allowing for the creation of a third attribute by filtering the name column in
Excel. Fifty nine of the names were corporate entities. One hundred twelve of the names
contained the titles “Mrs” or “Miss”, identifying the proprietor as female (two additional names
lacking titles were identified as female). The remaining two hundred names were labeled as
male. The resulting file was imported into ArcMap as a list for geocoding.
The 1904 edition contained three hundred seventy one addresses. Of these, four addresses
were duplicates, making three hundred sixty eight unique addresses. The table of addresses was
geocoded using the Sanborn based geocoder. The geocoder located three hundred seven
addresses, found multiple possible candidate matches for forty-two addresses, and failed to
identify locations for nineteen addresses. However, geocode matches identified by the locator do
not necessarily correspond to the reality on the ground. By comparing the resulting geocodes to
the Sanborn maps, it is possible to verify the presence of bakeries at the mapped locations, and
clarify some of the reasons for errors in locating addresses.
Following the geocoding process, the map sheet for each of the geocoded address was
inspected to confirm that the address could be found. The Sanborn maps provide additional
information that was used to classify the addresses into four categories: addresses with ovens,
addresses labeled as a store or saloon with no oven, addresses labeled as dwellings, and
addresses that were not found.
4.2 Assessing Geocode Errors
Roughly 16 per cent of addresses listed in the directories had tied candidate matches or failed
to match addresses at all. Of the nineteen unmatched addresses, ten were intersections that could
not be recognized in this locator, because line segments lack connectivity. Seven of the errors
resulted from problems with optical character recognition and could be located once the

47

addresses were corrected. Just two addresses could not be found at all. Problems with tied
candidate address matches were more difficult to resolve. They resulted from errors in numerical
ranges provided in the insurance maps index, introduced in the transcription process or from
problems with distinguishing odd from even street ranges.
4.2.1 Range Overlaps
Numerical ranges provided in the index sometimes overlapped. For example, a bakery listed
at 1587 Market Street matched to segments numbered 1501-1685 on sheet 144 and 1401-1599 on
sheet 146. Market Street was numbered inconsistently. The block of Market depicted on sheet
144, shown in Figure 4.3 was also numbered 1201-1345. Both ranges potentially reflected the
correct address. By examining the map sheets themselves, it was possible to identify the correct
location, which appears on sheet 144, shown in Figure 4.4.

Figure 4.3: Detail of Map sheet 144 shows inconsistent numbering on Market Street.

Figure 4.4: Detail of map sheet 146 shows 1587 Market Street.
Other instances of range overlaps were the result of OCR errors that were not identified
through quality control, which included visual inspection of OCR results. A bakery listed at 2402

48

Folsom Street matched the range 2000-2748 Folsom. In this case, the numeral “1” was
misrecognized as numeral “7”. Address ranges on most map sheets are rarely larger than 200.
Scrutiny of disproportionate ranges can help to reduce these errors. It is possible to make
changes to the geocoder as such errors are identified, to improve its accuracy.
4.2.2 Renumbering of Houses
A larger set of geocode errors resulted from house renumbering that took place between
1899 when this edition of maps were first published and 1905. Evidently, street indexes were
developed by finding the upper range number for each street on a map sheet, as opposed to the
theoretical address ranges of a block. Therefore, the even numbered 900 block of Valencia
Street, which appears on sheet 640, appears in the index as 900-960. Having accurate upper
ranges is advantageous in linear interpolation address locator, but can result in more false
negatives (Zandbergen 2008). A bakery listed at 992 Valencia Street, shown in Figure 4.5, was
previously numbered 948.

Figure 4.5 Detail of sheet 639

49

Errors related to changes in addresses illustrate just how unreliable a street address was in
precisely identifying a location. Two buildings on one map sheet could be labeled with the same
address. A bakery owned by Frank Wing at 1552 Howard Street may have been on the corner of
Lafayette Street or the corner of Twelfth Street, depending if the newer of older address range
was being observed. Figures 4.6a and 4.6b show two properties labeled with the same address.

Figures 4.6a and 4.6b Details of sheet 145
In total, fourty of the the listings for bakeries in the directory could not be matched to an
address labeled on the maps. It is unclear if these ommisions reflect errors in the directories or
errors in the Sanborn maps. One particularly confounding listing is that of Mrs. W. Waldeck of
1209 Larkin Street. Her place of business, (or residence) is shown as a block-wide salt water
baths, in both the 1899 and 1904 versions of the Sanborn maps. It is possible that an address
existed at that location prior to the construction of the baths, or that there was a typesetting error
in the listing.

50

4.2.3 Classifying Bakeries
The directory listings make few distinctions among the bakeries, except for the few bakeries
where Pies or Crackers were baked. However, visual inspection of the map sheets reveals
important distinctions among the bakeries. While the presence of an oven might appear to be a
requisite for the running of a bakery, some ovens were not shown at addresses listed in the
directory. Figure 4.7 shows Christopher Aker’s Bakery at 1794 Haight. A baking oven is shown
in red in the back room, but the address is also classified as a store and lunchroom (denoted with
“S. & Lunch”), and a single family dwelling (denoted with a “D.”).

Figure 4.7 1794 Haight Street, Detail sheet 425
One hundred forty of the bakery listings are identified as a store, saloon or restaurant on the
Sanborn maps, with no visible oven. Such listings may have sold bread produced off site, or their
ovens may have been too small for the attention of the insurance surveyors. Two hundred and
five listings matched addresses with ovens illustrated in them. Such listings corroborate the
evidence that the location held a bakery. Figure 4.8 shows the distribution of bakeries throughout
San Francisco based on these categories, which are not exclusive. Stores are illustrated with a
small blue circle. Squares indicate culinary functions: red for lunch counters and cafeterias, teal
for bakeries, blue for restaurants and green for saloons. Listings with ovens illustrated on site are
circled in orange.

































































































 
























































































" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
! !
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
! !
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
±
!
Stores
" )
Bakeries
" )
Lunch/Cafeteria
" )
Restaurants
" )
Saloons

Ovens
Figure 4.8 Distribution of Bakeries within San Francisco
51
52

4.3 Female Proprietorship
Thirty percent of the total listings had female proprietors. Eighty two, or 58 percent, of
bakery listings depicted as stores on the Sanborn maps were headed by female proprietors. Just
twelve of the two hundred five bakeries depicted with ovens on the maps were headed by
women. Assuming that the directory correctly listed locations of bakeries, this discrepancy
suggests that women were more likely to run smaller, storefront bakeries, while large bakeries
were headed by men. Eleven of the listings corresponded to domestic buildings, with no store or
factory visible on the Sanborn maps. Six of these listings, were headed by women. It is
conceivable that such listings corresponded to home businesses.
4.4 Temporal Reliability of Sanborn Maps and Business Directories
The latest update present in this Sanborn maps edition were dated to 1905, but updates to the
maps were conducted incrementally. Areas of the city with greater density and more change
were surveyed carefully to reflect changes, while peripheral areas seemed to languish. Most
updates were undated. Therefore, the maps are properly dated to a range of years—1899 to 1905.
Not a snapshot, but a long exposure. The Crocker Langley directory, on the other hand, is
temporally continuous. A record exists for each year, and differences between directories from
year to year might represent change over time. Comparing the 1904 listings of bakeries to the
1905 listings might complicate that assumption. While they contain the same number of listings,
mapping them demonstrates a dramatic change. One hundred nine bakeries were found at
locations that were not found in the previous year’s listings, shown in Figure 4.9 on the
following page. Much of this expansion pushes towards the western and southern edges of the
city. It is nevertheless difficult to assert on the basis of this evidence that the presence of bakeries
changed so rapidly.

53

Figure 4.9 New bakeries found in 1905 directory
4.5 Comparison of Sanborn-based Geocoder and Contemporary Address
Locator
A common approach to historical address geocoding is to modify contemporary street
centerline files to comport with historical names, ranges and geometry (see, for example, Debats
2007). In order to compare the insurance map geocoder to a centerline geocoder, a street address
locator was constructed using publicly available data from the City of San Francisco website.
The shapefile contains all relevant attributes necessary for a functioning modern locator, but
some of these attributes must be edited to function with historical addresses. Incorrectly
identified historical addresses can indicate where the underlying data must be edited to create a
functioning historical locator.

54

The centerline geocoder identified the correct map sheet for two hundred seventy bakeries,
or 73 percent of the total. Centerline geocoded addresses are more precise than the insurance
map based geocodes, when they can be verified, but this precision corresponds to an increase in
plainly incorrect results. Ninety-seven were incorrectly identified. These are discussed below.
4.5.1 Renaming Errors
The simplest approach to developing a historic address locator is to create an alias table that
joins renamed streets to corresponding present-day street segments using a modern centerline
file. This approach is easy to accomplish, because lists of renamed streets are compiled
frequently in directories and guidebooks. Sixteen of the erroneous geocodes resulted from streets
being renamed. However, in many cases, the address ranges of streets changed along with the
names. These errors cannot be resolved by an alias table alone. Seventeen additional errors were
the result of ambiguity between numbered streets and avenues. Numbered avenues in the
Richmond and Sunset Districts were in early stages of development in 1904.
4.5.2 Renumbering Errors
Fifty geocoding errors resulted from street address ranges changing, eighteen resulted in tied
matches. To correct for such errors, individual line segments would have to be identified and
edited to fit a broader range, or edited to match numerical ranges found on the Sanborn map
sheets. Either approach would be relatively laborious.
4.5.3 Changes to Streets
Physical changes to streets sometimes resulted in unmatched geocodes when using the
centerline geocoder. Certain blocks of streets were removed over time as the result of freeway
construction, urban renewal, and other forms of development. Just five of the geocode errors are

55

attributable to physical changes to streets. These errors could be ameliorated by drawing new
street centerlines segments.
4.5.4 Intersections
The centerline based geocoder is able to correctly identify intersections. This is impossible
using the Sanborn geocoder. However, the locator does not currently distinguish between
directional indicators (e.g. “SW cor Mission and 11
th
”). This makes it necessary to take an extra
step to match intersection geocodes to map sheets, because an intersection could potentially
match to up to four distinct sheets.
4.6 Assessment of Geocoders
The performance of the two geocoders was comparable. In raw numbers, the centerline
geocoder correctly identified two hundred seventy map sheets, while the insurance maps
geocoder identified three hundred seven. Table 4.1, below, compares the match rates of the two
geocoders. The Sanborn based geocoder found seventy one bakeries missed by the centerline
geocoder. Thirteen bakeries missed by the Sanborn locator were correctly identified by the
centerline geocoder. An additional ten intersections could only be located using the centerline
based locator. The results suggest that the two methods could work in tandem, serving as a basis
to check one result against another.
Table 4.1 Comparison of Insurance Map and Centerline Geocoders

Insurance Map
Geocoder
Centerline
Geocoder
Correctly Matched 307 (83.4%) 270 (73.3%)
Multiple Candidate Matches 42 (11.4%) 27 (7.3%)
Unmatched 19 (5.1%) 71 (19.3%)
Total Listings 368 (100%) 368 (100%)
Matches missed by other
geocoder
71 13

56

4.7 Summary of Geocoding Test
Mapping the Crocker Langley Directory using the insurance maps geocoder demonstrates
the utility of the tool and uncovers interesting limitations in the both the directory and Sanborn
maps. While both sources purport to inventory business and industry in San Francisco more or
less comprehensively, neither source appears to be exhaustive. Checking the directory listings
against the Sanborn maps suggests that the bakery category was more heterogeneous than the
directory suggests. The insurance maps help not only to validate the geocode, but to provide
visual context that informs how they can be interpreted.
The insurance map geocoder modestly outperformed the centerline geocoder in identifying
the correct map sheet for bakeries listed in the directory. The centerline geocoder did not match
or incorrectly identified seventy one listings, while the insurance map geocoder incorrectly
identified just thirteen listings. The relatively high mismatch rates of both geocoders confirms
the need to employ reference data to confirm geocodes. It is important to note that the directory
listings of bakeries do not represent a random sample of historic addresses. Bakeries were
commercial enterprises, and tended to be listed on major streets. It is possible that a different set
of private addresses would perform better using the insurance map geocoder.

57

CHAPTER FIVE: DISCUSSION
Employing fire insurance map indexes to develop a geocoder demonstrates how the structure of
historical data sources can be exploited within a GISystem. This process also sheds light on the
implications of overlaying large-scale historical maps with modern data, and their value in
providing context for historical data. Sanborn maps are widely used by researchers, but it is
important to understand their peculiarities to evaluate suitability as a source document. A historic
geocoder can serve as a means to make better sense of the relationship between maps, texts and
the urban environment they represent.
This chapter reviews the insights gleaned through the process of developing a geocoder from
the Sanborn map indexes. Section 5.1 explores the implications of viewing insurance maps with
a GISystem. Section 5.2 articulates how Sanborn maps of San Francisco represent the turn-of-
century built environment, delineating the types of data that can be extracted from them. Section
5.3 assesses how the geocoding approach allows for more direct access to data represented on the
insurance maps. Section 5.4 suggests how the process of developing insurance map data into a
geocoder can be improved and what additional steps must be taken to improve its functionality.
Finally, Section 5.5 answers the research questions posed in Chapter 1.
5.1 Reading Insurance Maps through a GISystem
The process of digitization fundamentally transforms the way that researchers interact with
historical sources. Insurance map volumes are heavy, awkward to maneuver and rare. Stooped
over a table in a library leafing through map sheets, a researcher mirrors the posture of countless
underwriters who used the same volume with different questions or intentions. The physicality of
this process cannot be duplicated scrolling through microfilms or zooming and panning images
on a computer screen. Manipulating fire insurance maps with a GISystem further transforms the

58

map viewing experience. GISystems spare users the task of identifying map sheets and assessing
orientation, tasks that required careful examination of index maps and street directories.
GISystems make overlaying insurance maps with other sources of data straightforward.
5.1.1 Overlaying Insurance Maps with Contemporary Data
Overlay is one of the most basic functions of a GISystem. Without a GISystem, comparing
two maps of the same region is a tedious, time consuming, and imprecise task (Gregory and Ell
2007). Within a GISystem, the process of data overlay is so fundamental that it almost escapes
notice. Nevertheless, the decision of which data to overlay informs the types of questions that
can be asked and the context for understanding. Temporal context is an important consideration.
Contemporary data about San Francisco are plentiful, while historical data are rare.
Contemporary base maps display landmarks that allow users to understand how the historical
maps relate to the physical world. Superimposing fire insurance maps on familiar contemporary
base maps focuses the user on questions of contrast and change. The juxtaposition of the
historical fabric of tenements, boardinghouses and factories with homogenous superblocks of
office buildings downtown underscores the ways that San Francisco was transformed over the
twentieth century. However, this contrast does not explain the process of development. Maps and
other historical sources from intervening stages of change are needed to construct a narrative.
Nonetheless, visual overlay of insurance maps can provide important insights. One can situate a
historical building within its past context, with clues about past uses and architectural additions.
Maps can also serve as a basis to discuss political and social issues like housing and labor. On
their own fire insurance maps tell a limited story, with a disproportionate emphasis on hazards.
Maps are an important part of a city’s cultural heritage, not simply as a data source, but as a
record of ways people once lived. Preservationists rely on fire insurance maps to demonstrate the

59

historical or cultural significance of a building. Maps illustrate historical context in a way that
other sources cannot. In areas of the city where buildings from before the earthquake still exist,
they speak to the integrity of the urban fabric. In tandem with an architectural survey or
inventory and archival sources, maps provide evidence of a building’s historical significance.
5.1.2 Overlaying Historical Data on Insurance Maps
Comparing fire insurance maps to the present day city puts them into a familiar spatial
context, making it easier to navigate the map sheets. However, such comparisons may tell us
more about the contemporary city than the city of the past. Situating our understanding in the
present day, we are confronted with data extraneous to the experience of early twentieth century
San Francisco. Insurance maps are a rich, yet limited source of information. They depict the built
environment, emphasizing hazards over other cultural or social phenomena that exist in an urban
landscape. Yet insurance maps can provide much needed context for other historical data
sources. Mapping historical textual sources containing addresses, such as directories, newspaper
articles and advertising and business records in the context provided by the Sanborn maps can
expand and complicate our knowledge of the historical city.
Ultimately, we read a fire insurance map in the present, finding significance and surprises in
details intended as mater-of-fact records. It is essential to keep the essential function of insurance
maps in mind. Sanborns weren’t used for navigation; they were a means of risk assessment. Such
limitations parallel concerns in modern GIScience literature regarding fitness of use. GISystems
allow users to overlay data from a variety of sources, without regard to methods of data
collection or classification. Metadata provide careful users with a means of understanding the
methodology employed to derive data. Yet historical sources require more careful scrutiny than
contemporary data, precisely because the origins and methods of development are obscure.

60

5.2 Extracting Data from Sanborn Maps
All maps have blind spots or phenomena that escape the attention of the mapmakers. The
limitations of insurance maps are difficult to articulate without a thorough comparison of the
maps to other historical sources. The omissions of insurance maps merit further investigation,
but it is important to begin by articulating the types of data evidenced in an insurance map. A
GISystem requires that data be classified into discrete categories. While insurance maps are not
as cleanly categorized as digitally derived data, there are attributes that can be identified on all
maps related to three broad categories: infrastructure, landscape and buildings. Hazards informed
the surveying and mapping process. Keeping this focus in mind can help to explain what is being
represented, and what is missing from the maps.
5.2.1 Infrastructure
Parcel and lot boundaries formed the basis of map sheets. In general, San Francisco’s
historical parcel boundaries correspond to modern boundaries, except in areas where streets were
widened, moved or rebuilt. Names of streets were written and their widths were noted, though
the maps do not show these dimensions at scale. Materials of streets are noted only when they
differ from materials described at the beginning of each volume. Water mains were illustrated at
street intersections with their widths. Water hydrants are marked. However, other major aspects
of urban infrastructure escaped their attention. Electrical wires, communications infrastructure,
and streetcar tracks were absent, while heavy rail tracks were carefully documented.
5.2.2 Environment
While Sanborn maps existed primarily to record the built environment, they illustrated
significant natural features, like shorelines and other bodies of water. Despite San Francisco’s
complex topography, however, elevation and slope was generally ignored. Steep hillsides, rock

61

outcroppings and other natural features were shown only occasionally, as shown in Figure 5.1.
Parks, squares, and cemeteries were rarely mapped.

Figure 5.1 Schematic depiction of hill contours, Sheet 175, Sanborn Insurance Map
5.2.3 Buildings
Insurance maps focused on structural elements of the city. Buildings were carefully
documented, and their uses carefully described. Buildings were drawn in great detail, with an
emphasis on projections like bay windows or porches. Color was used to indicate building
materials, and some materials and construction techniques were indicated in writing. However,
some architectural elements were more carefully documented than others. Dimensions of light
wells and alleyways were often recorded, but windows and doors were rarely indicated.
The attention paid to building uses depended on their relevance to fire insurance. Residential
functions were categorized, and the number of housing units present in a building usually
indicated. The scrutiny paid to commercial and industrial uses varied. Entertainment facilities
like theaters, amusement parks and concert halls were named and documented, but other
commercial functions outside of the hazard focus were portrayed with a simple abbreviations.
Shops and restaurants were rarely expressly named or detailed. However, certain industrial
functions, such as confectioners or candy making, evidently had significance for fire insurance.

62

5.3 Assessment of Geocoding Approach
Sanborn maps prove themselves to be a valuable source of reference information to
contextualize historic addresses and provide a more holistic view of the architectural
environment of pre-earthquake San Francisco. However, this context could be provided by
simply vectorizing the index maps, without taking the additional step of digitizing the text-based
street directories. The insurance map geocoder modestly outperforms the centerline geocoder in
identifying the correct map sheets, but it has a distinct advantage over the centerline geocoder, in
that it does not introduce extraneous modern data into the process. It situates addresses within
their historical context.
Before endeavoring to create a historic geocoder, it is important to test and understand the
types of errors for addresses being geocoded. Some errors can be addressed simply by editing
and modifying centerlines. However, insurance maps are an unparalleled source to verify
historical addresses, despite their limitations.
5.4 Further Work
A historical geocoder functions as a piece of the spatial data infrastructure necessary to
develop a spatial understanding of the historical urban environment. I have shown that insurance
index maps and the associated street indexes contain the principal components necessary to
develop a functioning geocoder that provides historical context. The process of employing street
indexes could be repeated for other cities with extensive Sanborn coverage, particularly in cities
like San Francisco where changes in street numbering make street centerline geocoders
unreliable.

63

5.4.1 Improving Geocoder Precision
While aggregating geocodes to map sheets provides context to evaluate whether or not a
geocode is correct, a more precise geocoder would create more flexible point data that can be re-
aggregated for sensitivity analysis. Given the variety of textual sources that can be mapped using
this method, discrete points corresponding to individual records are often preferable, as opposed
to a homogenous aggregated count of observations occurring on a sheet. As outlined in section
4.2, the suitability of existing centerline data can be evaluated by geocoding an address with both
a centerline-based geocoder and the insurance maps geocoder. As long as both geocodes fall
within the same map footprint, the centerline geocode can be assumed to be more precise.
By mapping larger sets of historical addresses using this method, it will be possible to
identify streets where centerlines need to be redrawn or modified. Large areas of San Francisco
were not mapped by Sanborn, but these unmapped areas exist mostly in low-density regions of
the city. By identifying addresses that were not mapped in the Sanborns, a more systematic
understanding of the surveyors’ blind spots and unmapped regions of the city can be developed.
Building Inspector, the public participation map vectorization project run by the New York
Public Library, will ultimately develop sufficient data to create a parcel geocoder for historical
addresses in New York City. To replicate this effort in San Francisco, the insurance map sheets
must be satisfactorily georeferenced, with attribute data, including street addresses, assigned to
each parcel. However, parcel geocoders have greater rates of false negatives, because they will
only match known addresses (Zandbergen 2008). The ambiguity of San Francisco’s addresses
during this period could mean that a parcel geocoder based on data derived from Sanborn maps
would fail to geocode addresses that existed but were unlabeled in the Sanborn maps.

64

5.5 Conclusions
This thesis asked three principal questions. First, could historical data from insurance map
indexes meet the technical requirements of a modern geocoder? Second, would a geocoder based
on the indexes of fire insurance maps perform better than other types of historical address
geocoders? Lastly, would the benefits of an insurance map geocoder outweigh the costs of the
undertaking? These questions are not answered definitively by this study, but the project has
demonstrated that the approach is feasible and can help to uncover valuable insights about early
twentieth century San Francisco and other historical cities.
5.5.1 Exploiting the Structure of Insurance Map Indexes
Adapting Sanborn map indexes to a GISystems geocoder demands an understanding of both
the historical data and modern technical requirements. The regularity of the insurance map
indexes made it possible to capture the necessary data with minimal manual editing. Not all
historical data sources offer structure that so readily lends itself to digitization, but the Sanborn
map indexes can be digitized using available OCR software due to their consistent format. Street
indexes and geocoders serve similar functions—finding locations on the basis of street names
and house numbers. As such, they share similar attributes. Understanding the navigational
elements of historical geographical sources can make it easier to exploit their structure within a
GISystem. Fire insurance map indexes differ from street directories because they associate street
attributes with a geographical object—the map sheet. Insurance map indexes are uniquely suited
to adaptation because of their structure.
5.5.2 Comparison of Approaches to Historic Geocoding
The principal approach to creating historical geocoders is editing centerline files. However,
editing modern centerline data is a laborious process, and depends on high quality reference data.

65

Small scale maps and street directories can be employed to this end, but this process cannot be
automated, and requires a thorough understanding of the ways that the modern street system
differs from the historical streets. Few resources parallel insurance maps in their level of detail
and reliability as reference data for historical addresses. Modern centerline files are filled with
historically irrelevant information that complicate the task of identifying address locations.
Centerline address geocoders offer more precision than a geocoder that identifies map sheets, but
such precision is moot if geocodes cannot be verified in their turn-of-the century context.
Insurance map indexes can be exploited more quickly than editing centerline data.
5.5.3 Costs and Benefits of an Insurance Map Geocoder
It is clear that transcribing and digitizing Sanborn indexes is time consuming, but this
process can be automated more readily than the task of researching and editing centerline files.
Insurance maps provide historical context to the geocoded address, allowing the users to verify
the presence of an address, and also to visualize the surrounding environment. The map sheet
footprints derived by digitizing the index maps provide this context, regardless of how the
location is found. However, because the street indexes were created at the same time as the
maps, they ensure that geocodes are historically accurate. Identifying a map sheet provides a
straightforward means to confirm the presence of an address visually. It also underscores the
inherent imprecision of geocoding historical addresses.
5.6 Final Remarks
This thesis demonstrates the feasibility of using the Sanborn map indexes for the
development of an address locator. Geocoding historical addresses is contingent upon the
specific temporal and spatial relevance of the reference information at hand. In many cases, a
modern geocoder can function well to identify addresses in cities that have not experienced

66

major changes. However, the performance of any geocoder must be checked against temporally
relevant sources to insure that the geocoded addresses are meaningful. Insurance maps are one of
the most reliable historical sources of information about the built environment of nineteenth
century U.S. cities. For this reason, they function as an ideal means to verify historical addresses.

67

REFERENCES
Bolstad, P. V., P. Gessler and T. M. Lillesand. 1990. "Positional uncertainty in manually
digitized map data." International Journal of Geographical Information Science, 4(4):
399-412.
Colten, C. E. 1991. "Illinois Sanborn Geographic Information System." In Proceedings of the
34th Annual Engineering Geologists Meeting, Chicago, IL. October 3, 1991.
Chiang, Y. Y., C. A. Knoblock, C. Shahabi, C. C. Chen. 2009. "Automatic and accurate
extraction of road intersections from raster maps." GeoInformatica 13(2): 121-157.
Esri. 2010. "Customizing Locators in ArcGIS 10."
http://www.arcgis.com/home/item.html?id=aeb00de638f3492a93308a4a03183c7d.
Accessed 18 January, 2015.
Esri. 2015. "Commonly used address locator styles."
http://resources.arcgis.com/en/help/main/10.1/index.html#/Commonly_used_address_loc
ator_styles/00250000000v000000. Accessed 18 January, 2015.
Gregory, I. N., and A. Hardie. 2011. "Visual GISting: bringing together corpus linguistics and
Geographical Information Systems." Literary and Linguistic Computing 26(3): 297-314.
Gregory, I. N. and P. Ell. 2007. Historical GIS: Technologies, Methodologies and Scholarship.
Cambridge, UK: Cambridge University Press.
Goldberg, D. W., J. P. Wilson, and C. A. Knoblok. 2007. "From Text to Geographic
Coordinates: The Current State of Geocoding." URISA Journal 19(1): 33-46.

68

Harris, T. M., L. J. Rouse, and S. Bergeron. 2010. "The geospatial semantic web, pareto GIS,
and the humanities." In The Spatial Humanities: GIS and the Future of Humanities
Scholarship, edited by. J. Bodenhamer, J. Corrigan and T. M. Harris, 124-142,
Bloomington, Indiana: Indiana University Press.
Hoehn, P. "Union List of Sanborn & Other Fire Insurance Maps."
http://www.lib.berkeley.edu/EART/sanborn_union_list. Accessed October 23, 2014.
Hicks-Judd Company. 1901. The San Francisco Block Book. San Francisco: Hicks-Judd
Company.
Karrow, R. and R. E. Grim. 1990. "Two examples of thematic maps: Civil War and fire
insurance maps." In From sea charts to satellite images: interpreting North American
history through maps, 3.ed, edited by D. Buisseret, 213-220. Chicago: University of
Chicago Press.
Keister, K. 1993. "Charts of Change." Historic Preservation 45(3): 42-49, 91-92.
Keller, W. B. 1993. "Collecting and using fire insurance and real estate atlases: an individual
perspective." Art Reference Services Quarterly 1(3): 31-48.
Leonard, A. E. and P. Spellane. 2013. "Using Old Maps and New Methods to Discover the Early
Chemicals and Petroleum Industries of Newtown Creek in New York City." Journal of
Map & Geography Libraries 9(1-2): 25-43.
Lamb, R. A. 1961. "The Sanborn map: a tool for the geographer." California Geographer 2: 19-
22.
Lutkenhaus, B. 2002. "Digital Sanborn maps, 1867-1970." Reference Reviews 16( 3): 51.

69

MacDonald, S. and N. Osborne. 2013. "AddressingHistory—Crowdsourcing a Nation’s Past."
Journal of Map and Geography Libraries, 9: 191-214.
Migurski, M. 2011. "Sanborn Atlas." http://www.maptcha.org. Accessed November 4, 2014,
New York Public Library. 2010. "NYPL Map Warper." http://maps.nypl.org/warper. Accessed
December 26, 2014.
Oswald, D. L. 1997. Fire Insurance Maps: Their History and Applications. College Station, TX:
Lacewing Press.
Page, M. C. K. Durante and R. Gue. 2013. "Modeling the History of the City." Journal of Map
and Geography Libraries 9(1-2): 128-139.
Patton, D. K., A. K. Lobben and B. M. C. Pape. 2005. "Mapping Cities and Towns in the Late
Nineteenth and Early Twentieth Centuries: A Look at Plat, Sanborn, and Panoramic
Mapping Activities in Michigan." Michigan Historical Review 31(1): 93-122.
Raymond, A. 2011. "Denny Regrade, 1893–2008: A Case Study in Historical GIS." Social
Science History 35(4): 571-597.
Ristow, W. W. 1968. "United States Fire Insurance and Underwriters maps 1852—1968."
Quarterly Journal of the Library of Congress 25(3): 194-218.
San Francisco Department of Public Works. 2014. "Historical Block Diagrams."
http://bsm.sfdpw.org/subdivision/assessor/. Accessed October 22, 2014.
San Francisco Planning Department. 2014. "San Francisco Property Information Map."
http://propertymap.sfplanning.org/. Accessed October 22, 2014.

70

Sommer, L. 2011. "Mapping Project Reveals Pre-1906 Quake San Francisco."
http://ww2.kqed.org/news/2011/08/16/mapping-project-reveals-pre-earthquake-san-
francisco. Accessed October 22, 2014.
Torget, A. J., R. Mihalcea, J. Christensen, and G. McGhee. 2011. “Mapping texts: Combining
text-mining and geo-visualization to unlock the research potential of historical
newspapers.” White paper, University of North Texas Digital Library.
Vershbow, B. 2013. "NYPL Labs: Hacking the Library." Journal of Library Administration
53(1): 79-96.
Wrigley, R. L. 1949. "The Sanborn Map as a Source of Land Use Information for City
Planning." Land Economics 25(2): 216-219.
Yale University Library. 2013. "Yale University Library - The Map Collection / Sanborn Fire
Insurance Maps." http://www.library.yale.edu/MapColl/print_sanborn.html. Accessed
October 23, 2014.
Yuan, M. 2013. "Mapping Text.” In The Spatial Humanities: GIS and the Future of Humanities
Scholarship. Edited by D. J. Bodenhamer, J. Corrigan and T. M. Harris, 109-123.
Bloomington, Indiana: Indiana University Press.
Zandbergen, P. A. 2008. "A comparison of address point, parcel and street geocoding
techniques." Computers, Environment and Urban Systems 32: 214-232.

71

ATTRIBUTIONS
Images of Sanborn maps of San Francisco from the David Rumsey Map Collection are used
under the Attribution-NonCommercial-ShareAlike 3.0 license,
http://creativecommons.org/licenses/by-nc-sa/3.0/.
Links to original files for each figure are included below:

Figure Page
Figure 3.1 19
Unmodified.
Index Map: San Francisco Sanborn Insurance Map Atlas, Vol. 1.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213889~5501108
Figure 3.4 23
Cropped, resolution reduced.
Index: San Francisco Sanborn Insurance Map Atlas, Vol. 2.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213950~5501175
Figure 3.5 23
Cropped, resolution reduced.
Index: San Francisco Sanborn Insurance Map Atlas, Vol. 2.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213950~5501175
Figure 3.6 24
Cropped, resolution reduced.
Index: San Francisco Sanborn Insurance Map Atlas, Vol. 1.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213888~5501107
Figure 3.7 24
Cropped, resolution reduced.
Index: San Francisco Sanborn Insurance Map Atlas, Vol. 4.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~214074~5501427
Figure 3.8 26
Cropped, resolution reduced.
Index: San Francisco Sanborn Insurance Map Atlas, Vol. 5.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~214133~5501488
Figure 3.9 28
Cropped, georectified, resolution reduced, vector overlay.
Index Map: San Francisco Sanborn Insurance Map Atlas, Vol. 2.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213951~5501176

72

Figure 3.10 28
Cropped, georectified, resolution reduced.
Index Map: San Francisco Sanborn Insurance Map Atlas,
Vol. 1 http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213889~5501108
Vol. 2 http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213951~5501176
Vol. 3 http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~214010~5501363
Vol. 4 http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~214075~5501428
Vol. 5 http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~214134~5501489
Vol. 6 http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~214199~5501554
Figure 3.11 30
Cropped, georectified, resolution reduced, vector overlay.
Index Map: San Francisco Sanborn Insurance Map Atlas, Vol. 2.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213951~5501176
Figure 3.16 36
Cropped, georectified, resolution reduced, vector overlay.
Vol. 1, Page 31-32. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213905~5501124
Vol. 1, Page 41-42. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213910~5501129
Vol. 1, Page 43-44. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213911~5501130
Vol. 1, Page 57-58. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213918~5501137
Figure 3.20 39
Cropped, georectified, resolution reduced, vector overlay.
Vol. 1, Page 41-42. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213910~5501129
Vol. 1, Page 43-44. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213911~5501130
Figure 3.21 40
Cropped, georectified, resolution reduced, vector overlay.
Vol. 1, Page 41-42. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213910~5501129
Vol. 1, Page 43-44. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213911~5501130
Figure 3.26 42
Cropped.
Vol. 1, Page 43-44. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213911~5501130

73

Figure 4.1 45
Cropped.
Vol. 1, Page 103-104. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213941~5501160
Figure 4.3 47
Cropped.
Vol. 2, Page 143-144. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213961~5501186
Figure 4.4 47
Cropped.
Vol. 2, Page 145-146. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213962~5501187
Figure 4.5 48
Cropped.
Vol. 6, Page 639-640. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~214207~5501562
Figures 4.6a and 4.6b 49
Cropped.
Vol. 2, Page 145-146. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213962~5501187
Figure 4.7 50
Cropped.
Vol. 4, Page 425-426. San Francisco
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~214101~5501454
Figure 5.1 61
Cropped.
Vol. 2, Page 175-176. San Francisco.
http://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~213977~5501202

74

APPENDIX A: R CODE
#Load csv file exported from ArcMap
fishnet <- read.csv("C:/R/f25n.csv", header=T)
#Sort fishnet features by sheet number
fish <- fishnet[order(fishnet$Sheet),]
# Dataframe "fishdf" with Object ID and sequence number concatenated to sheet no.
fishDF <- data.frame(
s = sequence(rle(as.vector(fish$Sheet))$lengths),
ObjectID = fish$OID,
uid = paste(fish$Sheet, fishDF$s, sep = "_")
)
write.csv(fishDF, file = "C:/R/f25d.csv" )
#Assign unique ID to Address Ranges.
ar10 <- read.csv("C:/R/addressranges.csv")
arZ <- ar10[order(ar10$Sheet),]
arDF <- data.frame(
s = sequence(rle(as.vector(arZ$Sheet))$lengths),
jid = arZ$JoinID,
uid = paste(arZ$Sheet, ardf$s, sep ="_")
ar10 <- read.csv("C:/R/addressranges.csv")
arZ <- ar10[order(ar10$Sheet),]
s = sequence(rle(as.vector(arZ$Sheet))$lengths)
arDF <- data.frame(
JoinID = arZ$JoinID,
uid = paste(arZ$Sheet, s, sep ="_"))
arM <- merge(arZ, arDF, by="JoinID")
write.csv (arM, file = "C:/R/arsort.csv")

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Integrating spatial visualization to improve public health understanding and communication

PDF

Address points and a maser address file: improving efficiency in the city of Chino

PDF

Crowdsourced maritime data: examining the feasibility of using under keel clearance data from AIS to identify hydrographic survey priorities

PDF

Surface representations of rainfall at small extents: a study of rainfall mapping based on volunteered geographic information in Kona, Hawaii

PDF

A comparison of address point and street geocoding techniques in a computer aided dispatch environment

PDF

Cartographic approaches to the visual exploration of violent crime patterns in space and time: a user performance based comparison of methods

PDF

Finding food deserts: a study of food access measures in the Phoenix-Mesa urban area

PDF

Selection of bridge location over the Merrimack River in southern New Hampshire: a comparison of site suitability assessments

PDF

Wake County District Overlay: an online electoral data visualization application

PDF

Mapping firing ranges as social capital generators in Houston, Texas

PDF

Mapping uniformity of park access using cadastral data within Network Analyst in Wake County, NC

PDF

Social media canvassing using Twitter and Web GIS to aid in solving crime

PDF

Exploring San Francisco's treasures: mashing up public art, social media, and volunteered geographic information to create a dynamic guide

PDF

GIS data curation and Web map application for La Brea Tar Pits fossil occurrences in Los Angeles, California

PDF

Guiding business oriented volunteered geographic information through geotigger services: a case study of CrossFit affiliates

PDF

Visualizing sporting event temporary flight restrictions

PDF

Creating a flood vulnerability index for Houston, Texas

PDF

Campaign financing for the U.S. House of Representatives: an interactive Web map

PDF

Modeling patient access to point-of-care diagnostic resources in a healthcare small-world network in rural Isaan, Thailand

PDF

Spatiotemporal visualization and analysis as a policy support tool: a case study of the economic geography of tobacco farming in the Philippines

Asset Metadata

Creator Rosen, Yonatan (author)

Core Title A fire insurance map geocoder for pre-earthquake San Francisco

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Publication Date 02/02/2015

Defense Date 01/13/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag digital humanities,geocoding,historical geographic information science,OAI-PMH Harvest,San Francisco,Sanborn Fire Insurance maps

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Kemp, Karen K. (committee chair), Oda, Katsuhiko (Kirk) (committee chair), Chiang, Yao-Yi (committee member), Warshawsky, Daniel N. (committee member)

Creator Email yrosen@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-527853

Unique identifier UC11297275

Identifier etd-RosenYonat-3159.pdf (filename),usctheses-c3-527853 (legacy record id)

Legacy Identifier etd-RosenYonat-3159.pdf

Dmrecord 527853

Document Type Thesis

Format application/pdf (imt)

Rights Rosen, Yonatan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA